Skip to main content

Full text of "Model Engineering By Peter Wright"

See other formats



Peter Wright 

Nexus Special Interests 

Nexus Special Interests Ltd 
Nexus House 

First published by Nexus Special Interests 1997 

© Text Pfeter WHght 1997 
O Photographs and illustrations Peter Wright 1997 

ISBN 1-85486-152-2 

Typesetting by Kate Williams, London 

hinted and bound in Great Britain by Biddles Ltd., Guildford & King’s Lynn 


Preface v 

1 Introduction I 

2 Information and supplies 27 

3 Introduction to engineering materials 42 

4 Setting up a workshop 59 

5 Measuring and marking out 82 

6 Basic handwork I 11 

7 Bending, folding and forming 136 

8 Metal joining I - mechanical methods 158 

9 Metal joining 2 - solders and adhesives 180 

10 Hole production 204 

11 Screw threads 238 

12 The lathe - essential machine tool 255 

13 Holding work in the lathe 271 

14 Principles of turning 294 

15 Basic lathe practice 316 

16 Adapting the lathe 353 

17 Buying a lathe 388 


Index 405 


but this is not a printing or scanning 
fault and no content is missing. 


Judging by the comments and questions one 
overhears at exhibitions and the like, or indeed 
the questions which one is personally asked at 
club open days, there are many who would 
dearly like to participate in the hobby of model 
engineering but who do not in the least under¬ 
stand how to start. 

Many of us who are now practising model 
engineers were fortunate in our younger days 
to have had the opportunity to follow what 
would nowadays be termed a craft apprentice¬ 
ship, or perhaps, as in my case to have followed 
a sandwich course which included a worth¬ 
while introduction to the processes of mechani¬ 
cal manufacture at a time when there was still 
great reliance on the manual skills of individual 

This seems not to be the case these days, as 
more and more the craft apprenticeship is 
becoming a short-term education programme 
which seems incapable of producing engineers 
with a wide range of skills who somehow know 
how the system of making things has to work in 
order for it to be successful. 

This is obviously not true of all industries 
since the assembly of aircraft, for example, is 

still largely a manual trade, but the introduc¬ 
tion of numerically controlled machine tools 
has revolutionised the manufacture of the 
component parts and in the field of small batch 
production, or the realm of development engi¬ 
neering which is my background, operators of 
these computer-controlled machines are gradu¬ 
ally taking over from the time-served men who 
were their forerunners. Consequently, people 
with the wide range of skills which operators in 
development workshops used to have are 
becoming a thing of the past. 

These wide-ranging skills are unfortunately 
just those that are needed by the amateur 
worker, who must of necessity be a Jack of all 
Trades, since he will need to carry out most of 
the processes of manufacture himself (but usu¬ 
ally not all of them) given only the modest 
workshop facilities which his hard-earned 
money will buy. It is to provide the essential 
general introduction to these processes that this 
book has been written. It provides an introduc¬ 
tion to the techniques which may be used in an 
amateur workshop to perform tasks for which 
more complex, specialised (and more expen¬ 
sive) equipment would be employed in industry. 


As a substitute for this more expensive 
machinery, the amateur usually develops 
manual skills, and not a little ingenuity, to en¬ 
able him to create in the home environment a 
miniature locomotive or a full-size long-case 
clock, or whatever, and it is hoped that this 
book will encourage the development of such 

Many of the workshop operations described 
and photographed are actual ‘work in progress’. 
This means that the workshop is not always tidy 
and the photographic backgrounds are some¬ 
times more cluttered than is desirable. I have 
also learned as I have progressed, and I hope the 
photographs (in particular) show an improve¬ 
ment, although it may not be obvious to you 
since the illustrations do not appear in chrono¬ 
logical order. 

My own interests and facilities naturally 
limit what can be shown, but I have been fortu¬ 
nate to have access to other people’s workshops 
and tools, and my thanks go to them for putting 
up with the endless questions, “Have you 
got...?” and “Can 1 come over and photograph 

My thanks go especially to my father-in-law, 
Peter Feast, not only for his extensive collection 
of hand tools and long experience of their use, 
but also for being available when experienced 
hands are needed for the photographs. My 
grateful thanks go also to his daughter for her 
support and encouragement, for her willing¬ 
ness to help, particularly as photographer’s 

assistant, holding up backgrounds or reflectors, 
or firing the shutter when my hands are holding 
the tools. She has also learned uncomplaining¬ 
ly, over the years, the true meanings of the 
phrases “I’ll just go and process a few words” 
and “I’m going to have ten minutes in the 

Before the book itself, a word of encourage¬ 
ment. Do not think, that because you have not 
previously been involved in this fascinating 
hobby you will not be successful. Do not think, 
having visited a local or national handicrafts 
exhibition and admired those superb examples 
of the craft, that you must desist because you 
will not be able to produce such fine models. 
Very few of us can! Certainly not at the first 
attempt. All that matters is that you should en¬ 
joy the work which you do. 

If, at the end of the day, you do not feel that 
your results should be displayed for others to 
see, there is no compulsion for them to be dis¬ 
played. In any event, you are much more likely 
to feel reasonably proud even of your first 
efforts, for to complete a large project is itself a 
pleasure and a triumph, and what does it mat¬ 
ter if, at the end you feel that you could have 
done it better? It will be better next time, and 
that’s when even greater pleasure will come. 

If the book encourages some of you to make 
a start when you have previously only dreamed 
of owning some hand-made mechanism, the 
efforts of its preparation will be amply re¬ 

Peter Wright 
March 1997 


This book is dedicated to the memory of JAB 

Although not immediately interested in the subject matter, 
he was always interested in the products of the workshop 
and full of enthusiasm for the completion of this book 


but this is not a printing or scanning 
fault and no content is missing. 


What is model engineering? 

Before commencing, it would be as well to 
define what is meant by ‘model engineering’, 
for a definition is not easily found. Looking 
through the list of competition classes at the 
International Model Show provides no real 
guide since the classes adopted for these com¬ 
petitions reflect the wide range of interests 
which modellers have, rather than assisting in 
forming a definition of what model engineering 
actually is. Thus, one sees on display large-scale 
model steam locomotives and road vehicles, 
internal combustion engines, model boats and 
aircraft, alongside which are to be found scenic 
model railways, farm carts, model soldiers, 
model cars and radio-controlled everything. So 
a true definition of model engineering is not to 
be found at the site of our annual pilgrimage. 

The term itself is so ingrained into the lan¬ 
guage that one hesitates to suggest a change, 
but if that opportunity did exist I should be 
inclined to propose ‘mechanical engineering in 
the home workshop’. But no-one would prefer 
to use that, even though it suits the purpose of 
this book admirably. 

Some such definition is required, however, 
for we cannot possibly regard the construction 
of a clock as model engineering unless the 
clock produced is indeed a model of an actual 
mechanism, but this is seldom the case. Clock- 
makers prefer to make a dining-room size 
clock which has some practical value rather 
than make a model, yet their activities are 
catered for by the International Model Show 
organisers and by the supporting trade stands, 
just as the builder of model steam locomotives 
is supported. So we clearly need some alterna¬ 
tive definition. 

The difficulty in defining the term arises 
from the very diverse nature of the items which 
are produced, and from the fact that some 
branches of the hobby are so well supported by 
the trade that bolt-together kits may be pur¬ 
chased. For some models, kits are sufficiently 
low in cost to allow volume sales to be achieved, 
and in some areas of interest almost all activity 
centres around commercially produced compo¬ 
nents. This applies particularly to electrically 
powered model car racing, for example, for 
which virtually all components can be pur¬ 
chased from specialist retailers. 



There are, however, other branches of the 
hobby in which the models produced comprise 
so many parts that commercial production is 
incredibly costly, and although bolt-together 
kits are available, they sell in relatively small 
quantities, and most models are amateur-made. 
This applies particularly to large-scale model 
steam locomotives. 

Poor support from the trade is also the case 
when certain minority interests are considered. 
These include the design and manufacture of 
hot-air engines and the building of scale model 
cars for static display. These aspects are not yet 
sufficiently popular to encourage trade support 
although hot-air engine design details are pub¬ 
lished from time to time in the model press, 
encouraged by the annual competition for these 
engines at the International Model Show. 

The models in the competition classes at the 
exhibition therefore embrace the whole range 
of modelling skills, from the complete design 
and manufacture of every part of a large-scale 
steam locomotive (including the production of 
patterns and castings) to assembly and detailing 
of a kit, either of the bolt-together type, or the 
type in which basic parts are provided but the 
builder must shape the parts, assemble and fin¬ 
ish the model, such as in model boat or model 
aircraft building. 

The common theme which links together 
many of the models seen at the exhibition is that 
extensive use of a machine tool has been a 
requirement in their production. This is not 
true of all of the items, for many fine models are 
constructed without the use of relatively expen¬ 
sive machinery, but for many the term model 
engineering implies the use of such a machine 
and for a large part, the use of machine tools 
forms the subject of this book. 

As far as possible, this book has been filled, 
not with pictures of models or a great deal of 
theory, but with practical information related to 
the making of things. To some extent, the 
information provided attempts to cover sub¬ 

jects with which I myself found some difficulty 
in the early days of establishing my modelling 
workshop, having only a rather hazy knowl¬ 
edge of some of the techniques employed. 

The making of things requires skill. Unfortu¬ 
nately, skills cannot be learned from a book, 
only by experience. It is therefore important 
that you have plenty of practice, but this must 
be based on sound techniques, and I hope that 
the basics have been covered here and in suffi¬ 
cient detail for you to approach the practice 
with confidence. I also hope that there is indeed 
sufficient information within this book to prac¬ 
tise with confidence, even if, at first, the prac¬ 
tice may not be perfect. 

In order to build a model you do not need to 
have the capability to design it. Indeed, design¬ 
ing models is equally as difficult as designing 
the prototype and is best left until you at least 
have some experience as a builder. Building is 
therefore started by the purchase of the rel¬ 
evant drawings, followed by the materials and 
castings, if these are required. Once these basics 
are available, building may start. Sources of in¬ 
formation and drawings, and advice on the pur¬ 
chase of an essential stock of material, is given 
in Chapter 2, while Chapter 3 provides an in¬ 
troduction to the materials used in the manu¬ 
facture of models, and to the subject of heat 

Chapter 4 describes the requirements for the 
workshop itself, and its possible locations. The 
ways in which a workshop might be fitted out 
are described, together with the requirements 
for lighting and the need for magnifying view¬ 
ers. The general types of small hand tools are 
described, and the equipment needed for sol¬ 
dering and brazing. 

Once manufacture commences, you will 
need measuring and hand tools for marking out, 
cutting and filing, the capability to drill and size 
holes and the machinery to carry out the two es¬ 
sential operations of turning and milling. The 
basic hand skills are covered in Chapters 6 and 



7, while Chapter 5 describes the measuring 
equipment which is normally utilised. 

Chapters 8 and 9 describe methods of join¬ 
ing metals, considering separately what can be 
described as mechanical methods (attachment 
by screws, riveting and pressing parts together) 
and other methods in which materials such as 
solders and adhesives are introduced into the 

The later chapters describe the essential 
machinery which is required, and introduce the 
basic machining processes of hole production, 
threading, turning and milling. Turning is car¬ 
ried out on a lathe, the essential features of 
which are described in Chapter 12. Its use, and 
adaptation to other processes are described in 
Chapters 13 to 16 and Chapter 17 contains a 
description of the major considerations to be 
borne in mind when purchasing a lathe and 
contains advice on how to inspect a second¬ 
hand machine and set it up. 

A lathe is absolutely essential for the sort of 
modelling envisaged. Although it is designed 
for turning, it can readily be adapted for mill¬ 
ing, provided that the work is not too large, and 
with the addition of a few accessories can 
become an extremely versatile machine. The 
amateur is frequently equipped only with a 
lathe, but this is usually supplemented by a 
drilling machine since much drilling is neces¬ 
sary, and although holes can be drilled on the 
lathe, it is not readily adapted for drilling holes 
in large plates, and this is frequently required. A 
brief description of the main features of a drill¬ 
ing machine is given at the beginning of Chap¬ 
ter 10. The subject of making holes is quite 
diverse and this occupies most of Chapter 10. 

Chapter 11 contains an introduction to the 
more commonly used screw threads, and to 
methods of cutting threads using hand tools. 
Thus, what might be described as the basics of 
handwork are described in Chapters 5 to 9, 
while that essential machine tool, the lathe, and 
its use, are covered by Chapters 10 to 16. 

Don’t be tempted to skip the basics. If you 
take the trouble to acquire the basic skills, there 
is much pleasure to be had even from the sim¬ 
plest task of sawing off a piece of material. Per¬ 
formed carefully and skilfully, sawing squarely 
and close to the line, the item you are building 
will progress rapidly, with little frustration. If 
you approach sawing as a necessary evil, racing 
to finish it, using an unsuitable saw, you will 
almost certainly not saw close to the line, will 
break more blades and will have more filing or 
machining to do in the end. And the acquisition 
of the basic skills, which only needs to be done 
once, saves time over and over again. 

There is never enough modelling time, so 
saving time is as vital as saving money. In some 
ways, time and money are interchangeable 
commodities, and money invested in a tool, or 
machine tool attachment, for example, may 
save valuable time. 

Branches of the hobby 

Steam locomotives 

For many, model engineering simply means 
messing about with live steam locomotives. In 
lots of ways, live steam models are very satisfy¬ 
ing things to be involved with since one gets the 

Figure I. I My 5-inch gauge pannier tank to LBSC’s Pansy 



pleasure of the workshop time, which satisfies 
the urge to make things which most of us seem 
to have, but it also allows the builder to take his 
model to the local track and turn what were 
once just lumps of metal in a store into a living, 
breathing thing. 

This branch of the hobby developed out of 

Figure 1.3 Vic Newport and his great grandson adminng a the earl y commercial model railways which 

Visiting locomotive at their local track. were generally built to a much larger scale than 



is currently the case. The locomotives were also 
normally steam powered, and to be interested in 
model railways in the early part of the century 
meant being immersed in the technicalities of 
steam, spirit and coal firing and suchliketopics. 

At that time, much model steam locomotive 
design and construction was most definitely in 
the toy train category. This meant that simple 
mechanisms and simple techniques were the 
order of the day and when repairs became nec¬ 
essary, quite simple tools and techniques were 
all that were required to effect a repair. Much 
that was needed for a model could, therefore, 
be hand made and it was largely for wheels and 
cylinders that a lathe was required. As the avail¬ 
ability of small lathes increased, there came 
into the hobby of model engineering many of 
the model railway enthusiasts. Naturally, the 
lathe made the manufacture of domes, chim¬ 
neys, buffers and so on that much easier and 
with quite simple facilities the amateur could 
develop his own locomotives or improve vari¬ 
ous of the commercial offerings, many of which 
appear to have had in-built shortcomings. 

In those days, 2Viin. and 3‘/ 2 m. gauge mod¬ 
els were quite common commercial sizes and it 
was natural, therefore, that enthusiasts should 
be drawn to the idea of making sufficiently 
powerful models capable of hauling passen¬ 
gers. Through the efforts of writers in Model 
Engineer, notably LBSC who started his ‘Live 
Steam’ column in that magazine in 1924, basic 
techniques were developed for the building of 
boilers, motion work, cylinders and fittings for 
workable passenger-hauling models and by the 
late 1930s the hobby was well established. 

As hobbyists have become more affluent 
over the years, particularly since 1945, models 
have tended to become larger. The support 
offered by the trade in the form of drawings, 
castings and general materials has increased 
and there is nowadays a marked trend towards 
larger models, made on larger machines, cost¬ 
ing relatively more than was once the case. 

Currently, the most popular scales for pas¬ 
senger-hauling locomotives are 3 Am. per foot 
(19.05mm per foot) running on 3Viin. gauge 
track, and V/mn. (27mm) per foot, running on 
5in. gauge track. The elevated track is still the 
most common arrangement for passenger haul¬ 
ing and most modellers join a local club in 
order to have access to a track. Although 3Viin. 
and 5in. gauges are the most popular, there are 
still club tracks that can accommodate 2 ‘/ 2 in. 
gauge models. 

Many clubs with more spacious grounds 
available have established ground-level tracks 
which provide for 5in. and 7!4in. gauge mod¬ 
els, these latter being built to 1 Vim. (38mm) per 
foot scale. Models designed for 7‘/iin. gauge 
are naturally more powerful than their smaller 
counterparts. Being larger, they are also capa¬ 
ble of being made with almost all details to 
scale and yet can still be driven by ‘over-scale’ 
drivers, whereas this is not always possible in 
the smaller gauges if reliable steam fittings, 
lubricators and so on are to be made. 

A further current trend is also evident in the 
increasing numbers of narrow-gauge steam 
models which are being made, influenced by 
the preserved lines. The advantage of a narrow- 
gauge locomotive is the increased size (and pas¬ 
senger-hauling capability) one gets for a given 
gauge. Thus, a narrow-gauge model designed 
to operate on 3'/2in. gauge track might repre¬ 
sent a 2ft gauge locomotive and is therefore 
built to a scale of l 3 /<in. (44.5mm) to the foot. 
It is consequently more than twice the scale of a 
regular 3 Vim. gauge model and can have larger 
cylinders, boiler and fittings. It is easier to han¬ 
dle on the track, more powerful and more 

There is, too, a charm about narrow gauge 
which makes an attractive model and there is 
the further consideration that, particularly on 
7%in. gauge, ground-level track, one easily gets 
the impression that one is running a real loco¬ 



At the other extreme, the last few years have 
seen a revival in interest in ‘O’ and ‘1* gauge 
live steam models. Again, this has arisen from 
the availability of commercial models and track 
in these gauges, many of the suppliers concen¬ 
trating on the production of ‘freelance’ and 
narrow-gauge models having simple mecha¬ 
nisms and boilers and selling at reasonable 
prices. The availability of these models has 
stimulated amateurs to improve on the com¬ 
mercial offerings and to develop more true-to- 
scale models and much interesting work has 
been done by the members of societies such as 
the Gauge “O” Guild and the Gauge “1” Model 
Railway Association. 

Figure 1.7 A double-ended, battery-powered locomotive 
from John Brotherton's stable. 

Battery-powered locomotives 

There has recently been a rapid growth of inter¬ 
est in battery-powered locomotives. The con¬ 
siderable adhesive weight provided by a 12-volt 
car battery, or perhaps two of them, ensures ad¬ 
equate haulage capacity, and if the type of 
‘boxy’ outline provided by a diesel shunter is 
adopted, there is plenty of room to house a 
large-capacity battery, which ensures that a full 

Figure 1.6 David Elen driving the Easthampstead Sewage 
Worta' Lady Muck. 

afternoon’s passenger hauling can be under¬ 
taken without the necessity to have a spare bat¬ 
tery available. 

Some of the simplest (and cheapest) arrange¬ 
ments utilise a simple chain drive, power being 
provided by a second-hand car generator ener¬ 
gised through a home-made electronic control¬ 
ler. If you are into electronics and like tinker¬ 
ing, this is a useful field for experimentation. 
However, some mechanical knowledge is also 
called for since skill is required in arranging a 
drive which allows some rise and fall of the 
driven axle without changing the centre-to- 
centre distance between the driven and driving 
shafts. In this respect, there is some merit in 
using a belt drive, which allows a little give, an 
internally toothed belt perhaps being ideal. 

One of the advantages of the battery- 
powered model is the general simplicity of the 
design. The number of components which must 
be made is reduced in comparison with even 
the simplest model steam locomotive and there 
is no boiler to construct, thus removing the 
need for the equipment which is necessary for 
boiler building. 

There are some advantages on track days 
too. There is no need for the lighting up, and 
afterwards the cleaning down of the dirtier 



parts of a steam model. A quick oil-round and 
connection of the battery usually suffices at the 
start of the session, and an equally quick wipe 
round with an oily rag to remove the general 
grime picked up from the track, together with 
disconnection of the battery, is all that’s needed 
at the end of the day. 

A further advantage of this type of model is 
the ready availability of kits of parts and ready- 
assembled models. The kits require finishing 
and painting but costly machine tools are not 
necessary and the simple ability to drill fixing 
holes, clean up, bolt together and paint is all 
that may be required. Provided that funds are 
available, the purchase and assembly of a kit 
can be a quick way into the hobby and certainly 
gets you down to the track, participating, in a 
short time. 

Traction engines 

Model traction engines are deservedly popular. 
They are normally coal-fired, steam models 
(unless built for the showcase or sideboard) and 
therefore provide that nostalgic coal-oil-and- 
steam smell which is so characteristic of the 

Figure 1.8 An example of the Minnie traction engine seen at 
a modelling exhibition. 

« .. 

Figure 1.9 An award-winning showman's tractor. 

steam era. They possess the decided advantage 
of not needing a track on which to run and yet 
are eminently suitable for doing some real 
work in the form of passenger hauling. The 
steam side of things needs managing correctly if 

Figure 1. 10 fine detail on the tender of a large-scale model 
traction engine. 



the engine is to perform satisfactorily and the 
exposed motion work gives much visual satis¬ 
faction unlike that on inside-cylindered loco¬ 
motives which is normally all but invisible. 

Strictly, what are colloquially termed ‘trac¬ 
tion engines’ comprise several distinctly differ¬ 
ent vehicle types. The most common is the 
general-purpose engine which is effectively a 
self-propelled prime power source. It is nor¬ 
mally capable of towing a trailer-mounted load 
but its prime purpose is to act as a local power 
source for agricultural and light industrial use. 
It can, therefore, drive threshing machinery, 
circular saws and other light-to-medium duty 

A self-propelled machine known as a tractor 
or road locomotive is used for other than local 
haulage. This provides weather protection for 
the crew in the form of a short awning above 
the footplate. Tractors may be physically some¬ 
what larger than the general-purpose machines 
but are otherwise broadly similar. 

The Rolls-Royce of the tractor world is what 
is usually called the ‘showman’s road locomo¬ 
tive’. This type has an all-over canopy sup¬ 
ported on polished brass ‘Olivers’ and is usually 
more ornately decorated in comparison with 
other engines. This type is generally the largest 
of the tractors, its length extended by the usual 
front-mounted platform for the generator 
which allowed the machine to provide the 
power for the electrically operated rides at the 
fairgrounds, and for lighting at the circuses, 
which the showmen ran. Models of showmen’s 
locomotives are enormously popular and draw¬ 
ings and castings are available from several 
commercial sources. 

Also on the massive scale were the agricul¬ 
tural engines which were used to perform 
heavy-duty tasks on the land such as ploughing, 
dredging and the cutting of land drains. These 
machines make interesting models, but for 
some, there is the disadvantage that they were 
frequently used in pairs thus requiring twice the 

work to model authentically. While this is true, 
it will not necessarily take twice as long to 
make a pair since machine setting-up time does 
not increase, but an alternative is for two mod¬ 
ellers to make one each and then to combine 
their efforts for the making of accessories such 
as plough, cultivator mole drainer or whatever 
seems to be appropriate. 

The popular scales for model traction 
engines (to use the colloquial term) range from 
1-inch scale to 4-inch scale (25 to 100mm per 
foot) or larger. In the smallest of these scales, a 
model engine is a relatively small model, about 
12in. high (300mm) and 18in. long (460mm). 
Such a model is capable of pulling the owner 
along, but it is a hard struggle on grass, and 
there has been a move towards the larger scales 
over the last few years. A popular size is the 
1 'Ain. to the foot scale (37mm), giving a model 
that is capable of real work and yet not too 
heavy to move about nor too large to store con¬ 

However, the trend is towards “bigger and 
better”and 2-, 3- and 4-inch scale models are 
now quite common and well catered for com¬ 
mercially. Before embarking on a model, your 
capacity to turn the hind wheels should be con¬ 
sidered. If, say, the prototype’s wheels were 6ft 
in diameter (and some were larger) you will 
have to be able to cope with a 24in. (600mm) 
diameter to machine the hind wheel rims in 4- 
inch scale and this will therefore limit the scale 
of model that you can consider. Large boilers 
may present some problems too, unless you are 
a capable welder and competent enough in the 
making of pressure vessels to consider steel for 
the boiler material, and this aspect is also 
worthy of consideration before a start is made. 

Stationary and marine engines 

Stationary and marine steam engines have been 
popular modelling subjects since the early days 



Figure 1. 11 A nicely displayed Stuart Turner beam engine. 

Figure 1.14 Testing an example of Tubal Cain’s Kitten before 
final polishing and assembly to its base. 

Figure 1.12 A horizontal steam engine to the popular 
Perseus design. 

Figure 1.13 A Stuart Turner triple expansion engine. 

of the hobby and the range of kits produced un¬ 
der the name of Stuart Turner includes designs 
which date from before the First World War. 
Kits and drawings are also available from many 
other suppliers for engines as simple as a single, 
oscillating-cylinder engine which requires no 
valve gear as such, to the more complex wind¬ 
ing, roller-mill and beam engines. 

The small, oscillating-cylinder types are 
intended for incorporation into simple model 
boat power units, when married to a suitable 
boiler, or may be incorporated into the types of 
toy steam plant which are always popular. 
Some simple designs are available also for small 
model cranes using this simple type of non- 
reversible steam engine. 

Some of the larger stationary models may be 
non-reversing but they are usually provided 
with a slide valve and single eccentric arranged 
for running in one direction but without the 
second eccentric and some kind of link 
arrangement to allow running in either direc¬ 
tion at will. These engines provide a useful 
introduction to the steam engine generally, if 
one is needed, but do not require the manufac¬ 
ture of large numbers of components and can 
therefore be completed in reasonable time. A 
variety of materials will be encountered during 
the manufacture (cast iron, brass or gunmetal, 



mild steel) which provides valuable modelling 
experience and only a small boiler is required 
to generate steam if a working model is wanted 
rather than a sideboard or showcase model. 

Model boat enthusiasts may well wish to 
build a small, reversing steam engine for use in 
a model boat or launch hull. This provides a 
valuable extension of the skills of model boat 
building which have perhaps already been 
acquired and takes the modeller into the quiet 
realms of slow-speed engines and a leisurely 
progress of the vessel. 

Internal combustion engines 

The building of internal combustion (1C) 
engines utilises quite different techniques from 
those used in the manufacture of steam engines. 
In comparison with an 1C engine, the steam 
engine is relatively crude. It is slow revving and 
its reciprocating parts are not normally bal¬ 
anced. Its motion work is not usually enclosed, 
and its lubrication is therefore most frequently 
by simple oil can. Also, pressures and tempera¬ 
tures existing in the steam engine cylinder are 
quite low, and simple (and rather crude) means 

Figure 1.15 A 15cc X/vw petrol engine. 

Figure 1.16 A four cylinder, water-cooled petrol engine. 

are usually employed in providing the seal 
between piston and cylinder bore (at least in 
models), graphited yarn still being the most 
common material for this duty. 

1C engines usually utilise finely lapped bores 
having lightweight pistons which are fitted 
with cast iron rings, although pistons are 
sometimes used without rings, especially in the 
smaller sizes. However, pistons are usually 
closely lapped to the cylinder bore. Con¬ 
necting rods are normally polished, and are 
matched for mass (weight) in a multi-cylinder 
engine and a rudimentary balancing exercise is 

Figure 1.17 Two air-cooled 1C engines. 



routinely undertaken to provide a smooth¬ 
running, high-speed engine. A high-pressure 
lubrication system may also be incorporated, 
but even if simple ‘splash’ lubrication is used, 
this is obviously more sophisticated than 
the locoman’s occasional squirt from the oil 

Drawings and castings are available from the 
trade for a very wide range of IC engines, 
embracing designs for petrol, glowplug and 
diesel types, both air and water cooled. Those 
strictly intended as model power range from 
less than lcc capacity up to 30cc or so, provid¬ 
ing power for a wide range of model boats and 
aircraft and adaptable to model cars or loco¬ 
motives should your interests lie in that direc¬ 
tion. Complementary designs for carburettors, 
contact breakers and magnetos are also avail¬ 
able and several suppliers of miniature ignition 
equipment and sparking plugs can satisfy these 
essential needs. 

If you do intend to use the engine in an air¬ 
craft, car or boat, be sure to enquire about the 
type and capacity which are allowed by the 
rules of the various competition classes, before 
starting, otherwise the model may not be usable 
or saleable. 

In common with other branches of the 
hobby, there is increasing interest in scale mod¬ 
els of particular IC engines, rather than designs 
intended specifically as power sources for mod¬ 
els, and beautiful (working) models of rotary 
aero engines are now quite common at club 
exhibitions. Working models of the Rolls- 
Royce Merlin engine have also been exhibited, 
at least one of which is capable of sustained 
running through the use of scale magnetos. 
There is thus much scope for fine workmanship 
in this field and a wealth of simpler designs on 
which to cut one’s teeth before progressing to 
the fine-scale end of the range. 

Clocks and scientific instruments 

At one time, principally between the wars and 
immediately after World War 2, there was 
much amateur interest in the making of scien¬ 
tific or semi-scientific instruments which were 
either not readily available, or impossibly 
expensive for ordinary mortals to acquire. 
There was also a modest interest in the home 
manufacture of various common devices such 
as electric clocks and harmonographs and 
designs for these devices were published by 
Model Engineer. Drawings for some of these 
devices are still available and they may still be 
made, together with other devices such as a 
microscope or slide projector, although these 
latter do require that you have available, or can 
obtain, the necessary lenses and other special¬ 
ised components. 

Figure 1.18 A lantern clock. 



Figure 1.19 A well-polished skeleton clock and Figure 1.20 A medal-winning cased clock, 

monogrammed key. 

Today, such items can readily be purchased, 
but there remains an underlying amateur inter¬ 
est in various types of scientific devices which 
were once produced, mostly by hand, to suit 
the scientific interests of the past. Amateurs 
having these interests tend to work alone, pro¬ 
ducing whatever they fancy, based on the origi¬ 
nals and the methods of manufacture employed 
to make them. Very little appears to be pub¬ 
lished in the model press but quite delightful 
and well-made examples appear regularly at 
various shows and exhibitions. 

Clockwork devices figure greatly among the 
items exhibited, and clockmaking generally 
appears to go from strength to strength. Much 
is published concerning the design and manu¬ 

facture of clocks, and drawing sets and books 
are readily available to assist the beginner to 
make a start. 

Clock work is generally quite different from 
most other branches of model engineering. 
Since a clock mechanism is naturally mostly 
wheels and pinions, equipment is required for 
accurate circular division and specially pre¬ 
pared cutters are needed to cut the various 
tooth forms which are required, so some initial 
investment is required in equipment for cutting 

Apart from the cutting of wheels and pin¬ 
ions, much of the work required in making a 
clock, is, or may be, carried out by hand. This 
includes the cutting and truing of plates, the 



crossing out of wheels, burnishing of arbors, 
and so on, and there is much personal satisfac¬ 
tion to be drawn from this kind of work since 
the builder feels that he (or she) has made a sig¬ 
nificant contribution by his or her own efforts. 

If you have a leaning towards woodwork as 
well as metal work, some sort of cased clock 
should satisfy your needs. If not, skeleton 
clocks or metal-cased types provide an alterna¬ 
tive and a transparent dome or cover can be 
obtained to protect the mechanism, once it is 
complete. One additional pleasure which may 
be derived from the making of a clock is that of 
engraving one’s name or initials into the 
mechanism, a practice which is so well estab¬ 
lished that it no longer seems pompous or 
affected to do this. If your engraving is not up 
to scratch (sorry!) there is always the alterna¬ 
tive of cutting the key in the form of your 

If you are content to assemble a clock in the 
first instance, without cutting plates and wheels 
yourself, there are some kits available which 
require the assembler to contribute by cleaning 
up and polishing commercially made compo¬ 
nents and then to assemble the clock. These kits 
are not inordinately expensive and can in any 
event sometimes be purchased in stages, thus 
spreading the cost. These kits provide an 
extremely useful introduction to the art of 
clock assembly and are a valuable means for a 
potential clockmaker to gain some experience 
without the necessity to establish a full work¬ 
shop at the outset. 

Workshop tools and accessories 

There are two views concerning the making of 
tools and accessories, one which regards their 
building as being complementary to, but an 
interruption of, the real business of model¬ 
making, and the second which maintains that a 
well-made tool is a joy to use and care should 

Figure 1.21 A small drilling machine, machine vice and 

Figure 1.22 A Quom tool and cutter grinder. 



Figure 1.23 A non-geared rotary table to the G H Thomas 

therefore be exercised in its manufacture. Both 
views are equally supportable, and no-one 
would totally disagree with the second. What 
the opponents of this view claim is not that 
their tools are poorly made, but that they are 
essentially simple and not therefore consumers 
of large amounts of workshop time in their 

Whatever your view, it is undoubtedly true 
that much of what is required in the workshop 
may be manufactured once a lathe has been 
purchased. Many of the available tool designs 
date from the days when most amateurs could 
not afford to purchase accessories and machine 
tools, beyond the basic lathe, since such were 
not manufactured for the amateur market and 
commercial (industrial) prices were simply too 
high. Commercial machinery was also often 
too big to be housed in the amateur’s work¬ 

There was also a general lack of availability 
of such lathe accessories as collet chucks (at 
least for the size of lathe in which the amateur 
was interested, or could afford) and vertical 
slides. Publication of design details for these 
items, suitable for manufacture on the ama¬ 
teur’s limited machinery, was therefore com¬ 
monplace and drawings, castings and those 
hard-to-get materials were readily available. In 
spite of the fact that there is today a much 

wider availability of some of these items, infor¬ 
mation and materials are if anything more 
readily available and there is significant choice 
if you need to reduce the cost, by making items 

Among the major items which may be manu¬ 
factured on a small lathe are bench grinders, 
drilling machines, sawing attachments or circu¬ 
lar saw tables for the lathe, and motorised 
hacksaw machines. It is common practice to 
make ready-finished components available 
whenever their manufacture presents difficulty 
to the amateur equipped with only limited 
machinery, and this allows relatively large items 
such as vertical milling machines to be com¬ 
pleted on a V/i inch lathe. Various modellers’ 
suppliers specialise in the supply of materials 
and information for tools and other accessories 
and those retail outlets which carry a wide 
range of items for the model engineer usually 
have some items for sale in this category. 

Model boats 

Model boats embrace such a wide range of dif¬ 
ferent classes that one must be careful whether 
they are covered by the general definition of 
model engineering adopted for the purposes of 

Figure 1.24 The African Queen. 



Figure 1.25 A more formal steam launch with two saloons. 

this book. Unless one wants to adopt a metal- 
plate construction for a model ship (and there 
are some who do) a hull constructed in wood or 
glass-reinforced plastic is likely to be used. Its 
construction is more likely to be allied to 
woodworking and other manual skills than to 
model engineering. Boat modelling therefore 
attracts those model engineers also having 
skills, or interests, in woodworking, or allied 

Given these prerequisites, there is much 
scope for experimentation and development 
work. If your interest lies in the building of 
high-speed power boats, together with the 
production of their power plants and drive 
assemblies, there can be a real engineering 
involvement and much room for private 
experiment, not only with the mechanics but 
with hull design, and the transformation of 
engine power into speed over the water. 

Such development work is encouraged by a 
strong network of local clubs and societies and 
by organisations such as the Model Power Boat 
Association (MPBA), in the UK, and the interna¬ 
tionally recognised Naviga which has estab¬ 
lished classes which define groups and sub¬ 
groups of models so that competition craft 
conform to certain basic formulae. 

The recognised classes number some 28,23 
of which are for working boats, as distinct from 
non-working, scale models. These encompass 

straight running and steering classes, radio- 
controlled models, with both IC and electrical 
power plants, and tethered (round-the-pole) 
hydroplanes in both airscrew- and waterscrew- 
driven types. There is a very wide range of 
choice should you be interested in the com¬ 
petitive side of the hobby, but it naturally 
pays to seek out a club whose activities corre¬ 
spond with your own interests since not all 
classes may be vigorously supported within all 

If your interest lies outside the area of com¬ 
petition boating, radio control, nowadays read¬ 
ily available even to those without knowledge 
of electronics, makes possible the sailing of 
power or sail boats (model yachts) as a relaxing 
leisure activity. There has consequently been an 
increase in all forms of pleasure boating, more 
than adequately supported by the kits, fittings 
and radio-control equipment available through 
the model trade. 

Electrically powered scale model boats are 
straightforward to build, either from kits, or 
scratchbuilt using commercially available 
plans. They are clean and quiet in operation 
and within reason can be operated on almost 
any stretch of water to which the public has 
access. Some mechanical design knowledge is 
required in arranging the drive, and the differ¬ 
ent movements which are controlled by radio, 
but these can often be arranged using commer¬ 
cial components and significant workshop 
facilities are not necessarily essential. 

At the other extreme, if steam is your scene, 
plans for hulls suitable for power plants as sim¬ 
ple as a single-cylinder oscillating engine, or as 
complex as a triple-expansion power plant, are 
readily available. There is thus, as in other 
branches of the hobby, a wide range of activities 
available for those who are interested. 

Radio control of live steam models is well 
established in the smaller model railway scales 
(principally gauge ‘O’ and gauge ‘1’) and this 
form of control is ideal also for steam-powered 



boats. Those afflicted with the three model 
‘diseases’ of live steam, model boats and radio 
control naturally find much of interest, and 
scope is available for trying out your own ideas, 
or just doing your own thing. 

Most sailing model boats are of the yacht 
type, having what is known as a fore-and-aft 
rig, as distinct from a square rig. The most 
popular models are of the yacht type, having jib 
and main sails, and sometimes even a spinnaker 
for use when sailing downwind. Again, there 
are internationally recognised standard classes, 
formulated to allow competitions to be staged, 
and many of the club activities revolve around 
the racing of class boats. 

Radio control is naturally widely used, 
and deservedly popular, but racing of vane- 
controlled models is still widely organised and 
if you are interested in sailing a model influ¬ 
enced only by the wind, once released, without 
the encumbrance of radio control, a local club 
is almost sure to exist to satisfy the need for 
facilities and to organise competitions. 

Radio-controlled sailing of a model yacht 
requires control of the sail settings and also of 
the rudder. Two servo-controlled winches and a 
tiller control are therefore required and a good 
basic knowledge of sailing generally is needed 
to obtain the best out of a particular model 
under the prevailing wind and weather condi¬ 
tions. Again, various recognised classes are 
established to allow competitive sailing to be 
organised, radio-control model racing being 
organised and run pretty much as is the full-size 

If the competitive element does not attract 
you, a more true-to-scale type of model may be 
more appropriate (racing models tend to be 
devoid of any non-functional parts). Once 
again, within reason, this type of leisure sailing 
can be undertaken on almost any publicly avail¬ 
able water. 

Model aircraft 

Like model boating, the flying of model aircraft 
attracts model engineers with an interest in the 
building of power units, and the production of 
engines of appropriate size and having the 
desired power:weight ratio can be a challeng¬ 
ing field for experimentation. 

Competitive flying is still the basis of club 
activities and models are classified into three 
major groups: free-flight, control-line and 
radio-controlled models. Control-line models 
are naturally all powered, but even so, several 
sub-groups are established including stunt and 
speed flying, team racing and combat. There is 
also an interest in the control-line flying of 
scale models but most of these appear in the 
free-flight group. Free-flight also embraces 
rubber-powered models, gliders and radio- 
controlled gliders, while powered models pro¬ 
vide opportunities for aerobatic competitions 
and pylon racing. 

The confirmation of the feasibility of mak¬ 
ing and flying model helicopters has also pro¬ 
vided a field for amateur development of any¬ 
thing from the mechanics to the complete ma¬ 
chine, although most modellers still prefer to 
purchase a kit or ready-assembled unit. How¬ 
ever, the possibility exists, provided that you 
have the requisite theoretical knowledge and 
the capability to design and manufacture the 
required parts. 

Model cars 

Two basic types of model car are produced by 
model engineers: the strictly scale model, and 
the functional, sporting or racing car, normally 
radio controlled and used simply for fun, or for 
competitive club racing. These racing models 
are most commonly electrically powered since 
this allows clean, quiet models to be produced 
which are suitable for racing both indoors and 



Figure 1.26 A fine model Bugatti. 

out, thus permitting all-year-round racing. 

Club racing is, however, well organised for 
IC-engined cars, the models being made to /% 
scale and powered by 3.5cc glowplug engines. 
Full radio control is fitted, as it is for electri¬ 
cally powered racing, modern radio equipment 
allowing several cars to race together, simply by 
using different crystals. Due to the noise and 
smoke, and the possibility of fuel spillages, IC- 
engined models are normally raced only on 
outdoor circuits. 

Electrically powered racing has developed 
since the introduction of fast-charge nickel- 
cadmium batteries. A battery pack of this type 
allows perhaps 7 or 8 minutes racing before 
exhaustion, but may then be fully recharged in 
20 or 30 minutes, allowing further racing. 
Clubs usually arrange a full racing programme, 
individual events lasting 5 minutes plus one lap, 
the day’s racing culminating in various group 
finals according to performance during the pre¬ 
liminary events. 

Due to the popularity of electrically pow¬ 
ered racing there is significant support from the 
trade and it is not normally necessary to be a 
fully-equipped model engineer in order to par¬ 
ticipate. Extremely comprehensive kits are 
available and a range of screwdrivers and span¬ 
ners is all that is required in order to assemble a 
commercial kit. Assembly instructions are usu¬ 
ally well written and a mechanically minded 12 

year-old can readily carry out the assembly, 
given a modicum of adult supervision. 

This is not to say that the models are simple. 
Four-wheel drive through differentials, to¬ 
gether with independent suspension and 
Ackerman steering, are normal, together with 
oil-damped shock absorbers. Variations in 
treads and wall rigidity in the tyres, and the 
type of shock absorber, allow experiments in 
relation to the suspension and road holding. 
There is also the never-ending search for the 
‘best’ motor (within the limits allowed by the 
class rules) and the selection of gear ratios to 
provide what one considers to be the best 
compromise in terms of top speed, maximum 
acceleration and battery duration. 

To contain these experiments within a for¬ 
mal framework, there are recognised racing 
classes, relating broadly to motor type, and 
operating voltage, and determining whether 
motor modification is permitted. Naturally, if 
you wish to join the club racing scene, you must 
adopt one of the recognised classes of vehicle. 
The choice lies between 1C engine or electric 
motor drive, the choice for this latter type 
being either a straight racer or a buggy. 

Unlike model car racing, scale model car 
building is not well supported by the trade and 
is therefore a scratchbuilder’s activity, usually 
based on a detailed study of particular proto¬ 
types. Such models may be built in any materi¬ 
als with which you are familiar, but the very 
best are made with the ‘correct’ materials as far 
as possible and may also be internally correct if 
built to a sufficiently large scale and level of 
detail. For this type of model, a lathe and a 
capability to carry out milling are essential. 

Modelling and making 

Modelling, in relation to model engineering, 
can simply be interpreted as ‘making mini- 



atures\ Modelling is thus distinctly different 
from making real things which are not mini¬ 
atures, and what will do for de-burring a hole 
in a bracket for hanging a shelf on will not do 
for de-burring a hole in the frames of a model 

If you are building a model which is one- 
tenth the size of the prototype, a .OlOin. cham¬ 
fer on the edge of a hole is equivalent to 0.1 in. 
on the prototype. This may be acceptable in the 
case of a hole, but surface imperfections of as 
little as .005in. are almost sure to show, and in 
any event represents .050in. on the prototype, 
at one-tenth scale. 

Surface blemishes must be considered for 
each part of the model. Many of the materials 
which are bought have surface imperfections of 
one sort of another. Bright-drawn steel rods 
and bars frequently have inclusions in the 
surface which can well render them unsuitable 
for use if the need is to model a polished com¬ 
ponent and it is often prudent to buy the next 
largest size and machine the item all over, 
rather than take the easier route of removing 
just the minimum of material and trying to 
polish out the blemishes on unmachined sur¬ 

When sheet materials are required, they are 
usually produced by shearing off the required 
size from a large sheet using a guillotine. Guil¬ 
lotined plates are often more seriously flawed 
than rods and bars. They frequently show con¬ 
siderable distortion along a cut edge since the 
cutting process tends to turn the edge over, as 
well as cutting it, and the result is sometimes 
very distorted. Considerable roughness can 
also be evident on the edge, and the very mini¬ 
mum which needs to be done is to file the edges 
smooth. This means that the plates need to be 
purchased a little larger than the required final 
size, and in view of this it is well worthwhile 
getting the sheets cut well oversize, afterwards 
cutting and filing the edges to the correct size 
and profile. 

In practice, many of the models which are 
built are smaller than one-tenth scale. A 3 Vi- 
inch gauge locomotive is one-sixteenth full size, 
and even a 5-inch gauge model is a little smaller 
than one eleventh of its full-size counterpart. 
This is an extremely important point to bear in 
mind if the aim is the production of a fine 
model. Everything must be in scale. 

On a perfect model, this means that the 
heads of bolts and screws are the correct type 
and size, that all fastenings are present, or 
appear to be so, and the model appears to be 
constructed as was the prototype. This means, 
for example, that the fastenings will be hexa¬ 
gon-head bolts and scale-size rivets, and the 
more convenient slotted-head screws should 
almost never be used as fasteners on a model. It 
means that there is a need for quite small 
fastenings. A 1-inch bolt scales at .088in. for 
1 '/i6 inch to the foot scale (a 5-inch gauge loco¬ 
motive) and this means using an 8 BA bolt on 
the model. If the model is built to 'A inch to the 
foot scale (3 Vi-inch gauge) a 1-inch bolt scales 
at '/win. which means something between 10 
and 11 BA. Similar considerations apply to 
rivets, and once again, small sizes which are in 
proportion with the scale of the model are 
essential to achieve correct appearance. 

Of course, this doesn’t mean that all of the 
rivets need to be present. They certainly do not 
all need to be holding the parts together. Pro¬ 
vided that there is sufficient security, some of 
the rivets can be just dummies, and it is possible 
to solder parts together and insert rivets, simply 
for appearance sake. The tanks of a model ten¬ 
der can often be better treated in this way, and 
it is usually simpler to put the dummy rivets in 
position in a plate, countersinking them on the 
underside, and leaving just one or two here and 
there for actual attachment, and then soldering 
the tank together. 

Care needs to be paid to the fit of parts. If 
you are making a plant stand or a firescreen in 
steel scroll work, a Vtsin. gap between parts 



which nominally touch one another is unfortu¬ 
nate, but not usually a disaster. On a 5-inch 
gauge locomotive (l'/is inch to the foot scale), 
such a gap represents over s /sin. and if it isn’t 
supposed to be there, its presence will severely 
mar the model and spoil its scale appearance. 

Achieving scale appearance also means pay¬ 
ing particular attention to surface finish, and to 
the thicknesses of parts, particularly the plate 
work, in situations in which the edges of the 
plates are visible. This means using thin plates, 
since a Vi-inch plate on a prototype locomotive 
must be represented by .022in. on a 5-inch 
gauge model, and by .0156in. on its 314-inch 
gauge counterpart. Thus, quite flimsy struc¬ 
tures can be produced and this means handling 
the complete, or half-completed model with 
suitable care. 

It is arguable that such a model is too fragile 
to take out on the track and this then raises the 
question of whether there are different stand¬ 
ards of models which are made for different 
purposes - practical working models, and those 
made for exhibition or for the domestic side¬ 
board. The working model may be more 
robustly made in certain parts, and may not 
strictly be a scale model. In this case, it still 
needs to look like the prototype it represents, 
and should aim to capture the spirit of the 
original rather than be a faithful miniature 
copy. It may be more simply made, have fewer 
rivets, be modified to make it easier for a ‘non¬ 
scale’ driver to handle, have parts omitted for 
access, and so on. Nevertheless, if it gives 
the impression of being a miniature version 
of the original, the designer and the builder 
have collaborated well in producing a good 

In truth therefore, absolute scale is not nec¬ 
essarily essential, but correct proportion is, and 
it is vital that this sense of proportion is carried 
through in every aspect of the model. It pays to 
have a clear idea of the intention right at the 
outset since it is disappointing if the finished 

model doesn’t quite come up to expectations at 
the end. 

Since large models naturally take a long time 
to make, and there are many parts, your expec¬ 
tations and ability at the end will be different 
from those with which you set out at the begin¬ 
ning. It is, therefore, important to choose some¬ 
thing easily achievable for your first model so 
that your skills and ambitions don’t have too 
long to change during its building. Provided 
that the aims don’t change during the building 
of a model, you should feel at the end that the 
initial objectives have been achieved. 

By keeping your mind firmly on the initial 
objective, which ought to be ‘build a simple 
steam engine*, ‘build a simple clock’ or what¬ 
ever, you shouldn’t divert yourself by becoming 
convinced that parts made at the beginning 
need to be remade. You should not keep remak¬ 
ing parts, and you should avoid abandoning a 
project on the basis that it doesn’t meet your 
current aims. The current aims ought properly 
to be applied to the next project. 

Workshop machinery 

The machinery with which the workshop is 
equipped is dependent on many factors. Your 
own inclinations and interests naturally play a 
part, but opportunities for acquiring machinery 
also vary from one individual to another and if 
you work in a situation in which secondhand 
items occasionally become available, you may 
be fortunate enough to obtain a wider range of 
basic machines and accessories than other mod¬ 

The operations which need to be performed 
in the workshop are essentially two-fold: cut¬ 
ting material from the work so that it assumes a 
circular cross-section, and the alternative of 
cutting material away so that a flat surface is 
produced. Cylindrical shapes, or those with 



circular cross-sections, such as cones, are pro¬ 
duced by rotating the work and bringing a cut¬ 
ting tool into contact with it. 

This process is illustrated in Figure 1.27 
which shows a steel billet which is rotating, 
while a cutting tool is used to reduce its diam¬ 
eter by being pushed progressively along its 
length. This process is known as turning 
(describing the motion of the work) and is the 
operation for which a lathe is designed. 

Although rectangular shapes can be formed 
on a lathe, flat surfaces are usually formed on a 
machine in which the cutting tool rotates and 
the work is fed into the path of the cutter in 
such a way that the desired shape is produced. 
One simple form of this operation is shown in 
Figure 1.28. A tool having cutting edges on its 
outside diameter is being used to profile the 
edge of a thick steel plate. 

This process is known as milling and the 
machine on which the process is carried out is a 
milling machine. These machines are produced 
in two basic types, known as vertical and hori¬ 
zontal milling machines, depending upon 
whether the cutter rotates about a vertical or a 
horizontal axis. The two types of machine differ 
fundamentally in the manner of their construc¬ 
tion, and a horizontal mill (to use the colloquial 

Figure 1.27 A straightforward turning operation-reducing 
the diameter of a circular steel billet. 

name) is not simply a vertical mill turned on its 
side. The uses of the two types of mill differ, and 
while a vertical mill will not do all of the things 
that a horizontal mill is capable of doing, the 
vertical mill is really the one to buy, since a hori¬ 
zontal mill is not suitable for much of the work 
which the modeller will wish to perform. 

Manufacturers of milling machines over¬ 
come the problem posed by the different capa¬ 
bilities of the two types of machine by making a 
vertical head which effectively converts a hori¬ 
zontal mill into a vertical mill. Such machines 
thus provide the major capabilities of both 
types of mill. 

Most real milling machines are large, floor¬ 
standing structures provided with a large table 
which can be moved by calibrated feedscrews 
in three direction: along and across horizon¬ 
tally, and up and down, vertically. The work is 
mounted on the table and can be positioned in 
all three directions for a cut to be applied. 

In recent years, small bench-mounted verti¬ 
cal mills have been introduced onto the market. 
Due to the bench-mounted arrangement, the 
table cannot be moved in the up-and-down 
direction. Instead, the spindle is provided with 
a down feed operated either by a lever or a cali¬ 
brated handwheel. The lever-operated down 

Figure 1.28 Milling the profile of a steel plate. 



feed arrangement is precisely like that on a 
drilling machine and a mill of this type is de¬ 
scribed as a mill/drill to distinguish it from the 
more usual type of industrial machine. 

Most of the operations for which a vertical 
milling machine is used industrially can be per¬ 
formed by adapting a lathe, using a rotating 
cutter in the position normally occupied by the 
work, and cutting material mounted in the 
position usually reserved for the tool. The lathe 
must be provided with additional accessories in 
order that this adaptation can be successfully 
applied, but this is far less costly than buying a 
milling machine or a mill/drill. It also takes up 
virtually no space, and is therefore frequently 
the way in which milling is performed in the 
amateur’s workshop. 

One way to obtain both a lathe and a milling 
machine is to purchase a combination. Several 
manufacturers produce lathes for which sup¬ 
plementary vertical milling attachments are 
available. Since the lathe is provided with a 
table which has calibrated and controlled 
movement in two directions, it is possible to 
perform milling by fitting a column-and- 
spindle assembly to the rear of the lathe. This 
saves space and money, both valuable com¬ 
modities, but brings the disadvantage that the 
workholding table on the lathe may be smaller 
than that on the equivalent mill/drill. 

Some manufacturers provide a solution to 
this problem by producing both a mill/drill and 
a lathe in modular form, so that the column 
assembly of the mill/drill can be fitted to the 
lathe by using an adaptor. The column assem¬ 
bly naturally also fits the sturdy base and large 
co-ordinate table of the mill/drill, so the two 
machines can be bought progressively, first the 
lathe, then the vertical head, and afterwards the 
co-ordinate table. 

Because the simplest adaptation of the lathe 
for milling, (without a supplementary vertical 
head) has its shortcomings, principally a 
restriction on the size of the workpiece, the 

amateur production of rectangular work is 
sometimes done on a shaping machine. In this 
type of machine, the work is bolted to a table, 
and a tool not unlike a lathe tool is passed over 
it, acting much in the way that a plane does for 
wood. Indeed, there is little difference in prin¬ 
ciple between a planer and a shaper, as far as 
metal cutting is concerned. 

The shaper seems to have fallen out of 
favour in recent years, but it was at one time 
readily available in the form of small bench- 
mounted models suitable for amateur use. 
Many of these were operated by hand, called 
hand shapers, but were often also available in 
motorised form. They are useful in allowing 
larger rectangular items to be shaped, and this 
is an operation which can be difficult or slow if 
only a small lathe is available. A shaper is less 
costly than a milling machine, takes up less 
space and represents a useful addition to the 
small workshop if space and money do not 
allow a mill/drill to be obtained. 

A third operation, which complements the 
production of circular and rectangular work, is 
that of drilling holes in items made by the other 
processes. If you are lucky enough to be able to 
obtain a mill/drill at the outset, this may well be 
used for much of the drilling which is required. 
It is not absolutely ideal for this work since in a 
mill the drill cannot pass through the table as it 
usually can on a small drilling machine, and this 
means that packing must be put below the work 
to allow penetration, unless use can be made of 
the clearance provided by the workholding 
slots. Even so, this doesn’t allow a drill or 
reamer to be put right through the work, and a 
drilling machine is certainly not a luxury even if 
a vertical mill or mill/drill is available. 

The requirement for a small drilling 
machine is considered at the beginning of 
Chapter 10. 

If workshop space and money are readily 
available, there is much choice of machinery 
available. For most people, a start is usually 



made with a lathe and a drilling machine, but 
some basic accessories are required to comple¬ 
ment these two items. The lathe requires 
workholding chucks and a faceplate, together 
with a drill chuck for the tailstock and the drill¬ 
ing machine (drill) requires a suitable machine 
vice. Tool bits for the lathe, and drill bits for 
both machines will naturally be required and 
there must be some means available for sharp¬ 
ening these. A small, general-purpose grinder 
thus complements the two basic machines. 

Once these machines are purchased and set 
up, the basics are available. 

If your needs are very specialised, and you 
habitually undertake much heavy milling, or 
need to cut gears and pinions frequently, it may 
be too time-consuming to utilise an adaptation 
of the lathe, or in the case of heavy milling, may 
unnecessarily strain the machine, or be beyond 
its capability. In these instances, you will need 
to consider the purchase of specialised machin¬ 
ery to suit your particular needs. Possible addi¬ 
tional items will suggest themselves as your 
modelling progresses. 

As noted elsewhere, time and money are 
usually in short supply, but in some ways they 
are interchangeable, since money spent on a 
specialised machine (and the space to house it) 
frequently leads to a saving in time, as it is not 
necessary to install (and later remove) an adap¬ 
tation to another device. A specialised machine 
might, in any event, be more appropriate and 
thus inherently more productive. 



Someone once said that there are only three 
types of tool: those that cut you, those that hit 
you, and those that fall on the floor and roll 
under the bench out of sight! Note that two- 

thirds of all tools are therefore dangerous. 
Many of the processes which are employed in 
the amateur’s workshop are governed by the 
Health and Safety at Work Act or by other legis¬ 
lation, when used in industry. Industrially, ma¬ 
chines too are governed by legal requirements 
which demand that guards be correctly fitted, 
and be used at all times, in order to try to 
ensure that the more obvious dangers posed by 
rotating machinery and work do not become 
sources of personal injury. 

Employees in industry are expected to con¬ 
tribute to their own safety by adopting a style of 
dress that avoids such obvious dangers as flap¬ 
ping clothes or flimsy shoes, or to avoid possible 
dangers caused by too-long a hairstyle, and it is 
obligatory for many operators to wear protec¬ 
tive spectacles or goggles as a matter of course. 
For some occupations, the wearing of protec¬ 
tive footware is obligatory and for amateurs, a 
reasonably stout shoe or boot is a sensible pre¬ 
caution: even a lathe chuck key dropped on to a 
toe inside a flimsy shoe can cause painful injury. 

We should take note of this. There are just as 
many potential hazards in the amateur’s work¬ 
shop. In fact, there may well be more, since our 
machines are, for the most part, manually oper¬ 
ated and we are therefore that much closer to 
the actual cutting processes than some of our 
industrial counterparts. In the home workshop 
there are also likely to be more processes car¬ 
ried out than in the average industrial ‘shop’ 
since we normally carry out machining, heat 
treatment, chemical processes such as pickling 
and etching, spray painting and even the cast¬ 
ing of metal. Greater, rather than less, care is 
therefore needed at home. 


Perhaps the greatest potential hazard to hands 
and eyes are the machining operations which 
are performed on lathes, drilling and milling 



machines. Either the workpiece or the cutter is 
rotating and there are consequently immediate 
dangers from the rotating item. So keep hands 
and fingers well out of the way. 

There is often a temptation to ‘feel’ the fin¬ 
ish on the work while either it or the cutter is 
still rotating. Don’t do it. Under some circum¬ 
stances it may be less dangerous than under 
others, but if the work or cutter is asymmetrical 
(has a ‘lump’ going round well off the axis of 
rotation) as is often the case, the ‘lump on the 
side’ will catch you a nasty bang and part of a 
finger will be lost, or even worse. So, once 
again, don’t do it. If you must remove swarf 
while drilling or milling, then use a brush. 

Swarf comprises the curls of metal that are 
removed from the workpiece by the cutting 
tool. This comes off in different forms depend¬ 
ing upon the tool, the cutting process and the 
material, but all swarf is potentially dangerous. 
It may be hot enough to burn, sharp enough to 
cut and flying about all over the place, more or 
less all at the same time. Some of the old-time 
machinists used to say that you hadn’t become 
a professional until you had burned a few hairs 
off your chest (ladies excepted, of course). This 
is amusing at first sight, but it is also sympto¬ 
matic of the kind of bravado that used to over¬ 
ride matters of safety at one time, especially for 
a generation of machine shop men who were 
brought up in the days of machines driven by 
overhead line shafting and totally unguarded 

It may be bearable if a piece of very hot 
swarf sticks to the forearm and produces a 
small burn, or goes down the front of an open- 
necked shirt to stick somewhere on the neck or 
chest. Unfortunately, such burns always come 
at an unexpected moment and what you might 
do to shake off the offending chip causes as 
much danger as the hot chip itself. It is not an 
affectation to wear a warehouse type of coat in 
the workshop. Its tapering sleeves, narrowing 
at the cuff, protect the arms from the abrasive 

and burning effects of the swarf which is pro¬ 
duced and, being tapered, are not likely to 
present a hazard by becoming caught up in the 

The danger to the eyes is obvious. Not only 
may the swarf be very hot, but the individual 
chips can be quite large under some circum¬ 
stances. They also have sharp and very ragged 
edges and will do much damage if they come 
into contact with the eyeball. All-enveloping 
safety goggles will protect the eyes completely 
from these dangerous particles of swarf, and 
will also provide complete protection should 
the unexpected happen, such as the shattering 
of a cutter. 

Safety spectacles may be more comfortable 
to wear, but they do provide less comprehen¬ 
sive protection than safety goggles, and it is 
vital that you keep your head as far as possible 
out of the way of flying swarf, particularly if the 
machining operation is producing its swarf as 
the result of an interrupted cut. In these cases 
the shortish cuts produce distinct chips which 
simply fly into orbit once they are removed 
from the workpiece. 

If the cut is continuous, swarf tends to come 
off in relatively long strands due to being still 
attached to the workpiece at the point at which 
the cut is being made. Even when detached, the 
longer lengths of swarf, having more mass, are 
not expelled from the work (or tool) with the 
same velocity and are therefore less dangerous 
on this account. The greater mass also tends to 
allow heat to dissipate more rapidly along the 
chip and, in any case, some parts of a long chip 
have been removed from the work for a 
relatively long time so they are cooler. 

The machining operations posing the great¬ 
est hazard are therefore when fly cutting or 
when facing the end of a rectangular bar in the 
lathe. Both of these operations produce short 
pieces of swarf which are completely separated 
from the work, the short curls of swarf pro¬ 
duced when cutting across the corners of a 



square bar being much hotter than when cutting 
continuously near the centre of the bar. They 
are also much more likely to fly about due to 
their complete detachment from the job. Pro¬ 
tective goggles are readily available from local 
DIY stores and should be used whenever there is 
a likelihood that swarf will be flying about. 

Drilling machines also produce their fair 
share of swarf which naturally presents the 
same hazard as that produced when milling or 
turning. The drilling of steel frequently pro¬ 
duces long curls of swarf that wind up the drill 
flutes and go round with the drill. These often 
become quite long and project from the top of 
the drill shank. Since they are rotating with the 
drill, they pose a threat to the hands, which are 
usually in the immediate vicinity of the work, 
and also to the eyes, especially if the drill is 
bench mounted as is often the case. 

Curls of swarf produced by a large drill are 
shown in Figure 1.29. 

The way to prevent the occurrence of these 
long curls is to ease the pressure on the drill 
marginally from time to time during the drill¬ 
ing process. This reduces the rate of cut at the 
drill point and therefore reduces the thickness 
of the curl, weakening it and causing it to break 
off. Reducing cutting pressure completely stops 
the cutting action and the curl becomes 
detached from the work, so the long curls can 
be prevented from forming by varying the cut¬ 
ting pressure. If the drill flutes do get jammed 

Figure 1.29 Curls of swarf. 

up with swarf, stop the machine and clear them 
before any damage or injury is caused. 

The machining of some forms of brass 
produces a quite different swarf from that pro¬ 
duced by steel. This comprises fine, needle-like 
chips, some of which are visible but many of 
which are not. The smallest chips are akin to 
cut hair, and being small in diameter they are 
extremely sharp and readily penetrate the skin 
and can be very painful, and very difficult to 
extract, if they lodge in the fingers. Once again, 
the danger to the eyes is obvious, but easily 
forgotten, and if you are turning brass rod (par¬ 
ticularly) in the lathe, take great care not to rub 
the eyes unless you are absolutely sure that no 
brass swarf is on the hands or fingers. 

Soldering and brazing (silver soldering) 

There are always some dangers posed by sol¬ 
dering and silver soldering, even the making of 
a simple electrical joint using soft solder. Mate¬ 
rials are being heated to around 200°C and a 
large, high-temperature heat source is in use. 
There is thus the potential danger of fire. It is 
therefore important that these operations are 
carried out in a safe environment, with ade¬ 
quate facilities to prevent accidents, and even 
the ordinary, electrically powered soldering 
iron must be provided with a protective stand if 
accidents are to be avoided. 

When brazing or silver soldering, the dan¬ 
gers are greater. Much higher temperatures are 
required - up to 750°C. A workpiece such as a 
locomotive boiler can be very large and a large 
source of heat is required to raise it to the cor¬ 
rect temperature. Heat is commonly provided 
by a propane gas torch having a large diameter 
and projecting its flame 12in. (300mm) or 
more from the burner head. The dangers are 
obvious. The workpiece becomes very hot and 
there is much radiated heat. Very hot air 
surrounds the burner flame and this can readily 



ignite any flammable material in the vicinity, 
even though this is not in the direct flame. A 
clean, dust-free location is essential if silver 
soldering is to be carried out as is some dedi¬ 
cated area containing a steel-framed hearth. 

There are also subsidiary potential dangers. 
Some silver solders can decompose and give off 
poisonous gas if maintained at high tempera¬ 
tures for some time in the liquid state. There¬ 
fore it is vital that adequate ventilation is avail¬ 
able so that any gaseous emissions are diluted 
by a good air supply. 

The dangers posed by leaks of the bottled 
propane gas are also obvious and valves, hoses 
and equipment generally must be kept in good 
condition and be inspected regularly to ensure 
that no accidental leaks occur. Hoses must be 
connected to the gas bottle through a hose fail¬ 
ure valve (as a minimum) which is designed to 
cut off the gas should excessive flow rates be 
detected. A regulator is the alternative connec¬ 
tion between hose and bottle and will normally 
also incorporate a safety device. 

It must also be borne in mind that many of 
the fluxes used during soldering are mildly poi¬ 
sonous, and care must be exercised during stor¬ 
age, mixing and use, that they are not accessible 
to children or pets and cannot contaminate 
food. It goes without saying that the hands 
should always be washed thoroughly after han¬ 
dling any chemicals. 

One further danger posed by brazing is the 
use of an acid ‘pickle’ for cleaning the work and 
removing flux residues. The pickle used is 
dilute sulphuric acid. Even in its dilute form, 
this acid is highly corrosive. For such activities 
as model boilermaking it is needed in relatively 
large quantities and safe methods of storage 
and use are absolutely essential. 

However, the greatest danger comes when 
preparing the dilute acid. It is supplied in con¬ 
centrated form, which is even more corrosive, 
and it must be mixed with water to provide the 
working solution. ON NO ACCOUNT MUST 

TRATED ACID. The first drops of water to 
touch the acid cause a chemical reaction not 
unlike an explosion, thus spreading acid in all 

Further information relating to all of these 
hazards is given in Chapter 9. 


Ordinary brush painting poses no particular 
hazards other than those posed by the flamma¬ 
bility of the solvents and thinners used, also not 
forgetting that prolonged contact with any 
chemicals can lead to skin problems in particu¬ 
lar cases. 

However, much model painting is today 
undertaken by the use of miniature spray guns 
and air brushes, when paint and the medium 
which carries it (the thinner) is atomised into 
fine droplets and sprayed onto the work. Rapid 
evaporation of the medium then occurs, as the 
paint dries, thus ‘loading’ the atmosphere with 
further amounts of chemical. 

Several inherent dangers are present. The 
very fine droplets of paint and medium can be 
inhaled readily and will contaminate nasal and 
bronchial passages. Also, the air in the immedi¬ 
ate vicinity of the operator can become so 
loaded with solvents that unconsciousness can 
occur, and the results can be fatal. To avoid 
these dangers it is vital to ensure that there is 
adequate ventilation available to prevent the 
build-up of chemicals in the atmosphere. It is 
helpful if the work is contained within a box 
and enclosed on all sides except the front. This 
allows excess paint and overspray to be con¬ 
tained within the box, and also helps to contain 
the airborne solvent medium. 

To prevent ingestion of paint droplets, a sim¬ 
ple fabric- or paper-based mask is generally 
adequate. Such masks are available from DIY 
stores, but it should be borne in mind that they 



are frequently intended to prevent the inges¬ 
tion of dust during cutting and grinding opera¬ 
tions and therefore do not provide a barrier to 
the inhalation of solvents so their use does not 
remove the need for adequate ventilation. 

To prevent excessive build up of solvent 
vapour in the atmosphere, it is helpful to carry 
out spray painting in short sessions, allowing 
plenty of time for the air to return to normal 
before continuing. If a great deal of spray paint¬ 
ing is likely to be required, a small booth venti¬ 
lated by an electrically driven fan should be 
considered as essential. 

Metric or imperial? 

Going metric presents a particular problem for 
the modeller, principally because many of the 
items one wishes to make will be based on 
designs which were prepared when the only 
conceivable measuring system to use in the UK 
was the imperial one. 

This means that the whole device will have 
been designed to use imperial materials. All 
dimensions will be in imperial units and it fol¬ 
lows therefore that imperial measuring equip¬ 
ment will be the most convenient to use. Unfor¬ 
tunately, in the UK we are now in a non¬ 
imperial, but at the same time a non-metric, 
environment. The preferred British Standard 
drill sizes are now metric; some, but only some, 
of the materials which we purchase are now 
specified in metric dimensions; many of the 
drawings which are used are totally imperial 
and many of us only have imperial measuring 

A further problem has also arisen, in that it is 
likely that many readers of this book will be 
totally unfamiliar with the imperial system, 
having passed through their school years at a 
time when the government was totally commit¬ 
ted to the conversion of industry to metric 

measurements. Consequently, their schooling 
was entirely metric and for them, the mere 
mention of sixteenths and sixty-fourths of an 
inch, or the sight of an imperial rule, makes the 
whole business of making measurements seem 
about as complicated as degree mathematics. 

There are thus likely to be problems for 
those unfamiliar with the imperial measuring 
system, since very few sets of drawings are 
available in metric scales. One basic problem 
which arises when using an unfamiliar measur¬ 
ing system is that it is not always possible to 
visualise a size simply by knowing what it is. 
Thus workers in imperial units may have no 
idea how big a 1 Omm-wide strip of metal is, and 
metric workers may not be able to visualise how 
big a 'Ain. diameter hole is. To avoid problems 
which might be caused by the adoption of one 
system rather than the other for use in this 
book, where dimensions are given, they are 
given in both systems, imperial first, followed 
by the metric equivalent in parentheses. 

This policy has, however, caused some 
inconsistencies to appear since some metric 
dimensions given are rounded equivalents of 
the imperial size, whereas others are the com¬ 
mercial equivalent size of the imperial. For 
example, the metric equivalent size for a Vs in. 
square high-speed tool bit is 10mm and this is 
referred to asVsin. (10mm). On the other hand, 
the rounded metric equivalent of Vain. is 
1.6mm, making it appear that '/sin. is the same 
as 9.6mm. It is hoped that the context will pro¬ 
vide sufficient clues for you to decide the inten¬ 
tion in each case. 

It should be particularly noted that there are 
no accepted metric equivalents of the larger 
model railway gauges, from 2 '/ 2 -inch gauge 
upwards. If you build in one of these gauges 
you must at least ensure that your wheel stand¬ 
ards (flange width, tread width, back-to-back 
measurement etc.) comply with the accepted 
standards or you may not be able to run your 
model on the available tracks. 



Information and supplies 


Some research and background reading is a 
necessary part of taking up any new hobby. The 
most obvious sources of initial (general) infor¬ 
mation are the bi-monthly, monthly or fort¬ 
nightly magazines which are now so readily 
available from newsagents. The periodical 
which most readily comes to mind is Model 
Engineer (published by Nexus Special Interests) 
due to the fact that it has been published since 
the very beginning of the ‘home workshop 
experimentation’ era and carries the name by 
which the hobby is now known. 

The editorial staff of Model Engineer have 
always tried to represent as wide a range of 
interests as possible in the magazine, although 
they are obviously dictated by readers’ interests 
in that they publish only the articles which their 
contributors write. Over the years, a tremen¬ 
dous amount of information has been published 
in Model Engineer and much of this is still avail¬ 
able in the form of sets of drawings. There is 
also a healthy trade in secondhand copies of the 
magazine and bound volumes are quite as col¬ 
lectable as are other types of secondhand books. 

In the period prior to and immediately fol¬ 
lowing the 1939-1945 war, Model Engineer 
reflected the gradual widening of interests in 
modelling activities and at that time published 
articles covering model power boats and their 
power plants, model yachts, aeromodelling, 
stationary steam engines, locomotives and trac¬ 
tion engines, together with electrical and opti¬ 
cal equipment such as microscopes and the 
then relatively rare slide projectors. 

The hobby was perhaps less well supported 
by the trade in those days and in any event, 
modellers probably had relatively less dispos¬ 
able income that could be devoted to a hobby, 
and much that would nowadays be available as 
castings had to be fabricated. Treadle-operated 
lathes were also still quite common until the 
1940s and this influenced the way that lathe 
work was approached since one tends to notice 
pretty quickly that the tool is blunt if the power 
for cutting is provided by the leg muscles. 
Reading some of these older issues of the maga¬ 
zine is quite instructive, and it is surprising how 
many ingenious dodges were developed at that 
time to overcome the limitations of the avail¬ 
able machine tools. 



Drawings for many of the major projects 
designed during the 1930s, and since then, are 
still available and access to these and to the 
more recent designs is described under the 
heading “Sources of drawings and designs” 

Currently, Model Engineer provides articles 
covering large-scale steam locomotives, trac¬ 
tion engines, agricultural machinery, tools and 
workshop items for home construction 
together with reviews of equipment, notes on 
the prototype and so on. The increased inter¬ 
est in small-scale live steam for garden rail¬ 
ways is also reflected in occasional articles 
covering this subject. 

A newer magazine which provides informa¬ 
tion on a similar range of topics is Engineering 
in Miniature. This magazine commenced publi¬ 
cation in 1978 and it is now well established as 
a monthly periodical which complements 
Model Engineer in many ways rather than just 
being a competitor by covering the same range 
of topics. 

As the various branches of the modelling 
hobby have become more specialised, new 
magazines have appeared on the bookstalls in 
order to allow the editorial content to expand. 
Aeromodeller has been published for many 
years now, but more recent titles are Model 
Boats, RCM&E, RJC Model Cars and Radio 
Modeller. Younger still, is Model Engineer’s 
Workshop. This was originally introduced as a 
quarterly magazine, but is now issued every 
two months. It covers all aspects of workshop 
practice and is an invaluable source of informa¬ 
tion for metal workers, of whatever persuasion. 
( Aeromodeller, Model Boats, RCM&E, RJC 
Model Cars, Radio Modeller and Model Engi¬ 
neer’s Workshop are all published by Nexus 
Special Interests.) 

If your local newsagent does not have addi¬ 
tional copies of these magazines for display on 
his shelves for casual sales, he will be quite will¬ 
ing to place an order for you for delivery with 

your daily paper or for collection, if you do not 
have a daily delivery. Another way to receive a 
magazine regularly is to take out a subscription 
and have it posted to you by the publishers. 

Moving away from strictly modelling activi¬ 
ties, clockmakers are catered for by the 
monthly Clocks magazine (also published by 
Nexus Special Interests). 

Becoming a regular reader of a modelling 
periodical brings one into the ranks of the arm¬ 
chair modeller. This is a desirable step forward 
since it means only a modest financial outlay 
yet it brings you into the learning phase. It 
should allow you to become aware of the exist¬ 
ence of local clubs, their open days and exhibi¬ 
tions, meeting dates and so on. Periodicals will 
also provide the opportunity to see what others 
are making, and the sort of techniques which 
are used, and will generally allow you to 
become familiar with model engineering draw¬ 
ings and, through published photographs, 
allow comparison between drawings and the 
items which are produced from them. 

Through the advertisements you will 
become aware of the retailers that can supply 
drawings, castings, materials, tools and general 
supplies and as your ideas take shape you will 
be able to obtain a few catalogues which are 
relevant to your interests. From these you 
should be able to judge the sort of financial 
commitment that will be involved so that you 
can plan the transition from armchair to work¬ 
shop. Some advice about the choice of a first 
project is given below. 

Sources of drawings and designs 

It is perfectly possible to prepare your own 
drawings for construction of a model, clock or 
scientific instrument, but there are pitfalls, even 
for the experienced, and as a beginner it is 
probably best to work to a published design. 



Fortunately, there is no shortage of drawings 
for items covering the whole range of activi¬ 
ties which go under the heading of model engi¬ 

Without doubt, the most important collec¬ 
tion of drawings is that associated with Model 
Engineer magazine. This periodical has special¬ 
ised over many years in presenting descriptions 
of the making of many different items, accom¬ 
panied by reduced-size drawings inset into the 
descriptive text. Most of the drawings utilised 
over the years are still available. Those likely to 
be of interest to model engineers are catalogued 
in Plans Handbook No. 2 (published by Nexus 
Special Interests) which lists the sets of model 
engineering and model boat drawings which 
can be purchased from the publishers. 

This Plans Handbook is effectively a cata¬ 
logue of all of the drawing sets which are avail¬ 
able, covering model locomotives, traction 
engines, stationary and marine steam engines, 
boilers, steam-powered toys, clocks, workshop 
tools and accessories, in addition to model 
boats of all sorts. 

These original drawings are normally pre¬ 
sented at full size for the model so that a large 
item such as a passenger-hauling, live steam 
locomotive occupies several large sheets of 
drawings. Prints of these are available from the 
Nexus Plans Service (Nexus House, Boundary 
Way, Hemel Hempstead, Herts HP2 7ST, tel: 
01442 66551) and will generally be found to 
provide all of the details necessary for the 
building of the item concerned. 

A useful feature of Plans Handbook No. 2 is 
that it lists the content of each of the individual 
sheets that go to make up the full set for a large 
model. Normally, one will need the full set of 
drawings in order to construct the model as 
designed, but if only the tender of a locomotive 
is of interest, for example, then the sheets can 
be purchased individually and there is therefore 
no unnecessary expense. This Plans Handbook 
is also useful in listing suppliers from whom 

castings are available, should the item need 
them, and this provides a valuable indication of 
the support offered by the trade in respect of 
particular designs. 

A further good feature is that the list contains 
references to the Model Engineer volumes in 
which the design was originally described, as¬ 
suming that it did originally appear in the maga¬ 
zine, thus providing a ready reference to what is 
usually described as the ‘words and music’. 

Many retailers are agents for the Nexus 
Plans Service and so provide an alternative 
local source of supply from this important col¬ 
lection of information. Some suppliers have 
developed their own designs over the years and 
are able to provide both the drawings and the 
castings that are required for these models or 
workshop items. The advertisements in the 
model press will provide details of these as will 
the individual suppliers’ catalogues. 

Choosing the first project 

Your first project must be chosen with care. It 
must be relevant to your interests, and interest¬ 
ing and challenging to make, but above all, it 
must be a project which has an achievable con¬ 
clusion. It is a mistake to commence your mod¬ 
elling with a too-detailed model as it may well 
take so long that you will become discouraged 
and the project will founder due to lack of 
progress. Better to choose a simple glow-plug 
single-cylinder engine than go immediately for 
the 9-cylinder radial that you really have your 
eye on. There is only disappointment to be 
found if a project founders due to lack of expe¬ 

There is no need for this to happen. First of 
all, it must be appreciated that a complicated or 
extremely detailed item naturally takes longer 
to complete than a simple, straightforward one. 
Secondly, as a beginner or inexperienced 



modeller, there will inevitably be techniques to 
learn which will slow down the ‘production 
rate’, making completion that bit further away. 
The learning process is naturally essential to 
your progress as a modeller, and it must there¬ 
fore take place, but it is best that it takes place 
on a simple model and that the project 
progresses to a conclusion. 

It may not be true for every aspect of the 
hobby, but it certainly is for model locomo¬ 
tives, and may well be for other fields, that the 
older the model design, the simpler it is likely 
to be. There is a current tendency for models to 
be designed more as scale models and less as 
broad look-alikes for the prototype which they 
represent. Designers are tending to attempt a 
much improved scale appearance for their 
models and they are consequently becoming 
more detailed. This means that there are more 
parts to make, more rivets to insert, smaller fit¬ 
tings to assemble and increased problems of 
making and assembling the parts. Conse¬ 
quently there are more drawings to purchase, 
and, naturally, an increase in cost. There may 
also be an increase in the cost of the special 
castings, and many more of them. 

On the other hand, locomotive designs by 
LBSC, and others, who popularised the hobby 
of model locomotive building, are intended to 
produce models without excessive detail. The 
drawings are simple and uncluttered, and 
although they may be lacking in detail concern¬ 
ing some actual dimensions and materials to be 
used, can generally be relied upon to produce 
workable models. They are ideal as a source of 
information for your first attempts. You will 
learn much from having to make your own 
decisions regarding materials, hole sizes and so 
on, and in addition you will be working with a 
set of drawings which were designed to pro¬ 
duce a straightforward working model, with¬ 
out too many frills, which can be built by a 
beginner in a reasonable time. 

Conventions used on drawings 

Projection drawing 

It is conventional to present the external out¬ 
lines of an object by drawing elevations and a 
plan view, as illustrated in Figure 2.1. The illus¬ 
tration shown is presented in what is known as 
‘third angle projection’. In this method of pres¬ 
entation, the views of the object are presented 
in such a way that each shows what would be 
seen by looking on the near side of the adjacent 
view. Thus, the plan is drawn immediately 
above the roof and is the view which would be 
seen by looking down on it. The view of the 
right-hand end of the house is placed at the 
right-hand side of the front view, or front eleva¬ 
tion, and shows what would be seen by stand¬ 
ing at the right-hand side of the house and 
looking towards it. 

An alternative method of showing the plan 
and elevation is also in use. This presents the 
plan and elevation of what can be called the 
‘garage’ side of the house in the juxtapositions 



Figure 2.1 Third angle projection. 




Figure 2.2 first angle projection. 

The example of a full set of views in third 
angle projection which is shown in Figure 2.3 
will serve as an example of drawing methods. 

Views of a rectangular block are presented 
which show all six faces. The plan has the two 
side elevations above and below it. Together, 
they show that the top surface has a circular 
recess in it, since the two side elevations show 
its outline in dotted form, indicating that it is 

The two end elevations appear on each side 
of the lower of the two side elevations. They also 
have the dotted outline of the circular recess in¬ 
dicated and show that a deep groove having a 
shallow groove within it is machined down the 
length of the block. The hidden edges of these 
grooves are shown dotted on the plan view. 

shown in Figure 2.2. Therefore, although the 
views presented are the same, the use of First 
angle projection causes the end elevation and 
plan to appear on the opposite sides of the front 
drawing elevation from those in which they ap¬ 
pear in third angle projection. 

Modern drawing practice requires the pro¬ 
jection which has been adopted to be shown in 
the margins of a drawing so that there can be no 
doubt of the draughtsman’s intention. A small 
symbol should accompany the statement, as 
shown in the figures. 

This convention is not generally followed on 
drawings for model engineers, so modellers 
must be aware that the two projections are in 
current use. Most model drawings known to 
me seem to adopt first angle projection, which 
was at one time the preferred projection in the 
UK. This is a pity since I was brought up on 
third angle projection, and as a consequence, at 
least one item in the workshop was made the 
wrong way round! The lesson to note is to be 
sure to use all of the information available on 
the complete drawing set to determine which 
way things are to be made. There is not usually 
a front door and roof to provide orientation! 

Figure 2.3 A full set of views in third angle projection. 

The side elevations also indicate the pres¬ 
ence of drilled holes in the downward exten¬ 
sions of the block, the dotted lines again indi¬ 
cating hidden features. 

One side elevation and the view of the 
underside show that there are seven holes: six, 



described as ‘blind’ since they do not penetrate 
the block, drilled in the bottom face, and the 
seventh drilled through from the side, into the 
central groove. 

In a practical case, the drawing would carry 
dimensions to specify the sizes of all of the 
features, details of the holes and so on. Dimen¬ 
sions cannot always be shown on a simple 
outline drawing and there is frequently the 
need to provide a cross-section to show details 
which would otherwise not be evident. 

The recess in the top surface of the block can 
be used as an example. This is revealed by 
drawing a section of the block, effectively slic¬ 
ing it along the line A-A and presenting the 
view which would be seen by looking in the 
direction of the arrows A. The portion of the 
block which has been ‘cut through’ by the sec¬ 
tioning process is hatched, whereas those parts 
not cut are shown without hatching, thus 
revealing the recess and the drilled hole which 
lie on the line A-A. 

The sectional view is used whenever internal 
or hidden features must be shown in detail, or 
when dimensions need to be specified in some 
detail. Clearly, this is not always the case, since 
it is possible to specify the dimensions of the 
recess in the top surface of the block as 3 /«in. di¬ 
ameter x Vwin. deep (19mm diameter X 
4.75mm deep) without the necessity to draw 
the section. 

Sectional views are much used on model 
engineers’ drawings, most commonly as what 
might be described as ‘sectioned assembly’ 
views, an example of which is shown in Figure 
2.4. This shows an assembled water gauge for a 
model locomotive. The drawing has been sec¬ 
tioned to reveal the internal features so that it is 
possible to see the internal passages, the fit of 
the gauge glass, the seals at top and bottom of 
the glass and details of the blowdown valve in 
the bottom fitting. 

On an actual drawing, the threads and sizes 
of the passages are specified, together with the 

Figure 2.4 A locomotive water gauge shown as a sectioned 

sizes of the hexagonal nuts and the general 
external dimensions of the parts, and the size of 
the glass tube. 

Detailed dimensions are not usually pro¬ 
vided for every last feature and it is sometimes 
necessary to measure the dimensions on the 
drawing to determine the sizes which are 
unspecified. To assist this process, the items are 
drawn at a suitable size to allow easy measure¬ 
ment. This is called the ‘scale’ of the drawing. 

The most usual scale is full size for the item 
shown, but if the features are very small, as is 
frequently the case, items are drawn to a larger 
scale, typically twice full size and will be identi¬ 
fied as such. Dimensions shown on such a detail 
are, however, given as the actual dimension 
required, and a dimension shown as 'Ain. 
(6mm) is drawn at twice this size, a fact which 
can be confirmed by measuring the drawing 
with a rule. It is thus necessary to work to the 



stated dimensions, but if a dimension is 
unspecified and needs to be measured, the size 
required is naturally one half of that indicated 
by the rule. 


If items having complex shapes are required, 
they are frequently produced as castings which 
embody the essential basic shape. A casting 
generally needs to be machined to produce the 
flat, circular and cylindrical sections which are 
its working or attachment surfaces, but it does 
not necessarily need to be machined all over. 

If some surfaces can be left ‘as cast’, the 
drawing may indicate this by identifying those 
which must be machined. Several different 
symbols have been used to indicate the 
machined surfaces and two may still be encoun¬ 
tered on model engineers’ drawings. 

The first method is illustrated in Figure 2.5 
which shows a section through a cylindrical 
sleeve or bearing. It is made from a casting, and 
only the bore and the registers at each end need 
to be machined. This is indicated by using a tick 
to show each machined surface. Where there is 

Drill 8BA Gear and Spot Face to 
clear Hex. Hd. Bolts 
(4 positions) 


Figure 2.6 Indication of surfaces to be machined on a casting. 

room, the tick is placed on the surface which is 
to be machined, but in cases where this is not 
possible, the tick touches a dimension line, or a 
similar line drawn especially for the purpose. 

An alternative method of indicating ma¬ 
chined surfaces is shown in Figure 2.6, which 
shows a cast shaft with a flange at one end. 
Only the end face and outside diameter of the 
flange need to be machined and these two sur¬ 
faces are identified by a letter ‘f’ which is 
drawn through the surface to which it relates. 

One further point about the use of castings is 
also shown in Figure 2.6. The flange of the 
shaft needs to be drilled through in four posi¬ 
tions for attaching bolts and the unmachined 
cast surface of the flange needs to be machined 
locally so that the bolt heads, or nuts, bed down 
on to a squarely machined surface. This opera¬ 
tion is called ‘spot facing’. It is performed by a 
flat-ended cutter which is generally known as a 
counterbore but which might equally be called 
a spot facer on account of this method of use. 
Counterbores are described in Chapter 10. 



Dimensions and tolerances 

General practice 

If you are familiar with the sort of conventions 
adopted on the formal drawings used in indus¬ 
try, you will find model engineering drawings 
surprisingly lacking in detail. First of all, very 
few tolerances will be found on the details, 
these only being shown where a press (interfer¬ 
ence) fit is to be used or a defined clearance is to 
be provided to allow Loctite or one of the other 
modern adhesives to be used to secure two 
items. The situation which immediately comes 
to mind is the fitting of locomotive wheels to 
their axles, or the fits required when building 
up a crankshaft. 

For running fits there is also seldom any 
indication of the sort of clearances which are 
envisaged, and it is usual to find simple instruc¬ 
tions such as ‘ream Vim.’ on the detail drawing 
of the bearing and an equally simple ‘‘/iin. 
diam.’ on the shaft which fits the bearing. 
There is thus a need to understand what one is 
trying to achieve when making the different 
parts. This subject is described in more detail in 
Chapter 10. 

It must also be appreciated that in many 
instances actual sizes do not matter and the 
amateur builder is expected to be reasonably 
competent in assessing the sort of fits and clear¬ 
ances that are required and to know which 
dimensions are critical and which are not. This 
is particularly true for running fits, where what 
matters above all is that there should be ‘nice’ 
clearances to suit the duty envisaged. 

In clockmaking, the builder approaches the 
provision of some bearings in exactly this way, 
the bearing holes in clock plates being polished 
by broaching until the shaft ‘feels’ right when 
mounted between the plates. Model locomo¬ 
tive builders naturally need to provide good 
bearings without significant play when consid¬ 
ering the valve gear components, but for some 

components of the running gear (axles particu¬ 
larly) sloppy fits are beneficial, within reason, 
and the model will ride the better because of 

The general lack of tolerances arises because 
our models and other items are one-offs for the 
most part, since not many of us will want to 
build a ‘production batch’. Therefore there is 
no need for interchangeability between two 
parts which are nominally identical since it is 
sufficient that each individual part fits those 
with which it immediately mates. This does not 
mean that there is necessarily scope for sloppi¬ 
ness in manufacture, only that the parts which I 
make do not have to be interchangeable with 
yours. Equally, if a 2-cylinder engine is being 
built, it is not absolutely vital that both bores 
are exactly the same size. It is sufficient that 
each piston fits the bore for which it is 

For this reason also, many minor or unim¬ 
portant dimensions are sometimes omitted 
from the drawings, and the builder is left to 
determine his own actual sizes. In model loco- 



motive work, this seems to occur particularly 
with small links and pull rods which have 
rounded ends. The main dimensions are given, 
together with hole sizes, but the end radii are 
not always specified. Blend radii are similarly 
not identified. Figure 2.7 shows typical links 
taken from one of LBSC’s designs of the 1950s. 
For these, and similar, links an end radius equal 
to the diameter of the hole will be found satis¬ 
factory and this two-to-one ratio in diameters 
may be used for all such links unless space or 
the shape of the prototype preclude this. 

‘Full’ and ‘bare’ dimensions 

Figure 2.8 shows the reversing lever and its 
stand for a model locomotive. The arrange¬ 
ment is typical. A long lever is pivoted near the 
bottom of the stand and is fitted with a latch 
and trigger which allow the lever to be locked 
at various positions throughout its full range of 

Figure 2.8 ‘Full’ and 'bare' drawing conventions. 

movement. The lever is connected to a reach 
rod which connects to the valve gear in some 
way, allowing full forward or full backward 
gear to be selected, or positions in between. 

The reach rod is '/sin. (3 mm) thick and is 
linked to the handle by a shouldered screw 
which allows the necessary rotation between 
the two parts as the handle rotates about its 
pivot. To provide the clearance for this to oc¬ 
cur, the shouldered screw is specified as being 
Vs FULL below the head, meaning that this di¬ 
mension must be slightly greater than the thick¬ 
ness of the reach rod. 

On the other hand, the fitted bolts holding 
the stand to the frame must not be longer in the 
unthreaded portion than the total thickness 
through which this part of the bolt passes. Since 
this is 7 /i6in. (11mm) in this particular case, the 
dimension on the bolt is specified as ‘ 7 /i6 
BARE’, i.e. barely (less than) 7 /i6in. 

Possible errors 

It is unfortunately true that errors and incon¬ 
sistencies do sometimes remain in a set of draw¬ 
ings when they go on sale to modellers. Model 
engineers seem not to be worried by the fact 
that drawings sometimes contain errors. 
Indeed, it is surprising that errors seem to per¬ 
sist for many years without them being reported 
to the publisher of the plans. There is doubtless 
much swearing when errors are discovered (and 
this is frequently after parts have been fully 
made) but after a couple of days one is usually 
reconciled to the fact that something must be 
remade, or perhaps a modification has already 
been schemed out that allows the incorrect 
parts to be used. Thus, what commences as a 
disaster is usually turned into a triumph and this 
somehow increases the pleasure of the hobby. 

But it obviously does pay to talk to other 
modellers who have built the item concerned as 
they may remember if there were any particular 



difficulties which could be corrected by amend¬ 
ments to the drawings. 

Problems don’t always become well known 
so it pays to look out for errors before it is too 
late. A certain amount of time must be allo¬ 
cated to just looking at the drawings, learning 
your way around them and trying to visualise 
how the various bits fit together. It is as well to 
look out for errors, for I fear that model engi¬ 
neers generally are not too good at reporting 
their findings to the designer or publisher. 
Nevertheless, it pays to buy an up-to-date set of 
drawings rather than use the set which Bill 
bought ten years ago. 

Hole sizes 

Drawings naturally contain details of the drill 
sizes that should be used to produce the general 
‘fixings and fastenings’ holes that the model 
will require. If particular items are to be 
secured by rivets, the drawing shows hole posi¬ 
tions or pitches and indicates the rivet size. For 
bolted or screwed together parts, the actual 
screw or bolt size is frequently not given but 
must be inferred from the hole sizes shown on 
the drawing. On the majority of drawings, 
these are given by reference to the series of 
number and letter drills which were once the 
standard drills for small holes. You will conse¬ 
quently find simple instructions such as ‘Six 
holes, No. 34’ or ‘Three holes, No. 27’ these 
being the normally accepted clearance size 
holes for 6BA and 4BA screws. 

As a newcomer to the hobby you will most 
probably not have these sizes of drill available 
since they are now becoming the more expen¬ 
sive, second choice, non-preferred sizes. You 
will consequently have to convert these sizes 
into the metric equivalents in order that you 
may utilise the preferred range of drills. This 
subject is covered more fully in Chapters 10 
and 11. 


In many instances considerable choice is avail¬ 
able as to the sort of materials that may be used. 
For example, in considering a steam engine 
(locomotive, traction engine or stationary 
engine) the cylinders may be made from brass, 
gunmetal or cast iron, dependent upon the duty 
envisaged, the size of the model or the size of 
the modeller’s pocket - or perhaps all three. 
There is consequently a choice of the material 
to be used for the pistons to work in the cylin¬ 
ders, it being necessary to choose something 
compatible. The same is true for eccentrics and 
their related eccentric straps. Gunmetal straps 
go with mild steel eccentrics and mild steel 
straps with cast iron eccentrics. 

Designers often acknowledge this by not 
specifying the materials to be used for such 
items, preferring to leave the builder to choose 
according to the size of his bank account or his 
personal wishes. For example, gunmetal is sig¬ 
nificantly more expensive than cast iron and 
modellers’ suppliers frequently offer castings in 
both materials for such items as cylinders and 
their accessories, axleboxes, keeps, hornblocks 
and the like, for model locomotives. 

Suppliers also offer materials cut specifically 
for the larger parts of a model, leaving the 
builder to decide only on the smaller items. 
Background reading will provide basic guid¬ 
ance, but there are some general rules which 
are quickly learned. If it needs to be strong, it 
should be steel, and if it’s in contact with water 
(at normal temperatures) it should be brass, but 
gunmetal or phosphor bronze must be used for 
their better wearing qualities, for the bodies of 
valves and so on. Stainless steel is used for con¬ 
trol rods and spindles for valves for both water 
and steam. Boilers are normally made in cop¬ 
per, with gunmetal or phosphor bronze bushes 
being fitted to the structure for the attachment 
of fittings. All of these metals are described in 
Chapter 3. 



Material sizes 

For many parts of a model the builder must 
decide for himself the size of bar material that 
will be used to make particular items. The 
drawings will show the critical dimensions of 
the parts and it is up to the builder to determine 
the method of manufacture and hence the size 
of material which is appropriate. 

Most older designs, indeed the majority of 
designs, were prepared to imperial scales, and 
parts are most easily made from imperial mate¬ 

At the present time, suppliers’ catalogues still 
list imperial sizes of round and rectangular bars 
in the common materials. In the longer run it is 
likely that only metric materials will be avail¬ 
able although it does seem that imperial copper 
tubes will remain in stock for the foreseeable fu¬ 
ture, thus making things relatively straightfor¬ 
ward for the boiler builders among us. 

For sheet materials and wires, older designs 
will be sure to use the Standard Wire Gauge 
(swg) nomenclature for thickness. This is the 
standard imperial method of specifying wire 
and sheet thicknesses which uses a simple 
number referencing system. This allocates a 
series of numbers to the different thicknesses or 

Imperial standard wire gauges 


in. mm 


in. mm 












































































diameters which are produced. The number 
series commences at zero and progresses 
upwards as the material becomes thinner or 
smaller in diameter. Several different standards 
were used but the British swg system was 
almost universally adopted in recent times in 
the UK, and the list is tabulated here. 

Only the even-numbered gauges were 
widely available and these are most likely to be 
met as specifiers for sheet thicknesses and wire 
diameters on imperial drawings. The most 
common range of thicknesses likely to be 
encountered will be from 10 swg, 0.128in. 
(3.3mm) down to, say, 30 swg, at .0124in. 
(0.3mm). Below this, thicknesses are definitely 
in the ‘shim’ category and are usually specified 
as a decimal fraction of an inch, or simply 
described as ‘ten thou’ (0.25mm) etc. 

The system may seem cumbersome, but in 
reality, most of us who are used to the system 
could readily reel off the decimal thicknesses 
for the very commonly used 16,18 and 20 swg 
and remember 10 swg for its closeness to ‘/sin. 
The apparent illogicality of the reversed num¬ 
bering system is explained by the fact that the 
thinner materials have been through the rolls or 
drawing-down dies more times than the thicker 


Obtaining supplies 

For most model engineers, material supplies 
are usually obtained from retail outlets which 
cater specifically for the amateur worker. Such 
retailers are in business to satisfy the customer 
who only requires material in relatively small 
quantities. Flats, rounds and hexagons will 
therefore most probably be available in 2ft 
(600mm) lengths, or perhaps up to 3 ft (lm), 
standard lengths being limited by the fact that 



much of their business is by mail order and 
longer lengths are not suitable for transmission 
by post. 

Sheet materials will also probably be avail¬ 
able in standard sized (small) sheets, but some 
suppliers offer a cutting-to-size service, espe¬ 
cially for the more expensive materials such as 
brass and copper. Cut-to-size pieces in these 
more expensive metals are often sold by weight. 

Retail outlets obtain their supplies from 
wholesalers who are normally described as 
metal stockists or metal factors. If you can 
obtain a factor’s list, you will see that every 
material which he sells is defined by a reference 
code of some sort. These codes define the avail¬ 
able materials quite precisely, and you could 
therefore purchase say, riveting and turning 
quality brass (CZ131) as distinct from drawn 
brass (CZ108). 

Retail outlets do not, however, normally 
indicate the precise type of material which they 
sell. This allows them some flexibility in the 
source of supply since they do not necessarily 
have to purchase specific alloys. You will be 
offered round brass rod in various diameters, 
phosphor bronze or cast gunmetal without ref¬ 
erence to the precise material which will be 
supplied. This is not a problem since you can 
rely on the experience of the retailer to stock 
material which is adequate for general needs, 
and unless you have a specific use for some¬ 
thing out of the ordinary, the materials you 
obtain from retail suppliers are entirely satis¬ 
factory. You do not need to know any detail of 
British Standard reference numbers, except to 
be aware that they do exist, and it might in due 
course be necessary to investigate further. 

As a beginner therefore, if the drawing indi¬ 
cates that brass is required, just get hold of 
some brass and make the thing. It cannot be 
critical or the designer would have mentioned 
the particular grade he had in mind. 

There are similar detailed specifications for 
other materials too, but the local supplier can 

be relied upon to stock a material that is likely 
to meet most common applications. So, for a 
start at least, just accept the colloquial names 
for these materials and purchase them as such. 

Scrap merchants also provide another possi¬ 
ble source of materials and should certainly be 
considered whenever you have no particular 
interest in the actual specification. 

For the purchase of steel however, it is best 
to avoid anything for which you do not have a 
reasonably precise definition. There seem to be 
large quantities of steel produced for industrial 
uses which have desirable characteristics such 
as cheapness and good weldability which suit 
them to the manufacture of particular types of 
goods. There is no doubt that they are entirely 
adequate for their intended purposes, but if you 
‘come by’ some of them you may find them 
hard to cut or impossible to turn to a good 
finish and high consumers of both good temper 
and time. 

As a learner lathe operator you will inevita¬ 
bly experience certain difficulties during early 
attempts to cut metal on the machine and can 
avoid some of the likely pitfalls by purchasing 
your mild steel rounds in a free-cutting variety. 
Steel is not, in any event, particularly expen¬ 
sive, and the cost penalty in buying new steel is 
not an onerous one to bear. You will also need 
to have some bright-drawn mild steel available 
for use when case hardening is to be employed 
since some free-cutting steels are not recom¬ 
mended for hardening by carburisation. 

If hardening and tempering of steel is 
required, for the manufacture of tools and cut¬ 
ters, either silver steel or gauge plate will be 
used. You will, therefore, need stocks of three 
or four distinctly different steels and should 
buy your requirements against the specific 
descriptions of bright-drawn (or black) mild 
steel, free-cutting mild steel, silver steel or 
gauge plate. This latter is also known as ground 
flat stock since it is supplied in rectangular 
cross-sections, having a ground surface finish. 



An introduction to these and other engineer¬ 
ing materials is provided in Chapter 3 which 
also describes heat treatment methods (anneal¬ 
ing, normalising, hardening, tempering and 
carburising) which may be performed in the 
home workshop. 

Having different materials in stock in the 
workshop, which might easily be misidentified, 
you should devise an indelible marking system 
so that you do not inadvertently pick out stain¬ 
less steel instead of silver steel, only to find that 
the beautifully made tool you have just com¬ 
pleted will not harden as you had expected. A 
simple system can be adopted in which one or 
two (generous) splashes of coloured paint mark 
the end of each bar according to the type of 

One point to bear in mind is that metal is 
bulky, heavy stuff which attracts a not inconsid¬ 
erable amount of postage and packing charges. 
Most of us have to order materials by post from 
time to time, but it is surprisingly easy to incur a 
charge of a few pounds in respect of transport 
costs. It is probably best, therefore, to assess 
your requirements ahead of time and plan a 
bulk-buying trip to a retailer who is likely to be 
able to satisfy your needs. If you are a member 
of a club, or otherwise in touch with other 
modellers in your locality, a joint visit in which 
you share the transport costs should allow 
access to establishments relatively far away, and 
if you can assess your needs for the next model¬ 
ling ‘season’, you are likely to find this 
approach quite worthwhile. Even if you travel 
alone, you might as well spend the money on 
petrol and take the family out for the day, as 
pay the costs of having someone else transport 
the goods for you. 

What materials to buy 

It is difficult to be specific about what a begin¬ 
ner should buy by way of building up a small 

stock of materials as a start in the hobby. In the 
first place, tastes vary and what would be suit¬ 
able for a clockmaker might remain in the 
workshop of a live steam man for a very long 
time before being of much use. In the second 
place, space to store the material will vary 
between individuals and so will the depth of 
one’s purse, and there is always the problem 
posed by the location of the nearest supplier. It 
is all very well running out of that vital bit of 
material if there is a supplier in the locality, but 
quite a different matter to have to order by post 
and endure the consequent costs and delays. 

Of course, with inflation being what it is, a 
little stock is a great comfort, especially if one is 
approaching retirement, but for a great many 
projects the bulk of the cost will be found to be 
in the castings and the special-to-model items 
rather than in the general materials and there¬ 
fore it does not pay to invest large amounts of 
money in ordinary stock items. Should you 
have spare money, it is best invested in the spe¬ 
cial items that your next model (or whatever) 
will require, given that you already have a basic 
stock of commonly used materials. 

So what should comprise the basic stock? In 
a way it is easier to be specific about what the 
basic stock should not be. It should not contain 
large quantities of the more expensive items 
such as brass, gunmetal and phosphor bronze. 
It is best to buy these in what appears to you to 
be sensible quantities, once a specific need has 
been identified. Naturally, you are likely to 
need some of the common sizes of round brass 
rod and might reasonably have stocks of the 
common sizes up to Wm. (10mm) but beyond 
this diameter the price begins to increase dra¬ 
matically and unless you are likely to use large 
quantities on a particular project it is certainly 
not worth buying the larger sizes initially. The 
same applies also to hexagon brass. Large sizes 
are quite costly and should only be bought as 
required, perhaps with a small amount added 
on for luck. 



Some hexagon brass will be useful, espe¬ 
cially for the live steam man, but it is likely that 
only relatively small quantities will be required 
and nothing above 7i6in. (11mm) or Vim. 
(12.5mm) across flats (AF) should be bought 
for the general stock. 

Similar arguments can also be used in rela¬ 
tion to phosphor bronze and gunmetal. Large 
amounts of these materials are not likely to be 
required and the acquisition of stocks can be 
made a low priority, certainly in the early days 
of establishing a workshop. 

One point to bear in mind in relation to 
these more expensive materials is the cost of 
using a larger size bar than is actually required 
for the job. Reducing a 'Ain. (13mm) bar to 
3 /8in.(10mm) in diameter is almost the equiva¬ 
lent of turning away the amount of metal in a 
9 /j 2 in. (8mm) diameter rod, so it pays to use the 
size you actually need and you may need to 
keep a larger number of diameters in stock in 
order to economise on the amount of metal 
used. But it does of course depend upon the 
total amount of metal which you personally 
use, and for the occasional job it might be ac¬ 
ceptable if it avoids buying a further size. 

Mild steel is the cheapest of the materials 
which you will need in the workshop. Para¬ 
doxically, this means that modellers tend to 
keep a wider variety of cross-sections and sizes 
than they do of the expensive materials, 
although with cheaper stuff it is obviously not 
so costly to reduce a larger diameter to the size 
required. However, the temptation is always 
there to buy some more because it’s good value, 
but do consider whether you can store it and 
also whether you are really likely to use it. 
Looking at my own stock of steel, there are cer¬ 
tainly some sections in which I seem to have 
more than a lifetime’s supply but since there is 
as yet no storage problem, no harm seems to 
have been done, except that money has perhaps 
been invested unwisely. 

Steels are available in several different varie¬ 

ties and also in different sections, although the 
full range of sections is only likely to be readily 
available in what is known as bright mild steel. 
This is nice, clean material but is not necessarily 
the best form in which to purchase your steel. 
Items which are to be machined all over are 
best made from black mild steel and, if this is 
available, it is most definitely the best mild steel 
to purchase. A stock of round bright mild steel 
in '/Sin. (3mm) increments from 3 /i6in. (5mm) to 
'Ain. (13mm) in diameter will be likely to 
satisfy your initial needs and the stock can then 
be built up to suit your activities once the usage 
rates become clear. 

Ground mild steel should be available from 
your supplier and this will provide a source of 
steel having an accurately sized finish. This can 
be used for shafts and axles for which an accu¬ 
rate diameter is required along the length and 
its use does reduce the amount of work that is 
needed when making such items. 

Steel is also available in flats and hexagons. 
A small stock of flats is likely to be generally 
useful and might include a range of thicknesses 
from '/isin. (1.5mm) up to 'Ain. (6mm) in !/i« in. 
increments, but not in too many different 
widths. Up to 'Ain. thick, material may readily 
be cut to size with a hacksaw and unless you are 
particularly averse to this activity you need not 
stock a great range of widths. 

There isn’t a very great demand for square 
and hexagonal mild steel, although this latter is 
useful for making the occasional nuts and bolts. 
This is not an activity that is likely to be 
required too often, but it is occasionally neces¬ 
sary to manufacture special fine-thread nuts 
and bolts for particular applications and a small 
stock to suit the range of spanners you use will 
be useful. General stock up to 'Ain. (12.5mm) 
will probably find ready use. 

It might also be found helpful to have stocks 
of steel hexagons in the BA nut and bolt sizes, 
particularly the odd-numbered threads, as 
these are useful for making the small-head bolts 



which find wide use in models. If you envisage 
that you will require hexagon-head brass bolts 
or screws with these smaller-sized heads, then a 
small stock of the appropriate sizes might also 
be useful. 

You are also likely to need a ‘hardenable’ 
steel for the manufacture of punches, scribers, 
cutters and other small tools. This material is 
that known as silver steel which is supplied in 
standard 13in. (330mm) lengths ground accu¬ 
rately to diameter. An initial stock of the 
smaller sizes, say up to 3 /sin. (10mm) in diam¬ 
eter is certain to be useful. Another hardenable 
steel, known as gauge plate, is available in rec¬ 
tangular cross-sections but this should only be 
bought as required. 

Chapter 3 contains an introduction to the 
materials that are likely to be of use in the ama¬ 
teur workshop, and in particular provides guid¬ 
ance concerning hardening and also describes 
the types of steel that are most suitable for the 
different applications that might be met. Steel 
is relatively cheap, as noted above, but comes in 
a range of different alloys, surface finishes and 
characteristics and it is as well to be aware of 
the type with which one is working. There is lit¬ 
tle point in buying ‘any old steel’ just because it 
is available and cheap, unless you actually need 
some rough old steel for a rough old job. 

Useful oddments 

You will frequently read about items ‘made 
from the scrapbox’, so it is obvious that many 
model engineers have one of these ubiquitous 
boxes. This is not to say that we produce large 
quantities of scrap, although this is regrettably 
an occasional activity. The scrapbox mostly 
contains material which is too good to throw 

away, but which is not considered to be part of 
the general stock. This may be because the 
material is not stocked in sufficient quantity to 
have a ‘bin’ of its own, or is an odd length or 
offcut that would easily be overlooked if kept 
with similar stock. It may also be that you are 
not exactly sure what the material is, but feel 
nevertheless that it may be useful. There will 
also be some of your mistakes which now 
simply represent material which might be 
useful, one day. 

What goes into the scrapbox, and how big it 
is, is entirely a personal choice. It is certainly 
useful to segregate the materials and separate 
ferrous and non-ferrous boxes should prove 
useful in shortening any searching time when 
you are looking for that odd bit of material. It is 
probably also helpful to segregate any large- 
diameter offcuts since these are usually held in 
small quantities, and it is a nuisance if small 
items are mixed in with them. 

If you make items which need lots of small 
parts, such as small-scale model locomotives, it 
is certainly a time-saver to segregate small 
offcuts of brass and nickel silver into a separate 
box, which need not, in any event, be very 

If you are in a position to obtain offcuts of 
materials at reasonable prices, it is probably 
wise to concentrate initially on the more 
expensive ones, basically the copper alloys, 
brass, phosphor bronze and gunmetal, espe¬ 
cially in the smaller sizes. Useful materials not 
widely sold by the modelling trade are alu¬ 
minium alloys in thicker sheets, say, from about 
'Ain. (6mm) upwards. There is not likely to be 
much large-scale need for thin alloy sheets, but 
the thicker pieces are excellent for the manu¬ 
facture of bending bars (around which material 
is bent - see Chapter 7) and for cutting out 
formers for flanging items in copper and brass. 



Introduction to 
engineering materials 


Iron is produced from iron ore by heating it to a 
high temperature, in the presence of coke and 
limestone, and subjecting it to an air blast. At 
the high temperatures achieved in the blast fur¬ 
nace (up to about 1800°C) chemical reactions 
take place that release the iron from the ore. At 
these temperatures, the iron is molten and it 
falls down to the bottom of the furnace where 
it is run off into open sand moulds, forming bil¬ 
lets of iron, known as pigs. The material is 
therefore called pig iron. 

As the molten iron falls through the furnace, 
it picks up impurities from the limestone and 
the coke, the most important of which is car¬ 
bon, and the iron normally contains between 
three and five per cent of this element. 

The rate of cooling of the iron determines 
the form of the carbon within the material, 
which may be present as graphite or as iron car¬ 
bide (cementite). Slow cooling produces graph¬ 
ite rather than cementite and produces a 
coarse-grained structure which shows a charac¬ 
teristic grey colour if the material is fractured. 
Pig iron containing nearly all of its carbon as 

cementite is called white iron because it shows 
a whiter, more close-grained structure when 
fractured. Iron containing most of its carbon as 
graphite is more easily machined than white 

Pig iron is not used as an engineering mate¬ 
rial but is further refined in a second furnace. 
By incorporating special additions to the melt, 
a variety of irons can be produced, the two 
main classifications being grey cast iron and 
white cast iron. Since cementite is intensely 
hard, white cast iron is hard and durable, 
but extremely brittle, whereas the graphite in 
grey cast iron is a good lubricant and this iron 
is therefore softer, less brittle and easily 
machined. Grey cast iron is also cheap and 
melts at the relatively low temperature of 
1200°C. It is also extremely fluid when molten 
thus enabling intricate shapes to be cast. 

Grey cast iron has a wide range of uses, such 
as machine frames and machine tool beds and is 
used for the production of a wide range of cast¬ 
ings, particularly for the modeller and home 

Virtually all of the carbon in cast iron may 
be removed by reheating it in a puddling 



furnace and mixing millscale (oxide) repeatedly 
into the molten metal. This forms chemical 
compounds with any carbon which is present 
and the iron may be reduced to an almost- 
carbonless state. Chemically, it is almost pure 
iron, usually greater than 99%, with small 
amounts of other elements. This material is 
known as wrought iron. It is very durable, 
bends easily, even when cold, but resists high 
shock loads without permanent damage. Virtu¬ 
ally all commercial puddling furnaces are now 
closed down, and it has been left to industrial 
museums to continue the production of 
wrought iron. 


Production of steel 

The difference between steel and iron is deter¬ 
mined by the amount of carbon which is 
present. Wrought iron has virtually no carbon 
whereas cast iron contains between 2 and 5 per 
cent. Steels lie between these two extremes, 
normally containing between 0.1 and 2 per 
cent carbon. Although other elements may be 
present, the carbon is by far the most important 

Consequently, plain (non-alloy) steels are 
normally classified according to their carbon 
content, the commonest material being mild 
steel, which has a carbon content of between 
0.1 and 0.25 per cent. Medium-carbon steels 
contain from 0.2 to 0.5 per cent carbon and 
high-carbon steels have more than 0.5 per cent. 

Black and bright-drawn mild steel 

The most-frequently used steel is that known as 
mild steel. This has between 0.1 and 0.25 per 
cent carbon. It is widely available in sheet and 

bar form, making rounds, flats (rectangles and 
squares) and hexagons all readily available, in 
addition to sheet, in a wide range of sizes and 

At the steelworks, the basic material is pro¬ 
duced in large billets which must be rolled, 
pressed, drawn or otherwise worked down to 
the sizes and cross sections which are required. 
This working down may take place with the 
material in a hot condition, in which case the 
finished steel has a characteristic black surface 
scale and relatively poor surface finish. Having 
been hot worked, the material is, however, 
relatively free from internal stresses. 

For the most part, model engineers’ suppli¬ 
ers stock what is known as Bright-Drawn Mild 
Steel (BDMS), sometimes simply referred to as 
Bright Mild Steel (BMS). This has a bright, 
scale-free surface since the sections are pro¬ 
duced by cold-drawing billets of steel progres¬ 
sively through a range of dies to produce the 
required sections. The cold drawing (cold 
working) stresses the steel and although what is 
called the tensile strength actually increases, 
the ductility (the flow capability, or the ability 
to bend) is considerably reduced. This means 
that BMS should be heat treated before any cold 
bending operations are performed, otherwise 
the material may be overstressed and a fracture 
will result. 

The locked-in stresses in BMS are also a 
problem if large parts of a rectangular cross- 
section bar are removed to form slots or holes, 
or if large sections are removed to produce a 
tapered and fluted connecting rod, for exam¬ 
ple. Although stressed by the cold-drawing 
process, the rectangular bar is in equilibrium 
and will remain in the same shape if no metal is 
removed. Taking material out of a bar which is 
in this stressed condition, removes a vital part, 
the initial presence of which balances the 
stresses in adjacent parts. The stresses in the 
remaining material are therefore unbalanced, 
and distortion is the result. 



BMS is not the ideal material for use when 
making parts of complex shape. However, the 
internal stresses can be relieved, and stress 
relieving of rectangular cross-section BMS (par¬ 
ticularly) should become the rule rather than 
the exception. Relieving the stresses in round 
or hexagonal BMS is not normally undertaken, 
since turning tends to remove metal in a sym¬ 
metrical fashion. Stress relieving, or normalis¬ 
ing, is described below. 

BMS was originally introduced for use in 
situations where a clean surface finish was 
required without great accuracy, so if a particu¬ 
larly accurate overall size is required, machin¬ 
ing of the outside will be necessary and the next 
larger standard size of steel will naturally have 
to be used. There is, therefore, no disadvantage 
in the use of black mild steel and it does have 
considerable advantages where large amounts 
of metal need to be removed from rectangular 
cross-sections or where cold bending will be 

If accurately sized circular rods of mild steel 
are required, Precision Ground Mild Steel 
(PGMS) is readily available from modellers’ sup¬ 
pliers. This is available in increments of '/sin. 
from 3 /sin. to 1 '/sin. and in some metric sizes, 
from 6mm to 40mm in diameter. This material 
is ideal for model locomotive axles, crankpins 
and the like. Many modellers seem not to 
appreciate the ready availability of PGMS and 
tend instead to use an alternative steel, also 
having a ground surface finish, known as silver 
steel (see below) but PGMS is preferable if load 
carrying is required. 

Free-cutting steel 

Mild steel is fairly easy to machine to a good 
surface finish, but the removed material usually 
comes off in long, spiral curls which fly all over 
the workshop. In industry, where high-speed 
turning and drilling are naturally adopted as a 

means to reduce costs, the removal of the long 
curls of swarf from the machine tools poses sig¬ 
nificant problems. Consequently, steels have 
been developed from which the removed mate¬ 
rial comes off as small chips rather than long 
curls. The breaking up of the curls is caused by 
adding 0.1 to 0.2 per cent each of phosphorus 
and sulphur to the steel. This induces a certain 
degree of brittleness which causes the removed 
material to come off in the form of chips. Steel 
which has been ‘doped’ in this way is known as 
Free-Cutting Mild Steel, sometimes abbrevi¬ 
ated to FC Mild Steel. It is easier to produce a 
smooth machined surface on free-cutting steel, 
hence its usefulness in the home workshop. 

Carbon steel 

Steels containing more than 0.5 per cent car¬ 
bon are known as high-carbon steels. They are 
extremely useful since they are supplied in a 
soft condition but may be hardened to different 
degrees of hardness without elaborate equip¬ 
ment. Two forms of high-carbon steel are of in¬ 
terest to the amateur because of their wide 
availability - silver steel and gauge plate. 

Silver steel is a high-carbon steel suitable for 
hardening, intended for the manufacture of cut¬ 
ting tools, punches, scribers etc. It is most com¬ 
monly supplied in 13-inch (330mm) lengths of 
ground rod manufactured by the firm of 
P. Stubs. These have a ground finish accurately 
sized to within .00025in. (.00635mm). The 
rods carry the imprint ‘STUBS’ at one end 
which distinguishes them from other bright 

Stubs silver steel has 1.1 or 1.2 per cent car¬ 
bon and small additions of manganese, chro¬ 
mium and silicon. It has a slightly higher tensile 
strength than mild steel but is much less ductile 
and it should not be used for load-bearing pins 
and suchlike. It is frequently recommended for 
such duties as crankpins due to its accurate 



grinding to size during manufacture, but it is 
more difficult to machine than mild steel and 
its wearing properties are no better. It therefore 
offers no mechanical advantages over mild 
steel and its higher cost is disadvantageous, to 
say the least. 

Silver steel should also not be considered for 
use in its hardened state for such load-bearing 
duties. If a harder, wear-resisting surface is re¬ 
quired, this may be obtained by case-hardening 
mild steel. This produces a hard surface with a 
softer, more-resilient core, whereas hardened 
silver steel is much more brittle (right through) 
and is therefore more liable to fail in service. 

Silver steel is very difficult to machine to a 
smooth surface finish and a free-cutting carbon 
steel is becoming available from some suppliers 
which has the same properties of ‘harden- 
ability’. This is not yet available in such a wide 
range of sizes as silver steel but offers a big ad¬ 
vantage in terms of ease of machining. It is also 
cheaper than the ground-finish Stubs silver 

Hardening and tempering of silver steel for 
the manufacture of tools, for which the mate¬ 
rial is intended, is described under the heading 
‘Heat treatment of steel’ below. Case hardening 
of mild steel is also described later. 

Gauge plate, or ground flat stock, to use its 
other name, is also a high-carbon steel, but is 
supplied in rectangular cross-section bars, 
18in. (500mm) long, accurately ground to 
thickness. Composition varies, but generally 
about 1 per cent of carbon is present, together 
with small amounts of manganese, chromium, 
tungsten and vanadium. 

As its name implies, gauge plate was intro¬ 
duced for the manufacture of gauges. Like sil¬ 
ver steel, the material is supplied in the soft 
state but may be hardened after machining to 
the required shape. It is also suitable for making 
cutting tools and is especially useful for making 
form tools, which can be a great help when a 
‘production run’ of small turned components is 

required. Gauge plate is very expensive (7 to 10 
times the price of mild steel) so should only be 
bought as required. 

Hardening and tempering of gauge plate is 
performed in a broadly similar fashion to that 
used for silver steel, but the required technique 
is subtly different and varies from one manu¬ 
facturer to another (see ‘Heat treatment of 
steel’ below). 

Silver steel and gauge plate are both avail¬ 
able in imperial and metric sizes. 

Stainless steel 

It is possible to create an alloy steel which is vir¬ 
tually rustless. This is accomplished by utilising 
chromium as an alloying component with iron, 
together with small percentages of other ele¬ 
ments. The chromium produces a bright, silver- 
coloured alloy and when the percentage of 
chromium exceeds 10 or 11 per cent, the range 
of materials known as the stainless steels is pro¬ 

Chromium is commonly used as the plated 
finish on steel or brass, either for decoration, as 
a means of avoiding corrosion, or to produce a 
hard surface. Chromium combines readily with 
oxygen and forms an oxide skin which effec¬ 
tively precludes further chemical action. 
Alloyed with iron, it confers its ‘rustless’ quality 
on the resulting alloy, but it is a hard material 
and this makes most stainless steels difficult to 

Stainless steels do however vary in their 
compositions. The principal additional compo¬ 
nent to these alloys is either nickel or carbon 
and if one is present, the other occurs only as a 
trace impurity. Quite large proportions of 
nickel may be used (up to 20 per cent) with 
between 8 and 20 per cent of chromium. How¬ 
ever, if carbon is used in the alloy it does not 
normally exceed 2 per cent. The presence of 
carbon is vital if the stainless steel is to be 



subjected to heat treatment but this is not nor¬ 
mally required by the amateur worker. 

Nickel forms one of the most versatile and 
important elements used in the production of 
alloy steels (of which the stainless types are 
only one variety) since, depending on the per¬ 
centage of nickel which is present, slow cool¬ 
ing, or annealing, can leave the steel in various 
conditions and special non-magnetic irons and 
steels can be manufactured. 

Many stainless steels are therefore either 
non-magnetic, or only mildly so, but some may 
require annealing to achieve the non-magnetic 
state since cold working tends to induce a 
degree of magnetism. 

Due to the presence of large percentages of 
chromium, stainless steels can be difficult to 
machine. To overcome this problem, free- 
cutting or free-machining alloys are available. 
These are created by the addition of small 
amounts of sulphur, as for free-cutting mild 
steel, but for stainless steels, selenium is some¬ 
times used instead of sulphur. 

Only very small percentages of sulphur are 
required, a typical analysis showing only 0.3 
per cent of this element. One particular free- 
cutting alloy is very close to the universally 
known 18/8 stainless steel (18 percent of chro¬ 
mium with 8 per cent nickel) having 17 to 19 
per cent chromium, 8 to 11 per cent nickel and 
up to 1 and 2 per cent of silicon and manga¬ 
nese. It is described as a non-magnetic, free- 
machining alloy and contains a maximum of 
0.12 per cent of carbon. 

Model engineers’ suppliers normally have 
available tubes in stainless steel in a small range 
of sizes up to about 0.5in. (12mm) in diameter 
and circular rods up to lin. (25mm). A small 
range of rectangular sections is also available 
from suppliers catering for the live steam 
enthusiast as it is used for superheater return 
bends and for the manufacture of fire grates. 
Ground rods are available from some sources. 

Heat treatment of steel 

Types of heat treatment 

Heat treatment is an operation, or combination 
of operations, involving controlled heating and 
cooling of a metal or alloy while in the solid 
state. The purpose of the heat treatment is to 
bring about some change in the physical prop¬ 
erties of the material with the object of chang¬ 
ing its hardness, increasing its strength or 
making it more ductile or malleable. Most 
metals and alloys react to some form of heat 

Heat treatment of steel is important since 
the carbon which is present in all steels is capa¬ 
ble of being transformed into hard or soft varie¬ 
ties by heating and cooling an item in a precise, 
but simple, manner. The main forms of heat 
treatment performed on steel are as follows: 


This is a process of softening a previously hard 
metal, usually so that it may be reworked in the 
soft state. 


Steel may be hardened to increase its resistance 
to wear and abrasion, or it may even be ren¬ 
dered sufficiently hard to cut other metals. 


Steel which has been hardened may be 
extremely brittle, making it useless for most 
applications. Some of the ductility may be 
regained by sacrificing some of the hardness, 
the process being known as tempering. 


Normalising is a process which restores the 
structure of a metal to its normal condition. If a 
metal has been worked (hammered, drawn or 
rolled) while in a cold state the grain structure 
becomes distorted. 



This can also occur if the metal has been sub¬ 
jected to prolonged heating, when grain 
growth, or swelling, takes place. Grain distor¬ 
tion is undesirable and normalising is used to 
restore grains to normal size. It is therefore 
sometimes referred to as grain refining. 

Hardening of carbon steel 

Steel is effectively pure iron to which small 
quantities of carbon have been added. This 
results in the formation of iron carbide, or 
cementite, making the steel progressively 
harder and tougher as the carbon content 
increases up to about 1.5 per cent. 

Steel which has 0.87 per cent of carbon is 
known as eutectoid steel. Its structure com¬ 
prises thin particles of cementite alternating 
with similar particles of ferrite, in the ratio of 
87 per cent ferrite to 13 per cent cementite. 
This particular structure is known by the name 

If eutectoid steel is heated to 700°C, the 
pearlite undergoes a change as the cementite 
and ferrite merge together to form a solid solu¬ 
tion of carbon in iron, called austenite. This is 
non-magnetic and the transformation may be 
confirmed by bringing a magnet close to the 
heated steel. 

A similar change takes place for all steels, 
except that, for other percentages of carbon the 
change takes place in two stages. At the lower 
critical point (700°C) the change in structure 
starts to take place but only for eutectoid steel 
does the change completely occur at this point. 
If there is more or less carbon than is found in 
eutectoid steel (0.87 per cent) the change to 
austenite is not completed until what is known 
as the upper critical point is passed. This varies 
according to the percentage of carbon which is 
present. For silver steel (1.1 or 1.2 per cent car¬ 
bon) the upper critical point is 800°C. 

The change to austenite, during heating, 

does take time to occur and the steel must be 
held above the upper critical point for a reason¬ 
able time. ‘Reasonable’ is generally taken to be 
roughly one hour per inch of thickness. This 
means 15 minutes for work which is 0.25in. 
(6mm) thick, or 7 to 8 minutes for work which 
is 0.125in. (3 mm) thick. These times should 
enable the maximum possible hardness to be 

The most important feature of austenite is 
that it can only revert to pearlite if it cools 
slowly. Rapid cooling prevents this change and 
instead ‘freezes’ the structure in a modified 
austenitic condition which takes the form of a 
glass-hard constituent called martensite. 

The more rapid the cooling of the austenite, 
the greater the quantity of the glass-hard 
martensite which is created and the harder is 
the resultant steel. However, extreme hardness 
brings with it extreme brittleness. 

Rapid cooling is arranged by plunging the 
heated material into water, or better still, brine 
( 3 /«lb of salt per gallon of water, 500g per 6 
litres) which is at room temperature. The item 
(tool) must be plunged into the liquid while still 
above the upper critical point, pointed or ‘busi¬ 
ness’ end first, and should be given a gentle up 
and down motion to encourage as rapid a cool¬ 
ing as possible. 

Brine is a better quenching bath than water 
due to the fact that, in water, bubbles tend to 
form on the surface of the tool as the water 
boils locally when the tool tip is plunged in. 
The bubbles insulate the metal from the cold 
water and therefore do not provide such a 
rapid quench as is possible with brine which 
does not exhibit this effect. Nevertheless, little 
trouble is likely to be experienced with water 
quenching if the tool is given a gentle motion, 
as described above. 

A similar procedure is also used for gauge 
plate except that this is described as ‘oil harden¬ 
ing’ and should therefore be quenched in oil. A 
clean engine oil, viscosity about 20SAE, can be 



used for quenching, but it is inclined to flame- 
up and special quenching oils are available 
which avoid this problem. 

Hardening temperatures for gauge plate do 
vary from manufacturer to manufacturer as the 
critical points are modified by the small addi¬ 
tions of manganese, vanadium etc. which are 
usually present. The manufacturer’s recom¬ 
mendations, which should be available with the 
purchased material, should be followed, but as 
a general rule, a temperature of between 810°C 
and 840°C will be found satisfactory for hard¬ 

Estimation of temperatures for hardening 

For hardening, temperatures of the order of 
700-800°C must be estimated; broadly in the 
‘red hot’ category. However, there are degrees 
of redness and although the degrees are highly 
subjective, they do give a guide to the tempera¬ 
ture which has been achieved. The discernible 
red colours commence at a temperature of just 
over 500°C with the ‘barely red’ description 
and continue through to ‘bright orange-red’ 
which occurs at 1000°C. The full list of tem¬ 
peratures, together with their usual descrip¬ 
tions, as seen in subdued daylight, is as follows: 


Temperature (°C) 

Barely red 


Dull red 


Blood red 


Cherry red 


Bright cherry 




Bright red 


Very bright red 


Bright orange-red 


The above descriptions are highly subjective 
but you should be able to experiment by heat¬ 
ing a steel sample and watching the colour 

change progressively as the temperature 
increases. For the hardening of silver steel and 
gauge plate, temperatures in the range 800- 
850°C are required, putting the colour range in 
the region of cherry red to bright red. 

Although it is necessary to ensure that the 
temperature of the whole sample has reached a 
value above the upper critical point, and has 
remained there long enough to convert all of 
the material to austenite, there are dangers in 
holding the work at high temperatures for long 
periods. The first problem is decarburisation 
which is the ‘burning out’ of the carbon from 
the heated surface of the metal. This reduction 
of the carbon reduces the hardness which may 
be achieved and therefore destroys the very 
quality which is desired. 

The second problem is caused by oxidation 
of the metal surface when held at high tempera¬ 
tures. This produces a scale on the surface 
which degrades whatever surface finish the 
item had prior to commencing the hardening 
process and may make it unusable in some 

If the tool is to be ground after hardening, 
then neither effect is likely to be a problem 
since the scaled or decarburised material will be 
removed. If this is not the case, the method 
used to overcome the problem is to coat the 
point (business end) of the tool in a protective 
material such as a paste made from chalk and 
water. Sometimes, it is recommended that iron 
wire should be wrapped around the sample in 
addition to the use of chalk paste, but the wire 
must be removed before quenching otherwise 
the surface may not be cooled sufficiently 
rapidly to produce the required degree of hard¬ 


Once the hardening has been performed, the 
material is extremely brittle and the tool may 



shatter if dropped. The material must therefore 
be Met down’ from this very hard but brittle 
state so that it is just as hard as is needed for the 
particular service. The letting down, or temper¬ 
ing, also relieves the internal stresses caused by 
the sudden cooling during quenching. 

Tempering consists of heating the hardened 
steel to a certain temperature and then cooling 
it. The hardened steel consists largely of mart¬ 
ensite, but if this is reheated, further changes in 
the structure occur to create softer, tougher 

The higher the temperature to which the 
martensite is raised, the softer and more ductile 
is the steel structure which is formed. Reheat¬ 
ing hardened steel to just below the 700°C 
lower critical point produces a modified 
cementite which has minimum hardness and 
maximum ductility. Reheating to lower tem¬ 
peratures leaves progressively harder com¬ 
pounds as the reheating temperature reduces. 
Clearly, no reheating leaves the martensite 
unaltered and therefore at maximum hardness. 
A range of structures of differing hardness can 
be created by reheating hardened carbon steel 
to a range of different temperatures. 

Estimation of temperatures for tempering 

For tempering, the actual temperature to which 
the steel is raised is very critical, ranging from 
200°C to a little over 300°C depending on the 
service to which the tool is to be put. When it is 
realised that this limited temperature range 
covers the full range of hardness which is likely 
to be required, from cutting tools to ‘softish’ 
springs, the degree of control of temperature 
can be appreciated. Fortunately, these tempera¬ 
tures can also be estimated by observing the 
colour of the tool as it rises in temperature. 

If steel is heated in an oxidising atmosphere 
such as air, a film of oxide forms on the surface. 
The oxide changes colour as the temperature 

increases and the colours are used as a means of 
judging the temperature. This is, again, a subjec¬ 
tive test, but the practice is so well established 
that little difficulty is likely to be encountered. 



Colour (°C) Uses 

Pale yellow 


Turning tools for brass 

Pale straw 


Turning tools for steel, taps 
and dies 

Dark straw 


Drills, milling cutters 



Cutters for hard wood 



Wood boring tools 



Twist drills, coopers’ tools 

Light purple 





Bone and ivory saws 

Dark purple 


Punches and cold chisels 

Full blue 



Dark blue 



Lists of tempering colours and their related 
temperatures do vary from one authority to 
another, as might be expected for such a subjec¬ 
tive test. Some lists are very abbreviated, show¬ 
ing only six or seven colours, but a relatively 
comprehensive list is given here, together with 
the approximate equivalent temperatures and 
the normal service for which the tempered steel 
is used. In normal references to tempering, 
little attention is paid to the actual tempera¬ 
tures required, the usual advice being to 
‘temper to dark straw’ or whatever. 

The best way to reconcile the list of colours in 
your mind, and to practise the procedure, is to 
carry out an experiment for yourself. Take a 
piece of mild steel, about 0.25in. (6mm) in 
diameter, and polish a couple of inches at one 
end with emery cloth until it has a nice bright 
finish. The colours of the oxides of steel do not 
develop correctly within a flame, so the end of 
the rod (point of the tool) cannot be heated 
directly. Instead, heat a point about 1.5in. 
(40 mm) from one end with a single small flame 
from a blowtorch. As the steel heats up, the 
range of oxide colours will be seen progres- 



sively, from pale yellow, through straw, brown, 
purple and blue as the temperature increases. 

If the flame is held stationary, 1.5in. (40mm) 
from the end, a band of the colours will be seen 
to progress along the rod, away from the flame, 
the coolest, pale yellow, leading this band 
towards the end. Within the flame, these col¬ 
ours cannot be seen, but they will be evident, 
travelling both ways from the ‘hot spot’ as the 
steel heats up. 

Take the flame away when the pale yellow 
reaches the end of the rod but observe that the 
travelling band of colours still keeps on mov¬ 
ing, the end of the rod becoming progressively 
straw, brown, purple etc. as heat from the hot¬ 
test part is conducted away. The tip of the rod 
thus continues to heat up even after the flame is 

If you are attempting to temper to dark 
straw, say, you need to take account of the fact 
that the end of the tool will continue to heat up 
after the flame is removed. You must stop this 
happening, otherwise the tool will be rendered 
too soft as the ‘hotter’ colours reach the end. 
The answer is to remove the heat in the tool as 
quickly as possible once the tempering colour 
has reached the end. This is achieved by remov¬ 
ing the tool from the flame and immediately 
quenching it in water or brine at room tem¬ 
perature, as for hardening. 

The above tempering procedure inevitably 
raises the body of the tool to a higher tempera¬ 
ture than the point, therefore producing a soft 
body and a hard tip to the tool. For punches, 
scribers and so on, this is quite acceptable, but 
if a cutting tool having cutting edges all along 
its length is being made, such as a broach or a 
tap, its whole length must be tempered. In these 
instances, it is most convenient to lay the tool in 
a small tray of dry sand and heat it indirectly by 
means of a blowtorch until it achieves the 
required colour. 

The alternative is to immerse the item in a 
liquid held at a suitable temperature. Molten 

solders can be used, or alternatively, an oil bath, 
for which a normal oven thermometer will 
provide a sufficiently accurate measurement of 
temperature. The tool should be allowed to 
soak for a sufficient time for it to heat right 
through. Once again, engine oil of about 
SAE20 viscosity may be used, but quenching oil 
is to be preferred. After tempering in oil or 
sand, the item should be quenched in water or 
oil in the normal way. 

Annealing hardened and tempered 
carbon steel 

Carbon steel tools which have been overheated 
in use can be rehardened and tempered again 
provided that the material is first restored to its 
unhardened (pearlite and cementite) state. This 
process is known as annealing. If annealing is 
not first carried out, there is great danger that 
cracking or distortion will occur when an 
attempt is made to reharden. 

For complete annealing of carbon steel, the 
tool must be heated slowly to just above the 
upper critical point (800°C) and held there for 
one hour per inch of tool thickness, as for hard¬ 
ening. The subsequent cooling must also take 
place slowly and if the annealing is performed 
in a furnace, the tool is normally left inside and 
the whole allowed to cool naturally. If the 
annealing is done in an open hearth, as is likely 
in the home workshop, the object should be 
covered with a good layer of hot ashes or sur¬ 
rounded with as much hot firebrick as can 
reasonably be arranged. Once annealing is 
complete, hardening and tempering may be un¬ 
dertaken normally. 

Case hardening 

Mild steel cannot be hardened in the same 
manner as carbon steel since it contains too 



little carbon and insufficient martensite is 
formed to give the material any significant 
hardness. However, the process known as case 
hardening can be used to create a hard surface 
on mild steel, thus providing an extremely hard 
(high-carbon) surface, but a soft, tough inner 
core. Case hardening is not recommended for 
free-cutting mild steel and components which 
are to be given a hard surface by this means 
should be made from black or bright-drawn 
mild steel. 

To create the hardened skin, a way has to be 
found to convert the surface mild steel into a 
high-carbon steel. This can be achieved by plac¬ 
ing red hot mild steel in contact with a high- 
carbon material and allowing additional 
carbon to be absorbed into the surface. Deep 
penetration by the carbon is not normally 
achieved, but a hard skin, or case, up to Vu in. 
(1.5mm) thick is perfectly possible. 

The high-carbon steel surface may then be 
hardened by heating above the upper critical 
point and quenching, as for silver steel or gauge 
plate. This produces a ‘dead hard’ surface but 
leaves a soft core and in many applications the 
ability of the material to flex under load is 
extremely advantageous, especially considering 
the very hard surface which can be produced. 

For case hardening of the small quantities of 
relatively small components likely to be of 
interest to the amateur, special carbon-bearing 
substances are available commercially. The 
object to be hardened is heated to bright or¬ 
ange-red (about 1000°C) and dipped into a tray 
of the compound. This is supplied as a powder 
which melts on contact with the heated sample 
and sticks to its surface. 

A good coating of the preparation (techni¬ 
cally known as a carburiser, but more com¬ 
monly called case hardening compound) is 
built up on the sample by repeated heating 
and dipping. When the article is adequately 
coated, it is heated once more to 1000°C and 
then quenched in water. The remaining carbur¬ 

iser may then be removed and the article 
polished or cleaned up to suit the application. 

Should it be necessary, the hardened case 
may be rendered soft by following the proce¬ 
dure described above for softening hardened 
and tempered carbon steel, except that there is 
no need to ‘soak’ the sample for such a long 
period since only the surface is hardened. 

Normalising mild steel 

Normalising is intended to ensure that the 
internal structure of steel takes on a uniform, 
unstressed condition in which the ‘grains’ of 
steel assume a uniform size. It is therefore a dis¬ 
tinctly different process from annealing which 
is designed to reduce high-carbon steels to a 
soft, or machinable, condition. 

Low-carbon steels consist of pearlite and 
ferrite, neither of which is intrinsically hard, so 
what is generally described as mild steel is 
always sufficiently soft to be in a machinable 
condition. However, the internal structure of 
the steel may not be in the ideal state. If the 
steel has been cold worked (drawn, rolled, 
hammered, guillotined from large sheets etc.) 
the structure of the metal has been distorted 
and there will be unequal stresses in the mate¬ 
rial. These stresses sometimes produce a distor¬ 
tion in the shape of the steel, but not always. 

For example, a piece of bright mild steel 
sheet, guillotined from a large sheet usually 
shows a large-radius ‘bow’ from end to end, 
and perhaps even a twist along the length. Sec¬ 
tions of bright mild steel, on the other hand, are 
normally quite straight as bought, but will dis¬ 
tort significantly if large sections of the mate¬ 
rial are removed by machining. 

The cause of the distortion is unbalance of 
the stresses that have been set up by working 
the material in the cold state. Normalising is 
the method adopted to remove the residual 
stresses and for low-carbon steel (mild steel) 



places the material in its best condition for 
machining. Removal of the stresses also 
reduces the likelihood of distortion occurring 
during case hardening. 

Normalising simply consists of heating the 
steel to just above the upper critical point and 
then allowing it to cool in still air at room tem¬ 
perature. For mild steel with 0.3 per cent car¬ 
bon, this temperature is about 800°C (bright 
red). The sample should be heated up slowly to 
ensure that an even temperature is achieved, 
and then allowed to cool in still air. It is vital 
that cooling is uniform, and draughts should be 

There is one other situation in which grain 
distortion can be caused - this is by prolonged 
heating of a steel sample, well above the upper 
critical point. This can occur, for example, 
where much hot forging of the material has 
taken place and several reheatings to quite high 
temperatures have been required. This causes 
grain growth or swelling to take place. 

Normalising - sometimes referred to as 
‘grain refining’ - tends to restore the grains to 
their normal size and refinement and therefore 
removes any internal stresses resulting from 
prolonged heating. 


The major source of copper is the ore known as 
Pyrites, which normally contains more than 30 
per cent copper. The copper is released from 
the ore by use of a blast furnace, but the ore re¬ 
quires various stages of refinement prior to 
smelting in order to produce a relatively pure 
copper. Even so, further refining is required to 
produce commercial grades of copper. 

One way of refining copper is to use elec¬ 
trolysis to isolate pure copper from ingots of 
impure smelted copper. This produces a high- 
purity copper, known as electrolytic copper, 

which is used for preparation of copper-based 

Pure copper possesses valuable properties 
which make it an extremely useful material. It 
has a high conductivity to both heat and elec¬ 
tricity and has the added advantage that it is 
both ductile and malleable. This means that it 
can be drawn down to produce very fine wire 
and can be pressed, forged or spun into compli¬ 
cated shapes without cracking. Copper is also 
corrosion resistant and may be joined by 
soldering, brazing or welding. 

Copper is supplied in sheet, bar and tube 
forms, bars being available as rounds and flats 
(rectangular cross-sections). Its hardness varies 
according to the way that it has been worked or 
heat treated. Basically, it may be in the hard or 
soft condition. Bars or tubes which have been 
drawn or extruded, without subsequent heat 
treatment, will be in the hard condition, but 
tubes (particularly) are also available in the soft 
state, and are much used, for example, in the 
refrigeration industry. Soft tubes are normally 
supplied in a coil, but hard copper tubes, e.g. 
for plumbing, are in the straight condition. 

Tubes are available having a specified bore, 
or alternatively having a specified outside 
diameter, so it pays to be careful when order¬ 
ing. Model engineers’ suppliers stock by out¬ 
side diameter. 

Copper strips are available in some sizes in 
the soft condition, particularly those with small 
cross-sectional areas, but sheet is not normally 
stocked in the soft state since large sheets are 
susceptible to damage and therefore difficult to 
handle. Sheets are therefore normally hard or 
are in an intermediate condition between hard 
and soft. The processes under which these vari¬ 
ous tempers are produced are described under 
the heading ‘Working copper and brass’. 

The principal model engineering use of cop¬ 
per is for boilermaking, for which its ductility, 
corrosion resistance and solderability make it 
ideal. The flanging and forming of copper (and 



brass) is described in Chapter 7. The softening 
of hard copper is described under the heading 
‘Working copper and brass’ below. 

Copper is also extremely important because 
of its wide use in the preparation of alloys, since 
it is the basis of brass, bronze, gunmetal and 
monel metal, among others. An alloy is a combi¬ 
nation of two or more metals which are smelted 
together to form a combination material. 

Alloys generally may be harder and stronger 
than the pure metals which are their constitu¬ 
ents, and may have melting points lower than 
any of their constituents. There is, therefore, 
much scope for development of new alloys hav¬ 
ing special characteristics. This is true also of 
steel alloys, but these are of less interest to the 
amateur than the copper-based materials. 

Alloys of copper 


The most common alloy of copper is brass. This 
word is used to describe many actual alloys but 
it is too vague a term to have any scientific 
meaning, although it is in general use. Most 
brasses are alloys containing only copper and 
zinc and are known as binary alloys since they 
contain only two components. However, there 
are some brasses which contain other metals, 
conferring special properties on these alloys. 

When less than 36 per cent of zinc is present, 
brass is very ductile and can be cold worked 
into complex shapes e.g. cartridge cases. Bras¬ 
ses with less than this percentage of zinc are 
termed Alpha Brasses. Nearly all sheet brass is 
of this type. 

As the zinc content of brass increases, it 
becomes more brittle (zinc is itself brittle) and 
brasses with between 46 and 50 per cent of zinc 
have a quite different structure from alpha 
brass, being harder but very brittle. These 

alloys are known as Beta Brasses and their 
properties make them suitable only for use as 
brazing materials. 

If the percentage of zinc is between 36 and 
46 per cent, the material is known as an Alpha- 
Beta Brass since it contains both types of alloy. 
These brasses are harder and stronger than the 
alpha brasses and most bar stock is of the alpha- 
beta type. Bar stock brasses frequently contain 
other alloying elements to improve the strength 
or machinability. Aluminium increases the 
strength, and high-tensile brass contains 3 per 
cent aluminium in addition to iron and manga¬ 
nese. A small percentage of lead may be added 
to improve the machinability of the brass, but 
large quantities tend to weaken the alloy sig¬ 
nificantly so only small amounts are usually 

Like copper, brass is available in sheet, bar 
and tube form, but square-section tubes are 
available in addition to circular section, and 
thick-walled tubes can also be obtained. These 
are described as ‘hollow rod’. In addition, brass 
is available in equal and unequal angles, tee, U 
and half-round sections. Brass suffers a pro¬ 
gressive breakdown at high temperatures and is 
not suitable for making boiler fittings, other 
than for ‘toy’ boilers operating at low pres¬ 

Manganese bronze 

A useful variant on the brass theme is a material 
known as manganese bronze. This is basically 
an alpha-beta brass to which up to 2 per cent 
of manganese is added together with up to 1 
per cent each of aluminium, iron and tin. 
Usually, 58 to 60 per cent of copper and 39 
to 41 per cent of zinc form the basis of the 
alloy. This material is as strong as mild steel, 
highly corrosion resistant and is excellent for 
bearings and steam fittings. It is also easy to 



Bronze and gunmetal 

Bronze is the general name given to copper-tin 
alloys but commercial bronzes are usually 
composed of copper, tin and zinc. Bronzes 
contain 70 to 90 per cent of copper, 1 to 18 per 
cent of tin and 1 to 20 per cent of zinc plus 
other elements in varying proportions depend¬ 
ing on the properties required of the alloy. 

The most common bronze likely to be used 
in the home workshop is that generally known 
as gunmetal. This name is given to a group of 
bronzes originally used for the making of can¬ 
nons. It resists corrosion, is relatively strong, 
casts well and is widely used for pump bodies, 
steam fittings and for castings that are subject 
to pressure or shock loads. 

The alloy is normally used as cast (with 
appropriate machining), specific castings being 
provided for steam fittings and valves, cast 
bushes for model boilers, regulator bodies and 
water pumps. Gunmetal is also frequently used 
for steam engine cylinders and pistons, but in 
this case two distinct alloys will be supplied, in 
order to ensure adequate life, since adjacent 
sliding or rotating surfaces formed of identical 
materials suffer high wear. 

Gunmetal is also available in continuously 
cast sticks, sometimes called continuously cast 
bronze. This is available as solid or cored (hol¬ 
low) round bars which are ideal for the manu¬ 
facture of bushes or bearings. 

Phosphor bronze 

The description phosphor bronze is generally 
applied to a bronze containing 10 to 14 per 
cent of tin with 0.1 to 0.3 per cent of phospho¬ 
rus. This alloy is stronger and harder than 
gunmetal and it makes a good bearing material, 
particularly for components which must sup¬ 
port heavy loads. Commercially, it is frequently 
used in the form of castings because it is very 

fluid when molten and may be cast into intri¬ 
cate shapes. In the home workshop it is most 
likely to be used in the form of drawn bars, for 
use as bearings, or for the manufacture of steam 
fittings for model boilers. Rounds, squares, 
hexagons and flat bars are widely available in 
this useful material, but it has the disadvantage 
that it is not easy to machine. 

Hard-drawn phosphor bronze wire can also 
be obtained. This is quite springy and yet can be 
bent relatively easily for the manufacture of 
non-corrosive springs without subsequent heat 
treatment. Rolled strip is also available, which 
also finds use as springs. It has relatively good 
electrical properties and is used for spring- 
contact electrical connections, as also is hard- 
drawn wire. 

Monel metal 

Monel metal is primarily a nickel-copper alloy, 
usually in the proportions of 2 parts nickel to 1 
part copper. Small percentages of manganese, 
silicon, iron, sulphur and carbon are normally 
present, together making up less than 3 per 
cent. This alloy came about because in Canada, 
deposits of ores were discovered containing 2 
parts nickel to 1 part of copper and a great deal 
of expense, and very elaborate smelting proc¬ 
esses were required to separate the two 
elements for use. A smelter, Ambrose Monell, 
experimented by smelting the combined ore 
directly, thus producing a natural alloy of the 
two elements. This was the inception of monel 
metal, a first quality, corrosion-resisting alloy. 
Monel is one of only very few alloys that are 
smelted directly from mixed ores. 

Monel is used extensively where relatively 
high mechanical strength is required at high 
temperatures (it will withstand temperatures 
up to 500°C) or where its high corrosion resist¬ 
ance is required. It therefore finds much use in 
hospital, laundry and food handling machinery. 



For the model engineer, it is used for valves and 
valve seats and for stays in model boilers. It is 
an extremely tough alloy but can be machined 
readily if sharp tools are used. 

Nickel silver 

Nickel silver is an alloy of copper, zinc and 
nickel. Normally, 20 per cent of zinc is present 
together with 55 per cent of copper and 20 per 
cent or so of nickel. Small quantities of cobalt, 
lead and iron are also usually present. 

Nickel and, to a lesser extent, zinc are shiny, 
white metals. When these are mixed with cop¬ 
per, the resultant alloy has a bright, silvery 
appearance. The alloy is generally known as 
nickel silver but is also referred to as German 
Silver, the names arising from the bright colour 
of the alloy rather than from the actual pres¬ 
ence of silver. 

Nickel silver has poor electrical conductivity 
and is used for making electrical resistance 
wire. It solders well, takes paint better than 
brass and has relatively good corrosion resist¬ 
ance. It is popular as a material for use in the 
construction of small-scale model locomotives 
and is available in sheet form, from ‘shim’ 
thicknesses such as .005in. (0.13mm) to '/win. 
(1.6mm) or so, in addition to rounds and flats. 
It machines easily also (somewhat like brass) 
and is useful as a substitute for steel for small 
model fittings which will not be painted but re¬ 
quire some better corrosion resistance than 
steel fittings would provide. 

Nickel silver also casts well and is used ex¬ 
tensively as a steel look-alike in the small-scale 
model railway field. In the larger scales it is fre¬ 
quently used for live steam locomotive cross¬ 
heads due to the ready machinability of the 
castings and steel-like appearance. 

Working copper and brass 

Work hardening 

Materials which are malleable and ductile have 
an internal structure which allows individual 
grains within the material to slide over one 
another without a great deal of energy being 
absorbed. Such materials are therefore very 
easy to bend, or squeeze into new shapes. The 
easy slippage between the grains does not con¬ 
tinue unrestrained however, as the material is 
worked. Only limited slippage of the grains can 
occur, since once a grain has slipped to take up 
the available adjacent ‘space’ it becomes locked 
to its neighbours and is no longer free to slide. 

As the material is worked in the cold state, 
the grains in the region being worked progres¬ 
sively take up their ‘free’ movement and even¬ 
tually will not slip relative to one another. The 
material therefore appears to become much 
harder as it is worked and the process is known 
as ‘work hardening’. 

Copper is one of the common materials 
which is malleable and ductile but which suf¬ 
fers work hardening. Since copper forms the 
major constituent of most brasses, they also 
exhibit this characteristic, but it is only the al¬ 
pha brasses (less than 36 per cent of zinc) which 
are ductile and likely to be formed into com¬ 
plex shapes. 

Any cold working of the metal distorts the 
grains within the structure so that simple cold 
rolling of copper and brass, to produce com¬ 
mercial sheets, also causes work hardening. 


Once the material has become hard it is very 
difficult to work and may fracture if further 
deformation is attempted, so it must be sof¬ 
tened before further work can be done. This 
process, known as annealing, consists of raising 



the temperature to a sufficient level to allow 
rearrangement of the individual atoms into 
their preferred shape and juxtaposition, thus 
allowing re-establishment of the ‘free spaces’ 
and the rounded and even grain structure. 
Grain refining also takes place, as it does for 
carbon steels, and since smaller grains give 
greater strength and better shock resistance, 
annealing brings the metal to the best possible 

The temperature required for the annealing 
of copper and brass is in the range 400-600°C. 
Industrially, where annealing is carried out in 
temperature-controlled furnaces, 400°C is nor¬ 
mally used. For the home workshop, something 
more ‘visible’ is required and heating to a dull 
red will be found satisfactory. This is equivalent 
to 700°C and therefore above the required tem¬ 
perature, and since the structural change only 
takes a short time to occur, excessive heating is 
not required. Viewing the heated metal in sub¬ 
dued daylight will normally allow redness to be 
seen at 600 to 650°C, which is ideal. 

There is no absolute need to quench copper 
or brass to achieve a satisfactory anneal, but it 
does help to remove any scale which forms. It 
does have a tendency to distort largish sheets 
since the rate of cooling is not usually uniform, 
but if this is not a problem, clean water at room 
temperature may be used. 

Overheating of brass and copper for long 
periods causes grain growth to occur, as it does 
for steel, thus rendering the metal weaker and 
less able to withstand shocks. This is not a 
problem when annealing but should be borne 
in mind when silver soldering is carried out as 
temperatures up to 800°C may be required. 

Degrees of temper 

Hot working of copper and brass does not 
cause work hardening since the atoms have suf¬ 
ficient energy to rearrange themselves into the 

preferred structure. It is, therefore, possible to 
produce copper and brass in the soft condition. 
This is available commercially in some forms 
(tubes, for example) but is susceptible to dam¬ 
age in sheet form and therefore only available 
in a limited range of sizes. 

However, since these materials are hardened 
by cold working, it is possible to complete the 
manufacturing process by cold rolling to final 
thickness. Depending upon the amount of cold 
working, the metal may be in any state between 
soft and fully hard. The degrees of hardness, or 
tempers, which are available are usually soft, 
quarter, half and fully hard and metal stock 
holders will be able to supply copper and brass 
to these degrees of temper. Half hard repre¬ 
sents a compromise for most purposes and is 
the most commonly available material. It is suf¬ 
ficiently hard to permit ease of handling, mark¬ 
ing out and cutting and yet may be rendered 
soft by a simple annealing process. 

Aluminium and its alloys 

Aluminium is an important engineering mate¬ 
rial due to its light weight, corrosion resistance 
and good electrical and thermal conductivity. 
Its most serious disadvantage is its relative 
weakness (it has only about one-third the 
strength of steel). This means that careful 
design is required to place appropriate sections 
in the correct orientation with respect to the 
anticipated loads. Fortunately, aluminium is 
very workable and the production of the 
required shapes is quite straightforward, what¬ 
ever the method of manufacture involved. 

Since aluminium in its pure state is soft, it is 
seldom used in this condition. It does, however, 
form the basis of many extremely useful engi¬ 
neering alloys, being alloyed with copper, man¬ 
ganese, magnesium, or silicon (or even all four) 
in various proportions, to produce specific 



characteristics. Aluminium has good corrosion 
resistance since, like chromium, it rapidly forms 
a tough oxide skin on contact with air, which 
prevents further corrosion. When alloyed with 
other metals, oxide formation is naturally not 
so readily induced, but nevertheless, most 
aluminium alloys do exhibit good corrosion 

Like copper, pure or near-pure aluminium is 
a work-hardening material and it is therefore 
available in different degrees of temper that 
characterise the sheet forms of copper and 
brass. Some alloys of aluminium, particularly 
alloys of aluminium and copper, are susceptible 
to a process of age hardening i.e. they may be 
rendered soft by heat treatment but will then 
reharden over a period of time. Rehardening 
may take place at elevated temperatures or may 
occur at room temperature. 

Perhaps the best-known name among alu¬ 
minium alloys is Duralumin (or Dural); an alloy 
with 4 to 5 per cent of copper and small amounts 
(less than 1 per cent each) of silicon, iron, man¬ 
ganese, magnesium and zinc. Duralumin was 
the original ‘strong’ alloy, having roughly twice 
the strength of commercial aluminium, or about 
two-thirds the strength of steel. 

Duralumin is available as extruded rounds, 
hexagons and flats from most factors or alter¬ 
natively these are available in alloys containing 
only 0.1 per cent of copper, but greater quanti¬ 
ties of magnesium, quite distinct materials be¬ 
ing available having 0.5 to 1,1 to 2.5 or 4 to 5 
per cent of magnesium. 

Extruded angles, tees and Us are normally 
only available in magnesium-rich alloys, rather 
than dural, and the same is true also of drawn 
and extruded tubes. 

Sheet, plate and strip are supplied in com¬ 
mercial aluminium (99 per cent pure), magne¬ 
sium-rich alloys or as the copper-rich dural. 

In model engineering, aluminium alloys tend to 
be used in just those situations in which their 
light weight is advantageous, especially since 

relatively strong alloys are available. Model 
aero engines use these alloys extensively as they 
are ideal for pistons, crankcases, cylinder bar¬ 
rels and cylinder heads, sumps, timing covers 
and so on. Since the use of these alloys is now 
well established, they are used for major com¬ 
ponents for small internal combustion engines 
generally. Where adequate strength can be 
achieved, these alloys are also used for castings 
for parts of some tools for home machining, 
mostly on account of the relatively low cost and 
ease of manufacture. 

Heat treatment 

Dural exhibits the age hardening referred to 
above and since this occurs at room tempera¬ 
ture, the material is naturally normally in the 
hard state. In this condition it cannot be bent 
through more than small angles without frac¬ 
turing and sheet dural must be annealed before 
any significant bending operations can be 
carried out. 

One of the problems of annealing dural is es¬ 
timation of the temperature of the sample. Since 
the alloy melts at about 650°C and a propane 
torch will comfortably exceed this, it is all too 
easy to melt the sample and reduce it to a pool of 
scrap. The usual way to assess the temperature is 
to heat up the sample (keeping the torch well on 
the move) and to test its temperature peri¬ 
odically by rubbing the end of a ‘dead’ match 
along the surface. When the dural is hot enough 
for the match to ignite, the annealing tem¬ 
perature has been reached (500°C) and the 
material must immediately be quenched in 
water at room temperature. This renders the 
dural fully soft and it will remain in this condi¬ 
tion for a while, progressively age hardening at 
room temperature over a period of a few hours. 
Any bending operations should therefore be 
carried out as soon as possible after completion 
of the anneal. 




For the production of castings, aluminium 
alloys having relatively high levels of silicon are 
used since this gives greater fluidity to the alloy 
in the molten state. If particularly thin sections 
are to be cast (where greater fluidity is required) 
more silicon is generally utilised. 

The addition of silicon to the ‘casting’ alloys 
of aluminium tends to increase the difficulties 
of machining since the silicon is itself abrasive 

and tends to form particularly hard compounds 
within the material. Cast aluminium is, there¬ 
fore, frequently much more difficult to 
machine to a good, bright finish. 

The low melting point of aluminium also 
makes it suitable for casting in the home work¬ 
shop. This makes it possible to make castings 
from one’s own patterns, for items that are not 
too highly stressed, and can provide a more 
than useful introduction to the patternmaker’s 
and foundryman’s trades. 


Setting up a workshop 


Much of what is needed in a workshop may 
seem at first sight to be enormously expensive. 
There are several points to bear in mind how¬ 
ever. First of all, it is highly likely that you 
already have some interest in manual skills and 
will have a ‘basic’ set of hand tools, perhaps 
even a bench and vice, and somewhere to work. 
Secondly, it is not necessary to acquire the con¬ 
tents of a complete workshop immediately, and 
thirdly, many of the smaller items (and even 
some quite large ones) can readily be made 
once the basic machinery is available. It is also 
worth bearing in mind that the best models are 
not necessarily made on the newest or most 
sophisticated machinery, and new items are 
not, therefore, vital to success. 

Given that a bench and vice will naturally be 
required, together with an area set aside for 
assembly work, the items needed can be 
broadly grouped under the following headings: 

• a drilling machine 

• a lathe and its attachments 

• measuring and marking-out equipment 

• hand-cutting tools (saws and files) 

• threading equipment (taps and dies) 

• some smaller hand tools (scribers, punches, 
clamps etc.) 

• some means of sharpening cutting tools. 

If much soldering or brazing (silver soldering) 
is likely to be required, at least a rudimentary 
brazing hearth and the necessary torch (blow¬ 
lamp) and gas supply will also be required. 

The provision of space for the workshop 
should be the first consideration. Once this is 
available, the purchase of a lathe is likely to be 
the next step, if only for psychological reasons, 
since it is the possession of this versatile tool 
which turns one from a ‘hobbyist’ into a ‘model 
engineer’. The general characteristics of small 
lathes are described in Chapter 12 and the pur¬ 
chase of a secondhand machine is considered in 
Chapter 17. 

Space for the workshop 

General considerations 

Space (floor area) is a fairly expensive com¬ 
modity, at least in the South-east, where a 3- 
bedroom house offering some 1500 square feet 



(150 square metres) of floor area is currently 
valued at around £70 per square foot (£700 per 
square metre). The provision of the space for a 
workshop is therefore likely to be the major 
item of expense, since some dedicated area is 
essential to house the machinery, benches, and 
so on which are required. 

Several possibilities exist, however, from 
the spare room to the garden shed, not forget¬ 
ting the use of part of a garage, or a purpose- 
built extension. In reality, a combination of 
locations can be more convenient and may well 
allow quite a small floor area to be used for the 
basic workshop, without the necessity for it to 
accommodate all of the items required. 

The essential requirements for the workshop 
are that it should be dry, warm in winter and in 
a location or situation which makes it available 
for the ‘odd half hour of modelling’ for, try as 
we might, many of us cannot ordinarily spend 
long periods away from family and social com¬ 
mitments. Above all, the workshop must be a 
suitable location in which to store relatively 
expensive machinery and valuable models, so it 
must be secure. 

To achieve these desirable characteristics, 
the workshop must be substantially built, well 
insulated and incorporate at least a modest 
degree of heating and ventilation, as a means of 
removing excess moisture from the atmos¬ 
phere. In these respects, a spare room in the 
house may seem desirable, but there are some 
disadvantages to this. First of all, there is the 
noise problem. Some processes, even simple 
sawing, are by their nature noisy, and a work¬ 
shop within the house may impose some 
restrictions in this respect. There is also the 
problem that the close proximity to the living 
quarters increases the likelihood that swarf, oil 
and general dirt will spread around the house. 

A spare room on the first floor is also not 
ideal for the installation of heavy machinery (it 
is fine once up there, but difficult to install in 
the first instance). The first floor is also unlikely 
to be suitable if the model you will build is 

likely to be heavy. Even a modest 0-6-0 tank 
locomotive in 5-inch gauge may well weigh 80 
to 100 pounds (36 to 45 kilos) when complete 
and may be difficult to get downstairs at the 
end of the day - or 10 years, depending upon 
how fast you work! 

If a spare room is available on the ground 
floor, perhaps a utility room or playroom 
which can be dispensed with, this is altogether 
a different proposition. It meets the essential 
requirement of being instantly available, will 
most likely have some form of heating already 
installed, and is certain to be provided with 
ventilation. Provided that noise is not likely to 
be a problem, it is socially more acceptable to 
be within the house, since you are available 
should the family need your attention. You can 
also readily be supplied with refreshment and 
can easily ‘down tools’ for a social break over a 
cup of tea or coffee. 

Bearing in mind the need to store steel and 
iron components and materials, a positive 
approach to rust prevention must be adopted. 
The essential requirement is to keep the atmos¬ 
phere dry. This can be by use of ordinary heat¬ 
ing and ventilation or by use of one of the small 
dehumidifiers which are now becoming avail¬ 
able. Unfortunately, these are expensive and 
also take up significant space, thus consuming 
commodities which are generally in short sup¬ 
ply. The alternative is likely to be an electrically 
powered fan heater or a tubular heater, allied to 

If the workshop is within the house, the 
domestic heating system might satisfy this 
essential requirement, which is very conven¬ 
ient, but even nice, dry heating provided by 
radiators cannot be expected to prevent rusting 
entirely since there are inevitably spaces which 
do not have sufficient air flow, and moisture¬ 
laden air can accumulate here and there. Under 
no circumstances should moisture-producing 
heaters such as paraffin burners or gas heaters 
be used, unless provided with a properly 
installed external flue. 



Space requirements 

It is difficult to be specific about the actual floor 
area required since it depends a great deal on 
the size and type of your machinery, the type of 
model engineering which is done and whether 
some additional space is available outside the 
main working area. As an example, my work¬ 
shop is roughly 8 ft (2.4m) square which is 
adequate for model locomotive building except 
for the fact that three of the walls include a 
door, all three of which are used from time to 
time, so models, some materials and brazing 
equipment have to be stored elsewhere. 

When originally established, the workshop 
housed a small bench, with 4-inch (100mm) 
vice and drilling machine, the lathe on its own 
bench, a small bench used more as a table, and 
two wooden bookcase-type units that were 
used for general storage. These were roughly 
the same height as the small bench and about 
9in. x 30in. (230mm x 760mm) in plan. Each 
bookcase unit was just about large enough to 
accommodate a 5-inch gauge 0-6-0 tank loco¬ 
motive. The basics of the workshop were thus 
housed reasonably well in 64 square feet, or 
about 5.75 square metres. 

Recently, the small bench/table has been 
replaced by a smaller one, and one of the book¬ 
case units has been discarded, permitting the 
introduction of another small bench which 
provides a mounting for a small mill/drill, but 
this has so cramped the workshop that a visitor 
is now difficult to accommodate, and the hori¬ 
zontal table on the mill/drill must be traversed 
to one end or the other of its travel, according 
to whether one needs to sit at the bench/table, 
or gain access to the adjacent garage. 

Even so, with only a single door, this space 
would also house the second bookcase unit, 
which would be valuable for assembly work 
and for further storage, or would accommo¬ 
date larger models. Even an 0-6-0 tender 
locomotive in 5-inch gauge would be an embar¬ 
rassment at present. 

Garden shed workshop 


For many, the possibility of an indoor work¬ 
shop is fairly remote, if only on the grounds of 
cost. For this reason, the garden shed is often 
the only choice and is the traditional location, 
at least in the UK. Such a location does bring 
with it certain advantages - noise and dirt are 
unlikely to be a problem to the rest of the fam¬ 
ily and the enforced walk between workshop 
and house provides an opportunity for dust and 
swarf to fall off clothing and footwear. This is 
very advantageous as it is amazing how far 
from the workshop ‘sticky’ swarf such as cop¬ 
per and phosphor bronze will travel before 
deciding to embed itself in the carpet. One 
acquaintance relates how wheel turning, or 
other cast iron machining, always produces a 
rusty patch outside his garden workshop, just 
where the dust tends to fall away. 

The cost of an 8ft x 6ft (2.4m x 1.8m) 
wooden garden shed is around £250 at the 
present time, depending upon the finish and 
quality (type) of timber which is used. It is a 
simple matter to design and build a workshop 
to one’s own specification and is a cost- 
effective way to provide the workshop space. 
The overall size can be tailored to suit the site, 
and the internal layout and timber sizes can be 
arranged to suit the type of insulation which is 
to be installed (and this is highly recom¬ 
mended). Positioning of the internal timber 
‘studs’ can also be arranged to suit the wall 
covering to be used (again, highly recom¬ 
mended) and the result will certainly not be in¬ 
ferior to commercial products. 

Commercially, frames made from 50mm x 
75mm timber are used, or sometimes even 
lighter material, and this is certainly strong 
enough, although if it is envisaged that the walls 
will be used for supporting large amounts of 
shelving, or heavy cupboards, 5 0mm x 100mm 
frames might be preferable. The basic structure 



can be arranged as a series of modules compris¬ 
ing a floor unit, two sides and two ends which 
are bolted together using coach bolts. Simple, 
bolt-together roof frames, one for each end and 
intermediate ones at one metre intervals can be 
used to support the roof, which can conven¬ 
iently be covered with pre-felted chipboard, the 
ridge afterwards being overlaid with roofing 

An examination of a few commercial prod¬ 
ucts will show the general type of construction 
which is adopted, and will indicate where the 
overlaps must be arranged in the external clad¬ 
ding in order to provide a weatherproof struc¬ 
ture. The only likely problems are in the provi¬ 
sion of windows and the door, but commercial 
items can be utilised. The need for ventilation 
and security should not be forgotten, but it is 
not advisable to purchase a door which is too 
heavy to be supported by the framing. 

The workshop must stand on a firm base. 
Ideally, this should be a solid concrete founda¬ 
tion, but it is acceptable if paving slabs are laid 
onto well-consolidated ground to provide the 
base. A floor construction comprising 75mm x 
50mm or 100mm x 50mm timbers, standing 
on edge, with floorboards or chipboard floor¬ 
ing nailed to the timbers will serve adequately, 
but if chipboard flooring is used, it is essential 
that it is a moisture-resistant type. 

External cladding to the walls and ends 
should overlap the flooring, but not reach the 
foundation level, so that good ventilation is 
provided below the floor. All timber used 
should be liberally treated with preservative 
during construction and an external finish 
adopted which will be easy to maintain. 


Provided that the workshop is well insulated 
(even the garden shed type), the provision of 
background heating need not be too costly, and 
a thermostatically controlled fan heater, for 

example, set to maintain the temperature at 
10°C will provide the most useful form of over¬ 
all heating as well as being capable of bringing 
the temperature up to a comfortable level when 
the ‘shop’ is occupied. The heat losses through 
the structure can easily be calculated, once its 
size, and the type and thickness of insulation 
are known, and the local library will provide a 
ready source of reference material to enable 
this to be done. As a guide, the theoretical heat 
loss from my workshop is approximately 700 
watts assuming inside and outside temperatures 
of 10°C and 0°C respectively. 

Nowadays, a wide range of materials is read¬ 
ily available for insulation, and the final choice 
often rests on those materials which are easy to 
use. For walls, expanded polystyrene is one of 
the most convenient, and is also one of the 
cheapest. It is a high-void material comprising 
very small pockets of air entrapped in poly¬ 
styrene. It is made in 25mm and 50mm thick 
sheets, in a grade which is treated with a fire 
retardant and therefore safe to use in buildings. 
It is light in weight, cuts easily and is a very 
good insulator. 

In addition to providing insulation, the 
installation also has to consider the effect of 
moisture in the air. If warm, moist air is in con¬ 
tact with a cold surface, it cools and releases 
some of its moisture and condensation occurs 
at the cold surface. In a modern house, conden¬ 
sation is most frequently seen on single-glazed 
windows, but seldom on walls, since their 
cavity construction is designed to mitigate the 
condensation problem. If the interior of a 
workshop is to be heated, the possibility of con¬ 
densation occurring must be considered. 

Figure 4.1 shows the problem diagrammati- 
cally. Warm, moisture-laden air on the inside of 
the structure is insulated from the cold exterior 
by insulating material and the external wall 
material. However, there is a temperature gra¬ 
dient across the insulation, and if moisture can 
pass through this, condensation occurs on the 
internal surface of the wall. This cannot be per- 



mitted, since the insulation will become satu¬ 
rated and degradation of the structure will 
occur. The answer is to install a vapour (or 
moisture) barrier, the most common material 
for which is polythene sheet. This must be posi¬ 
tioned between the insulation and the warm 
interior in order to prevent water vapour com¬ 
ing into contact with the cold external wall. 

One important point to remember; the 
vapour barrier can only be effective if it is com¬ 
pletely impervious. This means that care has to 
be taken to ensure that no tears or cuts are made 
in the polythene and if joints need to be made, 
they should have adequate overlaps and be dou¬ 
bled over and sealed with an impervious tape. 

The use of a vapour or moisture barrier 
between the moist room air and the cold wall of 


Exterior wall 



Condensation occurs here 


Exterior wall 



Figure 4.1 The need for a moisture barrier in an insulated 

the structure means that a wall covering of plas¬ 
terboard or insulation board is required. Plas¬ 
terboard is preferable since it presents a harder 
and more durable surface and is more readily 
decorated than insulation board which has a 
soft and very absorbent finish. If the vertical 
timbers are positioned to correspond with the 
joints between the sheets of board, it is a simple 
matter to cut the sheets to the correct height 
and nail them to the timber. It is customary to 
nail plasterboard into position with a Viin. to 
%in. (12mm to 20mm) gap between the floor 
and the lower edge of the sheet. This is to pre¬ 
vent the absorption of liquid which might oth¬ 
erwise degrade the plaster core, should the 
floor be washed or any liquid be spilled, and 
the same consideration applies also to the use 
of insulation board. The gap must be allowed 
for when cutting the sheets and is readily 
arranged when fixing a board by use of suitable 
packing to stand the board on when it is being 
nailed into position. 

Several methods can be adopted for conceal¬ 
ing the gaps between the boards. A paper strip, 
with central or longitudinal corrugations, or 
ribbing, can be stuck over the joints with heavy- 
duty paperhangers’ paste - this is perhaps 
the old-fashioned way. Alternatively, a non- 
corrugated strip can be ‘plastered’ over the 
joint with a shallow wedge of board-finish plas¬ 
ter applied to each side thus making a wide 
ridge which is virtually invisible after painting. 
A variation of this method is to use a ceiling 
texture material such as Artex to form a 
feather-edge strip about 12in. (300mm) wide 
spanning the joint between two plasterboard 
sheets. The jointing should be reinforced using 
a 4in. (100mm) wide cloth scrim which is first 
applied to a thin coat of Artex which spans the 
joint, the whole being afterwards covered with 
the feathered strip. This is the method used for 
joints in ceiling board and this, or the plain 
paper strip, is also used in houses which have 
what are called ‘dry liner’ wall finishes. 

Once the plasterboard has been installed, 



and the joints made good, a standard skirting 
can be nailed into position to protea the lower 
edge of the board and close the gap between 
board and floor. 

The positions of the timber studs to which 
the plasterboard is fixed should be evident, or 
have been marked, and elearical outlets can 
therefore readily be installed for the machine 
tools, and for a wall-mounted fan heater. This 
can conveniently be of the type which is fitted 
with a thermostat to allow seleaion of the set 
temperature, and while this may seem extrava¬ 
gant, it certainly provides a cosy working envi¬ 
ronment. Good insulation allows reasonable 
temperatures to be maintained at not too great 
a cost, both for storage and use, and the insu- 
lated-and-heated environment affords protec¬ 
tion to what are becoming valuable machine 
tools and models. 

Garage workshop 

Isolation of the workshop 

Unless a garage can be dedicated exclusively as 
a workshop, it does not come very high up the 
scale of desirable locations. Firstly, it may very 
well be damp, due to the lack of a damp-proof 
membrane and damp course in its floor struc¬ 
ture, but even if this has been attended to, 
which is unlikely, the regular presence of a hot 
and damp vehicle is sure to create a humid 
atmosphere which rapidly promotes rusting. 
An unmodified garage cannot be recommended 
as an ideal location in which to store ferrous 
parts, and this includes many types of model, 
most machinery and much stock material. 

Unless the machinery you have, or intend to 
purchase, is exceptionally large, a garage also 
represents a very large space and a better envi¬ 
ronment may well be more readily provided by 
partitioning off one end to create an isolated 
space. The partition between garage and work¬ 

shop can be construaed by using a timber frame 
of 50mm x 75mm or 50mm X 100mm pre¬ 
pared timber incorporating a standard commer¬ 
cial frame and door, if one is required. Garage 
side cladding may be by nailing on a thin 
plywood cladding, using an exterior (water¬ 
proof) grade if dampness is likely to be a prob¬ 
lem. It is a good idea, in any event, to use a damp- 
proof material below the timber frame and a 
standard roll of damp proof course (DPC) felt 
can be used for this. 

Some form of heating for the workshop is 
thoroughly recommended and this becomes 
more effective (and cheaper to run) if the walls, 
floor and ceiling are insulated. Expanded poly¬ 
styrene, as recommended for the garden shed 
workshop, should be used for the isolating par¬ 
tition, and also for the walls, if possible, and the 
installation of a ceiling, with a glass fibre or 
mineral wool blanket above, is also advanta¬ 
geous in reducing the heating costs. For the 
walls and isolating partition, a moisture barrier 
must be installed, and a plasterboard lining, as 
described above will complete the installation, 
as for the shed workshop. Provided that there is 
a large air space above the ceiling, with good air 
circulation, condensation will not be a problem 
in the roof space. 

Floor insulation 

If a workshop can be provided by segregating 
part of a garage, some floor treatment is desir¬ 
able since a garage floor does not normally have 
a damp-proof course. The choice lies between a 
professionally laid damp-proof membrane, 
with concrete screed floor, or a floor laid on 
wooden joists resting on the existing concrete, 
with a damp-proof course between. A construc¬ 
tion of the type shown in Figure 4.2 is suitable, 
in which 50mm x 75mm joists are used, with 
the gaps between the joists filled with mineral 
wool or glass fibre insulation. This type of con¬ 
struction loses relatively little heat in compari- 



son with a concrete floor (in a ratio of about 
1:4). It therefore assists retention of heat and 
reduces heating costs. 

Two moisture barriers are required - one to 
prevent the ground-borne moisture from per¬ 
meating the insulation (and the supporting 
timbers) and another to prevent water vapour 
in the room air from penetrating the structure 
and causing condensation on cooling at the 
cold concrete surface. However, the timber and 
insulation are not sealed into a cocoon since the 
upper and lower polythene sheets are not 
sealed at the edges. The lower sheet is laid on 
the floor and carefully folded at the corners to 
lie up the wall a few inches above the damp- 
proof course. Prior to laying the polythene, 
strips of conventional DPC must be placed on 
the oversite concrete, below where the joists 
will lie, to prevent the polythene being punc¬ 
tured by the rough finish. 

Once the joists and insulation are in posi¬ 
tion, the upper moisture barrier is positioned in 
the same way, to lie up the walls, but is not 
sealed to the lower layer. The below-floor 
space can therefore ‘breathe’. Both moisture 
barriers must extend above the DPC to prevent 
moisture entering the enclosed space. 

The type of floor described is frequently rec¬ 
ommended where good insulation is needed, 
not only with false joists, as described, but also 
with floors having expanded polystyrene as the 
insulation, either screeded over with concrete 
or covered with flooring-grade chipboard. Both 
surface treatments act to spread the load 
applied to the floor so that it may be supported 
by the polystyrene, but the construction is suit¬ 

able only for relatively light loads. Although 
polystyrene is normally impervious to moisture, 
the fact that it must be laid in relatively small 
sheets means that it does not provide an unbro¬ 
ken moisture barrier, and two polythene sheets 
are recommended for use in floors. 

Workshop electricity 

The workshop will need several electrical out¬ 
lets (sockets) so that the machine tools, lamps, 
soldering irons, and so on, can be connected to 
a source of power. The electrical supply avail¬ 
able in the home is what is known as a single¬ 
phase supply, having just live and neutral con¬ 
nections, with an earth. To match this supply, 
the machine tools which are bought should be 
fitted with single-phase motors. This makes 
them suitable to be plugged into a normal 13- 
amp outlet. Lathes, milling machines and mill/ 
drills of light construction, or those intended 
for amateur use, are likely to be fitted with sin¬ 
gle-phase motors. 

If larger machine tools are purchased, or 
those intended for industrial use, they may be 
fitted with three-phase motors which cannot be 
connected to a domestic 13-amp outlet. In these 
cases, a special three-phase supply will need to 
be wired into the workshop, or an adaptor pur¬ 
chased which produces a three-phase supply (or 
its equivalent) from a domestic supply. An adap¬ 
tor is the cheaper alternative to the provision of 
a three-phase supply, since the connection of a 
three-phase supply requires a new electrical 

20mm T & G chipboard flooring Figure 4.2 A suggested method 

115mm wide DPC below 
each joist 

In-fill of 100mm mineral wool 
or glass fibre 



feed into the property from the electricity com¬ 
pany’s cables in the locality. 

If the workshop is established in a purpose- 
built location, or is an adaptation of part of a 
garage, for example, it may require a com¬ 
pletely new electrical installation. Any electrical 
work, even minor modifications, should only be 
carried out by a competent electrician. It is 
important that all of the tools are correctly 
installed, from an electrical point of view, and 
are correctly earthed. Only a competent person 
can ensure that this has been done, and only a 
person with the appropriate knowledge and test 
instruments can check an installation to con¬ 
firm that it conforms to the regulations and is 
safe to operate. It should be particularly noted 
that the earthing requirements for a garden shed 
workshop may require special consideration 
due to the distance from the house, and this 
again dictates that the installation is inspected 
professionally, prior to use. 

Fitting out 


Like a good kitchen, a good workshop needs 
plenty of work surface, plenty of storage and 
somewhere to sit down to drink a cup of coffee. 
The work surfaces are generally the benches, 
the storage spaces are boxes, cupboards and 
shelves, and the place to sit for a cup of coffee 
might also equate with another bench. Benches 
are thus required in profusion and the follow¬ 
ing specific areas are probably desirable, if you 
can find the space: 

(1) a dirty bench, very sturdy, and fitted with a 
large vice. This is where sawing and filing 
processes are performed. 

(2) a fairly clean bench which can be used for 
fine sawing and filing, polishing and gen¬ 
eral finishing off. 

(3) a clean assembly area, remote from sawing. 

filing and machining. 

(4) a storage bench for items under construc¬ 
tion. This might need to be large if a large- 
scale tender locomotive is being built. 
Needless to say, few workshops will have the 
space for all of these separate areas and some 
compromises must inevitably be made by com¬ 
bining functions. Unless very large models are 
being built, assembly work generally amounts 
to the equivalent of ‘instrument making’ in in¬ 
dustry. Many people find that this is most con¬ 
veniently undertaken in a sitting position, and 
this also applies to the finer sawing, filing and 
polishing activities, so a desk-type bench can 
conveniently satisfy the need for both the clean 
and fairly clean activities provided that they 
can be separated in time. A small clamp-on vice 
and clamp-on vee-notch table for the piercing 
saw (see Chapter 6) allow an ordinary table or a 
secondhand desk (provided it is rigid enough) 
to be used for smaller work of this type. 

For heavier sawing and filing, it is generally 
recommended that the top of the vice should be 
at elbow height, so that the forearm is horizon¬ 
tal during these activities, therefore it is a good 
idea to obtain a vice (or at least measure the 
height) before designing and building the 
heavyweight bench. For a work surface at table 
height, where only light work will be per¬ 
formed, a comfortable height for writing is 
probably the major consideration, but if you 
have particular eyesight problems, you may 
need to consider these. The height required thus 
depends upon you and the chair you will use. 

Since drilling produces volumes of swarf, 
the bench to which the vice is attached might 
also accommodate the drilling machine if it is 
of the bench-mounting type, thus combining 
two of the essentially dirty processes. 

Unless the lathe is provided with its own 
stand, it too will need a bench. This must be 
rigid and substantially built in order that the ma¬ 
chine can be set up correctly and maintain its ac¬ 
curacy. Naturally, as a swarf producer, the lathe 
needs to be located at the dirty end of the shop. 



The bench vice 

The main vice in the workshop should be the 
largest you can afford or accommodate, and it 
must be mounted to a substantial bench. A good 
quality, 4-inch (100mm) vice will generally be 
sufficient for the general run of modelling 
work, something smaller being provided for use 
at the table bench, in the form of a clamp-on 
vice, or by making or purchasing a small-piece 
vice for clamping into the bench vice. 

The important feature of all vices is that they 
should have smooth jaws which remain parallel 
throughout their gripping range. Most bench 
vices are manufactured with hardened, serrated 
jaws, but unless you are interested in gripping 
only full-size, agricultural items, such jaws 
are useless. They naturally grip well, but only by 
making impressions of their jaw serrations in 
the work, and this is definitely not required in 
the modeller’s workshop. Serrated jaws should 
therefore be removed and replaced by home¬ 
made versions made from ordinary mild steel. 

An alternative to replacing the jaws is to 
purchase or make up slip-on soft jaws. These 
comprise steel pressings to which are riveted 
strips of soft, composition material, making the 
jaws suitable for holding soft materials without 
marking them. Something of this sort, or 
loosely inserted pieces of soft packing, is fre¬ 
quently required when working with soft 
materials, since even mild steel jaws will mark 
or squash soft materials such as copper, brass, 
nickel silver and aluminium alloys. Soft jaws of 
the type described are not good for holding very 
small items, since they compress due to the lo¬ 
calised loads. 

A clamp-on vice for the table bench does not 
need to be larger than a 2-inch (50mm) jaw. 
Some vices of this type are provided with bolt- 
on jaws, frequently soft plastic mouldings, but 
unless you envisage a great deal of work in soft 
materials, or with work that is very easily 
damaged, such as printed circuit boards, it is 
better to use the type without removable jaws, 

Figure 4.3 A small vice, with integral clamp, fitted to my 
second bench. 

such as that shown in Figure 4.3, since very soft 
jaws are easily damaged by gripping small 

Seeing and looking 

Good lighting is necessary to most of the opera¬ 
tions carried out in the workshop. It is vital for 
all machine work, since there is a repeated need 
to bring the cutter up to the work to detect the 
point at which it first starts to cut. When cut¬ 
ting or filing to a line, good lighting is essential. 
It is vital too, during marking out or any opera¬ 
tion which requires that a rule or vernier be 
read to the smallest divisions on the scale. 

Since the locations of these different opera¬ 
tions are likely to be different, it follows that 
several small sources of light are required. 
These must be movable so that the light can be 
brought to bear from different angles and are 
ideally provided by the common type of coun¬ 
terbalanced reading lamp. The only disadvan¬ 
tage of these lamps is the large size of reflector 
fitted to those types which use conventional, 
tungsten-filament bulbs since they tend not to 
be very convenient when used on the lathe. The 
type of lamp which uses a low-voltage tung¬ 
sten-halogen light source is preferable, since 



this offers a significant reduction in the size of 
the bulb, and hence, the size of the reflector. 
There is a cost penalty, however, and this may 
dictate the choice. 

Once the lighting is installed, attention can 
be turned to the problem of seeing. It may be 
quite impossible to separate the adjacent 
engravings on a rule with the naked eye, and so 
some assistance is necessary in this direction. 
The first tradesmen to meet the need for such 
assistance were the watch, clock and instrument 
makers involved in the development of ever- 
smaller precision mechanisms. These craftsmen 
were doubtless associated with early optical 
experiments and were thus quick to understand 
the usefulness of a magnifying glass. 

However, the ‘Sherlock Holmes’ type of 
magnifying glass was impossibly large to cast, 
grind and polish and early lenses were conse¬ 
quently very small. The habit grew of using 
these small magnifiers close to the eye, and they 
were mounted into holders small enough to be 
held in the eye socket and were known as 

The type of lens used is the same as those 
used by schoolboys for focusing the rays of the 
sun onto a piece of paper and burning a hole in 
it. The lens is a bi-convex, and it brings the 
sun’s image to a sharply defined spot on the 
paper when it is placed the correct distance 
away. This distance is known as the focal 
length. The magnification which is provided 
when a bi-convex lens is placed between the 
eye and an object is related to this measure¬ 
ment, and the lenses are described by their focal 

The shortest focal length gives the greatest 
magnification and the range of lengths sold is 
from IV 2 inches (38mm) to 4‘/2 inches 
(115mm). A 4Vi inch glass does little more than 
concentrate the view, but what it does do is to 
allow the eye to focus on something roughly 4 
inches away (100mm), or 3 inches (75mm) in 
front of the lens. A 1 Vi inch glass allows the eye 

Figure 4.4 Watchmakers' eye glasses. 

to approach to within about 3 inches (the 
object is 1 Vi inches in front of the lens) but has 
the disadvantage that the field of view is nar¬ 
row and only a limited range of objects is in 

The glasses are sometimes described by the 
nominal magnification which is provided, a 4Vi 
inch giving 2 times, a 3 inch giving 3.5 times 
and a 1 Vi inch giving 6 or 7 times. A 3 times is 
a useful size for general use when marking 
out and centre punching and the only other 
consideration is whether you can hold it con¬ 
veniently in the eye socket. A AVi inch and a 6X 
(described as a 42mm) are shown in Figure 4.4 
together with one type which clips on to specta¬ 
cle frames. The two glasses are significantly dif¬ 
ferent in size and this might be the determining 
factor when making a purchase. 

There are several types which fit onto specta¬ 
cle frames. The one shown in Figure 4.4 can be 
swung out of the way when not required, but 
others may not incorporate this feature - which 
type to buy depends on how you envisage its use. 

If relatively long-term use of a glass is 
needed, for example when painting small, intri¬ 
cate work, a binocular magnifier which fits to a 
headband might be preferable. These are again 
available giving different magnifications and 
are essentially designed to work at different 
ranges, from something like 1 Vi times magnifi¬ 
cation at 20 inches (500mm) to 3 Vi times at 4 
inches (100mm). A compromise of something 



like 2 or 2Vi times working at 8 to 10 inches 
(200-250mm) is correct for an initial trial. 

A quick and simple way to check the focal 
length when purchasing any magnifier is to 
project a picture of a nearby window onto a 
sheet of paper. If you retreat a little from the 
window and hold the lens up, with the paper 
behind, facing the window, and vary the dis¬ 
tance from the lens to the paper, you will see an 
image of the window appear on the paper. 
When this is sharply focused, the lens is 
approximately at its focal length from the 
paper. If a sheet of paper is not available, it is 
quite possible to project onto your hand, and if 
you are out of doors, then you can focus the 
sun, as the schoolboys do, but in this case, use 
someone else’s hand! Indoors, without a bright 
window to use, the image of a light can be 
focused in the same way, but a reasonably dis¬ 
tant one needs to be used or the result is not 
very accurate. 


General considerations 

In comparing the requirements of the work¬ 
shop with those of the kitchen, the general suit¬ 
ability of secondhand kitchen units should be 

Figure 4.S The traditional type of toolmaker's cabinet. 

borne in mind. While base units are hardly 
sufficiently robust to make an ideal bench for 
sawing and filing, they are provided with sub¬ 
stantial chipboard work surfaces which are eas¬ 
ily cleaned, and have drawer units, shelves and 
so on that provide under-cover storage which is 
readily organised to suit the modeller. Wall 
cupboards and midway shelf units might also 
find ready uses. 

One advantage of being able to utilise 
kitchen units is that much storage will be under 
cover and therefore protected from dust, swarf 
and general dirt which is apt to accumulate. 
This is immensely preferable with one excep¬ 
tion - ventilation is a problem inside a closed 
cupboard, and this can lead to corrosion on 
iron and steel due to stagnation of moisture¬ 
laden air in proximity to cold metal. This is a 
serious problem, even inside a centrally heated 
house, and it can also affect cutting tools such 
as taps, dies and reamers if they are left for long 
periods in closed boxes, without being dis¬ 
turbed. So, you need to open the cupboards 
regularly, and perhaps even move the stock 
about. As far as tools are concerned, the solu¬ 
tion is obvious - just get out to the workshop 
and use them! 

Tool storage 

Tools divide broadly into the large-and-dirty 
and small-and-handy categories and their stor¬ 
age can be considered in these two ways. The 
larger tools are those associated with sawing 
and filing. In a small workshop, nothing is ever 
likely to be far from where it is needed, but it is 
naturally convenient if saws and files are 
located adjacent to the bench vice. The dirty 
bench should incorporate drawers to hold 
these, together with hammers and mallets. 
Other large hand tools such as pliers, pincers, 
tin snips and screwdrivers can also be located 
here. If drawer space is at a premium, a rack on 



the end of the bench, or on the wall behind, 
might be preferred. 

The traditional storage for a craftsman’s 
small tools is the toolmakers’ cabinet, one type 
of which is illustrated in Figure 4.5. These cabi¬ 
nets provide one or two long drawers and six or 
eight small ones, according to the design, and 
some, like the one illustrated, have a top sec¬ 
tion covered by a lift-up lid. They are designed, 
and therefore ideal, for the storage of small 
tools which might be used at the assembly or 
fitting stages, and can store the smaller taps, 
dies, reamers, tap wrenches, spanners, pliers 
and cutters which are used at those stages. If a 
shallow, long drawer is provided, this is ideal 
for the storage of rules, protractors, a depth 
rule or perhaps combination square and vernier 

When determining methods of storage, the 
other scarce commodity (apart from space and 
money) should also be borne in mind - time. 
There is never enough of this, so that which is 
available must be maximised. One way to 
achieve this is to reduce searching time by stor¬ 
ing small items so that they are immediately 
available and identifiable. The storage of drills 
is a case in point. It is all very well having a 
large box full of drills covering the range from 
'Ain. to Viin. (6mm to 12mm), but quite a dif¬ 
ferent matter to find the one to drill a hole 
.256in. or 6.5mm in diameter. The selection 
process can be long and tedious, measuring 
drills which look right, repeatedly, with a 
micrometer until the particular one is found. 
Drills are much better stored in drill stands or 
boxes of the types illustrated in Figure 4.6, 
although the open stands are dust collectors if 
they are stored in the open. 

A similar argument can be applied to small 
taps. A drawer full of BA taps, which are in any 
event small, might contain over 30 taps if you 
have three of each of the common sizes (the 
normal set). The 0 and 2 sizes are readily distin¬ 
guishable, perhaps also 10 and 12, but if the 

need is for a particular type of size 8, several 6s, 
7s and other 8s are sure to be examined before 
the one required is found. 

One way to overcome this is to make up 
drilled blocks designed to accept the three taps 
in a set, perhaps also with the tapping and 
clearing drills for the particular size. These 
blocks are pleasant to handle if made up in hard 
wood, but oak should not be used since it 
causes corrosion. Depending upon the shelf or 
cupboard storage which is available, larger 
blocks holding several sizes might be more con¬ 
venient. This type of storage is not required for 
dies since, in any event, there are fewer of them 
for a given thread form (only one of each size) 
and they are generally more readily identi¬ 

Stock storage 

Everyone has his or her own ideas about how 
much stock to lay in and how best to store it. 
However, one soon begins to build up a large 



amount, and its storage can become a problem. 
One way to solve this is to store rod and bar 
materials in lengths such that they can be 
accommodated within the depth of a standard 
bench and to build into one of them a long 
pigeon-hole rack. My rack has holes roughly 
2ft (600mm) long and 2in. x 2in. (50mm x 
50mm)or2in. x 4in. (50mm x 100mm) cross- 
section. Accordingly, the rack holds the stand¬ 
ard 2ft (600mm) lengths which are sold by 
most modellers’ suppliers, and it is only the 
occasional longer lengths of bar which cannot 
be accommodated. These stand in some con¬ 
venient corner until actually needed. 

A simple solution to the problem is to utilise 
square-section rainwater downpipe, as some 
retailers do for their exhibition stands. If you 
are a woodworker, and have some suitable ply¬ 
wood available, a rack can easily be con¬ 

The pigeon hole or tube arrangement works 
very well provided that the lengths of material 
are reasonably long, but short ends and odd 
stubs of material tend to get lost among the 
long pieces and inevitably get pushed to the 
back. A box in which to store these oddments is 
a useful adjunct to the main store and can possi¬ 
bly equate with the ‘scrapbox’. 

Storage of sheet materials is not so likely to 
require much space, since a large number of 
different types is not required and most sheet is 
purchased for specific jobs. A single shelf in a 
cupboard, or a stand-up slot in a bench might 
conveniently hold sheets up to 2ft (600mm) 
square and quite a lot can be packed into a 
space 9in. (230mm) high, or wide, depending 
on whether the sheets lie down or stand verti¬ 
cally. Once again, smaller offcuts tend to get 
overlooked among the larger pieces, and an 
oddment box or subsidiary store is very useful. 

Any castings which have been purchased for 
a particular model can be stored in suitable 
boxes, but it is preferable if these can be closed 
up so that the contents remain clean and dust 

free. Cardboard boxes are perfectly satisfactory 
for this, provided there is not a need to carry 
them about frequently and nothing elaborate 
(or expensive) needs to be arranged. The boxes 
do need to be fairly strong since the wheel and 
cylinder castings, even for a modest 3 Vi-inch 
gauge locomotive amount to quite a weight. 

Storage of cutting tools 

Many of the items in the workshop are 
designed for cutting, a process which can only 
be performed satisfactorily by a sharp tool. 
Tools which become blunt must therefore be 
resharpened, or be replaced if sharpening is not 
possible. Cutting tools naturally wear away and 
become blunt through use, but they can also be 
blunted by careless storage. If all of your BA 
taps are kept in one box and carelessly taken 
out or dropped back in, or the box is handled 
so that the taps rattle against one another, they 
become blunted even though they are not in 
use. This can seriously shorten a tap’s life, 
putting more pressure on that rare commodity, 
money. If loose storage is more convenient, 
then the taps can be protected by slipping on 
short lengths of plastic tubing. 

Reamers, end mills and slot drills, which 
have cutting edges all along their working 
lengths, are especially prone to damage if care¬ 
lessly stored, and are best kept in compart- 
mented wooden trays or drawers so that they 
cannot damage one another. Many of these 
items are nowadays supplied in plastic boxes, 
and if bought new, will have ready-made pro¬ 

Some cutters are supplied with protection in 
the form of wax-dip sleeves. These can be 
removed by cutting carefully down the length 
of a flute with a sharp knife and removing the 
sleeve by a combination of peeling and 
unscrewing. The sleeve can usually be removed 
intact by this method and can be replaced on 



the cutter by reversing the action. Since the cut¬ 
ters are coated with oil before immersion in the 
wax, the retention of the sleeve is also useful as 
a rust-prevention measure. If the sleeves 
become damaged, plastic tube can be used as an 

Storage of small items 

For the storage of small items such as nuts, 
bolts, washers, rivets and so on, an extensive 
collection of small boxes is required. From the 
point of view of time saving, it is vital to be able 
to access a given size of nut or bolt, or what¬ 
ever, rather than having to scrape about in a 
box which contains all sizes. It pays, therefore, 
to segregate different sizes into their own boxes 
and it might even be beneficial to store nuts, 
bolts and washers separately, for a given size, or 
to segregate hexagon-headed bolts and screws 
from other types. An extensive collection of 
small boxes is therefore required. 

Traditionally, small-item storage has been in 
tobacco tins, but with the reduction in the 
number of smokers and the use of plastic pack¬ 
aging, this source of small boxes has almost 
dried up. However, if you know a smoker, (or 
an ex-smoker), particularly if he favours (or 
favoured) a pipe, cultivate him. There may be a 
stock of tins which is no longer required and 
they can readily be labelled and will stack 

Failing the availability of tobacco tins, the 
modern equivalent has to be the plastic box. 
You may be lucky and be able to locate a source 
of supply, since much packaging is nowadays 
simply discarded, but failing that, an extremely 
useful range of interlocking boxes is available 
from suppliers of electronic components. The 
system is based on a standard module compris¬ 
ing a plastic sleeve, about 2in. x 2in. (50mm x 
50mm) in cross-section, into which is fitted a 
clear plastic drawer. The sleeves are moulded 

with external dovetails so that they can be 
interlocked to form a nest of units and they 
may be wall-hung by using keyhole slots 
moulded into the outer sleeves. 

The range of drawers uses one-module and 
two-module wide drawers, but other sizes are 
also available from some sources, for example, 
2 wide X 2 high and 4 wide X 2 high. While 
not perhaps the cheapest available, there is 
great merit in being able to rearrange the unit, 
or add to it as time goes on, and the ability to 
hang the sleeves on the wall is also a useful 
alternative to providing shelves. 

Hand tools 

General tools 

Many of the hand tools used by the modeller 
are not different in any way from those used 
ordinarily and anyone interested in things prac¬ 
tical is sure to have available pliers, screwdriv¬ 
ers, hammers and the like, perhaps also wire 
cutters, soldering irons, some drill bits and a 
hand drill to use them in. With the general 
availability of power tools these days, a DIY 
electrically powered drill is almost certain to be 
available also. 

What may not be available are the tools 
needed for small-scale modelling; small screw¬ 
drivers, usually called watchmakers’ screwdriv¬ 
ers, smaller spanners, especially for the BA sizes 
of threads, including the odd numbers, which 
are quite widely used in modelling, and some of 
the more specialised items such as pin chucks 
and pin vices, centre punches and riveting 
tools. Taps and dies for the common threads 
used in model engineering may also need to be 
obtained, together with the relevant tapping 
and clearing drills. Precision measuring equip¬ 
ment will also most likely be needed. 

The more specialised items which are 



required are described in succeeding chapters. 
Use of the important hand tools for marking 
out, sawing, filing and cleaning up or polishing 
is described in Chapter 6 and the more usual 
devices for performing the important measur¬ 
ing functions are described in Chapter 5. Hole 
drilling is described in Chapter 10 and the pro¬ 
duction of screw threads using hand methods, 
in Chapter 11. Riveting, and the tools needed, 
are described in Chapter 8. 

Soft-faced hammers 

One operation which occurs very frequently is 
the mounting of work into holding devices 
associated with machine tools. These are prin¬ 
cipally machine vices and workholding chucks 
for the lathe. When placing work in these 
devices, it is usually essential that squareness is 
achieved, and that the work ‘seats’ correctly 
against the relevant reference faces, when this 
is appropriate. Correct seating is achieved by 
partially tightening the workholding clamp(s) 
and then tapping the work down onto the seat¬ 
ing face. 

One way to do this without marking the 
work is to use a normal (hard-faced) hammer 
but to interpose a soft block between it and the 
work. It is more convenient to use a soft-faced 
hammer, however, and two or three of these 
useful tools should be included in the tool kit. 
Several types (and different weights) are avail¬ 
able, and the governing factor in making the 
choice is the size of the machinery and the 
likely size of the work. Figure 4.7 shows three 
types; a type with fairly-hard, screw-in plastic 
heads, a type with an all-rubber head and a 
third variant in which a metallic head is pro¬ 
vided with plastic inserts. Types having screw- 
in heads are also available, which offer a choice 
of head shape, but they offer little advantage if 
the need is simply to seat work, as described. 
Shaped soft heads are supposed to be useful 

Figure 4.7 Three types of soft-face hammer. 

should panel beating be required but if there is 
a need to deform metal, there is no substitute 
for a hard-faced hammer and a good solid 
block to support the work. 


There is a frequent need for clamping things 
together for trial assembly, for brazing or sol¬ 
dering together or for drilling or machining 
two or more components simultaneously. For 
soldering or brazing, it is worthwhile making 
up some clamps especially for the task since 
they will inevitably be heated during these 
operations, perhaps to high temperatures, and 
are sure to become contaminated by flux, and 
will ultimately corrode. 

The most useful form of clamp for other 
purposes is that known as a toolmaker’s clamp, 
examples of which are illustrated in Figure 4.8. 
Each clamp comprises two similar jaws, united 
by opposing screws, one pulling the front of the 
jaws together, and the other pushing the rear 
ends apart. It is possible to maintain the jaws 
parallel in use, and the relatively large, flat jaws 



Figure 4.8 Some small toolmakers' clamps. The largest 
shown is a 2'/S inch (65mm). 

do not mark the work, even though good 
clamping pressure is applied. 

Initial adjustment of the gape of the jaws is 
achieved by grasping both screw heads and 
winding the clamp in the appropriate direction 
to open or close the jaws. To assist a smooth 
adjustment during this operation, the head of 
the front screw is sometimes grooved to receive 
a cranked retaining strip which is secured by 
screws to one jaw. This ensures that the front 
screw takes this jaw with it when being un¬ 
screwed from the other jaw. 

Figure 4.9 A toolmaker's clamp on a length of steel bar. The 
rear screw needs to be tightened to bring the jaws to parallelism. 

Final adjustment to parallelism on the work 
must be by adjusting both screws, but is best 
achieved by adjusting the front screw until the 
jaw gape is too large at the front, as shown in 
Figure 4.9. This allows the clamp and work to 
be positioned correctly, after which the rear 
screw can be tightened to apply pressure and 
achieve parallelism, this being tested by rotat¬ 
ing the clamp slightly to check whether it is 
gripping only at the ‘toe’ or ‘heel’ or all along 
the jaw. 

A useful exercise is to make up a pair of 
clamps and Figure 4.10 shows a drawing of the 
parts and gives a range of suggested sizes. The 
retaining strip for the front screw is by no 
means essential and the clamps can be made as 
shown in Figure 4.10 without any real disad¬ 
vantage, except for an occasional tendency for 
the jaw to jam during initial adjustment. 

The material to use for the clamps is bright 
mild steel (BMS). The jaws can be shaped by 
hand and after completion should be polished 
with fine grade emery cloth and afterwards 
case hardened, as described in Chapter 3. After 
hardening and quenching, the bright finish can 
be restored by cleaning off the scale and re- 
polishing with emery cloth. 

The screws are a straightforward turning 
and threading exercise. If you don’t have a set 
of knurls, simply cross drill the screw ends so 
that a tommy bar can be used for tightening 
and releasing. This might be done anyway, fol¬ 
lowing commercial practice. 

Scribers, squares and protractors 

A preliminary to almost all work, even the sim¬ 
plest of components, is the marking-out stage. 
A description of this process is contained in 
Chapter 5, but the subjects not covered there 
include a consideration of the marking-out tool 
(the scriber) itself, and the means for establish¬ 
ing lines on the work at the relevant angles, this 






2/. (58) 

3 (75) 

3K (95) 

4M (115) 


* (3) 


'/. (6) 

y. (8) 

M (9.5) 


V. (19) 

1 A (30) 

1 A (40) 

\’A (50) 

2% (57.5) 


'/. (6) 

A (10) 

A (12.5) 

M (16) 

M (19) 


'/« (6) 

* (19) 

1 (25) 

1* (32) 

1K (40) 



Mr (2.5) 

A (3) 

Mi (4) 



W (10) 

VI. (15) 

K (19) 

'M. (24) 

\A (30) 














Figure 4.10 Drawing for toolmakers' clamps up to 4'/i inch 

function being performed by squares or pro¬ 

A scriber has a hardened, sharp point which 
is used to scratch the work to establish a refer¬ 
ence line or position. For accurate marking out, 

the scriber must have a fine point and this is 
usually ground on a finely tapering end. Figure 

4.11 shows scribers of two basic sorts, an engi¬ 
neers’ or machinists’ type comprising a rod 
with centralised, knurled grip, and a pocket 
scriber which comprises a body having a collet 
fitting which accepts the scriber point either 
way round. A clip fitted to the body creates a 
pocketable scriber. Provided that the point is 
fine, there is no functional difference between 
the two types. 

Although usually just called a square, the 
correct term for the tool for establishing the 
90-degrees, or square, condition, is an engi¬ 
neer’s try square. A square comprises a stock 
into which is set a thinner blade, centrally at 
one end. The completed square usually has a 
ground finish and both edges of the blade are 
nominally at 90 degrees to the stock. 

Squares are available in different sizes, speci¬ 
fied by the length of the blade, in inches. Figure 

4.12 shows a 6-inch and a 3-inch. Most needs 
are satisfied by a 6-inch and something smaller, 
say a 2-inch or 3-inch, but occasional access to 
something larger can be helpful. Failing the 
availability of a larger square, a straight length 
of stock rectangular bar can be clamped to the 
work, and its squareness achieved by using a 



Figure 4.12 3 inch and 6 inch try squares. 

Figure 4.13 The components of a combination set, and the 
protractor mounted on to the rule. 

smaller square. A 6-inch square is just about 
adequate to set up a 24in. (600mm) bar and 
this method will serve for the odd occasions 
when something larger is required. 

A longer reach is frequently provided by a 
combination square or combination set, the 
foundation for which is usually a 10-inch or 
12-inch (300mm) rigid rule. The rule is 
grooved on one side so that sliding heads can be 
located and locked on it. The available heads 
are of three types. A square head has faces 
which lie at 45 and 90 degrees to the edge of 
the rule and thus forms a 12-inch square or a 
similarly sized, 45-degree bevel. A protractor 
head allows any angle to be set against an angu¬ 
lar scale which is typically graduated every 

The third type of head is known as a centre 
head. It comprises a two-faced stock, the angle 
between the faces being bisected by one edge of 
the rule. If the stock of the centre head is placed 
on a circular workpiece, the rule edge lies on a 
diameter and a line scribed along the rule must 
therefore pass through the centre of the work. 
The intersection of two diametral lines thus de¬ 
fines the actual centre. The components of a 
combination set are shown in Figure 4.13 and 
the centre-marking operation in Figure 4.14. 

Figure 4.14 Centre marking using a centre square. 



Figure 4.15 A simple protractor. 

A good quality combination set might cost 
about £100, of which £50 or so is accounted 
for by the protractor head. If less-frequent 
measurement of angles and bevels is envisaged, 
and such expenditure is not justified, a simple 
protractor of the type illustrated in Figure 4.15 
will be satisfactory. This comprises a slim rule 
held in a pivot on a plate that allows setting to 
1 -degree engravings on an angular scale. One 
advantage of this type is that the rule can be 
used for depth measurements when set to the 
90-degree position. 

Small vices and pin chucks 

For the instrument making kind of work which 
much of model engineering comprises, there is 
a frequent need to hold small items, either tools 
or workpieces. Sooner or later, holding devices 
for these will be required, several of which are 
appropriate to hand use. 

The most frequently needed devices are pin 
chucks and pin vices, intended principally for 
holding small-diameter, circular items. Two 
types of pin chuck are shown in Figure 4.16. 

On the left, a standard type having a knurled 
body and 4-jaw split collet, intended for hold¬ 
ing small drills, up to about .040 in. (1mm) in 
diameter. The body is bored through, however, 
and the chuck will accept long rods, should 
these need to be held. The upper item in Figure 
4.16 is a double-ended pin chuck having a dif¬ 
ferently sized collet at each end. This accepts 
items from virtually zero diameter up to about 
'/sin. (3mm) which is sufficiently large for it to 
be used as a handle for needle files. 

Figure 4.17 shows a pin vice. This comprises 
a small, springy vice jaw unit mounted to a hol¬ 
low stem, the jaws being cut with vee-shaped 
grooves to allow small rods to be gripped 
coaxially with the stem. The otherwise plain 
jaws may naturally be used for holding small, 
rectangular work and the device will thus func¬ 
tion as a simple, hand-held vice. 

A hand vice is usually more substantial than 
the pin vice of Figure 4.17 and may have jaws 

Figure 4.16 Two types of pin chuck. 

Figure 4.17 Apinvice. 



up to l'Ain. (30mm) wide, opening to s /sin. 
(16mm) or so. Both types of hand-held vice are 
useful when working with small parts, but an 
adaptation which can be held in the bench vice 
is also convenient for some work, since it 
allows both hands to be used for filing, allow¬ 
ing more precise control to be exercised. 


The most important of the punches is probably 
the centre punch. The ‘business’ ends of three 
from my stock are shown in Figure 4.18. A cen¬ 
tre punch has a hardened and tempered point, 
usually ground to a 60-degree included angle. 
Centre punches are used for forming dimples in 
the work in positions which define the centres 
of holes. The purpose of the dimple is to allow 
the drill to find the location and commence 
drilling in the correct position. 

A drill does not strictly have a point but ter¬ 
minates instead in a vee-shaped line on the end 
of the web, or central core of the drill. The 
larger the drill, the larger is the core, and the 
larger the centre-punched dimple needs to be. 
This means a wider angle to the punch point 
and a harder blow from the hammer. Unfortu¬ 
nately, a wider point and harder blow are not 
conducive to accurate positioning, so it is useful 
to have two punches available, one for spotting 
the location and a second for widening and 
deepening the dimple. Too sharp a point can¬ 
not be used since it lacks strength and quickly 
breaks down in use. 

Small punches such as those shown can eas¬ 
ily be made up using silver steel rod, the 
pointed end being hardened, and afterwards 
tempered to dark purple, as described in Chap¬ 
ter 3. 

An alternative to the basic punch is the auto¬ 
matic centre punch shown in Figure 4.19. This 
comprises a short, hardened point mounted 
into a holder which incorporates a spring- 

Figure4.l8 A close-up and shadowgraph of three centre 
punch points. 

loaded latch. Pressing the point onto the work 
through the holder, compresses the spring pro¬ 
gressively until the latch releases, suddenly 
applying the pressure directly to the point. A 
knurled cap, or ring, on the holder allows the 
latch release pressure to be adjusted, thus 
allowing a variable blow to be applied to the 
point. Initial spotting of the hole position, 
using only light pressure, is thus possible, fol¬ 
lowed by setting the punch for a heavier blow 
and going round the hole positions for a second 

Centre punch points do become blunted in 
use and they should be examined under a 
watchmakers’ glass from time to time since a 
damaged or worn point makes it difficult to 
position the dimple accurately. 

Figure 4.19 An automatic centre punch. 



Parallel-shank punches, known as pin 
punches, will be needed for inserting and 
removing taper pins or any parallel fixing pins. 
These can be purchased from commercial 
sources but can also readily be made up from 
silver steel, again being hardened and then tem¬ 
pered to dark purple, as described in Chapter 3. 

Silver soldering and brazing equipment 

Gas blowtorches 

There is, strictly, a difference between brazing 
and silver soldering, but the processes are so 
similar that they are frequently just described 
under the one heading of brazing. Both proc¬ 
esses require the work to be heated to 600- 
800°C which is a ‘comfortable’ red heat and it 
is essential to have appropriate and adequate 
equipment available. 

The first need is a proper source of heat and 
if brazing, as distinct from welding, is required, 
the most convenient source is a propane 
blowtorch. These torches are usually designed 
on a modular basis so that one buys a handle 
(incorporating the shut-off valve), a neck tube 
and a burner to acquire the components of a 
basic torch, afterwards adding other burners 
and neck tubes to suit the size of the job. A han¬ 
dle, two neck tubes and a selection of burners is 
shown in Figure 4.20. 

To complete the blowtorch outfit, a large 
bottle of propane is required. This must con¬ 
nect to the handle through a regulator and a 
hose suitable for use with propane. A regulator 
is employed to deliver a constant gas pressure 
to the handle-and-burner assembly but also in¬ 
corporates a detector to cut off the gas supply 
should the hose fail. An alternative to a regula¬ 
tor is to use a hose failure valve which does not 
regulate the pressure but does guard against 
hose failure. The outlet of a small propane cyl- 

Figure 4.20 A propane blowtorch outfit, showing handle, 
neck tubes and burners. 

inder fitted with a hose failure valve is shown in 
Figure 4.21. 

When obtaining a torch outfit of the type 
shown in Figures 4.20 and 4.21, it should be 
borne in mind that a range of different burners 
is required to cope with differently sized jobs, 
since a large structure needs a larger input of 
heat to achieve and maintain the same tempera¬ 
ture, due to its ability to radiate a greater quan¬ 
tity of heat. A larger quantity of heat is 
obtained by burning more gas, and the burners, 
although given simple reference numbers, are 
classified according to the gas consumption, in 
ounces or grammes per hour, when working at 
the standard pressure. The heat radiated from 

Figure 4.21 A hose failure valve fined to a propane cylinder. 



the work increases with its size, as noted above, 
so it is necessary to stand farther away from the 
larger jobs and a longer neck tube is required 
for use with larger burners. 

The amount of heat which is radiated from 
burner and work sets a practical limit on the 
size of burner which can be used, since the 
operator must be close enough to the work to 
apply the silver solder or brazing alloy, and this 
limits the maximum length of neck tube. The 
limit is represented by a gas consumption of 
about 70 ounces (2kg) per hour, but this is cer¬ 
tainly adequate to carry out the silver soldering 
on a model boiler having a 5-inch (127mm) di¬ 
ameter barrel and being about 18 or 20 inches 
long overall (around 500mm). Larger work can 
also be managed by a burner of this capacity 
provided that the work is well packed around 
with heat retaining material. If more heat than 
this is required, it is best to limit the burner size 
but to use two torches by calling in the services 
of an assistant. 

Two particular problems occur with bottled 
gas blowtorches, one associated with the bot¬ 
tled gas and the other with the basic torch 
design. When the gas burns, it consumes oxy¬ 
gen, and a good supply of this is vital to the 
combustion process. Conventional burners, of 
the type shown in Figure 4.20, mix gas and air 
within the burner, which is provided with a 
group of air inlet slots adjacent to the gas jet. 
The burner must always be able to draw in 
enough air through these slots to support com¬ 

If the torch is directed into a closed space, 
the products of combustion blow back towards 
the rear of the burner. This prevents air being 
drawn in, and the flame thus extinguishes itself. 
One way to prevent this is to redesign the 
burner so that the air is drawn in away from the 
burner head and one manufacturer makes 
burners which operate on this principle. The 
separate neck tubes and burners shown in Fig¬ 
ure 4.20 are replaced by an assembly which 

Figure 4.22 A Sievert Cyclone burner and its handle. 

combines both functions, the air holes being 
positioned at the handle end of the assembly, so 
removing them from the vicinity of any blown- 
back combustion gases and placing them in a 
location having a freely available air supply. A 
handle and burner of this type are shown in 
Figure 4.22. 

Brazing hearth 

Bearing in mind the above description of the 
amount of heat likely to be radiated during the 
brazing of a large structure such as a copper 
locomotive boiler, it is clear that a safe area, 
free from combustible material, must be avail¬ 
able in which to carry out the process. This is 
provided by a specially built brazing hearth. 
This should ideally be located outside the area 
containing the machine tools but in any event 
must be positioned where there is adequate 
ventilation since poisonous fumes can be pro¬ 
duced by the brazing processes. 

A common method of making a hearth is to 
use one end of a circular, 50-gallon (225 litre) 
steel drum. Ifthe drum is cut off 12or 15 inches 
(300-375mm) from one end an open-topped 
cylinder is produced. If part of the cylinder wall 



is cut away to provide access, the remaining up¬ 
standing wall around the base prevents the 
spread of flame and creates a safe brazing 
hearth. A set of legs will be needed to create a 
free-standing unit, and it is usually recom¬ 
mended that a central hole should be provided 
so that one end of a long workpiece such as a 
boiler can protrude through from below, thus 
maintaining the work low down in the hearth. 

If you are likely to have to deal with fairly 
long items, the circular shape may not be con¬ 
venient. Figure 4.23 shows a rectangular hearth 
constructed from commercial slotted angle 
with the top and back formed by two parts of a 
shelf unit from the same manufacturer. This has 
a central hole in the base which is ordinarily 
covered by a loose plate of mild steel. 

In order to minimise on the heating time 
(and save on gas) large work must be packed 
into a ‘cocoon’ of insulating material. Tradi¬ 
tionally, the material used was coke, but a 
cleaner alternative is firebrick, or some other 
type of refractory material. Secondhand 
bricks can sometimes be obtained when boiler 
or flue linings are renewed. Whole bricks, as 

Figure 4.23 My brazing hearth showing its central hole ana 
some firebricks obtained when a boiler flue was being rebuilt. 

Figure 4.24 A low-cost alternative to a small brazing hearth. 

illustrated in Figures 4.23 and 4.24, can be 
used to support the work, or form a barrier 
behind it, while broken brick, or small pellets 
of refractory material, may be used inside 
hollow work, or for packing around the work 

For small work, a large hearth of the type il¬ 
lustrated (about 30in. x 14in. or 750mm x 
350mm in plan) may not be required. Some¬ 
thing smaller can be created by utilising a few 
firebricks in a biscuit tin, as shown in Figure 
4.24. Two layers of firebrick provide the basic 
insulation below the work, but this is supple¬ 
mented by a piece of commercial insulation, 
known asSindanyo. In addition to being a good 
insulator, it provides a flat surface on which the 
work can rest. Since only small blowtorches are 
used with this hearth, two or three firebricks 
behind the work are adequate to provide insu¬ 
lation and prevent spread of the flame. 

For silver soldering of very small work, such 
as might be undertaken by a jeweller, very small 
flames are ordinarily used and a small block of 
Sindanyo, or similar, will serve most needs, 
provided that care is taken. 



Measuring and marking out 

Measuring equipment 

The humble rule 

Much ordinary measuring is done with a rule, 
especially when marking out or checking the 
size of work in the flat, as for sheet metal work. 
But there are rules and rules, and what will do 
for making a garden fence will not do at all for 
most operations which might be encountered 
in model engineering. 

First of all there is the question of accuracy. 
If you are making a scale model of about one- 
twelfth full size, one inch on the prototype will 
be represented by .083in. (2.12mm) on the 
model. While you may not think that it is 
important to work to high accuracy, if there are 
two items on the model which should be the 
same size, it is often noticeable if they are not. 
In any event, things need to fit where they are 
supposed to, and if they do not, that also is very 

You will need at least two rules - one for 
accurate measurements and the other for decid¬ 
ing whether you have picked up a piece of 1 x 
Vi or Vi x V «BMS (or whatever). This latter can 

be regarded as the ‘hack’ rule and it may be left 
lying about the workshop in a casual way, or 
thrown into a drawer when not in use. This rule 
can, therefore, be the older of the two. Its 
corners may have become rounded as a conse¬ 
quence of much use (or abuse) and it may there¬ 
fore not be suitable for use when marking out, 
or when real accuracy is required. 

The accurate rule should be the best you can 
afford and it should be stored and used care¬ 
fully so that it remains in good condition. It 
must have finely engraved scales so that the 
lines are significantly smaller than the smallest 
division on the rule. If you model to imperial 
standards, you will need a rule engraved down 
to Vuin. (.0156in.). The most useful type is per¬ 
haps a double-sided rule engraved with '/», ‘/is, 
Vn and Zm inch divisions on its four edges, but 
an alternative is the type which is engraved in 
'/32in. increments along one edge, and VMn. on 
the other. Both types allow measurements of 
Main. increments along the full length of the 

An alternative imperial rule is that engraved 
in '/iooin. divisions, this type being useful in 
instances where the scale dimensions have been 



‘rounded’ to decimal fractions rather than the 
more usual imperial (vulgar) fractions such as 
'/Min. The .OlOin. markings are difficult to see 
without use of a watchmaker’s glass, but since 
estimation of sizes is frequently performed us¬ 
ing a glass, this is not necessarily a serious dis¬ 

For modellers using metric standards, 
0.5mm engravings provide slightly larger incre¬ 
ments than the imperial '/kin. (.0197in. com¬ 
pared with .0156in.) so that 0.25mm divisions 
may seem more desirable. However, rules 
engraved to this level of precision are not read¬ 
ily available and present the same problems of 
visibility as those imperial rules engraved with 
'/iooin. divisions. 

When purchasing a rule, fine engravings and 
good visibility are the prime requirements, 
hence the popularity of satin-chrome finish 
rules and black-filled engraving. The other con¬ 
siderations are length, width and thickness. 
Length is obvious and most needs are satisfied 
by 6in. or 12in. (150 or 300mm) rules but for 
work on the lathe (for length measurements) 
there is real advantage in something shorter 
and a 3 - or 4-inch (75 or 100mm) rule is prob¬ 
ably more useful than a 6-inch unless you par¬ 
ticularly wish for one which fits conveniently 
into a top pocket. A dual, imperial and metric, 
4-inch rule is illustrated in Figure 5.1 together 
with a number of other rules. 

Figure 5.1 A selection of imperial rules and a 4 inch (100mm) 
having metric and imperial mailings. 

Once length is determined, the choice lies 
between rigid or flexible types. Personal prefer¬ 
ence dictates the decision, the two types being 
thin and narrow, and therefore flexible, or wide 
and thick, and hence rigid. Flexible rules are 
generally '/ain. or 12.5mm wide, while the rigid 
type is usually lin. or 25 mm wide, although 
some rules 3 /«in. (19mm) wide are available. 
Both flexible and rigid types are shown in Fig¬ 
ure 5.1. 

If you intend to make a large-scale model 
steam locomotive, you will need a longer 
rule; either 36in. or 1 metre long, according 
to your preferred measuring system, for mark¬ 
ing out the longer items such as the frame 
plates, water tanks, boiler barrel and running 


A rule is not suited to making really accurate 
measurements since it is engraved only in '/win. 
or !/imm increments. A rule is also not ideally 
adapted to measuring the diameter of a circular 
rod, or the bore of a circular hole. Neverthe¬ 
less, before the widespread introduction of 
more accurate measuring devices, rules were 
generally the only measuring instruments avail¬ 
able, and parts were made to roughly the right 
size and then finished by hand to achieve the 
required fits and clearances. 

For measuring thicknesses, and the sizes of 
holes and recesses, simple devices were devel¬ 
oped in ancient times for use by carpenters and 
stonemasons, and these naturally came to be 
used by metalworkers. 

These instruments are essentially ‘compara¬ 
tors’ rather than ‘absolute’ measuring devices, 
since they are not engraved with scales and 
their settings must be checked using a rule. 
They are known simply as calipers, two basic 
types being used, intended for outside or inside 
measurements and consequently known as 



Figure 5.2 Inside and outside calipers. 

inside and outside calipers. A group of calipers 
is shown in Figure 5.2. 

A caliper essentially comprises two stiff arms 
with hardened tips which are either turned out 
or turned in to create an inside or an outside 
caliper. The joint between the arms may be 
formed by overlapping two (flat) arms and riv¬ 
eting them together, with substantial washers 
on each side, to form a stiff, but movable, joint. 
This is the firm-joint caliper. Alternatively, two 
arms are united by a circular spring and roller 
and fitted with a screwed stud and adjusting 
nut. This is the spring caliper. 

Figure 5.3 Measuring the setting of an outside caliper using a 

For making a measurement, the caliper is 
adjusted to just touch either the outside or the 
inside faces of the work, after which its setting 
is measured using a rule. 

In setting the caliper, the correct sense of 
touch needs to be used to ensure that the 
caliper is not set to such a size that it is ‘sprung’ 
when fitted to the work. A description of the 
method of setting an inside caliper into a bore 
or recess is given below. 

Figure 5.3 shows the setting of an outside 
caliper being checked against a rule, one leg 
resting on the end of the rule and the position 
of the other leg (the setting) being read off on 
the rule. 

If the setting of an inside caliper is to be 
measured, the rule and the caliper must abut a 
square block in order that the setting can be 
measured accurately. A simple offcut of steel bar 
makes a suitable block, and the operation of 
checking the setting is illustrated in Figure 5.4. 

Figure 5.4 Measuring the setting of an inside caliper using a 
rule. Note the use of a square block of steel to provide an 



Adjustment of a spring caliper is obvious, 
but adjustment of the firm-joint type, for small 
movements, is best made by tapping either the 
back of the joint, or one of the arms, on a 
wooden block. Tapping the back of the joint 
causes it to spring apart by a small amount, 
while holding one arm and tapping it on a 
block closes up the caliper. 

Jenny calipers and dividers 

An alternative form of caliper is the odd legs or 
Jenny caliper. Two pairs of these are shown in 
Figure 5.5. This instrument is basically an ad¬ 
justable caliper with one side holding (or 
shaped into) a scriber point, and the other pro¬ 
vided with a turned in or recessed foot which 
abuts the edge of the rule, for setting, or the 
edge of the work, in use. During setting, the 
plain foot rests against the end of a rule and the 
scriber point locates in an engraving on the 
rule, thus making setting to a standard, en¬ 
graved value simple and straightforward. Once 
set, the caliper is used to scribe a line parallel to 
a reference edge, and provided that the caliper 
is always held squarely during setting and use, 
accurate marking off results. 

Related to the caliper is the divider, two sizes 
of which are illustrated in Figure 5.6. Dividers 

Figure 5.5 Jenny calipers, or odd legs, once a common item 
for apprentices to make. 

Figure 5.6 Two pairs of spring dividers, the larger is a 6 inch 

are used for measuring the distance between 
two points or parallel lines on a flat surface. 
They are also used for transferring measure¬ 
ments from a rule to a flat surface and are natu¬ 
rally used for scribing arcs or circles. Most 
dividers are of the spring type, but firm-joint 
types are still sometimes seen. 

The vernier caliper 

An improvement in the measuring precision of 
a linear device such as a rule can be obtained by 
using a vernier scale. This device, which is 
extremely simple in concept, allows an indica¬ 
tion of sizes between the smallest divisions on 
the scale to be obtained directly. Figure 5.7 
shows a portion of a rule engraved in '/ioin. 
divisions. Adjacent to this is a short scale, 
engraved with 10 divisions similar to those on 
the rule. However, this scale is engraved so that 
it has 10 divisions in the same length occupied 
by 9 on the rule itself, as shown in Figure 5.7a. 
Thus, each division on the auxiliary, or vernier 
scale is 9 /io of the size of those on the rule. With 
the scale positioned as shown in Figure 5.7a, 
only two lines on the vernier scale correspond 
with engraved lines on the rule, these being the 
scale’s ‘0’ and ‘10’ lines. This is the position of 
the scale whenever the measured value is a 



W • 

Figure 5.7 Two settings on a vernier scale, .01 inches 
(0.25mm) apart. 

whole number of inches and tenths of an inch, 
the position of Figure 5.7a corresponding with 

If the scale moves relative to the ‘2.2’ mark 
on the rule, one tenth of the distance towards 
the ‘2.3’ engraving, its own engraved ‘1’ mark 
exactly coincides with the ‘2.3’ line on the rule, 
as shown in Figure 5.7b, since the distance 
between the lines on the vernier scale is 9 Ao of 
that on the rule. The position taken up by the 
scale is thus 2.2in. plus 'Ao of the distance 
towards 2.3in. or 2.21in. Thus, by use of the 
vernier scale, a rule engraved in '/ioin. incre¬ 
ments can yield measurements which can be 
read to within Viooin. 

A simple rule fitted with a vernier scale is not 
a practical proposition and the device most 
commonly found is that illustrated in Figure 
5.8. This is generally simply known as a vernier 
although it is correctly called a vernier caliper. 
This comprises a rule which incorporates a 
fixed jaw and a sliding jaw, integral with a 
frame which slides on the rule. The sliding jaw 
frame is engraved with a vernier scale and its 
position relative to the fixed jaw may be 
assessed accurately. Both jaws are hardened, 
ground and lapped so that they lie parallel to 

each other and at right angles to the rule. 

The sliding jaw frame connects to a clamp¬ 
ing frame through a short screw which engages 
with a knurled nut running in a slot in the 
clamping frame. Since the clamping frame can 
be locked to the rule by a screw, the knurled nut 
allows fine adjustment of the position of the 
sliding jaw and its frame along the rule. A sec¬ 
ond locking screw is fitted to the sliding frame 
so that it, too, may be locked to the rule. 

The imperial scale on this vernier is divided 
into tenths of an inch, and each tenth into four 
divisions, each of which represents a quarter of 
one tenth, or .025in. The vernier scale is gradu¬ 
ated so that the distance occupied by 49 
divisions on the rule is divided into 25 parts. 
Thus, each vernier division represents 24 /ioooin. 
and since each corresponding division on the 
main scale represents 2S /ioooin., the distance 
between any two vernier and main scale divi¬ 
sions is '/loooin., or .001 in. This type of vernier 
caliper is used to indicate a measurement 
within ±.001 in., generally called ‘one thou’. 

When using the vernier caliper, the knurled 
clamping screws are first loosened and the slid¬ 
ing head and clamp adjusted approximately to 
the correct position. The clamping frame is 
then locked to the rule using the screw and the 
measurement taken by bringing the sliding jaw 
into contact with the work to be measured, 
making fine adjustment of the sliding jaw and 
frame with the knurled nut. For accuracy, the 

Figure 5.8 A vernier caliper. 



Figure 5.9 Close-up of the vernier caliper set to 1. 185/1.186 

sliding jaw needs to be brought ‘nicely’ onto 
the work, so that its final position is just touch¬ 
ing the work with neither slop nor excessive 
tightness. There is a need to develop a certain 
touch in using the vernier. This sense of touch is 
important, too, in respect to other measuring 
methods, and it sometimes pays to repeat the 
measurement to be sure that consistent results 
are being obtained. 

Vernier calipers are used for measuring 
external diameters and sizes. In some types, 
including the one shown in Figure 5.8, an addi¬ 
tional pair of jaws is fitted on the back of the 
fixed and moving jaws and these can be used 
for measuring internal diameters and sizes. A 
rod is also fixed to the moving jaws and can be 
used for depth measurement. 

To illustrate the use of the vernier scale, Fig¬ 
ure 5.9 shows a close-up of rule and scale. The 
setting portrayed is between l.lin. and 1.2in. 
this distance on the rule being divided into four 
increments of .025in. The position shown is 
thus slightly greater than 1.175in. The vernier 
scale is engraved with 25 divisions, each repre¬ 
senting .001 in. (one thou) and it is either the 10 
or 11 engraving on the vernier which is aligned 
with an engraving on the main rule. The setting 
is 10 or 11 thou greater than 1.175in. or some¬ 
where between 1.185in. andl.l86in. 

On the metric scale which is also engraved 

on this vernier, the setting is between 30.01mm 
and 30.012mm which may be taken as 


An alternative way in which measurements can 
be taken, not reliant on direct measurement 
along a rule, is by use of a screw thread. The 
instrument has essentially the same parts as a 
vernier, that is, one fixed and one movable 
‘jaw’, but this latter is attached to, or in one 
with, a screw thread which allows the distance 
between the fixed and moving elements to be 
measured indirectly. The instrument is known 
formally as a micrometer, but is frequently 
referred to as a mike. 

The components of a typical instrument are 
shown in Figure 5.10. It consists of a steel 
frame carrying a fixed anvil, A, on one horn, 
and opposite to it, an adjustable spindle, B, to 
which is fixed a thimble, C. The spindle passes 
through a boring in a fixed sleeve, D, which is 
effectively in one with the frame, although it 
may be held by a stiff, frictional grip, and be 

Ratchet collar 



Figure 5.11 0-1 in. micrometer set to 0.397in. 

Both the fixed sleeve and the spindle carry 
mating screw threads so that the distance 
between the anvil and the spindle end may be 
varied by rotating the thimble. On an imperial 
micrometer, the thread used for the sleeve and 
spindle has 40 turns per inch (tpi) so that each 
turn of the thimble moves the spindle through 
.025in. The thimble is accordingly engraved 
with 25 divisions, as shown in Figure 5.11, 
each of which thus represents .001 in. or one 

The actual measurement set on the microm¬ 
eter is obtained by reading the scales on both 
the sleeve and the thimble. The sleeve is 
engraved with lines at .025in. increments 
which correspond with whole numbers of turns 
of the thimble and spindle, from the zero posi¬ 
tion. Every fourth line is identified as 1, 2 etc. 
up to 9, corresponding with 0.1 in., 0.2in. up to 
0.9in. Obtaining the measurement requires 
three positions to be noted and the setting of 
Figure 5.11 is interpreted as follows: 

Highest sleeve engraving visible = 

3 -».300in. 

Number of .025in. increments above this = 

3 -».075in. 

Reading on the thimble scale = 



This procedure may seem laborious, but in 

practice one soon becomes proficient in inter¬ 
preting the readings without effort. 

A metric micrometer is perhaps easier to 
interpret since the spindle thread has a pitch of 
0.5mm and the thimble is engraved with 50 
divisions, each representing .01mm, as shown 
in Figure 5.12. Again, to save counting turns, 
the sleeve is engraved every 0.5mm; whole mm 
above the index line, and half mm below, the 
whole mm lines being identified every 5mm. To 
read the measurement it is necessary to read the 
sleeve and the thimble, and the setting of Figure 
5.12 is read as follows: 

Highest sleeve engraving visible = 6.5mm 
Reading on thimble scale = 0.47mm 

Figure 5.12 0-25mm micrometer set to 6.97mm. 

Reading this micrometer is more straightfor¬ 
ward than the imperial type. 

Two further points need consideration in 
relation to the micrometer. First of all, the 
sleeve is not usually firmly fixed to the frame 
but is held by a stiff friction grip. It is therefore 
adjustable and can be rotated to bring its index 
line to correspond with the ‘0’ engraving on the 
sleeve when the spindle end is closed against 
the anvil. To allow this adjustment, the sleeve is 
provided with a shallow drilled hole and the 
micrometer is supplied with a ‘C’ spanner 
which allows the sleeve to be rotated. These 
features are shown in Figure 5.13A. 




Figure 5.13 Close-up of a metric micrometer at what should 
be its zero position, with the 'C' Spanner which is used to adjust 
the position of the sleeve. 

From Figure 5.13B it will be seen that 
the index line on the sleeve does not quite 
correspond with zero on the thimble, and a 
small adjustment of the sleeve is required. 

A second point to consider is how firmly the 
spindle should be screwed into the anvil to 
define the zero. Clearly, a sense of touch is 
required, since the pressure needs to be just so 
much - neither more nor less. To remove the 
need to develop this required sensitivity, most 
micrometers are fitted with a ratchet on the 
thimble end which slips when a certain turning 
force (torque) is applied to it. Using the ratchet 

thus ensures that a standard torque is always 
applied, both when setting the zero and when 
making measurements. 

However, most craftsmen prefer to develop 
the required sense of touch for themselves and 
the most common way in which the microm¬ 
eter is held is shown in Figure 5.14. The ratchet 
is contained within the small-diameter thimble 
end, but is studiously not in use since the third 
and fourth fingers are holding the micrometer 
frame, and the remaining digits cannot reach 
the ratchet. This method is the most convenient 
when the work must also be held, since the 
other hand is freed for this duty. 

If several items need to be made to the same 
dimension, the micrometer may be used as a 
gauge by being set to the required size and 
machining each piece until it will (just) pass 
between anvil and spindle. To assist its use in 
this manner, the micrometer may be provided 
with a spindle lock. One type of actuation for 
this is the small lever visible on the frame of the 
metric micrometer in Figure 5.13. The imperial 
micrometer of Figure 5.11 is provided with a 
knurled ring set into the frame for the same 

The micrometers illustrated and described 
above, cover only the limited ranges from zero 
to 1 inch or zero to 25mm. Reducing the meas- 

Figure 5.14 Holding a micrometer. 



Figure 5.15 Three micrometers capable of measuring from 
zero to 3 inches, with setting bars I inch and two inches long. 

uring capability to this fairly short range places 
a lesser requirement for accuracy on the screw, 
since it must be accurate to within, say, .0001 in. 
in one inch (.0025mm in 25mm), rather than 
this amount in 2, 3 or more inches. Microm¬ 
eters are consequently usually made to cover 
just a lin. or 25mm range, standard microm¬ 
eter heads being attached to different frames to 
provide 0-1, 1-2, 2-3, or even 11-12 inch 
(10-25, 25-50, 50-75 or even 250-275mm) 

Figure 5.15 shows three micrometers cover¬ 
ing the range from zero to three inches, 
together with setting rods 1 and 2 inches long 
that allow the two larger instruments to be cor¬ 
rectly zeroed. 

Gauging the size 

It is naturally possible to reverse the process of 
using a caliper, first setting it to a required size 
by use of a rule and then removing material 
from the part until it fits the caliper, allowing 
you to make a part having a particular dimen¬ 
sion without measuring it directly. In these 
cases, the caliper is being used as a gauge to 
determine when the correct size has been 
reached. Using an outside caliper in this fashion 

allows the thickness, width or diameter of the 
work to be brought to the required dimension, 
whereas an inside caliper is used when a hole or 
recess of a specific size is required. 

This is a useful technique to adopt when 
diameters are being turned on a lathe, for 
example. In these operations, the work is 
reduced in diameter (or a hole increased in 
diameter) in discrete steps, thus progressively 
approaching the required size. If an outside 
diameter is being turned, a caliper can be set a 
little larger than the final size. It is then possible 
to continue machining until the caliper ‘goes’ 
over the outside diameter, which indicates that 
the size is approaching that required. This 
allows the approximate dimension to be 
achieved without making any measurements 
directly on the work. 

The same method can be adopted for circu¬ 
lar bores, by setting an inside caliper just smaller 
than the size required, continuing to increase 
the bore progressively, without measurement of 
the size, until the caliper will enter, after which, 
the actual size of the bore needs to be measured 
after each cut, until the size is reached. 

Any item which is being progressively 
brought to size, by whatever means, can be 
sized using a gauge, but the technique is espe¬ 
cially valuable if several items need to be made 
to the same overall dimension, but the size is 
larger than the capacity of your really accurate 
measuring instrument, which might only be a 
0-1 in. or 0-25mm micrometer. 

When building a model locomotive, the 
frame plates are held the correct distance apart 
by spacers, or stays, which all need to be the 
same length if the frame plates are to be paral¬ 
lel. There are usually several stays, and the 
attachments to the buffer beam at each end, 
and in some locomotives, the cylinder block is 
also mounted between the frame plates. There 
are, therefore, several items which need to be 
brought to the same length when making up the 
frame’s components. 



Distance between frames 
) (20mm) _J\ 

) I (25mm) 
Figure S. 16 A gauge for measunng locomotive frame stays. 

If a gauge of the type illustrated in Figure 
5.16 is made up at the outset, it may be used to 
test each item as it is machined, and each 
reduced in width or length until it just enters 
the gauge without noticeable clearance. All 
items then conform to the standard dimension 
defined by the gauge. 

The gauge itself can easily be made by hand, 
and its gauging dimension checked adequately 
using a rule, thereby providing an accurate 
gauge for making a group of similarly dimen¬ 
sioned components. 

If rectangular recesses or slots are being 
produced, a simple gauge for the size can be 
made up by hand and used for testing as the 
enlargement proceeds. For example, if a stand¬ 
ard gauge is available which measures .500in. 
(precisely), it may be used to determine (gauge) 
the width of a slot which is being machined, 
since, if the gauge will enter the slot, its width 
must be just greater than .500in. 

If small gaps need to be gauged, the usual 
tool is a feeler gauge. These are manufactured 

Figure 5.17 Standard set of imperial feeler gauges. 

in standard sets ranging from .0015in. (.04mm) 
to .025in. (25 thou) (0.6mm), the usual form 
being that in which ten blades are mounted to¬ 
gether into a folded metal ‘handle’, or plastic 
moulding, as shown in Figure 5.17. The blade 
thicknesses (in inches) provided in the usual 
imperial set are: 

.0015, .002, .003, .004, .006, .008, .010, 

.012, .015 and .025 

The individual blades are etched with their 
thicknesses. Two or three can be used together 
to make up a gauge of the desired size. Similar 
metric sets are also available. 

The usual way to employ feeler gauges is by 
employing a ‘go’ or ‘no go’ technique. If a gap of 
.015in. is required, and the gap is settable, the 
correct setting is such that a gauge of .014in. 
thickness should enter freely, or ‘go’, whereas a 
gauge of .016in. should not. By implication, this 
means that the gap is set to .015in. 

Plug gauges 

The technique for gauging the width of a slot, 
can also be used when machining a bore. This is 
achieved by employing a precisely sized cylin¬ 
drical gauge, known as a plug gauge, which is 
made to the required size. 

Industrially, plug gauges are available in 
closely spaced standard sizes, enabling a wide 
range of sizes to be gauged. As these gauges 
have a ground and polished finish they are 
consequently expensive and not usually avail¬ 
able to the amateur, but it is quite easy to make 
them up to suit particular jobs. 

The important characteristics are size and 
surface finish, so the home-made gauge needs 
first of all to be finish turned with a correctly 
ground and sharp tool. Secondly, it must be 
accurately sized but this can usually be 
arranged if a micrometer or vernier caliper is 




-.001 in. +0.001 in. 

(.025mm) GO NOGO (,025mm) 

Figure 5.18 'Go' and 'no go' plug gauges. 

available, and is not difficult if approached 

A single gauge may suffice if a hole is being 
bored progressively to size, but an existing hole 
may be measured if there are two plug gauges 
having a known small difference in diameter. If 
the smaller of the two enters the bore freely, but 
the larger one does not, the bored dimension 
lies roughly midway between the sizes of the 
two gauges. 

The ‘go’ and ‘no go’ gauges can often be 
made conveniently on a single plug, as shown 
in Figure 5.18A, although this does limit the 
use of the ‘no go’ gauge to bores which are deep 
enough to accommodate the ‘go’ gauge also. 
The alternative is to arrange a plug at both ends 
of the gauge, as shown in Figure 5.18B, which 
naturally avoids this problem. 

The gauges should ideally be hardened and 
polished, but to avoid spoiling the dimensional 
accuracy, it is best to aim for as good a turned 
finish as possible, afterwards hardening the 
gauge and then polishing. The simplest way to 
do this is to use bright mild steel (BMS) to make 
the gauge, afterwards case hardening it, as 
described in Chapter 3, and then polishing it to 
a fine surface finish by using emery cloth and 
paper to remove the scale. 

As noted above, a single gauge often serves 

the purpose, since, if a hole is being bored pro¬ 
gressively, it is only necessary to reach the point 
at which a correctly sized gauge ‘goes’ into the 
bore. However, to assist this type of operation, 
it is useful to make two gauges - a correctly 
sized one and another which is .005in. or 
.010in. (0.1mm or 0.25mm) smaller than the 
size required, thus allowing the approach to the 
desired size to be detected. 

Measuring practice 

Using a rule 

The method of using a rule may seem obvious, 
and so it is if the task is just to determine 
whether an item is 3 /+in. or lin. (19mm or 
25 mm) wide. However, the rule is not ideally 
suited to measuring accurately the width of an 
item, and the task is different if the need is to 
determine whether two scribed lines are 
6.75mm apart. Firstly, a metric rule having 
0.25mm engravings may not be available, and 
secondly, even if it is, you may not be able to 
discern the precise size without the use of an 
eyeglass. It therefore becomes necessary to esti¬ 
mate the position because it falls between two 
engraved divisions, and it is also necessary to 
use a magnifying aid. 

If the rule is engraved in 14+in. divisions, it is 
possible, with a glass, to estimate the distance 
between two lightly scribed lines with an accu¬ 
racy that is better than this i.e. it is possible to 
decide whether a scribed line actually coincides 
with the rule’s engraving, or is displaced, and if 
so, by how much. This means that an estimate 
of actual position can be made to better than 
'/usin., or better than .008in. A metric rule with 
'/ 2 mm divisions, allows estimation of the posi¬ 
tion of scribed lines to better than 0.25mm. 
This is illustrated by Figure 5.19 which shows 
two views which approximate to those pro¬ 
vided by a 4 l/ 2 -inch watchmaker’s glass. 



Figure 5.19 Close-up views of the two sides of a rule which 
has imperial and metric engravings, equivalent to the views 
provided by a 4'/S inch (I I Smm) watchmaker’s glass. 

Angle of View 

r "\ 


Figure 5.20 The possibility of parallax error should be 
remembered when checking or setting measurements on the 

There is, however, one general point which 
should be made concerning the reading of any 
kind of scale, and that is the phenomenon 
known as parallax error. This arises if the posi¬ 
tion of the eye, when making the reading, is not 
exactly over the engraving on the rule. The 
situation which can arise is shown in Figure 
5.20. An angled view of the scribed line and the 
engraving on the rule introduces an apparent 
error and the total error may be significant if it 
is in different directions at the two ends. 

The thickness of the rule affects the size of 
the error, given a particular point of view, and a 
thin rule is therefore beneficial in this respect. 
However, in some circumstances, it may be 
possible to stand a thicker rule on its edge, thus 
avoiding the problem, but care must be taken 
to ensure that it is not bent since an error is 
introduced by the curvature. 

Checking of marked-out positions must be 
performed as a matter of course. This requires 
careful use of a watchmaker’s glass in order to 
allow the divisions on the rule to be seen 
adequately, in addition to care in positioning 
the eye to avoid the error which parallax may 
introduce. Used carefully, the rule is an excel¬ 
lent means to check marked-out work. It is 
slightly less well suited to the measurement of 
the size ‘over’ something and relatively poor 
when used to measure diameters. It is also not 
good for measuring bores and there is naturally 
a limit to the smallest division which can be en¬ 
graved on the rule. 

The use of calipers to measure inside and 
outside dimensions, in association with a rule, 
is described above, and the technique to adopt 
when setting the caliper is described below. 

Measurement of depth 

The ordinary rule is quite readily utilised for 
the measurement of the depth of a bore or 
recess in the work, its major drawback being 
that its width restricts its use to relatively large 



Figure 5.21 A depth rule is simply a narrow rule which can 
be clamped into a shaped stock. 

bores. A narrower-than-normal rule is benefi¬ 
cial and is usually produced in the form shown 
in Figure 5.21, in which a '/tin. (6mm) rule is 
provided with a sliding stock which can be 
clamped to it. The instrument shown has its 
rule engraved with both metric and imperial 
graduations thereby satisfying the needs of 
both systems. 

The stock has a waisted shape which pro¬ 
vides both long and short ground faces which 
lie at right angles to the rule and it can also be 
used in ‘tight’ situations as a substitute for a 

Figure 5.22 This protractor uses a similar rule which can be 
clamped at an angle to the engraved plate. If the rule is set 
loosely to the 90-degree position, it might also serve as a depth 
rule, if used with care. 

An alternative to the depth rule, although 
not generally so useful due to its larger ‘stock’, 
is the type of protractor shown in Figure 5.22. 
This uses a narrow, engraved rule mounted to a 
screw clamp which allows it to be locked to any 
angle set on the scale. If the rule is set to the 
90-degree position, it may naturally be used to 
provide a depth measurement. However, the 
‘regular’ depth rule is generally superior for it 
slides in a machined groove in the stock and is 
always maintained in the 90-degree position, 
even when not clamped. The depth rule facili¬ 
tates the actual measurement in that the stock 
can be held on the work’s reference surface and 
the rule slid down until it contacts the bottom 
of the bore or recess. The rule can then be 
clamped to the stock and the tool removed 
from the work and the actual depth read on the 

Since the protractor does not automatically 
hold the rule in the 90-degree condition, there 
is always some uncertainty when trying to 
make a measurement in this way. The protrac¬ 
tor’s rule may nevertheless be useful in con¬ 
firming that a bore or recess is under or over 
the depth required, by being set to the measure¬ 
ment and then clamped at 90 degrees. The 
depth can then be tested by probing the hole 
and confirming whether or not the rule reaches 
the bottom before the stock contacts the refer¬ 
ence surface, or vice versa. 

More accurate measurement of depth may 
be achieved by utilising an adaptation of the 
micrometer, but with the sleeve engraved in 
reverse i.e. showing zero at what would con¬ 
ventionally be the 1-inch position, the spindle 
then extending downwards into the bore or 
recess to show the increasing depth. Figure 
5.23 shows a depth micrometer which meas¬ 
ures depths from 0 to lin. 

The micrometer comprises a standard 
0-1 in. head but with the sleeve graduations 
increasing as the thimble is screwed towards 
the stock. A spindle is inserted through the top 




Figure 5.23 The components of a 0 to I inch depth 

micrometer shown in Figures 5.24 and 5.25 is 
properly described as an outside micrometer. 

The use of the micrometer, and the ‘stand¬ 
ard closure’ that is provided by the ratchet 
normally fitted to the end of the thimble, is 
described above, and the necessity to develop 
the correct sense of touch when using a vernier 
caliper is also mentioned since a standard 
technique is important in achieving repeat¬ 

Figure 5.24 illustrates a potential problem 
when using a micrometer to measure circular 
work. If the work is small, as shown at A, the 
micrometer naturally measures the true diam- 

of the thimble and is clamped into place by a 
screwed cap. 

The spindle has a screwed bush at its upper 
end which can be set so that the spindle is pre¬ 
cisely the correct length. The micrometer is 
zeroed by standing the stock on a flat surface 
and screwing the spindle down until it, too, 
touches the surface, when the sleeve should be 
set to show zero. 

Depth micrometers are usually provided 
with alternative spindles, typically allowing 
them to measure from 0 to tin., from lin. to 2 
in. and from 2in. to 3in. Such an instrument is 
described as a 0 to 3in. depth micrometer. The 
equivalent metric instrument measures from 0 
to 75mm. A fitted case is normally supplied to 
house all of the items - the one shown lacks 
these refinements, having been rescued from a 
secondhand tool store. 

Outside measurements 

Measuring the external dimension of a piece of 
stock material, or of the work, is carried out 
simply by use of a micrometer or vernier 
caliper. To distinguish it from instruments 
designed for measuring bores, the type of 

Figure 5.24 There is a possible inaccuracy when measuring 
round material using a micrometer, if the material is not passed 
through the gap between spindle and anvil when the setting of 
the micrometer is adjusted. 



eter if the work is roughly centred over the 
anvil of the micrometer. However, a microm¬ 
eter applied casually to a larger bar can give a 
false reading if it is not applied precisely to a 
diameter, and it is essential to swing the bar 
through the gap between spindle and anvil, 
after closing up the micrometer on the work, to 
confirm that the setting does correspond to the 
true diameter. 

One further point is also worth emphasising. 
The making of outside measurements is only 
straightforward provided that the work has a 
uniform shape. Unfortunately, this is not true of 
some circular-section cutters, notably drills, 
reamers and end mills. Consideration of the 
measurement of a drill diameter serves to illus¬ 
trate the problem. 

A drill is machined with two flutes that allow 
the cut material to be expelled from the drilled 
hole. The cutting edges of the drill are actually 

on the end and not on its diameter, but the out¬ 
ermost part of the drill is naturally machined to 
the correct size. This sized part of the drill is, 
however, confined to a small ‘land’ behind the 
front edge of the flute, and it is between the two 
lands that the measurement of diameter must 
be made. 

Figure 5.25A shows an imperial micrometer 
measuring a Vtin. diameter drill. The micro¬ 
meter is not correctly positioned on the lips of 
the flutes, however, and is reading 0.351 in. or 
about 24 thou too small. In Figure 5.25B, the 
drill is correctly positioned between the anvil 
and spindle of the micrometer so that the 
maximum diameter is being read, and the 
micrometer is now showing 0.375in. This type 
of measurement can be a problem with all- 
fluted cutters and care needs to be exercised, 
particularly with drills, since they are manufac¬ 
tured in very small size increments. 

Figure 5.25 Since the end of a drill is not truly circular, an 
error in measurement can be made if the micrometer does not 
touch the lands just behind the edges of the flutes when the 
measurement is made. For similar reasons, end mills, slot drills 
and reamers are also difficult to measure correctly. 



Measuring the bore of a hole 

The vernier caliper shown in Figure 5.8 is 
arranged with two sets of jaws so that it may be 
used for both inside and outside measurements. 
The necessity to ‘feel’ the jaws onto the work 
when making measurements has already been 
referred to, and this is naturally also required 
when making use of the inside jaws. Equally 
important is the need to position the jaws on a 
true diameter of a circular bore and the tech¬ 
nique of Figure 5.26 should be adopted to 
ensure that the correct position is found. 

Rotate caliper gently to 
check setting 

True diameter 

Line of caliper jaws 
(incorrect position) 

Caliper set too large 
cannot rotate 

Line parallel to 
longitudinal axis 

Correctly set caliper 
can just rotate 

Figure 5.26 When using calipers, or the inside jaws of a 
vernier caliper, to measure a bore, care must be taken to ensure 
that the jaws, or tips of the caliper feet, are correctly aligned. 

In a circular bore, an incorrect position 
results in the caliper jaws being set too close 
together. A rotation of the caliper about one 
jaw will reveal if an incorrect position has been 
selected. If so, the caliper jaws must be opened 
slightly and the check repeated until the correct 
position is found. This corresponds with the 
‘no free play’ condition when the caliper is en¬ 
tered into the bore. 

In a rectangular bore, incorrect positioning 
of the line of the caliper jaws may result in too 
large a setting of the jaws, as shown by the cen¬ 
tral sketch in Figure 5.26. Rotation of the 
caliper reveals this condition since movement is 
possible only on one direction and the jaws 
need to be brought closer together and the 
check repeated, until the caliper can just be 
rotated through the correctly aligned position. 

When making the rotations, it is essential to 
keep one jaw in contact with the bore and 
rotate the caliper about it. 

A suitable firm-joint or spring caliper may 
also be set to the diameter of a bore by adopting 
the same technique. If more accuracy is 
required than can be achieved by comparing 
the setting of the caliper with an ordinary rule, 
a micrometer or vernier caliper may be used to 
measure the caliper setting. Figure 5.27 illus¬ 
trates this operation when using a micrometer. 

Figure 5.27 The setting of a caliper can be measured using a 
micrometer, carefully closing the micrometer until the caliper will 
not swing between spindle and anvil, under its own weight. 



The caliper has been set carefully to the 
diameter of the bore and its setting is in turn 
being measured by the micrometer. One arm of 
the caliper is supported in contact with the 
micrometer spindle and the other arm is 
allowed to rotate past the end of the anvil. As 
the spindle is advanced successively towards the 
anvil, the point is reached at which the caliper 
arm will not rotate past the spindle under its 
own weight. At this point, the micrometer has 
been set too small and needs to be opened 
slightly to produce just the right feel to the 
movement of the caliper between spindle and 

Two touch-sensitive settings need to be 
made to assess a bore diameter by this means, 
but a little practice should produce consistent 
results, and good repeatability will reveal when 
an appropriate level of proficiency has been 

If a direct measuring method is required for 
bores, an inside micrometer is the correct 
instrument to use. However, these are expen¬ 
sive, and their utilisation may not be sufficient 
to justify the expense. A compromise, falling 
somewhere between the caliper and an inside 
micrometer in ease of use, is the telescopic 
gauge, three of which are illustrated in Figure 

Figure 5.28 A set of telescopic gauges for measuring bores 
between 0.5 and 2.125 inches (12.5mm to 54mm). 

Figure 5.29 Maximum and minimum settings of the middle 
gauge of the set of three. 

Each gauge is made in the form of a tee¬ 
shaped assembly having two bars set at right 
angles to a plated and knurled handle. One side 
of the head of the tee is fixed to the handle 
while the other is spring-loaded within the first 
but lockable by turning a knurled ring on the 
end of the handle. Figure 5.29 shows the full 
range of adjustment of the middle size of the set 
of three, which, overall, cover the range of 
bores from 0.5in. to 2.125in. (12.5mm to 



To measure a bore, the gauge is allowed to 
expand (gently!) into the bore under the action 
of the internal spring. The movable plunger is 
then clamped using the locking ring and the fit 
in the bore tested by rocking and rotating the 
gauge and gently easing it along the hole. When 
the fit is correct, the gauge is carefully removed 
from the hole and its setting measured using a 
micrometer or caliper. 

Being somewhat larger in diameter than the 
arms of a caliper, a telescopic gauge is more 
readily established on a true diameter and its 
polished and domed ends make the setting of 
the micrometer more straightforward than 
when assessing the setting of a caliper. 

Various forms of small-hole gauges extend 
the principle of the telescopic gauge in the 
direction of decreasing hole size and an expan¬ 
ding gauge operating on a slightly different 
principle is shown in Figure 5.30. There are 
also other types of small-hole gauges which 
utilise a tapered mandrel which is entered into 
the hole, the depth of penetration indicating 
the diameter. 

The possibility of using plug gauges to judge 
the diameter of a hole should not be forgotten 
if the size of a hole needs to be measured. 

Figure 5.30 A gauge for small holes which is adjustable by 
drawing a shaft with a tapered end into the bore of a ball-ended 
split rod. 

Ground steel rod (either silver steel or ground 
mild steel) might be used, but the shanks of 
reamers may provide accurate gauges for stand¬ 
ard sizes. The shanks of drills may be used, but 
these are usually finished to a slightly smaller 
diameter than the drill itself, and they should be 
measured carefully before use as gauges. 
Nevertheless, the availability of an outside 
micrometer makes possible the measurement 
of inside dimensions, albeit by indirect meth¬ 
ods, but that all-important sense of touch does 
need to be developed to allow adequate accu¬ 
racy to be achieved. 

It should also not be forgotten that it is not 
always necessary to make an actual measure¬ 
ment, since, if a bush is required to fit a shaft, or 
vice versa, the one can be made until it is a good 
running fit on the other, as judged by its feel. 
Naturally, it is helpful if one of the items is a 
standard size, but exact dimensions may not be 

It is thus possible to machine the bore of an 
engine cylinder until it has reached the correct 
diameter as measured by spanning a rule across 
one end. This will mean that it is correct to the 
nearest Vt* in. or Vimm, say. When the piston is 
machined, it can be turned to provide the 
required fit in the bored cylinder and will then 
match the bore for which it has been made, 
although neither item has been measured accu¬ 
rately. Once again, however, that important 
sense of touch comes into play. 

Marking out 


Except in the case of a very simple component, 
or when dimensions simply do not matter, the 
workpiece needs to be marked to show the 
positions of its outlines and to define important 
points within it. This is generally true whether 



the work is to be finished by hand, or is to be 
machined. Marking out consists simply of 
using a sharp-pointed tool to scribe the outline 
of the work which needs to be cut out, or to 
define the positions of holes or the centres of 
arcs or curves, according to the requirements of 
the job. 

Scribing may be performed in a number of 
ways, but if hand methods are used, is normally 
performed using a scriber and straightedge, for 
straight lines, or dividers for arcs or circles. 
Dividers are also ordinarily used if several 
equal pitches need to be marked off. 

Hand methods of marking out are usually 
based on the creation of two edges (or surfaces) 
at right angles to each other, from which meas¬ 
urements may be set off to define relevant 
points on, or within, the outline. 

An alternative method of marking out uses a 
flat surface as a reference from which to set off 
measurements, the work being supported in 
some way on the surface. If the work can be 
rotated precisely through 90 degrees, it is possi¬ 
ble to scribe lines on the work in two directions 
at right angles, which satisfies the basic need. 
The method does not necessarily require a true 
edge or surface to be created in advance of 
marking out, although it can be helpful. The 
method uses a flat reference known as a surface 
plate, the use of which is described below. 


Scribing the work removes a small groove of 
material, producing a bright line. Removal of 
material means that the surface is permanently 
marked and it is therefore necessary to scribe 
only lightly. The visibility of the line can be 
improved by coating the material with an 
appropriate colour, darkening the surface of 
light materials (bright steels, brass and alu¬ 
minium alloys) and lightening the darker ones 
such as cast iron or black mild steel. 

Felt-tipped pens are a useful source of col¬ 
our, although their effect is not long-lasting. 
Commercially produced coatings are available, 
usually called marking-out fluid or layout fluid 
(blue and green are the popular colours) which 
comprise a solution equivalent to a rapid-dry¬ 
ing paint. A quick brushing over of the work 
quickly produces a reasonably hard coating to 
improve the visibility of scribed lines although 
it is important not to apply too thick a coat or 
there is a likelihood that it will flake off. Once 
the work is completed, the remaining coating 
can generally be removed using a simple sol¬ 
vent such as methylated spirit. 

For use on cast iron, a white coating is the 
most useful and this can be provided by one of 
the quick-drying correction fluids used by 
typists, but it is again necessary to apply a thin 
coating otherwise it flakes off. Rubbing over 
the casting with blackboard chalk and tapping 

Figure 5.31 The effect of coating the work with a contrasting 
fluid before marking out. 



off the excess is also a useful way to improve 
the contrast on cast iron. 

No particular surface preparation is neces¬ 
sary on inherently clean materials, but castings 
usually benefit from a cleaning up with a wire 
brush to remove any sand adhering to the sur¬ 
face and may even require attention from a file 
in some instances. To illustrate the benefit of 
using a marking fluid, Figure 5.31 shows an 
aluminium casting and an iron casting which 
have been partially coated and then scribed. 

Checking measurements 

All marked-off lines on the work must be 
checked for correct dimensioning before cut¬ 
ting out or centre punching for holes. This 
means using the rule directly in its measuring 
role, but with sufficient precision to determine 
the actual dimensions marked, to the relevant 
degree of accuracy. In model work, this usually 
means working to the smallest division on the 
rule (or perhaps better) which will require an 
accuracy of ± Yu in. or ‘/ 2 mm, depending upon 
the engravings on the rule. 

By using a watchmaker’s glass (loupe) of 
about 3 in. (75 mm) focal length there should be 
little difficulty in estimating positions more 
accurately than the smallest divisions on the 
rule. For an imperial rule, this readily allows 
measurements to be checked to one-third of 
'/win. or about 5 thou (.005in.). Figure 5.19 
shows typical views provided by a watchmak¬ 
er’s glass. 

Centre punching for holes 

For positions which define the locations of 
drilled holes, it is convenient to check the meas¬ 
urements at the time of centre punching a hole 
position. It is essential to use a watchmaker’s 
glass during this operation since it is the only 

way in which the centre punch point can be 
accurately placed on the intersection of the 
lines. Since the loupe is used during the dimen¬ 
sional check, it is convenient to associate the 
punching operation with the check. If the 
scribed lines do not exactly define the required 
position, it is possible to compensate for this 
when placing the punch, checking the dimen¬ 
sions once more after only lightly punching the 
dimple. If there is still an error, it is usually 
possible to ‘push’ the dimple in the required 
direction, using the punch at an angle, progres¬ 
sively checking with the rule until the correct 
position is achieved, and then punching verti¬ 
cally to ensure a symmetrical displacement of 
the material. 

Hole positions in mating or matching 

If sufficiently accurate marking out and drilling 
can be performed, it is possible to mark and 
drill two items independently for the common 
fixing holes which will be used for their attach¬ 
ment to one another. This is the method used 

In modelling, interchangeability is frequen¬ 
tly not essential. It is sufficient that we have two 
clock plates with matching holes, they do not 
need to be identical with another pair of plates. 
In locomotive work, it is sufficient that there 
are two cylinder, steam chest, piston-and-rod 
and cover assemblies which go together, even if 
individual components from the right- and left- 
hand sides are not interchangeable. 

If several holes are required in two or more 
items so that they may be bolted or riveted 
together, it is preferable if only one item is 
marked out, centred and drilled. The two (or 
more) parts can then be clamped together and 
one used as a jig to drill the other(s). 

It is often desirable to perform all the drill¬ 
ing for a set of items at one operation, for 



example if two plates are being drilled and 
reamed for bearings and fixings which must be 
in line, on assembly. The vital point in these 
cases is to ensure that both (or all) parts are 
clamped together sufficiently tightly for them 
to drill as one, otherwise burrs form at the exit 
and entrance of the holes in the individual 
items. These burrs force the items apart and 
this naturally spoils the location accuracy of the 
holes - the holes may not be in line when the 
parts regain their natural shape. 

If a large number of nominally identical 
parts is required it is usually expedient to make 
a drilling jig which has all of the required holes 
carefully marked and drilled. The jig can then 
be used to drill, or centre, all of the holes in 
each component. It is usually convenient to 
drill all of the jig holes to the same small size, so 
that rapid spotting of the holes is possible. 

It is unlikely that a jig will receive much 
repeated use in the amateur’s workshop so it 
can be made from a suitable offcut of mild steel. 
Only if extensive use is envisaged will it be nec¬ 
essary to fit hardened bushes into the jig holes. 
The jig should, of course, incorporate some 
form of positive location for the part to be 
drilled so that a simple placement and clamping 
operation positions the work correctly. 

If two identical (or more-or-less identical) 
items are required in sheet material, the ‘truing’ 
of the edges should be performed with the items 
clamped or bolted together firmly. They should 
then be retained in this condition throughout 
marking, drilling and profiling so that one 
operation at each stage will produce the pair. 

Marking out using hand tools 

Use of rule and scriber 

Having established two reference edges at right 
angles to each other, the most common way of 

marking out work in sheet form is simply by 
use of a square, a rule and a scriber. The 
method has the advantage of simplicity, but 
unless it is carried out carefully, and the mark¬ 
ing afterwards checked with equal care, signifi¬ 
cant errors can be introduced. 

The general requirement is to measure and 
mark a dimension, either from a prepared 
edge, or from an already scribed line. Two 
approaches are possible, as shown in Figure 
5.32. The rule can be laid on the work with the 
engraving for a required measurement placed 
over a reference line, or aligned with an edge, 
as shown at A, and the scribed line made on the 
end of rule. The alternative, shown at B, is to 
use only the edge of the rule. Personal prefer¬ 
ence will dictate the choice. 

Both methods have their disadvantages. If a 
line is scribed on the end of the rule, this usually 
means that the line is displaced due to the thick¬ 
ness of the scriber point, the resultant line being 


Figure 5.32 Two ways in which a scnbed position can be 
marked off using a rule. 





Scnber point 


Scnber point 

Figure 5.33 The effect of the size of the scriber point on the 
marked position, when using the end of the rule. 

Figure 5.34 The square from a combination set provides a 
ready means to mark a position relative to a prepared edge. 

too far from the reference, as shown in Figure 
5.33. So, a scriber with the finest possible point 
is needed in order that the error is small. 

In practice, two considerations apply; the 
error may be small enough for it not to be criti¬ 
cal in any event, or the error may be significant 
but can be accommodated (removed) by offset¬ 
ting the rule in the first instance. This is 
because, once a standard technique has been 
adopted, the error introduced by the scriber 
point tends to be very closely the same for each 
marking operation. So, provided that you 
know what the error is, you can readily com¬ 
pensate by displacing the rule slightly prior to 
scribing the line. 

Scribing against the edge of the rule can also 
introduce an error due to the parallax effect, 
which is illustrated in Figure 5.20. This results 
from the eye not being immediately above the 
engraving on the rule. In this respect, it helps if 
a thin rule is used and this is one reason they are 
often preferred to the rigid type. 

Use of a depth rule 

Measurements relative to a prepared edge can 
most readily be set off by using the simple 
depth rule, if one is available, or a combination 
square. In both cases, the fact that the rule can 
be adjusted in the stock and then clamped into 
position, allows accurate marking to be under¬ 
taken. This operation is illustrated in Figure 

If a line is required parallel to the long edge 
of work which has only been prepared along 
one edge, the process of marking off can be 
repeated at two points and a straightedge and 
scriber used to mark a line parallel to the 
reference edge. Although it should make 
no difference, it is useful to scribe at three 
points as a check on general straightness and 
accuracy - it is only too easy to introduce an 



Jenny calipers (odd legs) 

If the distances are small, lines parallel to a 
prepared edge can be scribed using a pair of 
odd legs or Jenny calipers. For setting, the plain 
foot rests against the end of the rule and the 
scriber point locates in an engraving on the 
rule. Setting to a standard, engraved value is 
simple and straightforward. Once set, the 
caliper may be used to scribe a line parallel to a 
reference edge, and the only point to note is 
that the caliper must always be held squarely 
during setting and use. 

Marking a centre line 

It is frequently necessary to mark the centre 
line of a component. For narrower work, this 
can readily be accomplished by using Jenny 
calipers since, to define the centre, it is only 
necessary to scribe two lines equidistant from 
opposite edges. The centre then lies midway 
between the two lines. It is naturally helpful if 
the two lines are close together, but there is no 
necessity to measure off the distance very 
accurately, since an estimate will do quite 

If the work is too wide to use Jenny calipers, 
other methods, such as the use of a depth rule 
from both edges, can be used to scribe two 
lines, nominally half the width of the material 
from the edges, to produce the same result, any 
errors in the actual measurements not being 
significant as long as they remain the same for 
each scribed line. 


For some marking operations, the rule may 
ultimately be abandoned in favour of some¬ 
thing more straightforward. This is particularly 
the case when a number of equal pitches are 

required, which is frequently for a series of 
fixing holes. Each position must be centre 
punched as part of the marking operation. 
Dividers have two hardened steel legs, united 
by a spring at the top and provided with an 
adjusting screw which allows the distance 
between the pointed ends of the legs to be 
adjusted. Both points are ground sharply to act 
as scribers and they may consequently be used 
for marking out a series of equal pitches, locat¬ 
ing the first hole in the series as dictated by the 
job, centre punching this and then using the 
dividers to mark the next, punching that, and 
so on. 

Dividers can be set to a measured pitch by 
standing one point in an engraved line on a rule 
and ‘feeling’ the second point into another 
engraved mark while adjusting the setting. 
Once set, the dividers may be used as shown in 
Figure 5.35, stepping off the pitches and centre 
punching progressively along the required line. 
One advantage of this is that all pitches will be 
equal (given care in punching the centres) but 
an increasing error can accumulate if the 
setting of the dividers is not correct, and it pays 
to mark in one or two absolute measurements 
on a long line of equal pitches, using a rule, 
starting the stepping off once again at several 

Figure S.35 Equal pitches are scribed by 'walking' the dividers 
progressively along a scribed line, and centre punching the 
positions progressively. 



Figure S.36 Arcs are scribed relative to the marked and 
centre punched centre, using dividers. Note the sparing 
application of marking fluid. 

intermediate points in order to minimise the 
errors. An alternative is to start at the centre 
and work outwards. 

Using the dividers, it is not necessary to use a 
standard measurement taken from a rule, but 
the dividers can be set to provide an equal 
number of pitches within an overall dimension, 
setting them by eye and then ‘walking’ them 
carefully along the desired line by way of trial, 
to determine whether the required pitch has 
been set, making successive adjustments until 
the setting is correct. 

Dividers are also used (naturally!) for 
scribing arcs in the work. For this operation, 
the centre of the arc or circle is marked and 
centre punched, and the arc then scribed, as 
shown in Figure 5.36. 

A practical example 

Figures 5.37 to 5.43 show examples of the 
methods described. A '/sin. (3 mm) mild steel 
plate for a model locomotive frame is being 
marked out. This plate, and its twin, have been 

Figure S.37 The first stage in marking out is to define a 
reference position from which other features may be marked. 

Figure 5.38 The scriber is being positioned relative to the line 
already scribed on the work, at the edge of the rule. 

Figure 5.39 The scriber is held in position and a square 
brought up to touch it so that the position can be marked. 



Figure 5.40 The measurement is checked using the rule. 

Figure 5.43 ... so that the rule can be positioned for scribing 
the position of the top edge of the frame. 

Figure 5.41 A short line is scribed at the required distance 
from the prepared bottom edge... 

bolted together through the nominal positions 
of the axle centres and the lower edges 
machined to create a reference edge. The plates 
have been separated so that a single one may be 
marked out, since it is easier to handle the one, 
rather than the pair with nuts and bolts 
protruding from both sides. Since the plate is in 
black mild steel, it has been given a very light 
coat of grey cellulose primer. 

The first operation. Figure 5.37, is to scribe 
a reference line through the leading axle posi¬ 
tion, right across the plate, nominally bisecting 
the already drilled hole. 

From this position, the distance to the front 
of the frame is marked off, as shown in Figure 
5.38, and the scriber held at the marked point 
while the rule is put to one side and the square 
brought up to the scriber so that the front end 
can be marked, as shown in Figure 5.39. 

The marked distance is checked with the 
rule. Figure 5.40. Notice how the rule which is 
not engraved to the desired precision along its 
full length is again not helpful to this task. 
Although the modeller doing the work thought 
the measurement was good, parallax error 
caused by the photographer’s viewpoint makes 
it appear that the distance marked may be in 

Following the check on the marked lines, the 
top edge of the frame is marked by scribing two 
short lines against the end of the rule. Notice 

Figure 5.42 ... and a second one further along... 


the angle at which the scriber is approaching 
the end of the rule, in order to minimise any 
error, in the views of this operation in Figures 
5.41 and 5.42. 

With two lines marked to define the top 
edge, the scriber point is placed in one of the 
lines and the rule brought up to touch it. The 
rule is then pivoted about the scriber point until 
it is aligned with the second mark also, and the 
line marked. A longer rule would be more help¬ 
ful here, since the 12in. (300mm) rule used 
spans only a short part of the frame, as Figure 
5.43 shows. 

The remainder of the marking-out process 
follows the same pattern, with the hole 
positions being centre punched prior to the 
pair of frame plates being bolted together 
once again, with their lower edges in align¬ 
ment, for hole drilling, sawing and shaping to 
be performed. 

Marking out using a surface plate 


Although the rule-square-and-scriber method 
of marking out is generally satisfactory and will 
serve if carefully performed, there are potential 
sources of error. Most of these can be removed 
by adopting an alternative method which uses a 
flat reference surface from which to mark off 
the required dimensions. 

The reference surface is provided by a spe¬ 
cially made surface plate which usually has a 
finely machined and hand-scraped surface 
which is as near flat as possible and is carried 
on, or integral with, a substantial cast base. 
Usually, the plate is made as a one-piece iron 
casting, but granite surface plates are now quite 
common commercially. How flat the surface is 
depends on how much you want to pay and 
how large a surface is required. A typical 12in. 

by 18in. (300mm x450mm) cast iron plate may 
be flat to within .0004in. (0.01mm) and cost 
between £250 and £300. A good surface plate 
is thus a highly valuable asset and if you have 
one it should be reserved for the precision tasks 
of marking out and measuring, and definitely 
not for supporting work for hammering, 
straightening, banging or centre punching. 

For amateur use, a less-costly alternative is a 
sheet of plate glass. This is ordinarily about 
'Ain. (6mm)- thick and is today usually made by 
the ‘float’ process which more or less guaran¬ 
tees flatness, at least to the sort of tolerances 
which generally apply. Suitably supported on a 
piece of non-warping material such as chip¬ 
board or blockboard, a sheet of glass is a per¬ 
fectly adequate substitute for its cast iron coun¬ 
terpart, the only disadvantage being its fragility 
should something be dropped on it. 

An expensive adjunct to the surface plate is a 
height gauge. This is effectively a vernier cali¬ 
per mounted to a substantial base and provided 
with a scriber which may be set precisely above 
the base by use of the vernier. 

The ability to set the scriber of a height 
gauge precisely above the base indicates the 
way in which the surface plate may be used for 
marking out. If the work has a true surface or 
edge which may be stood on the surface plate, 
the height gauge may be used to scribe lines 
parallel with that edge. The vernier-setting 
capability of the height gauge confers accuracy 
on the process and therefore represents the best 
marking-out practice. 

A height gauge is, however, expensive, being 
a precision instrument with ground and 
polished base, which precisely matches the 
depth of the scribing edge, an offset naturally 
existing between the underside of the base and 
the zero position on the vernier. A 12-inch 
(300mm) height gauge might cost between 
£100 and £200. 

A less-expensive alternative to the height 
gauge is a scribing block, or surface gauge, 



illustrated in Figure 5.44. This comprises a steel 
base, ground flat on the underside, which 
houses an L-shaped arm, to one end of which is 
attached a cylindrical pillar. Fitted to the pillar 
is a two-part clamp which houses a double- 
ended scriber, usually bent at one end and 
straight at the other. The cylindrical pillar is 
adjustable in the end of the L-shaped arm and 
can take any position from the horizontal to the 
vertical. For marking out, its normal position 
is, however, roughly vertical, as shown in Fig¬ 
ure 5.44. 

The L-shaped arm is hinged in the base and 
fitted with an adjusting screw so that fine angu¬ 
lar adjustment of the position of the pillar is 
possible, thereby raising or lowering the scriber 
fixed to the pillar. 

Used on a surface plate, the scribing block is 
capable of scribing lines parallel to the refer¬ 

ence surface, and provided that the scriber 
point can be set to the correct height, and one 
edge of the work established parallel to the 
surface, accurate marking out is possible. 

For most purposes, an ordinary rule pro¬ 
vides adequate accuracy for marking out and 
may be used for setting the scriber point. For 
accuracy, the rule must have its zero in contact 
with the surface plate and it must be vertical (in 
both directions). A simple way to ensure cor¬ 
rect setting up of the rule is shown in Figure 
5.45. Here, the rule is simply clamped to an 
angle plate standing on the surface plate. This 
automatically achieves squareness in one direc¬ 
tion and a square can be used to check square¬ 
ness in the other. Use of a watchmakers’ glass 
allows accurate setting of the scriber point. 
Figure 5.45 shows a 12-inch rule clamped to an 
angle plate. 

Figure 5.44 A scribing block, or surface gauge. 


Figure 5.45 An angle plate used to hold a rule for setting the 
scribing block scriber point. 


Marking sheet materials 

If a flat surface and a scribing block are avail¬ 
able, together with some means of setting the 
scriber point to the required height, the mark¬ 
ing-out operation may be carried out with 
respect to a prepared straight edge on the work 
which is in contact with the surface plate. To 
bring the item to a convenient height, it can be 
placed on parallel packing of known height, as 
shown in Figure 5.46. This shows a ground 
parallel in use as packing, but anything will do, 
provided that it is parallel and of some conven¬ 
ient (and known) height. 

Figure 5.46 shows a small brass plate being 
marked out using a scribing block. The plate 
has been filed to prepare two straight edges 

Figure 5.46 Marking out a small brass plate on a surface 

which are square with one another. One of the 
prepared edges is standing on the ground steel 
parallel, of known height, to bring it to a more 
convenient working position. Lines can be 
scribed on the plate parallel to this edge, and 
the plate afterwards turned through 90 degrees 
to stand on the other prepared edge for the re¬ 
maining lines to be scribed. The marked-out 
plate is illustrated in Chapter 7. 

There is no doubt that this method of mark¬ 
ing out is by far the most satisfactory and accu¬ 
rate. The point of the scriber can be set very 
accurately to the required height against the 
rule by observing with a watchmaker’s glass, 
and the scribed lines are always parallel to the 
reference edge, conferring real accuracy on all 
of the marked positions. The best type of rule 
for use as the reference is clearly one in which 
the whole length of the rule is engraved in the 
smallest divisions with which one needs to 
work, either !4dn. or 0.5mm. The rule illus¬ 
trated in Figure 5.45 has the sole benefit that it 
photographs well, since it has a satin chrome 
finish. It was bought for the metric scale which 
is engraved on the reverse side. 

Marking castings 

By their nature, castings are three-dimensional 
and therefore do not lend themselves to mark¬ 
ing out using a square and rule since they 
frequently do not incorporate sufficiently large 
reference surfaces or edges adjacent to the 
surfaces to be marked. This is illustrated by the 
cylinder block casting shown in Figure 5.47. 
This casting has been prepared by machining 
the underside of the base to produce a refer¬ 
ence surface, having first determined how 
much should be removed to bring the base 
flange to the design thickness. 

Having established one true surface, a sec¬ 
ond has been created by machining the front of 
the block, again bearing in mind the amount to 



Figure 5.47 A part-machined cylinder block casting being 
marked out on a surface plate. 

and then decide how much needs to be 
removed from the first surface to be machined. 
As a preliminary to the machining, it is usually 
necessary to flatten at least one face (filing is 
usually sufficient) to produce a sufficiently 
accurate base from which to set off the first cut. 
A case in point is shown by the two steam chests 
in Figure 5.48. Since machining at the first 
stage was simply to bring them to thickness, no 
preliminary marking out was performed, once 
the machining allowance was determined by 
use of a rule. 

be removed to bring the casting to the correct 
overall length by ultimately machining the rear 
end. With two adjacent faces, at right angles 
to each other, machined flat, the casting can be 
set up as shown in the figure for marking out 
for hole centres, overall length and height, and 
so on. 

As surfaces on the casting are machined pro¬ 
gressively, further marking out can follow and 
the process becomes one of marking and ma¬ 
chining alternately. 

Some machining operations on castings may 
not actually require preliminary marking out, it 
being sufficient to ‘run a rule over’ the work 

Figure 5.48 A pair of steam chest castings which have been 
machined on both sides without benefit of marking out. 



Basic handwork 


In the amateur’s workshop there is relatively 
more handwork carried out than would be the 
case in industry. Much of this occurs because 
the amateur is not equipped to deal with larger 
items which, in industry, would be profiled by 
machining. These larger parts include such 
items as locomotive frames and bodywork, the 
larger clock plates, long rods or levers which 
need to be profiled, sheet materials for boilers 
and so on. Anything large enough to take it 
beyond the capacity of the amateur’s machines 
is worked by hand, even cutting a small piece 
out of a large sheet. 

Conversely, small items are also frequently 
cut out and profiled by hand, perhaps due to 
the difficulty of holding such items for machin¬ 
ing, or simply because the machining capability 
is again not available. For these reasons, under¬ 
taking the basic shaping by hand is often the 
only option which is available, making it essen¬ 
tial to acquire reasonable skills in the basic 
handwork of sawing and filing. 

The normal progression of work is naturally 
from selection of the basic material, through 

marking out, to sawing, filing and final clean¬ 
ing up or polishing, and the processes can, 
therefore, be dealt with in this order. Before 
these processes are performed on some types of 
stock steel, it is highly desirable to carry out 
some initial heat treatment, however, and this 
can be considered first of all. 

Initial preparation of materials 

Heat treatment of steel sections 

Chapter 3 provides a description of the more 
common materials and includes a description 
of the way in which internal stresses may be 
built up in the material during its manufacture 
or preliminary cutting. Provided that the mate¬ 
rial is not ‘worked’ in any way, internal stresses 
are not usually evident so that a bar of bright 
mild steel (BMS) although internally stressed, 
will not distort further as long as it remains 

Removal of material, particularly from 
rectangular-section BMS, removes part of the 



structure that is essential to the stability of the 
bar’s shape, thus allowing it to distort. This is a 
nuisance to say the least, but it can be avoided 
by ‘normalising’ the material before any opera¬ 
tions are carried out. Normalising equalises the 
internal stresses and allows material to be 
removed without distortion of the remainder. 

In addition to normalising BMS, it is fre¬ 
quently useful to carry out the same process on 
sheet materials which have been cut off by guil¬ 
lotining. This process usually causes some 
bending of the cut strip since the guillotine 
blade normally acts like a pair of scissors, 
cutting progressively from one end to the other 
and bending the sheared strip downwards, 
away from the fixed blade. The result is a bow 
in the material along its length, and perhaps, 
also, a degree of twist. 

The problem frequently occurs when 
sheared plates are provided for locomotive 
frames. Since the frames are to be paired, it is 
often recommended that they should be used so 
that the bowed shapes complement one 
another, both bowing outwards or inwards, 
whichever produces the straightest result. This 
can be tried by bolting the two plates together 
through the axle centres, for initial marking 
out, drilling and shaping. Unfortunately, this 
doesn’t always produce frames which are 
straight, and the technique cannot in any event 
be used if you need a single piece of plate for a 
frame stay, or whatever, so the best way is to 
normalise the plates before commencing work. 

The process is very easy to perform, simply 
requiring the material to be heated up and then 
left to cool slowly. A fuller description of the 
normalising process is given in Chapter 3. 

Preliminary preparation of sheet materials 

Some preliminary preparation of sheet materi¬ 
als is generally required prior to marking out 
the work - this applies also to castings. This 

preparation is usually to bring at least one edge 
(for sheet materials) or one surface (for cast¬ 
ings) to a straight and/or flat condition. Since 
castings are distinctly different from sheet 
materials, they are considered separately 

The first point of note concerning sheet 
materials is that they are ordinarily cut into the 
required sizes using a guillotine. While in 
theory, the guillotining process provides a clean 
cut, in practice there is normally some distor¬ 
tion of the material along the cut edge. This 
renders it useless for modelling since the guillo¬ 
tined edge usually has the appearance shown in 
Figure 6.1, one surface having a rounded 
profile and the other showing a ‘pulled’ or 
extruded appearance. In model work we are 
representing a scale appearance, and any gross 
distortion of the edge of the sheet is generally 
not tolerable, except perhaps for ‘hidden’ work 
such as model boilermaking, when much of the 
structure will be hidden by cleading, and even if 
not, some out-of-scale appearance may be 
acceptable. In general, therefore, the edges of 
guillotined sheets must first be cut back to 
obtain a good edge which is then machined or 
filed to create a square and straight edge prior 
to marking out. 

If sheet work has a reasonable length in both 
dimensions, it is usually convenient to prepare 

Cut away this material before use 

Figure 6.1 Deformation on the edge of a guillotined sheet. 



two adjacent edges, by machining or sawing 
and filing, to establish them at 90 degrees to 
each other with both edges straight and square 
to the surface of the sheet. The major dimen¬ 
sions of the work can then be specified with 
respect to the two adjacent prepared edges, 
which are used as references. If the two edges 
are at 90 degrees, a square placed on one edge 
allows a line to be scribed parallel to the other 
edge, and measurements along such scribed 
lines will define the required positions. 

If sheet work is large, or long and narrow, a 
sufficiently large square may not be available to 
span the job and alternative means must be 
adopted. Large work can be accommodated by 
clamping a straight edge (or suitable length of 
stock rectangular bar) to the work, using as 
large a square as possible to establish the square 

For long and narrow work, such as locomo¬ 
tive frame plates, it is usually best to work 
solely from one of the long edges, establishing 
lines parallel to this using one of the methods 
described in Chapter 5, and scribing lines at 
right angles to the edge by use of a square. The 
short ends can then be brought to the finished 
size and shape after marking out. 

Saws and sawing 


The basic metal-cutting saw is the familiar hack¬ 
saw fitted with a replaceable blade. Saw frames 
are produced in a range of different styles, three 
of which are shown in Figure 6.2. Which type to 
choose really depends on personal choice, but if 
you have to make do with only one, make sure it 
is adjustable to accommodate blades of differ¬ 
ent lengths. If a single saw must suffice, the 
traditional type with a handle in line with the 
blade is probably most useful. 

Figure 6.2 A selection of hacksaw frames. 

The replaceable blades for these saws are 
punched with holes which engage pins in 
square-section fittings at each end of the saw 
frame. Blades are manufactured in a range of 
different lengths and types, some being all car¬ 
bon steel (the cheapest), others all high-speed 
steel, some with hardened teeth on a relatively 
flexible (and soft) backing, some even with 
tipped teeth. Lengths vary from 9in. to 12in. 
(230mm to 300mm) and tooth pitches from 
around 14 to 32 teeth per inch (25mm). 

Since there is such a choice of blade avail¬ 
able, even the simple hacksaw needs to be set 
up and used in the correct manner and since 
this fundamental tool also comes in other 
forms, a general introduction is in order. 

Choice of tooth pitch 

First of all, a consideration of the relationship 
between the saw blade and the material being 
cut is necessary. Figure 6.3 shows two views of 
a saw blade in the process of cutting the work. 
At A, the pitch of the saw teeth allows three 
points to be in contact with the work, whereas 
at B, the tooth pitch is so large that the material 
being cut can be accommodated between two 
points. In this case, the saw can fall down into 
the work, causing it to jam against the leading 



Figure 6.3 The relationship between the saw teeth and the 

edge of the material. The resulting ‘collision’ 
may be sufficient to break off a tooth, and thus 
ruin the blade, but even if this does not occur, 
sawing is very hard work and the jam-ups may 
well bend the material if it is thin and/or soft. It 
is essential that the pitch of the saw teeth is 
chosen to suit the thickness of material being 
cut. It is usually recommended that three teeth 
should be in contact with the cut material, and 
the pitch of the teeth (and the method of 
approach to the work) must be chosen with this 
in mind. 

Material thickness is not the sole considera¬ 
tion, however. When a material is cut, the form 
of the chippings, or swarf, is dependent on the 
way the material cuts. Brittle materials tend to 
chip, whereas greater ductility produces long 
‘shavings’ of material which may clog up the 

spaces between the teeth if these are small. The 
quantity of material cut on each stroke is also 
significant since even small chippings may clog 
the saw if they are produced in large quantities. 
A tooth pitch to suit the material is also 
required, the usual recommendations being: 

• for cast iron 14 teeth per inch 

• for mild steel 18 teeth per inch 

• for brass and copper 20 to 24 teeth per 


Harder steels can be cut with a saw having a 
finer tooth pitch than that used for mild steel. 

The above recommendations are not univer¬ 
sally applicable however, since the work may 
need a fine-pitch blade in order to maintain 
three teeth in contact with the cut surface, and 
material thickness must be considered, rather 
than material type, in making the selection of 
tooth pitch. 


Figure 6.4 Good and bad ways in which a saw blade can 
approach the work. 



The approach to the work must also be 
varied to maximise tooth contacts and Figure 
6.4 shows both good and bad approaches to 
three types of work. The solution for angles is 
obvious; they should always be approached so 
that the cut face presents its long dimension to 
the saw. The problems posed by thin materials 
when supported vertically in the vice are well 
illustrated, and this suggests that cutting sheet 
materials is normally better served by holding 
them down on to the bench surface, rather than 
in the vice, when the only good approach is 
from a kneeling position below the vice, which 
is hardly conducive to comfort! 

Tubes, particularly if thin-walled, present 
particular difficulties. The saw may jam on the 
thin material in any event, but additionally, 
there is always one position in which the saw 
jams on both sides at once. If the tube is small 
and in a soft condition, this usually means a 
bent tube, but whether soft or not, sawing does 
not proceed readily beyond this position. 
The remedy is to saw around the tube rather 
than across it, but if the tube is suitable, a prop¬ 
erly designed tube cutter is usually a better 

Sawing action 

The sawing action may appear self-evident, 
after all, it just needs to be pushed backwards 
and forwards. And so it does. But how can you 
ensure that the cut material is expelled from the 
saw and from the slot (called the kerf) which it 
cuts? At what speed should the saw move and 
how can blade life be maximised? In answering 
these questions the sawing action is seen to be 
not so simply defined after all. 

First, the question of blade life. To minimise 
wear, all teeth should do their fair share of 
cutting. This means using long strokes to 
ensure even wear throughout the length. A saw 

blade ideally needs a backing for the teeth 
which is slightly narrower than the teeth them¬ 
selves so that the body of the blade slides freely 
through the kerf (slot) which the teeth produce. 
To simplify the manufacturing process, this 
effect is produced either by ‘setting’ alternate 
teeth outwards in opposite directions or by 
forming a wavy edge to the line of teeth so that 
some project to the left, and some to the right, 
progressively down the blade. 

Teeth naturally wear where they do most 
cutting and if the saw is habitually used by 
making short strokes using just the centre of the 
blade, this naturally wears the centre but not 
the ends. Once the blade has worn near its cen¬ 
tre, a stroke longer than usual causes unworn 
teeth at the ends to jam in the kerf and the cut¬ 
ting process is interrupted. This process repeats 
itself frequently and sawing becomes a frustrat¬ 
ing business. The prime requirement is, there¬ 
fore, for full strokes from the saw. 

The next questions are the speed of sawing 
and the pressure to be applied. Firstly, the saw 
must be allowed to cut for itself and sufficient 
pressure should be applied only to prevent the 
saw from sliding over the work on the forward 
stroke (saws of this type should always cut 
going away from the operator). Concentration 
should be applied to guiding the saw along the 
line rather than forcing it to cut by the applica¬ 
tion of pressure. Secondly, the saw should cut at 
about 60 feet (18 metres) per minute. Thus, if 
the blade is 12in. (300mm) long, a full stroke 
should occupy one second - sawing is not a 
rapid process and should not be hurried. 

Finally, there is the question of clearing the 
teeth, and the kerf, during sawing. If the work 
is thin, this usually presents no problem, but for 
thicker work the sawing action should be such 
that the ‘driving’ hand is lowered slightly 
towards the end of the stroke. This lifts the 
front of the blade within the work and allows 
room for the swarf to be expelled. 



Fine-pitch saws 

Replaceable blades for the types of hacksaw 
frame shown in Figure 6.2 are available with 
tooth pitches up to 32 teeth per inch. For very 
thin materials, or intricate work, something 
much finer is required, preferably mounted 
into a lighter frame, two types of which are 
shown in Figure 6.5. The first of these is the 
familiar junior hacksaw with its shaped 
‘springy’ frame which can be squeezed up by 
hand to remove and replace the blade. Only a 
single type of standard blade is available for the 
junior frame - this is 6in. (150mm) long and 
comes in a standard pitch of 32 teeth per inch 
and is ideal for the small, light jobs around the 

The second type of saw frame in Figure 6.5 
is a jewellers’ piercing saw. This is the metal¬ 
workers’ equivalent of the woodworker’s 
fretsaw, but intended for very light work. It 
employs replaceable, fine-pitch blades which 
are only a little deeper than they are wide and 
they are capable of sawing very small radius 
curves or even being turned sharply through 90 
degrees during the sawing process. 

The piercing saw shown is adjustable and 
blade tension is achieved by sliding the handle- 
and-clamp along the frame and then locking it 
by use of the wing nut. The blade should not be 
adjusted ‘bow-string tight’ but just sufficiently 

Figure 6.5 Junior hacksaw and piercing saw frames. 

to take up any slack. One advantage of the 
adjustable frame is that blades with broken-off 
ends can still be utilised. Less costly, non- 
adjustable frames are also available in which 
blade tension is adjusted by ‘springing’ the 

The method of use of the piercing saw is 
quite different from that employed with other 
types of saw. The very fine blades, which range 
from 32 to 80 teeth per inch, are not strong 
enough to support the weight of the saw and 
the force of cutting if the saw is used in the 
conventional horizontal position. The saw is 
consequently used with the blade vertical and 
the work supported on a table having a vee- 
shaped notch cut in it, as shown in Figures 6.6 
and 6.7. The blade carries the stress of cutting 
along its length and is inserted into the frame so 
that it cuts going downwards, the table 
supporting the work to resist the force of cut¬ 
ting. No saw should be forced into the cut, but 
this is especially true of the piercing saw due to 
the extreme fineness of the blades. An illustra¬ 
tion of work which can be produced is shown 
in Figure 6.8. 

Piercing saw blades are manufactured with 
tooth pitches from about 32 to 80 teeth per 
inch, in 12 grades referenced from No. 6 (the 
coarsest) to No. 6/0, the blade thickness 
decreasing as the tooth pitch decreases. A No. 6 
blade has 13 teeth per cm (33 per inch) and is 
0.5mm, or .020in. wide, whereas a No. 6/0 has 
29 teeth per cm and is 0.22mm, or .09in. wide. 
In the centre of the range, the Nos. 1 and 1/0 
have 21 and 22 teeth per cm (54 and 56 per 
inch) and a blade width of about .013in. 

When making abrupt changes of direction 
with a piercing saw, the saw must be kept on the 
move but with absolutely no pressure applied 
in the cutting direction. Instead, the saw is 
rotated slowly to widen the kerf until it has 
changed position to face in the required dir¬ 



Figure 6.6 A piercing saw on the notched table with which 
it is used. 

Figure 6.7 A piercing saw in use. 

Figure 6.8 Locomotive frames for a 4mm scale model, cut 
from I mm brass using a piercing saw. 

Pierced work 

The name of the piercing saw comes from the 
fact that it can produce ‘pierced’ work. A hole 
may be drilled in the work to allow the blade to 
be passed through and afterwards clamped into 
the saw frame to allow the cut to commence 
within the work. 

Pierced work is frequently required in more 
substantial material than that for which the 
piercing saw is designed. The hacksaw can be 
adapted for this work by fitting to it a saw blade 
of circular cross-section which can cut in any 
direction. The most commonly available type is 
the Abrafile, a trade name now used universally 
for this type of saw. 

Figure 6.9 shows one end fastening of the 
file in a hacksaw frame. An adaptor clip fits to 
the blade mounting pin and allows a button on 
the blade end to be retained in a slotted angle. 
The file can then be inserted into the clips and 
tensioned in the normal way. 

The files are around Zttin. (1.5 mm) in diam¬ 
eter and may be used for any intricate work in 
substantial material. Files are available in a 
range of tooth pitches, usually described as 
coarse, medium or fine, and are produced in 
different lengths, although the most common is 
the 9in. (230mm). Like the normal hacksaw 
blade, the files should always be inserted into 

Figure 6.9 An adapter clip for fitting an Abrafile blade into a 
standard hacksaw frame. 



the frame to cut when going forwards, the 
direction of the cut being determined by sliding 
the file through the fingers (gently!) to decide 
which way the teeth face. 

The files are hard, but relatively brittle, and 
can easily be broken if too much pressure is 
applied, or if they jam in the kerf. The sawing 
action needs to be gentle and great attention 
must be given to keeping the saw absolutely on 
the same alignment during the whole stroke if 
blade breakages are to be avoided. 

Cutting large sheets 

The hacksaw frames shown in Figure 6.2 utilise 
square-section bolts with inserted pins as the 
means of holding the blade into the frame. The 
bolts may be inserted in four distinct positions, 
allowing the blade to lie in the plane of the 
frame, or at right angles to it. Long work may, 
therefore, be sawn conveniently as shown in 
Figure 6.10, provided that the cutting line is 
within reach of the frame depth. 

If a saw with the blade turned over in this 
way cannot reach the position of the cut, a spe¬ 
cial holder, known as a pad saw handle, may be 
used. This attaches to one end of the blade, 
leaving the other free to enter a sheet of mate¬ 
rial at any position. Since a standard blade is 

Figure 6.10 A hacksaw blade turned through 90 degrees for 
cutting parallel to an edge on the work. 

not designed to be pushed into the cut (it is not 
stiff enough) the blade must be inserted so that 
it cuts when pulling. A regular hacksaw frame is 
used with the points of the blade teeth pointing 
away from the handle and is sufficiently robust 
(or should be) so that a ‘push’ on the handle 
reaches the blade as a ‘pull’ from the front. Cut¬ 
ting with a pulling action with a pad saw is very 
awkward, but it is sometimes the only alterna¬ 
tive which is available. 

To enable the saw blade to enter the sheet for 
commencing the cut, a slot is created by drilling 
a line of closely spaced holes adjacent to the 
cutting line and joining them by filing away the 
material between them so that the saw blade 
will enter to commence the cut. If holes are 
drilled all along the cut, the amount of work 
which the saw must do is reduced, and this 
helps to reduce the awkward work of sawing 
with the cut taken on the pull stroke. 

If a line of holes is drilled all along the 
cutting line, it may well be possible to complete 
the cut by using a cold chisel to cut through the 
remaining material. This is carried out by 
supporting the work on a substantial block and 
hammering the chisel down into the webs of 
material remaining between the holes. For soft 
materials such as brass and aluminium alloys, a 
chisel having a narrow angle at its tip may be 
used, but for steel sheet, a larger tip angle is 
required (to give adequate strength) and the 
distortion that this produces should be borne in 
mind when drilling the initial holes. Do not 
despise the humble chisel - it is a very effective 
tool and may well solve the problem of cutting 
large sheets of thin materials. 

Alternative cutters, known as nibblers, are 
available in both hand- and power-operated 
form. These remove a narrow strip of material, 
up to '/sin. (3mm) wide, depending upon the 
design, by utilising a ‘pecker’ or ‘beak’ which is 
drawn up between the two jaws of an anvil. 
Beak and anvils thus act as two guillotines, 
removing the narrow strip of material. 



Provided that the cutting edges are not worn, 
there is little distortion of the parent material. 
The cutter is capable of cutting curves and 
changing direction quite sharply, and is very 
versatile in use. Types sold as attachments for 
DIY electric drills will usually cut steel up to 18 
swg (.048in. or 1.25mm) and light alloys up to 
'/i6in. or 1.5mm. 

Reducing noise 

Noise is sometimes a problem when cutting 
large sheets since the small area gripped by the 
vice allows most of the sheet to vibrate. This 
can sometimes be overcome by clamping the 
sheet to the apron of the bench, if it has one, or 
by hanging a heavy cloth such as a towel over 
the unsupported part. It is frequently useful, 
however, to clamp a heavy bar to the work, 
parallel with, and adjacent to, the cutting line. 
The bar then provides sufficient mass to damp 
any vibration and the work and bar may be 
quickly repositioned in the vice as the cut 

Files and filing 


The ability to file surfaces flat, or nearly so, is 
an undoubted asset, and some of the ‘old boys’ 
brought up in the days when hand fitting was 
more widely practised than it is today, can 
produce reasonably flat surfaces, seemingly 
with little effort. In those days, files were used 
as the intermediate stage in a chiselling, filing 
and scraping process which produced both flat 
and curved surfaces which matched one 
another or matched a standard reference, such 
as a flat surface plate. 

Figure 6.11 The tang end of a double-cut flat file. 

Files were also at one time used for filing 
hexagons, squares and approximately circular 
shapes out of forged or cast (but roughly 
shaped) billets. A great deal of expertise in the 
use of the file was needed and this was the first 
skill which the apprentice was expected to 

Files are available in a range of grades and 
shapes to suit the finish required and the size 
and shape of the work. The normal grades are 
known as coarse, bastard, second-cut, smooth 
and dead smooth, these terms describing the 
size of the teeth on the file. 

File teeth are cut by raising ridges across the 
body. The ridges lie at about 25 or 30 degrees 
to the long axis of the file and if only a single set 
of ridges is cut, each forms a long cutting edge 
across the file. Most files are double-cut, how¬ 
ever, two lines of ridges lying at 50 or 60 
degrees to one another. This is easily discerned 
at the handle, or tang, end of the file, as shown 
in Figure 6.11. 

The double-cutting of the file produces indi¬ 
vidual teeth on the body. This allows swarf to 
clear the teeth more readily and produces a 
cleaner cutting action which allows more mate¬ 
rial to be removed. Most files are of the double¬ 
cut type. 



Standard file types 

The commonest file shapes (cross-sections) are 
flat (rectangular), square, three-square (trian¬ 
gular), round and half-round. Most files are 
tapered in their length, being smaller in the 
point than in the heel (tang end), but rectangu¬ 
lar-section files known as hand files normally 
have the same cross-section throughout. A 
rectangular cross-section file with a taper in its 
length is described as a flat file. A variant of the 
flat file, known as a warding file, is thinner in 
relation to its width and more suitable for 
working in narrow slots. These are sometimes 
known as pillar files. A three-square file not 
having equal sides, but having teeth cut on two 
sides having a very shallow included angle, is 
known as a knife shape. 

Square files have teeth cut on all four faces 
and are intended for cutting into corners or 
cutting square shapes. A flat file also generally 
has teeth on all four edges but a hand file is 
normally provided with one ‘safe’ or uncut 
edge so that it may be used in a corner without 
significantly removing material from the side. 
With the exception of the knife shape, other 
shapes have teeth cut all round, or on all sur¬ 
faces. The more common file shapes are shown 

in Figure 6.12. 





/' //4 \ 
l J 

**//f !>' 




Figure 6.12 Common file shapes. 

Needle files 

The files described above are available in 
lengths from 4in. to 12in. (100mm to 300mm) 
in increments of 2in. (50mm). A 12-inch hand 
file is typically l!4in. (32mm) wide and %an. 
(7.5mm) thick and a 6-inch (150mm) might be 
s /8in. (16mm) wide and about s /32in. (4mm) 
thick. A 4-inch (100mm) file is somewhat 
smaller than this, but still not suitable for 
extremely fine work. 

A range of miniature files is needed, which, 
because of the form of some shapes, are 
described as needle files. These are available 
having the same or similar shapes to those 
described above. Each file is a one-piece forg¬ 
ing incorporating both the file blade and a 
small-diameter handle. The overall length is 
usually about 6in. (150mm) divided 50:50 
between handle and blade. Although different 
grades are available these are often not speci¬ 
fied, so it pays to browse around a little when 
purchasing these files. 

On some needle files the handle is plain and 
the file may be held in a long pin vice, which 
makes for a more comfortable grip than the 
small-diameter forged handle. On other files 
the handle is provided with knurled rings, 
either to improve the grip directly or to permit 
a push-on handle to be fitted. 

Figure 6.13 A group of needle files compared with a 4 inch 
(100mm) smooth file. 



With the exception of the hand shape, nee¬ 
dle files taper to a fine point. This is illustrated 
in Figure 6.13 which also shows a 4-inch 
(100mm), dead smooth file. Needle files are 
commonly referred to as Swiss files, whether 
originating in that country or not. 

Specialised files 

Related to the needle files, there are various 
other types of small file, described as precision 
files, escapement files or rifflers. Riffler files 
comprise a square-section forging, about 6in. 
(150mm) long, with file teeth cut on both ends, 
each of which is forged into a curved shape. 
Different sections are available, such as round, 
half-round, square, three-square and flat, 
which, with the short, curved ends allow access 
with a file to otherwise difficult places. 

Precision files are more finely machined and 
cut than their general counterparts, and are 
available in a very wide range of cuts at the finer 
end of the scale, from about 23 to 300 teeth per 
inch. Being this fine, they are not available in the 
larger sizes, but something up to an 8-inch, hav¬ 
ing 250 teeth per inch can be obtained. 

Escapement files are a variant of the stand¬ 
ard type of needle file which have square 
handles rather than the more usual round type. 

Some specialised shapes of file are also avail¬ 
able in the smaller sizes (up to 6in. or 150mm). 
A crossing file is radiused (like a half-round) on 
both sides, usually a different radius on the two 
sides, making it suitable for working into 
corners. Also useful for this task is the barrette 
shape which is something like a half-round, but 
having teeth cut only on the flat face. 

Using a file 

Before use, a file must always be fitted with a 
handle. Plastic handles are nowadays available 

which incorporate a device for gripping the 
tapered tang. The traditional handle, and still 
the cheapest, is a wooden turning fitted with a 
metal ferrule. An appropriate size should be 
chosen from the available 3-, 4- or 5-inch types. 
As supplied, these are bored centrally with a 
pilot hole and are fitted to the tapered tang by 
being wedged into position. 

The simplest way to do this is to heat up the 
tang (it doesn’t need to be red hot) insert it into 
the handle and allow a tapered hole to develop 
as the tang burns into the handle under pres¬ 
sure. If the tang burns its way into the handle 
almost, but not quite, far enough, the two can 
be separated to allow tang and socket to cool. 
Afterwards, the two can be jammed firmly 
together by fitting the handle and then drop¬ 
ping the assembly, handle end down, onto a 
solid support such as the bench top, the vice or 
the floor. 

Once a file is fitted with a handle, it is ready 
for use, but it needs to be applied to the work in 
the correct way. On a double-cut file, the 
cutting edges form lines of teeth on the faces. If 
the file is pushed straight across the work, suc¬ 
ceeding teeth tend to follow the grooves cut by 
those that lead and the result is a scratched and 
poor quality finish. The correct filing action is 
obtained when the file moves both forwards 
and sideways, as shown by Figure 6.14. At A, 
the file moves forwards and rightwards, and at 
B, forwards and leftwards. In practice, it is best 
to alternate the directions shown at A and B, 
making a few strokes leftwards followed by a 
few strokes rightwards. 

It is naturally necessary that the file should 
always traverse the work in the same plane and 
the action must be practised to acquire the skill 
to apply the correct amount of pressure to the 
driving and leading hands during each stroke, 
to keep the file level. The natural tendency is to 
produce a ‘crowned’ surface and this trait 
should be corrected until a flat surface can be 



Figure 6.14 The correct filing action. 

Like sawing, filing should be slow and lei¬ 

Draw filing 

Since the cuts in the working faces are generally 
made at 25 to 30 degrees to the axis of the file, 
cutting can take place even if the file is moved 
at right angles to the long axis, because the 
teeth are partly facing the edge of the file. If the 
file cuts in this manner, the teeth marks in the 
work are less obvious and the finish on the 
work is finer. The technique of using the file in 
this manner is known as draw filing. 

Figure 6.15 Draw filing. 

To use it for draw filing, a file is grasped in 
both hands, around the point and the handle. It 
is then stroked backwards and forwards along 
the work, as shown in Figure 6.15. If the file is 
double-cut, it removes material going in both 
directions and cuts more effectively than might 
be expected. Since the usual aim in filing is to 
produce a fine finish, draw filing is usually 
performed using a smooth or dead smooth file, 
as the finishing filing stage. 

Clogging of file teeth 

When working with soft materials such as alu¬ 
minium and copper, the chippings can become 
embedded in the file teeth. This reduces the 
cutting efficiency and also produces scratches 
as chippings become trapped between the teeth 
and the work and get pushed over the surface. 
Should a build-up of particles occur, they must 
be removed, for which a file card is used. 

A file card comprises a substantial fabric 
backing into which stiff steel tines are inserted. 
The material is ordinarily used for carding 
wool prior to spinning, but the sharp tines illus¬ 
trated in Figure 6.16, will also remove most 



Figure 6.16 A file card. 

embedded chippings from a file. To prevent the 
tines being pushed out of the fabric backing, 
the piece of card should be mounted on a 
wooden backing, but my sample continues to 
lack this refinement. 

Any chips which adhere stubbornly to a file 
after the use of the card can be removed using a 
sharp point to pick them out, or by filing an 
offcut of sheet brass. 

One way to prevent the build-up of swarf in 
the file is to fill up the teeth with some readily 
removed substance. If a block of French chalk 
is rubbed up and down the file it tends to pack 
into the teeth and thus prevents the swarf or 
chippings becoming embedded. Filling up the 
teeth spaces in this way does also affect the cut¬ 
ting action of the file since the teeth project less 
and therefore take a finer cut. 

Filing straight and flat work 


One of the most common tasks is that of clean¬ 
ing up a sawn edge and filing down to a scribed 
line. If the sawn edge to be filed is broadly par¬ 
allel to the scribed line, and the work is wider 
than the file itself, there is no need to force the 
file, it can be allowed to lie naturally on the 

work as the file is worked progressively along 
the edge. The ends of the edge do need to be 
approached with care, however, since the file 
tends to ‘fall off’ the end and produce a 
rounded, rather than flat, surface. 

Narrow work which the file overlaps needs a 
slightly different technique and the file may 
need to be twisted slightly to correct any lack of 
parallelism with the scribed line, during each 
cutting stroke. 

If the sawn line is a series of humps and 
hollows, work must start by attending to the 
high spots. These may be quite localised, and 
the File needs to be held very firmly to prevent it 
lying on the sloping side of a high spot thereby 
removing material from the adjacent hollow. At 
this stage, it is not desirable to use the normal 
forwards-and-sideways motion until the edge 
has been roughly brought to straightness. 

Filing ordinarily starts with a coarse file and 
progresses towards a fine file. Coarse and fine 
are relative terms however, since what will do 
for removing'/i«in. (1.5mm) from the edge of a 
'Ain. (3 mm) mild steel plate is not at all appro¬ 
priate for cleaning up the edge of a sawn-out 
component in 20 swg (1mm) brass. 

It does help in producing a straight surface if 
a wide file is used rather than a narrow one, and 
a selection of long (and wide) files down to at 
least a second-cut, will be found to be useful. 
One or two 6-inch (150mm)dead smooth files 
should, however, meet the need for the finest 

Testing the work 

As filing progresses, it is necessary to check that 
the surface produced is at right angles to (or the 
correct angle to) the already prepared surfaces, 
and is straight, or curved appropriately in its 
length. Curved work is dealt with below so the 
initial consideration can be of straight and 
square work. 



Squareness of the filed surface to an existing 
face should be checked as filing proceeds. 
Squareness is checked by using an engineer’s 
try square. The check is performed by placing 
the stock of the square firmly against the refer¬ 
ence face, as shown in Figure 6.17, and resting 
the blade on the edge or surface being pre¬ 
pared. Holding the work and square up to the 
light, reveals any out of squareness and shows 
which way the filing action needs to be adjusted 
to achieve a square edge. 

The stock of the square should be undercut at 
the root of the blade, as shown in Figure 6.17, 
so that any burr at the filed edge does not pre¬ 
vent the stock from lying in full contact with the 
reference face. In this respect, it is usually best 
to use as the reference the face with the marked 

Figure 6.17 Testing squareness using a small try square. 

line scribed on it. This should be nearest to the 
operator when filing (so that the line is visible). 
The major burr occurs on the far side since it is 
there that the file tends to ‘turn over’ the edge. 

Even though the filed edge is at right angles 
to the reference surface, it may not actually be 
straight in its length. Straightness should prop¬ 
erly be judged using a straightedge, but if one is 
not available, the blade of a try square will 
serve for most purposes, or even a rigid 12-inch 
(300mm) rule. It goes without saying that the 
straightedge, or its substitute, must be longer 
than the work. 

To test straightness, the reference edge is 
placed along the filed edge and it and the work 
held up to the light. It is then necessary to note 
or mark the high spots and file these down as 
part of the process of using finer files progres¬ 
sively to achieve a straight and square surface 
or edge. 

As the line (and squareness and straightness) 
are approached, finer files are used, and the 
burr consequently reduces in size. Even so, as 
completion is approached the remaining burr 
should be removed by draw-filing, first of all, 
along the edge which is being worked on, and 
then along the surfaces at right angles to it. 

A single operation along the two surfaces 
doesn’t always remove all of the burr which 
remains but can simply ‘turn it over’ onto the 
adjacent surface, so the process has to be 
repeated once or twice until the burrs are com¬ 
pletely removed. A burr is also produced on the 
edge nearest to the operator, but this naturally 
cannot be removed until the scribed line can be 
dispensed with. 

Flattening and testing large surfaces 

The filing and testing of large surfaces can be 
accomplished by adopting the technique used 
for edges of sheet materials, as described above, 
provided that there is a suitable reference 



surface against which the stock of the square 
can be placed. Without such a reference, it be¬ 
comes difficult to detect any twist in the filed 
surface since this does not become evident sim¬ 
ply by checking with a straightedge that the 
work is flat along particular lines. 

If the surface of a casting is being prepared 
as a preliminary to machining, there may be no 
reference available from which to make an 
assessment of the surface being worked on, so 
the surface must be judged against some other 
reference. For flat surfaces, this is ordinarily a 
surface plate, but the alternative sheet of plate 
glass can be used, or even the table of the drill¬ 
ing machine. 

Assuming that a casting is being filed to 
produce a flat surface on which the item may be 
clamped to a machine, the first stage is to file 

Contact areas 


Reference surface 

off any surface bumps or moulding ‘flash’ to 
produce a smooth and clean surface. This 
surface can be tested by standing it on some¬ 
thing which is nominally flat, and checking for 
any instability, or rock. This test will reveal any 
significant high spots, which can be attended 
to, progressing at least to a second-cut file as 
the work becomes flatter. 

Even when the work will stand on a flat sur¬ 
face without rock, this does not guarantee flat¬ 
ness, since the situations shown in Figure 6.18 
may apply. Three points of contact between the 
work and the reference surface provide a stable 
condition although pressing on an unsupported 
corner should reveal its ‘un-flat’ state. 

If the high spots on the work are more 
numerous, and do lie in the same plane, the 
work may stand on its four corners, and 
although it is not flat may appear to be so, due 
to its stable condition on the surface plate. The 
actual contact areas need to be highlighted in 
some way. The method to use depends on the 
accuracy required, and that is naturally dep¬ 
endent on the quality (flatness) of the reference 
surface and the required result. 

Figure 6.19 shows a feeler gauge in use to 
test the surface of a casting. Gaps between the 
underside of the casting are detected by prob¬ 
ing with a thin feeler and the magnitude of the 

Figure 6.18 Illustrating how an apparently flat surface may 
have high spots. 

Figure 6.19 Using a thin feeler gauge to locate high spots and 
hollows on the underside of a casting. 



gap can be assessed by determining the largest 
feeler gauge which will enter the gap. 

The above supposes that there is not com¬ 
plete contact between the casting and surface 
plate all round the edge, thus allowing the 
feeler blade to enter. If there is complete 
peripheral contact, the method cannot be used, 
in which case, a marking medium must be 
employed. This requires a thin coating of a 
greasy marking medium to be applied evenly to 
the surface plate before standing the work on it. 
High spots on the work then pick up the 
marker colour, revealing the areas from which 
material needs to be removed. 

A thin coating of marking medium (generally 
called marking blue, or engineers’ blue, since 
this is usually the colour used) is necessary, par¬ 
ticularly as flatness is approached, otherwise 
the medium bridges the gaps and shows all over 
the surface, giving a false indication. 

The method is usually employed when a 
large surface is to be brought to flatness by 
hand methods, the process traditionally pro¬ 
gressing from chiselling (chipping), to filing 
and finally, to hand scraping. This produces a 
fine and relatively decorative finish. Since a 
scraper is capable of removing very fine shav¬ 
ings from the work, the accuracy achievable is 
also very high, and with care and patience it is 
possible to match two surfaces fairly exactly. 

If it is simply necessary to make the surface 
of a casting (or whatever) sufficiently flat to 
avoid distortion when it is bolted to a machine 
table, a straightforward filing operation is 
usually sufficient. 

Filing curves 


The filing of curves is generally more difficult 
than the creation of straight and square sur¬ 

faces and unless great skill with a file has been 
achieved, some form of aid is a necessity. 

The creation of external (convex) curves is 
undertaken using flat files. Internal (concave) 
curves must be created using round or half- 
round files. Using these is similar to using flat 
files; the cut is taken by using both forward and 
sideways motions, when this is possible, but it is 
often necessary to impart a rotary motion, in 
addition, as this helps to eject the swarf and cre¬ 
ate a smoother finish. 

The sort of curves likely to be needed are 
blend radii between adjacent straight edges, 
radiused ends to levers and arms and com¬ 
pound curves such as are found at the bases of 
domes, chimneys and safety valve covers on 
locomotives. The first two of these may possi¬ 
bly be produced by machining, concave blend 
radii being created by drilling and lever end 
radii being produced by milling, if a rotary 
table and milling facilities are available. 

Compound curves must generally be pro¬ 
duced by hand. 

Radiused lever ends are the easiest of the 
curved shapes to produce and so this technique 
is described first. 

Using filing buttons 

Small levers frequently take the form shown in 
Figure 6.20, having a tapered form and being 
radiused at both ends. They are usually drilled 
or reamed at the ends for the attachment of an 
operating or control rod, the end radius being 
drawn from the centre of the end hole. 

The presence of the hole makes possible the 
use of a pair of guides which allow the radius to 
be produced easily. These guides are known as 
filing buttons and may take the form shown in 
Figure 6.21. Each button is simply a turned cyl¬ 
inder, the correct diameter for the lever end, 
having a through bore which fits a suitable bolt 
or plain pin. 



Figure 6.20 Typical small levers with radiused ends. 

If the buttons, mounted on a bolt, are fitted 
one on each side of the lever end, and the bolt is 
nutted and lock-nutted as shown in Figure 

6.21, the lever end may be filed to the correct 
radius by filing along the lever and around the 
end so that the buttons act as rollers to guide 
the file at the correct radius. If the buttons are 
made from mild steel, the file will mark them, 
as will be seen from the illustrations, but this is 
really not a problem and hardening is not es¬ 

The alternative to the rather ‘posh’ form of 
button shown here is simply to utilise two thick 
washers rotating on a pin which fits the lever’s 
end hole directly. Provided that the file is 
worked squarely around the end of the lever, 
there is little tendency for the washers to wind 
themselves off the pin and the plain type will be 
found satisfactory. Either type of button is very 
easily made up and once a range is available the 
filing of end radii is straightforward as long as 
the lever is provided with an end hole. Two 
washers and a plain pin are shown in use to 
radius the corner of a square plate, in Figure 

6.22, and Figure 6.23 shows a small lever and 
the buttons and pin which were used to round 
the end. 

Figure 6.22 The comer of a cast baseplate being rounded. 

Figure 6.21 A pair of small filing buttons fitted to a rectangular 
rod for rounding the end. 



Figure 6.23 A part-made lever with the filing buttons which 
have been used to round one end. 

Radius gauges 

If convex curves without a coaxial hole are 
required the radius must be produced entirely 
by hand and some form of gauge is needed to 
judge the shape of the curve. If working in sheet 
materials, the required curve can be marked 
out in the usual way to provide a basic cutting 
line. Once the filing stage is reached however, 
judging the shape by eye is extremely difficult 
since the scribed line seems to deceive the eye 
into believing that the curve is good. It is usu¬ 
ally beneficial to observe the curve from the 
unmarked side when its true shape will usually 
be discernible. 

The use of a radius gauge is the best method 
of judging the shape. Sets of these are available 
commercially, each gauge being provided with 
two radiused edges providing both a convex 
and concave surface of identified radius. Sets of 
gauges are shown in Figure 6.24. 

If commercial gauges are not available, they 
can very easily be made up from the plastic 
(polystyrene) card which is available in sheet 
form in various thicknesses. This is used exten¬ 
sively for plastic model building and small-scale 
model railways and is available from shops 
catering for these activities. 

For a radius gauge, .020in. or .030in. thick 
(0.5mm or 0.75mm) card is satisfactory. If the 
radius is small, and convex, a drilled hole will 
suffice, but larger radii or concave curves must 
be marked out and cut by hand. The material 
takes pencil markings very well and curves can 
be marked easily with compasses, or dividers 
can be used to cut a shallow channel at the re¬ 
quired radius. A sharp craft knife is then used to 
cut along straight lines up to the curve (only a 

Figure 6.24 Three sets of radius gauges. 



deep score mark is required, not cutting right 
through) and then the curve is worked round 
with the point of the knife. Once again, it is not 
necessary to cut right through, just to score the 

Once the plastic card has been scored it can 
be broken off along the line by simply bending 
in the direction which opens the cut. For a 
curved line, it is necessary to work around the 
curve progressively, but it will break off quite 
cleanly. The initial cutting does raise a burr but 
this can quickly be removed with a fine file. A 

Figure 6.25 Purpose-made radius gauges cut out of 
polystyrene sheet. 

Figure 6.26 Checking the radius on the front of a smokebox 
casting using a purpose-made gauge. 

pair of radius gauges for concave curves is 
shown in Figure 6.25. 

Figure 6.26 shows a plastic card gauge in use 
for radiusing the front of a large casting. In this 
situation, a gauge represents the only alterna¬ 
tive to simply judging by eye, a technique which 
is not generally very satisfactory unless both the 
eyes and the hands are well practised. 

Compound curves 

Compound curves are frequently found around 
the bottom of chimney, dome and safety valve 
castings for model locomotives and a chimney 
base will serve as an example of this type of 
work. Before the upper surface is shaped, it is 
best to finish the bottom radius on which the 
base will ‘sit’. Since it is important that this is 
square, both side-to-side and front-to-back, 
some preliminary marking out of the casting is 
required. The best method to adopt for such 
items is to mount the casting in the lathe and 
bore it through to the correct size and then to 
re-mount it in the lathe and turn the basic 
diameters at top and bottom. Ways of doing 
this are described in Chapter 13. 

The result of these operations is a partially 
turned casting which can be marked out with 
four reference lines. At front and rear, the 
correct height can be scribed, and at the sides 
an approximate position can be marked which 
will act as a ‘witness’ as the underside is filed to 
fit the smokebox on which it will be fitted. The 
marking out is most easily performed on a 
surface plate, using the scribing block, with the 
base casting upside down, standing on the 
turned top. The marking-out operation is 
shown in Figure 6.27. 

Once the part-machined casting has been 
marked out, the underside can be filed to fit the 
smokebox. Front and rear marks will show 
when the correct centre height is reached and 
the side marks will confirm that side-to-side 




Figure 6.27 Scribing reference lines on a part-turned 
chimney base casting prior to filing the skirt to shape. 

squareness has been (is being) achieved. For 
holding the casting, the method adopted for 
turning the outside diameters may be used, the 
casting being wedged on to a stub of material 
held in the vice. 

When filing curves such as these concave 
shapes, a round or half-round file is naturally 
used. The radius required is certain not to 
match that of the file and it must be moved side¬ 
ways and forwards during each stroke in order 
to remove metal evenly since it is difficult to 
persuade a half-round file to cut on the top of a 
high spot. If the opposite happens, and hollows 
form, the file tends to remove yet more metal 
from them, making matters much worse and in¬ 
creasing the difficulties of effecting a cure. 

Once the base is filed to fit the smokebox, 
attention can be turned to the upper profile. A 
wide range of curves is required in this case, 
most of which are difficult to determine, but it 
is possible to estimate the shapes required for 
the front and back radii and also for those on 
each side. Two radius gauges can be made up, as 
shown in Figure 6.25. Filing can start at the 
front and back and then progress to the sides, 
removing metal where it is really proud of the 
desired line, and keeping away from the edges 
as far as possible. 

A start should be made with a coarse file to 
get under the skin of the casting and remove 
metal from front, back and both sides sepa¬ 
rately but progressively. As the shape develops, 
metal can be removed from the ‘quarters’ filing 
around the casting in whatever way seems easi¬ 
est. The aim should be to get as good a shape as 
possible with the coarse file before changing to 
something finer, continuing to keep away from 
the edges as far as possible. 

The problem with the edges is that they eas¬ 
ily become rounded (convex). The casting is a 
convenient way to make the model, but the 
prototype chimney skirt is almost certainly a 
thin steel spinning or a thin-wall casting and the 
filing of the model base should aim to give this 
appearance to the skirt. Figure 6.28 illustrates 
the problem. At A, the required shape is drawn, 
suggesting the separate skirt of the prototype, 
while B shows the shape which is likely to result 
if too much pressure is placed on the rim of the 




skirt when filing. The ‘hump’ is difficult to 
remove but in any event the haphazard filing of 
the rim edge might already have reduced it to a 
knife edge which is not desired until the body 
of the casting has attained the required shape. 

The same problem can occur where the 
narrow top collar meets the body of the chim¬ 
ney and there can be additional difficulties if 
the chimney is an all-in-one casting since the 
chimney barrel will be scored should the file 
overrun the skirt. In these cases the barrel 
should be protected with masking tape or 
something similar. 

Once the basic shape has been achieved, it 
should be worked down by progressively finer 
files and then emery cloth to give the desired 
finish. Figure 6.29 shows a partially completed 
base. The chimney barrel has been protected by 
a few turns of masking tape and reference lines 
have been drawn on the tape to define the posi¬ 
tions at which the radius gauges should be used. 
The required shape has been achieved using 
coarse files and the surface now needs to be 
taken down to reduce the thickness of the base 
rim, and reduce the upper diameter to meet the 
already turned ring at the top of the base. The 

Figure 6.29 A chimney base casting roughly filed to the shape 
required. Having achieved the shape, it needs to be smoothed 
and reduced to the desired size. 

radius gauges for the base are illustrated in Fig¬ 
ure 6.25. 


Historically, the products of forge or foundry 
were brought to final size by hand. This meant 
that once parts had been made, they were 
assembled together, and during this process, 
any inaccuracies in the manufacturing proc¬ 
esses were accounted for by ‘fitting’ the parts, 
the fitter using hand methods to correct the fit 
when this was required. Traditionally, the final 
hand-finishing process was the scraping of the 
fitted surfaces. 

If skilfully carried out, scraping produces 
well-fitting parts and finely rippled surfaces 
which have both a decorative appearance and 
good oil-retaining properties. The acquisition 
of the levels of skill possessed by the experts is 
doubtless a long process, and not having 
scraped more than the odd big-end bearing in 
the days when a pre-193 9 car was the only type 
I could afford, I am not equipped to impart that 
knowledge here. 

Scrapers are useful for general purposes and 
the tool kit should at least be provided with a 
three-square. Craftsmen traditionally ground 
up scrapers from an old file, and indeed, files 
can be ground or forged into useful shapes, but 
it is easier to buy the useful three-square and 
flat forms readymade as they are supplied with 
handles and are ground to the correct shape. 

Figure 6.30 shows the cutting ends of the 
two most useful shapes of scraper; three-square 
(triangular) and flat. The other common shape 
is the half-round, but this finds little use in my 

The three-square scraper is slightly hollow- 
ground and designed to cut on its edges, which 
have a short straight portion nearest to the 
handle, but which otherwise curve towards the 



Figure 6.30 Small 3-square and flat scrapers. 

end so that the scraper can be applied in such a 
way that only a narrow line of contact is pro¬ 
vided with the work. The scraper can be 
applied to a high spot and narrow shavings 
removed locally until the high spot is removed. 

Scraping can be used to provide the required 
clearances in bearings, for example, but is 
equally suitable to achieve the correct flatness 
and fit for machine slideways, by hand. Since 
slideways mostly present flat surfaces, a flat 
scraper is used. The tool is designed to cut on 
its end, and is normally used by making succes¬ 
sive passes over the part of the work being 
scraped in two directions at right angles to one 
another. This equalises the removal of material 
and helps to improve the appearance. 

A flat scraper is held by both hands, pointing 
away from the operator, and held at a shallow 
angle (about 15 degrees) so that one edge cuts 
the work and removes a small shaving, when 
the scraper is pushed away from the operator. 
Large flat scrapers have huge handles which are 
long enough to be held between the upper arm 
and the body. The sort of work which a model 
engineer might undertake needs a 5- or 6-inch 
(125mm or 150mm) blade which is fitted with 
something like a small file handle. 

Both types of scraper may not be required, 
but a small three-square is essential. It is useful 

for de-burring work, whether finished by hand 
or by machine, for scraping off paint, marking 
fluid, general muck, or even light rust. 

Provided that it will go through them, a 
three-square is good for de-burring holes, but 
some specialised scrapers, known as de-burring 
tools, have been introduced during the last few 
years, especially for this purpose. The business 
end of one of these is shown in Figure 6.31. It 
comprises a handle, into which is inserted a 
hardened, cranked rod which has been ground 
away to form a cutting edge. Being about ‘/sin. 
(3mm) in diameter, the tool can be used in 
holes down to about '/-tin. (6mm), or a little 
less. Its shaped end allows the cutting portion 
to lie on the drilled surface of the hole and it 
removes the burr without producing an 
unsightly chamfer to the hole. It can naturally 
also be used to de-burr straight edges which 
have been filed or machined, although these are 
better de-burred by filing a small chamfer with 
a smooth file. 

Figure 6.31 The cutting edge of a de-burring tool. 

The final polish 


A polished finish is not always required, but it is 
generally necessary to smooth the surface left 
by the file, or even by machining, to produce a 
clean and bright finish. It is also frequently 



necessary to smooth the unmachined surfaces 
of stock material, which sometimes show 
surface blemishes or inclusions of dirt, as a 
result of the manufacturing process. 

Surfaces turned in the lathe can be carefully 
polished while they are rotating, provided that 
there are no ‘lumps’ going round on the work 
which pose a special danger. Turned outside 
corners should be de-burred using a small, dead 
smooth file, in the way described in Chapter 15 
and the outside surface of turned cylinders can 
be smoothed using the same file. Flat-machined 
surfaces should also be smoothed with the same 
file and burrs on the edges removed in the same 

For ordinary machined surfaces, the above 
operations may be sufficient, but there are cases 
where a proper polish is required. The process 
of polishing is progressive, starting with coarse, 
followed by fine, files, after which finer and 
finer grades of abrasive cloths and papers are 
used until the stage is reached at which very 
fine polishes may be applied. If carried out 
correctly, it should not be necessary to return, 
at any stage, to the use of a coarser abrasive or 
file, the progression always being towards a 
finer surface finish. 

Abrasive papers and cloths 

Abrasive papers and cloths are used for surface 
cleaning and preparation. These materials 
comprise abrasive particles cemented to a cloth 
or paper backing, various grades or ‘grits’ being 
available. The grit number relates to the mesh 
of the sieve used to select the abrasive, the 
higher numbers representing the smaller parti¬ 
cle sizes and hence smoother finish produced. 
The two main types of material are the emery 
cloths and papers and the so-called wet-and- 
dry papers. For these latter, a waterproof 
cement is used to attach the abrasive to the 
backing and they can be used in the presence of 

water under circumstances in which the paper 
would otherwise rapidly become clogged with 
abraded material, for example, when rubbing 
down painted surfaces. 

One particular advantage of wet-and-dry 
paper is that it is available in very fine grades, 
up to 1200 grit. It is a natural successor to the 
coarser emery cloths and papers, although it is 
available from some sources in grades as coarse 
as 80 grit. 

A cloth backing provides greater wear resist¬ 
ance, and for general use at the coarser end of 
the scale, an emery cloth is more satisfactory. 
These cloths tear very easily going ‘down’ the 
sheet, but less readily ‘across’, and they should 
be cut in this direction. Paper-backed abrasives 
need to be cut in both directions if tidy strips 
and smaller pieces are required, as is generally 
the case. 

Backing up the abrasive 

There is a great temptation to use emery cloths 
in the hand, but unless its use is simply to clean 
up stock material which will later be machined, 
the method should be avoided. Lack of support 
for the abrasive allows the cloth to ‘fall over’ 
the edges, which rapidly become rounded, and 
some kind of solid backing must therefore be 

On large surfaces, a hand-size wooden block 
can be used, or a woodworker’s cork block, 
and a strip of emery cloth or paper wrapped 
around it. Provided that care is taken to keep 
the pressure firmly on the main surface and 
away from the edges, there should be no degra¬ 
dation of the corner profile. It is important to 
ensure that the abrasive material is stretched 
tightly around the block, otherwise any slack 
can roll itself over the edge and spoil the 

Support for the cloth or paper is also natu¬ 
rally required for work on smaller items. One 



convenient method is to select a file whose size 
suits the job and to tear off or cut a piece of 
material long enough to lie along the blade and 
fold over the end. If the file is held as for draw 
filing, the emery can be gripped firmly to the 
file and the work cleaned up without damaging 
the corners. This operation is shown in Figure 


An alternative to the hand-held emery strip 
is to glue a strip of abrasive paper to a strip of 
wood, to produce what are known as emery 
sticks. A group of these is shown on Figure 

6.33. The sticks are commercially prepared 
wooden strips, about lin. x Van. (25mm x 
6mm) to which emery paper has been glued, 

Figure 6.32 A strip of emery cloth can be wrapped around 
the blade of a file for the later stages of smoothing a surface. 

creating in effect, an emery paper file. These 
sticks are a much more satisfactory arrange¬ 
ment than the emery-around-file method 
shown in Figure 6.32, and can readily be made 
up, or the emery paper replaced, using double¬ 
sided tape. Paper is normally used, rather than 
cloth, since it creases sharply when folded, and 
the sticks can be ‘wrapped up’ completely 
making both wide and narrow edges available. 

These methods of providing support for the 
abrasive are very convenient for the edges of 
small levers and rods, but are less useful when 
their faces require treatment, particularly if 
they are tapered and cannot readily be held in 
the vice. More flexibility is provided by turning 
the process upside down, laying the abrasive 
material on a flat surface and rubbing the work 
up and down on it. This must, however, be 
done carefully, ensuring that the cloth or paper 
remains flat and taut, otherwise ripples develop 
ahead of the work as it moves to and fro, and 
this tends to round the corner profile. 

A useful alternative is to stick the abrasive 
paper down onto a board using heavy-duty, 
double-sided tape, preparing three or four 
boards with different grades of abrasive. The 
boards can be sufficiently large to accommo¬ 
date a complete sheet of cloth or paper and will 
last very well, particularly the finer grades, 

Figure 6.33 Several grades of emery sticks. 

Figure 6.34 A sheet of emery paper or wet-and-dry can be 
stuck to a piece of board and used for smoothing small parts. 



which do not, in any event, remove much 

On the subject of the wearing capabilities of 
emery cloth, it is worth noting that the larger 
grit particles tend to become dislodged from 
the backing first and the emery thus tends to 
become finer as it wears out and can be down¬ 
graded to the later stages until it is finally 

A board to which a sheet of emery paper has 
been glued is shown in Figure 6.34. 

Direction of working 

If cleaning up is being performed as a prelimi¬ 
nary to painting, no great fineness of finish is 
generally required, it being necessary to ensure 
that the overall surface has a good shape and 
provides a key for the paint. Shape is deter¬ 
mined by the filing and machining processes 
and emery cloth is used to provide the key. A 
medium grade of emery is usually appropriate 
for the larger models and a good key for the 
paint is created by adopting a circular motion. 

For work on levers and long rods which are 
not to be finished to a high state of polish, but 
do need to be clean and bright, it is best to rub 
along the length using progressively finer grits, 
until the desired appearance is achieved. For 
models, the finish required is usually such that 
no scratches are discernible with the naked eye, 
since the work is, after all, only a small propor¬ 

tion of its prototype size. How much polish is 
required (or desirable) is dependent on the 
model, the item and the builder’s notion of 
what is needed. 

If the work is required in a highly polished 
state, which is generally the case for clock plates 
and wheels, great attention needs to be paid at 
all stages to ensure that surfaces are not spoiled 
during the initial stages of marking out, sawing 
and filing. Great attention also needs to be given 
to the polishing processes. First of all, the work 
must always progress from coarse to fine, the 
aim of each stage being to remove the scratches 
produced by the previous (coarser) stage. 

Scratches created by the previous stage can 
best be seen by carrying out each stage at right 
angles to the previous one, continuing to use 
finer and finer abrasives until the scratches are 
so small that the required degree of polish has 
been achieved. At this stage, very fine abrasives 
are being used and the direction in which they 
are applied is not usually significant. 

Polishing progresses through the finer 
grades of emery cloth and paper towards the 
use of compounds which are applied by being 
loaded onto a cloth pad. These compounds in¬ 
clude various grades of rouge, which can be 
purchased from modellers’ suppliers who spe¬ 
cialise in providing a service for clockmakers, 
but cans of liquid polishes such as Brasso are 
readily available from DIY and hardware 
stores, sometimes simply described as ‘metal 


Bending, folding and forming 


Preparation of the material 

The bending of metal is a frequently required 
operation in model work. Since thin materials 
are normally involved, bends are made with the 
metal in the cold condition, rather than by forg¬ 
ing at red heat. If the material is to be bent cold, 
it may require preliminary heat treatment to 
render it sufficiently ductile (soft) to bend easily 
and without fracturing. The process is known 
as annealing, and it may be applied to steel, 
copper, brass and aluminium alloys, although 
not always for the same reasons. Annealing is 
performed by heating the material (usually to 
red heat, although not for aluminium alloys) 
and allowing it to cool, although the manner of 
cooling can vary, depending upon the material 
and the aim of the process. 

Since copper and brass are ductile metals, 
they will usually bend readily in the ‘as bought’ 
state, although they actually become harder as 
they are worked, a process known as work 
hardening, and do need to be annealed at inter¬ 
vals through the forming process if significant, 

repeated shaping needs to be performed. This is 
usually when these metals are being flanged or 
formed into complex shapes. 

The most common form in which steel is 
purchased is that known as bright mild steel 
(BMS). This is drawn down into standard bar 
forms (rounds, squares, hexagons and flats) 
using a cold process which produces a smooth, 
bright surface finish to the bar. The bar has, 
however, been pulled and squeezed into shape 
and the process leaves the material with an 
elongated grain structure which is quite brittle. 

The result is that a bend across its width, 
formed cold, frequently produces a fracture. 
This can be avoided if the steel is normalised 
before bending. The need for normalising, and 
a more detailed description of the process, are 
given in Chapter 3. 

The bending operation 

Before considering the bending process in 
detail, it is useful to consider what happens 
to the bent material, as this helps to avoid 
likely problems. In making a bend, the work is 



Figure 7.1 Stretching and compression of the material in a 

Stretched material pulled inwards 

Compressed material pushed outwards 

Figure 7.2 Deformation which occurs on a bend, due to 
stretching and compression. This effect is particularly noticeable 
on narrow strips. 

distorted, as shown by Figure 7.1, the material 
being compressed on the inside of the bend and 
stretched on the outside. The effect of this on a 
narrow strip is illustrated in Figure 7.2. Com¬ 
pressed material on the inside of the bend is 
expelled outwards and stretching on the out¬ 
side tends to draw material in from the edges. 
The result is that the material is significantly 
distorted on the edges, as well as on the outside 
of the bend, and this produces a distinctive 
‘turned-up’ appearance. It is important that the 
technique adopted for forming the bend tends 
to mitigate these problems. 

The distortion which occurs on the edges 
means that they need to be ‘dressed’ (filed or 
machined) after the bend is formed. A narrow 

strip which is to be bent needs to be wider than 
the required size in order to allow the final 
shaping to be performed after bending. 

Local distortion also means that holes close 
to the angle cannot be drilled until the bend has 
been formed, otherwise the hole distorts with 
the material (more strictly, the material around 
the hole distorts) and the hole is ‘pulled’ 
towards the bend on the outside. The general 
procedure must therefore be: bend, shape the 
edges and then mark off for, and drill adjacent 

Cold bending 

Use of a bending bar 

In the absence of a specialised bending or fold¬ 
ing machine, the operation is most conveniently 
performed by ‘pushing’ the metal around a 
former which approximates the shape required, 
nothing more elaborate than a stout bar being 
required. Unless the material is very thin, and 
very ductile, ‘dead sharp’ internal bend radii are 
not ordinarily achievable, and the former needs 
to be provided with a suitable radiused edge, 
around which the material can be folded. 

For general use, a simple length of BMS 
rectangular bar can be retained, especially for 
the purpose, and a 12in. (300mm) length of 
3 /«in. or lin. (20mm or 25mm) square bar, 
having its longitudinal corners radiused, can be 
kept specially for the purpose. Ordinarily, an 
internal bend radius equal to the material thick¬ 
ness can be applied for general use, unless there 
is a specific need to represent a particular 
radius from scale considerations. A short bar of 
the sort shown in Figure 7.3 serves adequately 
for short work. This example has lived in the 
workshop for many years and has been used 
both for bending and as a holder for riveting 
dollies, hence its chequered appearance. 



Figure 7.3 A much-used bending bar produced from a length 
of 3 /<in. (19mm) square steel. 

This particular bar is about 9in. (230mm) 
long and has one long edge filed to a radius of 
about '/win. (0.8mm). It suits the making of 
bends in materials up to 20 or 22 swg (up to 
about 1mm thick). At one end, one edge is filed 
to a radius of'/sin. (3mm) for a length of lin. or 
so. This allows narrow materials about '/sin. 
thick to be bent satisfactorily and was originally 
prepared many years ago for bending mild steel 
strip for model locomotive guard irons. 

In use, the bar and material to be bent are 
clamped into a vice with the upper surface of 
the bar just proud of the top of the vice jaws, 
and the bending line, already scribed on the 
work, carefully aligned with the bar’s upper 
surface. The angle which the body of the work 
makes with the bar’s upper surface must be 
checked, if this is possible, although this natu¬ 
rally depends on the shape and size of the work, 
principally the length of the upstanding part. 
Figure 7.4 shows the essentially different types 
of work. 

When making the bend, it is not usually 
adequate simply to pull on the upstanding part 
of the material, even assuming it is long, as 
shown in Figure 7.4A. It is essential to push the 
material closely around the bar, and to do this 
requires pressure to be exerted on the material 

just above the bending bar’s corner. This pres¬ 
sure can best be applied by using a block of soft 
material, such as aluminium alloy, holding it 
firmly against the work and hammering it 
down to press the material around the radius 
on the bending bar, as shown in Figure 7.5. 


Figure 7.4 A brass plate positioned in the vice with the 
bending bar. If the plate has a large upstand. a small square can 
be used to check for squareness. 



Figure 7.S An aluminium alloy block being used with a 
hammer to form a bend over the edge of the bending bar. 

The aluminium alloy block serves two essen¬ 
tial purposes: it allows a large hammer to be 
used without fear that the work will be marked, 
and it spreads the load so that the full width of 
material moves together around the bend. 

It might be thought that the same result 
could be achieved by using a soft-faced ham¬ 
mer to apply the force directly. Unfortunately, 
this simple technique doesn’t work, and for 
several good reasons. Firstly, if the upstanding 
material is long and the blow is applied as 
shown in Figure 7.6, the inertia of the upper 
part of the upstanding material tends to make it 
want to remain undisturbed. The consequence 
is that the end lags behind the remainder of the 
material as the bend is formed, and a distortion 
of the work results. 

Secondly, the localised load, even from a 
soft-faced hammer, turns the material over 
unevenly, and can produce an uneven bend. By 
applying the hammer blow through a block, the 
localised load which the hammer applies is 
distributed over a larger area, consequently 
reducing the actual force applied locally and 
smoothing out the unevenness. 

The third problem caused by the individual 
hammer blows is that the bending force is not 
necessarily applied to the material immediately 

above the edge of the bending bar, and its shape 
cannot therefore be imposed on the work. A 
block of soft metal can be positioned and 
angled so that the force of hammer blow is 
applied exactly where required. 

If the upstanding portion of the work is 
short and fairly narrow, it may be possible to 
tap it over the former with a hammer since a 
squarely aimed blow may effectively contact all 
of the material. For a short upstanding part on 
wide work, some means of applying the bend¬ 
ing force over a wide area is again beneficial, 
and the use of a block some two or three inches 
(50mm to 75mm) long through which to apply 
the force is helpful. In these cases, it is prefer¬ 
able to bend progressively, deflecting the mate¬ 
rial through a small angle along the whole 
bend, and then repeating the process in this 
manner until the required angle is formed 
along the complete edge. 

Forming large-radius bends can naturally be 
by utilising an appropriately radiused corner to 
a bending bar or by using a circular bar as the 

Bowed effect due to 

Vice jaws 

Figure 7.6 If an attempt is made to hammer a long strip over 
the edge of the bending bar, a bow is usually produced due to 
the inertia of the workpiece. 



former. The difficulty comes in locating the 
bar, since the scribed line, which usually shows 
the start of the radius, must be positioned such 
that it is not visible. It is frequently best, there¬ 
fore, to put the bend in at the approximate 
position on a larger-than-necessary piece of 
material, afterwards trimming material on both 
sides of the bend to the correct length. If a little 
extra is left on the width, this might also allow 
for some lack of squareness in placing the bend¬ 
ing bar which can be taken out when the sides 
are trimmed to size. 

If a narrow strip is being bent, it must 
initially be wider than that which is finally 
required, in order that the turned-up edges can 
be removed afterwards. 

Using a ‘proper’ industrial bending or fold¬ 
ing machine the problem of lack of visibility 
does not occur, since the scribed line remains 
visible, as shown in Figure 7.7. The work is 
clamped to the machine bed such that the 

scribed line coincides with the front of the 
clamp. The bend is formed by rotating a sub¬ 
stantial angle or blade against the work, but at 
such a position that it rotates at the required 
radius. The blade rotation forces the work 
around and upwards, forming the bend radius 
as it does so. If the correct allowance has been 
made, the (now) upstanding part has both the 
correct radius and the correct amount of 

Material in the bend 

It is obvious that some of the material is 
‘consumed’ in forming the bend. It therefore 
becomes a question of what allowance needs to 
be made for this. For large-scale, production 
items it is convenient if material can be cut, and 
the bend formed, leaving both sides of the angle 
exactly the correct length. Most engineers’ 


Figure 7.7 The arrangement of a bending and folding machine allows a bend to be created with any required radius. 



reference books, even quite abbreviated ones 
intended for the workshop rather than the 
library, provide tables showing the allowances 
which need to be made in laying out the work 
‘in the flat’, both in relation to the material in 
the bend and the overall size of the work, bear¬ 
ing in mind also the thickness of the material. 

In model work, materials are usually thin 
and bend radii very small, often amounting to 
dead sharp with little concession to normal 
allowances. Published data are not therefore 
especially relevant to model work and if only a 
single bend is required, it is usually more con¬ 
venient to form the radius or bend and then 
mark out the two sides individually for cutting 
and finishing to length. The bend allowance 
then takes care of itself. 

If items of the form shown in Figure 7.8 
need to be made, the use of a conventional 
bender, together with calculation and applica¬ 
tion of the correct bend allowances, permits 
accuracy to be achieved quite simply. Lacking a 
bender, other techniques will yield acceptable 
results, if carefully applied, but much naturally 
depends upon the accuracy which is desired in 
the completed item. 

If several identical items need to be bent into 
the sort of shape shown in Figure 7.8, they 
should certainly all be marked out together if 
possible, and all bent up in the one working 
session thereby readily allowing the same tech- 

Start of bend (scribed line) 

Figure 7.8 Allowances for the material needed to form the 

nique to be applied to them all. If the span of 
the U shape is not too large, it is useful to 
cut out a former of the required width, radius 
the two edges and bend each piece closely 
around the former. In this case, the upstanding 
arms of the U can be marked out and cut off 
after bending. 

Folding up a box 

Small boxes frequently need to be bent up, 
together with flanged lids to fit them. These 
items can be bent using the techniques 
described above to form the actual bends, but 
since two adjacent sides will be bent up, the 
corners need special treatment. This is usually 
to cut away material at the corners to form a 
notch, but before this is done, the intersections 
of the lines defining the internal corner posi¬ 
tions must be centre punched and the corners 
drilled. Following this, the corners are notched 
so that the bent edges will abut nicely at the cor¬ 
ners. The purpose of the holes is to allow the 
bent material to expand at the edges of the 
bends, and if this space is not allowed, the cor¬ 
ners of the box ‘bunch up’ and distort. 

Folding up the box is straightforward once 
the technique is understood. The initial bends 
should be formed on two parallel sides, it 
usually being convenient to bend the two long 
sides on rectangular work. This results in a 
U-shaped channel, as shown in Figure 7.9. 

Examination of Figure 7.9 will show that the 
brass sheet for the box lid has been drilled and 
notched at the corners, operations which are 
described elsewhere. The two sides have been 
bent using the methods illustrated in Figures 
7.4 and 7.5 and the lid is ready for the remain¬ 
ing sides to be bent. 

To bend the remaining two sides, a bending 
bar is needed, to fit within the width of the box, 
sufficiently thick to project above the top of the 
already folded sides. This is then clamped into 



Figure 7.9 A half-folded box in thin sheet brass. 

Figure 7.10 The box of Figure 7.9 fitted into the vice with a 
short bending bar which allows the third or fourth side to be 

Figure 7.11 A completed box lid. 

Figure 7.12 Large sheets can only be bent by arranging the 
bending bar and material outside the jaws of the vice. This 
illustration shows a pair of specially made bars which have been 
dowelled together and clamped to the work, support being 
provided by a wooden dowel held in a vice standing on the floor. 

the vice as shown in Figure 7.10, and the 
remaining sides bent over the bar. 

The completed box lid, with the drilled-out 
corners in evidence, is shown in Figure 7.11. 
The material of the lid is .022in. (0.6mm) brass 
sheet and the corner relief was put in using a 
s /64in. (2mm) diameter drill, which has created 
ample clearance for material pushed outwards 
from the bends. 

Bending a large sheet 

Bending rolls (see below) might be used to bend 
large sheets, but they cannot produce small 
radii, and it is usual to bend material over a 
former of the required diameter in order to 
form such bends. Round bars immediately 
suggest themselves as suitable formers, and if a 
round bar can be clamped to the sheet material, 
and the ‘assembly’ put into the vice, the bend 
can be formed by pressing the material over the 
bar to achieve a bend of the required radius. 

With a long sheet it may not be possible to 
clamp it centrally in the vice, in which case the 
sort of arrangement shown in Figure 7.12 has 
to be used. In this case, a sheet of brass, 1 lin. 
(280mm) wide and 24in. (600mm) long has 
been bent to form the outer of a water tank. 



The corner radii are 3 /sin. (10mm) and the sheet 
is sandwiched between two rectangular steel 
bars which are located to one another by dow¬ 
els. One bar has a cut-down 3 /sin. diameter steel 
bar screwed to the top surface and this has been 
used to form the bends. The brass sheet was 
annealed along the bend lines before bending, 
fitted to the pair of bars on the bench, for easier 
handling of an awkward assembly, and tool- 
makers’ clamps used to hold the sheet firmly. 

Since the sheet could not be accommodated 
within the vice it was held in the vice at one end 
and supported by a length of lin. (25mm) 
wooden dowel at the other. The dowel was 
held in a vice standing on the floor of the work¬ 
shop in order to provide a safe support, and 
the bend put in by hammering on a length of 
wood placed on the sheet, to force the material 
down into contact with the V»\n. diameter bar 
attached to one of the rectangular bars. 

Rolling metal 

Even though a bend radius is large, it may still 
be put in by pressing the material over a suit¬ 
able piece of circular bar, but this is ultimately 
likely to become impossible, simply due to the 
non-availability of suitably large bar material. 
In these cases, something akin to an adjustable 
former is used. This takes the form of a set of 
bending rolls. The operating principle of these 
is that the material is guided through a set of 
rolls which includes one deflecting roller which 
is adjustable to alter the radius formed. This is 
illustrated in Figure 7.13A which shows the 
usual form of these rolls. 

The two lower rolls can be rotated manually 
so that the material passes through the rolls and 
is bent into a curve as it passes beneath the cen¬ 
tral, deflecting roll. The vertical position of this 
roll is adjustable and the curve produced in the 
material can be varied. Repeatedly passing the 


Adjustable position 

Figure 7.13 Two forms of bending roll. 

material through the rolls and progressively 
lowering the deflecting roll, causes a progres¬ 
sively decreasing radius to be formed. 

The disadvantage of this form of rolls is that 
the material cannot be curved right to the end 
since there is inevitably some material which 
lies on a straight line between the contact 
points on the deflecting roll and each of the 
lower rolls. This can be overcome by rearrang¬ 
ing the rolls into the form shown in Figure 
7.12B. Here, the deflecting roll acts directly 
and independently and the arrangement allows 
a cylinder to be rolled from flat sheet. Provided 
that this roll can approach the upper roll 



Figure 7.14 A nicely made pair of bending rolls seen at an 

closely, the radius achieved can be very closely 
that of the upper roll. A set of bending rolls of 
this type is shown in Figure 7.14. 

An example of the type of work carried out 
using the rolls is the rolling of the outer wrap¬ 
per plate for the firebox of a model traction 
engine boiler. The simplest type of wrapper has 
straight, parallel sides and a semi-circular top. 

The length of material required for the 
wrapper can be determined by calculating the 
length of the semi-circular arc and adding on 
the height of the two sides. The semi-circular 
arc is naturally one half of the circumference, 
which is calculated from the expression 2n (pi) 
multiplied by the radius, or, 7t multiplied by the 
diameter. The ratio 7t may well be available 

Figure 7.15 The initial position of a sheet of copper in the 
rolls. The position of the deflecting roll has been adjusted so that 
the work is nipped and the rolls are parallel. 

these days on a scientific calculator, but if not, 
can be taken as 3.142. The commonly accepted 
approximation for 7t of 22 divided by 7 is actu¬ 
ally closer to 3.143 than 3.142, but will make 
no difference in practical terms. In any event, it 
is wise to commence the rolling operation with 
a sheet which is slightly over the required 
length, allowing, say, an inch (25mm) on each 
side, for cutting off once the shape has been 

When the annealed sheet of material is avail¬ 
able, a centre line should be marked using a 
felt-tipped pen and the rolling operation 
started by inserting the sheet into the rolls and 
tightening the adjusting screws until the sheet is 
just ‘nipped’ and the adjustable roll is parallel 
to the others. This condition is illustrated in 
Figure 7.15. 

The rolling operation should be started from 
the marked centre line, progressively decreas¬ 
ing the bend radius until the desired curvature 
is achieved around the central mark but not yet 
extending over the full arc. As rolling and bend¬ 
ing continue, the work must be taken out of the 
rolls to check the radius which has been 
formed, in this case by testing its fit to the 
boiler barrel, as shown in Figure 7.16. 

Once the correct radius is being rolled, the 
material can be inserted into the rolls and 

Figure 7.16 Testing a part-rolled wrapper on the boiler 



Figure 7.17 The almost-complete wrapper in the rolls. 

passed progressively further through the rolls 
on each side of the centre line until an arc of the 
required length is achieved. Since the rolls exert 
only a friction grip on the material, it is neces¬ 
sary to push the material firmly into the rolls 
while turning the handle. 

This operation is shown almost complete in 
Figure 7.17. Starting with a sheet which is 
slightly over-length, and then cutting to size 
after rolling and fitting to the flanged plates. 

Figure 7.18 A commercially made firebox tubeplate. 

removes the need for accurate centring of the 
rolled arc. This naturally allows concentration 
on the important aspect of ensuring a good fit 
without worrying about the bend’s actual posi¬ 
tion on the sheet. 

Naturally, if using the type of rolls shown 
in Figure 7.13B to form a cylinder, the length 
of sheet must exactly match the diameter 
required, and this should be calculated from 2n 
multiplied by the radius, as described above. 

When rolling work using the rolls, it is vital 
to ensure that the work is inserted squarely into 
the rolls, and this should be checked carefully 
with a square before commencing. It is also 
important that the deflecting roll remains 
parallel to the driving rolls at all times, other¬ 
wise tapered work is produced since different 
radii are forced into the material on the two 
sides. Indeed, this is the way in which a tapered 
boiler barrel is produced. 

For some rolled work, the shapes required 
are more subtle and complex. The flanged plate 
shown in Figure 7.18 is the firebox tubeplate 
for a small locomotive-type boiler. This plate is 
4 5 /i«in. (110mm) high and 2 s /i6in. (59mm) wide 
at the bottom. The top is curved at a radius of 
3'/4in. (83mm) and the corners have a radius of 
’/lsin. (14mm), little larger than the diameter of 
the rolls shown in Figure 7.15. 

Since the wrapper plate is formed from 
copper sheet, it must be annealed before com¬ 
mencing, to render it soft and ductile. Since 
unworked, annealed copper is extremely soft, it 
can be formed by manual pressure over suitable 
bars (for the corners) and a wooden block, for 
the larger radii. An alternative is to shape up a 
wooden block to the form required, clamp the 
sheet to one bottom corner of the block and 
then to use manual pressure or a soft-faced 
mallet to force the sheet down into contact 
with the former, applying further clamps to 
hold the part-formed sheet as work progresses. 

Due to the inevitable spring-back which 
occurs whenever metal is bent, a wrapper 



formed in this way inevitably ends up being 
slightly bigger than the former, whose size 
might need to be adjusted to compensate. 

Although a wrapper to fit the tubeplate 
illustrated in Figure 7.18 (and its associated 
backplate, which is the same size) can be bent 
up by hand methods, the bending rolls do offer 
the most precise control over the shape pro¬ 
duced, although a little hand intervention is 
sometimes useful. The process must be progres¬ 
sive, however, and the larger radii must be put 
in first, followed successively by the smaller 
ones. This means beginning with the large top 
radius, then putting in the corner bends and 
afterwards attending to the sides. 

Rolling the top radius is straightforward, the 
sheet being first marked to identify the centre 
and the rolling starting on each side of this ref¬ 
erence mark. Since the top corner radii are very 
close to that of the rolls themselves, and also 
because they represent arcs of less than 90 de¬ 
grees, the normal rolling action of the rolls is 
not helpful in producing these curves. 

Since model boilers are generally made in 
copper, which is kept soft by repeatedly anneal¬ 
ing it during bending, it can actually be bent by 
hand. A circular bar of the appropriate diam¬ 
eter can be used as a former, and if the size of 
the job allows, it can be clamped into the vice 
with the circular bar correctly positioned 
against the sheet and the bend put in by hand 
pressure. If the corner radius is similar to the 
diameter of the bending rolls, it may be possi¬ 
ble to form the bend while the sheet is in the 
rolls, simply by pulling on the upstanding part 
of the wrapper. This at least solves the problem 
of holding the work, which can otherwise be 

Although the bending can be done by hand, 
direct manual pressure is not satisfactory since 
it cannot be applied evenly across the work 
without something interposed to spread the 
load. For soft materials a piece of wooden 
board is satisfactory. This can be used to pull or 

push the material over the bar, say, through an 
angle of 35 or 45 degrees, before checking that 
the bend is progressing evenly at the two edges 
and the material will fold squarely over the bar. 

If one edge is not folded over to the same 
extent as the other, gentle hammering on the 
required side of the board should be used to 
bring the work to squareness before proceeding 
to pull or push the bend into its final position. 
During the hammering, remember that soft 
material can be easily be crushed if hammered 
too hard against the roll. Provided that good 
contact with the circular bar is achieved 
throughout the operation, there should be no 
need to re-bend already formed material, 
which is, in any event, harder than the remain¬ 
der, due to its already having been worked. 
There should be little spring-back effect in the 
bend itself, but the sheet will need to be pulled 
through a slightly larger angle than is needed 
for the bend, since that is where any spring 
becomes evident. 

A trial fit against the flanged plate is neces¬ 
sary to confirm the correct form and fit of the 
bend and the sheet can be marked for the sec¬ 
ond bend at the same time by using a felt-tipped 
marker to show the line which indicates the 
starting point of the radius on the other side. 
Getting the starting point in the correct place is 
the difficult part of the operation and the initial 
bending of the second corner should concen¬ 
trate on this aspect before too tight a bend is 
put in. Once the radius is started in the correct 
place, the work almost always beds down 
nicely onto the bar at the correct position and 
the final part of the radius can be put in without 

If things do go wrong, don’t despair! The 
copper can be rendered soft again by annealing 
it once more, and there is no limit to how often 
this may be done. Once it is soft, it can be 
straightened out again, bringing to near¬ 
straightness by hand and then rolling it flat in 
the bending rolls. This doesn’t always bring the 



material to absolute flatness, but this is seldom 
absolutely necessary since it is in any case going 
to be rolled again into a new shape. 

Flanging and forming 

Perhaps the cart has already come before the 
horse, since the rolling of the boiler wrapper 
which is described above has assumed that the 
flanged plates are already available for use as a 
pattern for the shape required. The flanging 
process, which at its simplest just means raising 
a turned-up edge around the outside of a plate, 
is simply an extension of the straightforward 
bending process, except that it allows ductile 
materials to be formed into bends which com¬ 
prise two curves at right angles to each other. 
That is, to be formed into the type of com¬ 
pound curve illustrated in Figure 7.19. 

The shape shown is the top corner flange of 
the tubeplate of Figure 7.18, the flange extend¬ 
ing around the two edges which are roughly at 
right angles. The corner is formed on a Vn\n. 
(14mm) outside radius and has an internal 
radius of about '/sin. (3mm) where it leaves the 
flat surface of the main plate. To form this 
corner, the copper sheet is annealed and 

Figure 7.19 A detailed view of the comer of a flanged plate. 

clamped to a thick former (the flanging former) 
in the vice, and the copper turned over the 
radiused edge of the former by hammering 
with a substantial hammer. Since the copper 
becomes hard again due to the hammering, a 
process known as work hardening, the anneal¬ 
ing must be repeated several times during the 
flanging to ensure that the copper remains soft 
and ductile. 

During flanging, there is little use in 
attempting to turn over the flange using a soft- 
faced hammer, in the hope that the flange can 
be formed without damaging the material. It 
must be hammered into the desired shape, and 
in a corner such as that illustrated in Figure 
7.19 must be compressed into the shape 
required. This means firm support for the work 
and a large enough hammer to effect the defor¬ 

Provided that the copper is annealed 
frequently enough, and the hammer blows are 
sufficiently heavy, there is little difficulty in 
turning over the flange except, of course, in a 
corner of the type illustrated. Material has to be 
forced into the corner from both sides so there 
is too much material in this location and it 
tends to bulge outwards as blows are applied to 
the top and the sides. The additional work 
which is being done on the material in the cor¬ 
ner hardens the metal more rapidly and this 
naturally makes it less amenable to being 
squashed into the desired shape, thus aggravat¬ 
ing the problem. 

When cutting out a blank sheet for flanging, 
it is naturally cut oversize to allow for the mate¬ 
rial which forms the flange. There is a natural 
tendency to want to make this allowance just 
that bit larger, on the basis that the excess can 
afterwards be trimmed away, but in the corners 
a smaller allowance is actually required since 
metal is pushed into this region from both the 
top and the side. It is not helpful, to say the 
least, if the allowance has been made too great 
as this actually makes the job of forming the 



Figure 7.20 A cut copper blank and the flanging former in the 

corner correctly that much more difficult. So, 
what is required is a somewhat sparing allow¬ 
ance, particularly in the corners. 

Figure 7.20 shows a former and a pre-cut 
and annealed copper plate in the vice, ready for 
the flanging. The plate is 3 /j 2 in. (2.4mm) thick 
and it will be formed using a 1 pound ball pein 

Figure 7.21 Progress achieved on a flanged comer at the 
stage at which a second annealing is required. 

hammer. The objective is to turn the flange 
over, all round the edge, progressively. The 
work should receive good solid blows, and any 
tendency for the corner to bulge outwards must 
be resisted by firm hammer blows to correct it. 
Figure 7.21 shows the corner shape achieved at 
the stage at which a second annealing is 

Figure 7.22A shows the appearance of the 
plate after further forming. The corner is now 


Figure 7.22 A flanged comer at the stage at which a third 
annealing is required. 



bulging noticeably, but the copper has become 
hard due to the hammering and a further 
annealing is required. Figure 7.22B shows that 
the flat part of the plate is bulging away from 
the former. This is due to the hammer blows 
being directed too much in a downward direc¬ 
tion (attempting to make the flange lie flat on 
the former) while the material in the corner 
was hard, but the flat still soft. This bulge can 
easily be corrected when the plate is again soft, 
but its presence can be discouraged by backing 
up the copper with a stiff plate. 

Figure 7.23 The final fit of the comer of a flanged plate. 

Figure 7.24 Excess material pushed into the comer dunng 

One more annealing, and some heavy blows 
on the corner, produce the final form shown in 
Figure 7.23. There is an adequate fit for this 
sort of work, and the outside shape has still to 
be produced by filing. Figure 7.24 shows how 
the extra material has been forced into the cor¬ 
ner, but this excess can be trimmed off when 
the plate is cut to its final shape. 

Formers used for the flanging process are 
most conveniently cut from aluminium alloy 
sheet which is sufficiently solid to allow 
annealed '/sin. copper to be flanged over it, yet 
is much easier to cut out and shape than mild 
steel. Wood can be used for light work, say, for 
the forming of thin brass sheets, around 18 or 
20 swg (0.9 to 1.25mm) but is definitely not ad¬ 
equate for flanging thick copper around tight 
corners, where really solid hammer blows are 

Apart from the additional work involved, 
there is naturally no objection to steel being 
used for the former, but it must not be thought 
that effort can be economised by using only a 
thin plate. Considerable force is needed to 
complete the forming of the flange, and if the 
flange depth is greater than the thickness of the 
former, the situation of Figure 7.25 usually 

Flange deformed locally 

Figure 7.25 Possible effect of using a thin former for flanging. 



Due to the overhang of the flange on the 
former, the flange is distorted locally by the 
hammer blows, since there is no support, with 
the result that the former is jammed into the 
plate and cannot readily be removed. The 
former is ideally a little thicker than the flange 
depth, plus any additional allowance, and the 
ease with which it may be removed from the 
flanged plate well repays the extra effort put 
into its production. The hammering process 
tends to produce an irregular flange depth, in 
any event, as exemplified by Figures 7.19 and 

It is sometimes suggested that a composite 
former can be used, comprising a wooden 
blank faced with a thin steel plate, but this too 
is often not satisfactory since the wood tends to 
compress and the flanged plate can easily 
become jammed on the former, as shown in 
Figure 7.25. 

Although copper boiler plates are common 
flanged items, the same technique can also be 
applied to brass, which is used extensively for 
various fittings for model locomotives. Typical 
of these is the locomotive tank filler shown in 
Figure 7.26. Being an alloy of copper, brass can 
be worked in substantially the same way, and 
the cover shown was flanged from 26 swg 
(0.45mm) brass sheet over the aluminium alloy 

Figure 7.26 A locomotive tank filler cap and the former used 
for its manufacture. 

Figure 7.27 Flanged end covers for small boilers made in thin 

former shown alongside the completed and fit¬ 
ted item. 

This cover is l 3 /i«in. by lVsin. (31mm by 
42mm) with radiused ends, and has a flange all 
round which isVuin. (2.4mm) deep. The cover 
required annealing three or four times during 
the flanging operation which was carried out 
using a 2oz hard-faced, ball pein hammer. This 
left the flange surface somewhat marked, but a 
suitable finish for painting was achieved by 
cleaning up the flange using a smooth file and 
medium emery cloth. 

The flanging former was in this instance 
made quite deep, due to the available material, 
and it was used also to hold the rough-flanged 
covers for sawing off the excess material from 
the flange which is evidenced by the saw marks 
visible in Figure 7.26. 

Figure 7.27 shows a group of brass discs 
similar in size and material thickness to the 
cover of Figure 7.26. These are end covers for 
small-diameter, toy boilers which need to be 
polished after silver soldering to the tube form¬ 
ing the boiler shell. Figure 7.28 shows the 
flanged rim after soldering to the shell. The 
damaged to the rim is evident in this view but 
this was polished out (well, filed and polished) 
to produce the result shown in Figure 7.29. 



Figure 7.28 An end cover silver soldered to the boiler tube. 

progress. It has to be accepted that simple 
flanging cannot produce such sharp and evenly 
turned over flanges as pressing the blank 
between a punch and a die, which is how such 
items as these are produced commercially. 

If you are an experienced coppersmith, the 
type of work described here is of the utmost 
simplicity. A skilled copperbeater can stretch 
and ‘pull’ annealed copper into almost any 
shape extremely easily by supporting it on a 
range of fairly standard shapes of former, 
known as stakes, and hitting it with a hard- 
faced hammer. The stakes resemble miniature 
anvils which can be supported in a vice or held 
in a ‘stake hole’ on a real anvil. 

It is a real pleasure to watch someone 
‘knocking up’ a model locomotive dome from a 
disc of copper with seemingly little effort, but it 
obviously requires a great deal of practice to 
achieve this degree of proficiency, and unless 
you are well beyond the stage of being shown 
how to beat up an ashtray from a copper disc, 
which always seemed to be the starting point 
for school metalwork classes, it is probably 
wise to keep it simple. Nevertheless, domes and 
suchlike can be formed, simply by use of a ham¬ 
mer and appropriate stakes. 

The model locomotive dome of Figure 7.30 
was formed from a length of 3-inch (75mm) 

Figure 7.29 An end cover which has been cleaned up and 

With this type of work, in which the flange 
material is being compressed all around the 
rim, it is frequently difficult to ‘persuade’ the 
rim to turn over through a full 90 degrees all 
around the disc, since the compression of 
material which occurs in the rim increases the 
problem of spring-back. There is also inevitably 
the problem that hammering one part of the 
rim inwards causes an adjacent part to bulge 
outwards so that a point is ultimately reached 
at which further hammering produces no 

Figure 7.30 A model locomotive dome. 



Figure 7.31 The flanging former for the bottom of a 
locomotive dome. 

diameter brass tube, the base being flanged 
outwards using the light-alloy former of Figure 
7.31, and the top being cut away to form a 
three-lobed tulip shape before being beaten 
over a wooden former to obtain the approxi¬ 
mate shape. Final forming of the top was 
carried out using the large stake shown in 
Figure 7.32. This not-very-elegant device is a 
billet of cast iron mounted eccentrically to a 
V4in. (19mm) diameter steel bar and turned 
‘freehand’ on the lathe by manipulating both 
feed screw handles simultaneously. It was 
finally smoothed by hand filing but could have 
been brought to a much better finish with some 
benefit. However, the upper part of the dome 
was finally shaped by hammering against the 
stake and the edges of the ‘tulip petals’ silver 
soldered together to produce an acceptable 
fitting without the need to utilise a casting. 

The stake of Figure 7.32 illustrates the prin¬ 
ciple of these items. They do not at all need to 
be shaped like the item to be formed, but are 
designed to provide firm support for the work 
so that it can be hammered into the desired 

Figure 7.32 A stake for forming the top of the dome. 

shape by striking the parts just overhanging the 
support, until the correct shape is achieved. 
Regular annealing of the brass or copper is 
essential to ensure softness and ductility and to 
prevent the material cracking or splitting 
during forming. As work hardening occurs, the 
material deforms less for a given hammer blow 
(force) and the sound and ‘feel’ of the opera¬ 
tion changes, making it obvious that another 
annealing is required. 

One point about annealing is worth making. 
Heating of the brass or copper to red heat 
encourages the formation of oxides on the 
surface, and discolouration occurs. Sometimes, 
quite large amounts of a black residue are 
created which flake off when cold or when the 
hot metal is quenched in cold water. Some 
oxides always remain on the surface, however, 
and if the forming process is continued without 
cleaning the deposits away, they can become 
embedded in the surface of the work, hindering 
or preventing proper cleaning. It is necessary to 
clean the material thoroughly after each 



Tube bending 

Preventing collapse 

Steam engines (generally) require numerous 
small tubes of various sizes for such functions 
as steam feed, exhaust, cylinder drains, lubrica¬ 
tor feeds and so on. In the scales normally used, 
these tubes range from '/win. (1.5mm) diameter 
up to V4in. (12.5mm) or so, for the exhaust 
pipes for the larger steam locomotive models. 
Quite tight bends are sometimes required in 
these pipes and it is usual to use copper tube 
since it is easily softened by annealing and takes 
both soft and hard solders extremely well. 

The use of soft copper tube makes bending 
easy but it also brings with it some problems. If 
a piece of copper tube is bent, either around a 
former, or just in the fingers, the result is inevi¬ 
tably that shown in Figure 7.33 - a collapsed, 
unsightly and blocked tube. As mentioned in 
the introduction, and illustrated in Figure 7.1, 
material on the inside of a bend must compress, 
and that on the outside must stretch. If the ma¬ 
terial is tubular, there is naturally no central 
core to prevent the material on the inside and 
outside of the bend from doing precisely what 
it wants. Material on the outside of the bend, 
therefore, fails to stretch sufficiently, and ‘cuts 
across’ the corner, while that on the inside fails 
to compress and excess material collapses 

Figure 7.33 A bent and collapsed copper tube. 

The solution to the problem is to ‘put the 
middle back’ into the tube during the bending. 
In the plumbing trade, only a strictly limited 
range of tube sizes is used, and standard bend¬ 
ing springs are available which are a close fit 
inside the tube. Being coiled, the spring is flex¬ 
ible, but since it is wound from square-section 
wire and hardened and tempered, it cannot 
readily be crushed. It therefore allows the tube 
to be bent without it collapsing. 

Unfortunately, the limited range of tube sizes 
used by plumbers means that the method is not 
widely usable by modellers, but if you need to 
bend a largish tube, and its actual size doesn’t 
matter within reason, one of the plumbing sizes 
should be considered, especially if you can bor¬ 
row a spring for an evening. 

An alternative to the fitting of an internal 
spring is to fit a ‘restrainer’ to the outside, a 
practice which suits the smaller tubes used by 
modellers quite well. Sets of small springs such 
as those shown in Figure 7.34 are available, this 
set covering tubes of '/i«, Vn, /», Vn and Va 
inches in diameter. Being of a somewhat softer 
temper, and therefore less rigid than a plumb¬ 
er’s spring, these small coils do not completely 
prevent collapse if very sharp bends are 
required, since they rely on preventing expan¬ 
sion of the tubes on the sides of the bend rather 
than actively preventing collapse on the inside. 
So, it is helpful if the bends can be as easy as 

Failing the availability of bending springs, 
the tube must be filled with a substance which 
is rigid enough to prevent collapse, but which 
can readily be introduced and afterwards 
removed. The material which is used is a low 
melting-point metallic filler, one commercial 
variety being known as Cerrobend. This is a 
metal having a melting temperature lower than 
that of boiling water, so that sticks or beads of 
the material can be placed in a tube (with one 
end plugged) and the filler melted into place by 
placing the tube in a saucepan of boiling water. 



Figure 7.34 Small bending springs for use with tubes. 

A tapered plug made from scrap material can 
be driven into the end of the tube, but differen¬ 
tial expansion sometimes allows the plug to fall 
out when the tube is heated and it can be help¬ 
ful to turn up a shouldered plug and soft solder 
it into the end. 

The filler material is frequently sold as 
Wood’s Metal since referring to it by this 
generic name allows a retailer to obtain his 
supplies from a variety of sources rather than 
being confined to one supplier. 

Forming the bend 

Small-diameter copper tubes are quite easily 
bent between the fingers, even if fitted with 
bending springs or filled with Cerrobend, but 
bending in this manner doesn’t always produce 
the best results, since it can mean that the bend 
is very localised. Tubes are best bent by drilling 
a suitable hole in a wooden board and well 
radiusing one edge with a round file, as shown 
in the section of Figure 7.35. 

If the tube is filled with Cerrobend, or with a 
spring, the hole can be a nice clearance size for 
the tube. The board must be sufficiently thick 
to provide adequate support at the top and the 
tube should be inserted and bent through a 

Bending board 

Figure 7.3S A section through a hole used for bending tubes. 

Figure 7.36 Nicely bent copper tubes on a model of Scamp. 



small angle by pressing down on the free end 
projecting from the hole. This work hardens 
the bent portion, to some extent, and if the tube 
is withdrawn slightly, downward pressure 
causes the next (still soft) section of the tube to 
bend, distributing the bend over a greater 
length and helping to prevent collapse. The 
process is repeated until the desired radius and 
angle are achieved, progressively distributing 
the bend along the tube. The radius formed can 
naturally be judged by eye but it is frequently 
helpful to draw out the curve on a sheet of 
paper or even cut out a template from card. 

If the tube is small in diameter and fitted 
with an external spring, bends can usually be 
formed by pressing the spring, with the tube 
inside, over a piece of round material held in 
the vice. If pressure is maintained evenly on 
both ends, this usually results in an evenly 
distributed bend. Figure 7.36 shows some 
neatly bent tubes on a model steam locomotive. 

Hot bending 

Annealed copper and brass are ordinarily 
worked in the cold condition, with repeated 
annealings being performed as the material 
work hardens during the forming process. With 
a proper technique, copper (particularly) can 
be stretched, dished, pushed and hammered 
into virtually any shape, within reason, and 
some simple examples are given above. 

Steel is not sufficiently ductile to allow it to 
be cold worked in the same way, although thin 
strips can be bent cold. If you have ever 
watched a blacksmith working at his anvil, or 
seen a large steel forging straightened while red 
hot, you will know what a wonderful ‘softener’ 
a little red heat is for steel. 

If larger steel bars do need to be bent, con¬ 
sideration should be given to doing this while 
the material is red hot. The process can be 

somewhat hazardous, however, unless the 
hearth and the vice are adjacent to one another, 
or can be arranged so. It is also necessary to 
have large pliers or tongs available for handling 
the work, and then there is the problem of plac¬ 
ing it over a former and finally bending it up, 
before the redness goes out of it. 

The need to maintain red heat means that 
there is little time to form the bend before the 
heat is lost from what will inevitably be a small 
workpiece. Some sort of jig or former is 
needed, which allows the work to be quickly set 
in position and bent immediately. 

Figure 7.37 shows a simple jig used for 
bending Visin. (5mm) diameter bright mild 
steel while hot. The jig consists of two '/iin. 
(12.5mm) steel dowels fitted into holes in a lin. 
(25mm) square steel bar which can be held in 
the vice. 

The steel rod to be bent needs an angular 
offset created by two bends in opposite direc¬ 
tions and the dowels in the bar are easily 
removable so that only one need be fitted for 
the initial bend to be put in. The square bar is 
clamped into the vice so that its top surface is 
below the top of the vice jaws and a steel bar 
used as packing to support the rod to be bent 

Figure 7.37 A bar fitted with pins which were used as the 
former for hot-bending a 5 /<jn. (5mm) steel rod. 



parallel to the vice jaw, thus permitting the rod 
to be placed quickly in the jig. 

The rod is initially cut oversize in its length 
so the position of the first bend is not critical. A 
line drawn on the square bar with a felt-tip 
marker serves to indicate the initial position. 

For the second bend in the rod, the two 
dowels in the square provide a positive location 
for the rod and also set the position of the bend. 

Remember the proverb and bend the steel 
while it is still red hot. Although there is some 
softening effect below red heat, it is a case of 
the hotter the better. It is possible to discern 
when the strength is returning to the steel, just 
as the work hardening of copper or brass 
becomes obvious during cold forming, and it is 
best to reheat and continue, rather than persist 
when the material is too cold. 


If you are equipped with an anvil and the means 
to bring the work to red heat, there is no reason 
why items cannot be forged into shape. How¬ 
ever, forging is not much practised in model 
work, due principally to the small cross- 
sections of material which are ordinarily used, 
and the consequent difficulty in maintaining 
them at a sufficient degree of ‘redness’ to allow 
the forging to proceed. It is a useful technique 
however, particularly for awkwardly shaped 
items which might otherwise need to be made 
from quite large cross-sections. 

An example of a small part which might be 
forged is the latch for the locomotive reversing 
lever shown in Figure 7.38. This requires thick¬ 
ness in two directions at right angles which or¬ 
dinarily would require much cutting away of 
material to create the shape required. 

Although small, this item is an ideal candi¬ 
date for forging since it can be cut out of a thin 
strip suitable to form the spade shape at the 

Figure 7.38 A forged latch for a locomotive reversing lever. 



bottom and then the top forged to produce the 
flat at right angles to the bottom spade. No 
large anvil is required for such a small item 
since a small, hardened steel block properly 
supported in the vice is all that is required. If a 
support for the blowtorch can be arranged near 
the vice, the work can be held in the flame with 
a small pair of pliers and immediately trans¬ 
ferred to the block when the end is nicely red. 

The amount of metal to leave when cutting 
the top needs to be nicely judged, something of 
about the same volume being required as is 
contained in the desired shape. With a job such 
as this, you may find that the bar tends to fold 
over on itself rather than be flattened, but this 
does not actually matter since the folded-over 
parts weld together when hammered at red 
heat and the result is, therefore, the same as a 
simple flattening of the work. But, for the 
welding to be successful, the steel must be 
clean, and it must also be well into the red hot 
range of temperatures. So, ‘strike while the iron 
is hot’ and make each hammer blow actually do 

some work on the material. It is no use tapping 
at it tentatively. 

Forging can equally be used for larger 
sections of material such as are required for 
locomotive coupling and connecting rods, 
which generally have small cross-sections in the 
centre portion but large bosses at each end. 

Such items can be made by ‘bumping up’ the 
ends of a small bar to form the bosses at the 
ends, or a large bar, suitable for the bosses, can 
be forged down to form the slimmer cross- 
section in the centre. 

In 5-inch gauge (VAtin. to the foot scale) 
coupling and connecting rods are relatively 
large items and forging becomes akin to ‘real’ 
blacksmith’s work. A proper hearth and 
adequate source of heat are required, and the 
work must be packed round with firebrick or 
ceramic blocks and allowed to ‘soak’ adequately 
to attain the necessary temperature. A large, 
well-supported anvil is required, together with 
large hammers, and the component must 
naturally be struck while it is still red hot. 



Metal joining I - 
mechanical methods 


The items forming components of an assembly 
or sub-assembly naturally require attachment 
to one another. The methods used for such 
attachments are many and varied. Items may be 
individually threaded and screwed together, 
may be fastened by nuts and bolts, be riveted 
together or joined by soldering, with or with¬ 
out the benefit of rivets. Welding is also possi¬ 
ble, and is, today, much more common in the 
amateur’s workshop although this highly spe¬ 
cialised technique must be excluded from this 
general guide. 

However, even with this exclusion, there 
remain the methods listed above, together with 
the method which might easily be described as 
gluing except that this process is not readily 
thought of as being applied to metals. Yet it is 
progressively replacing what would previously 
have been accomplished by ‘press’ or ‘interfer¬ 
ence’ fits. In many ways it is much more con¬ 
venient than the method it replaces since it 
simplifies machining by allowing wider toler¬ 
ances and is therefore to be recommended on 
these grounds alone. 

Since the methods are many and varied, the 
subject matter is divided into two. What I have 
called mechanical methods - screwing, bolt¬ 
ing, riveting etc. - are described here, and 
soldering and gluing, both of which introduce a 
binding material into the joint, are described in 
Chapter 9. 

Screwed fastenings 

Common fastenings 

The most frequent forms of engineering 
fastenings are those known colloquially as nuts 
and bolts in which male- and female-threaded 
items provide a clamping action to fasten items 
together. The externally threaded items are 
known either as screws or bolts, according to 
whether they are threaded to the full length of 
the shank (screws) or are only partially 
threaded (bolts). 

Screws and bolts are ‘headed’ devices, the 
most common plain types of head being those 
shown in Figure 8.1. With the exception of the 





Cheese head 


I ( 

Countersunk Hexagon 

head head 

I I 

Figure 8.1 Screw and bolt head types. 

hexagon-headed type, all are provided with 
machined heads for use with a screwdriver, the 
most common heads having the simple single 
slot for the conventional screwdriver rather 
than the more recently introduced cross- 
pointed types. In industrial instrument work 
the newer (in the UK) metric fastenings are per¬ 
haps more commonly available with cross- 
point heads but the older-established threads 
such as BA, BSW and BSF are still most likely to 
be found with the single-slot head. 

With the exception of the hexagon head, 
common fastenings of the types shown in Fig¬ 
ure 8.1 are generally available in plated finishes 
which suit general applications rather than spe¬ 
cific modelling needs. They will be found to be 
available principally in brass and steel, but 
some threads and head types are available in 
stainless steel. 

Hexagon-headed bolts and screws are not so 
readily available in these materials and finishes, 
although they will be found in some threads. 
These are, however, less useful generally than 
hexagon-headed items prepared especially for 
modelling but are sometimes needed for instru¬ 
ment making or repairs generally. 

Whenever possible, washers should be used 
immediately below a nut in order to spread the 
load and prevent the sharp corners of the nut 
hexagon from marking or damaging the job 
(nuts are not always chamfered on both sides). 
A washer should also be used under a screw 
head whenever a screw has to be turned to 
tighten the fastening, for example when the 
screw enters a tapped hole. This must include 

instances in which a countersunk head screw is 
used to provide a low-profile fixing, but with¬ 
out countersinking the hole in the component 
which is being attached. In these cases, a cup 
washer having a countersunk hole is used 
below the head. 

Locking washers are also available - these 
take the form of a tempered spring washer, 
either a single coil or a double coil, or are made 
as a hardened steel washer with internal or 
external radial ‘fingers’. These latter types, 
known as internal or external shakeproof wash¬ 
ers, are designed to ‘bite’ into the surface of 
both the nut and the item being clamped, 
thereby preventing nut release except when sig¬ 
nificant torque is applied. Damage to the 
clamped item is therefore to be expected with 
shakeproof washers but is likely to occur with 
spring washers also. For this reason spring 
washers are frequently used in conjunction with 
a plain washer which both spreads the clamping 
force, and prevents damage, except to the un¬ 
derside of the screw or bolt head. 

For models, spring or shakeproof washers 
are not often appropriate and thread locking by 
one of the other methods (described below) is 
more common. 

Short, headless versions of screws are also 
available in the more common threads. These 
are known as grub screws. They are intended 
for use as internal fastenings which will screw 
into or locate in a shaft, for example, locking 
two items together by passing through a radial 

Socket-headed screws 

In the smaller sizes, the common fastenings de¬ 
scribed above tend to be produced in easily ma¬ 
chined, low-strength materials, hence the fre¬ 
quent use of plated finishes which suit the deco¬ 
rative rather than engineering uses envisaged. If 
a high-strength fastening is required, this need 



is most likely to be satisfied by a socket-headed 
screw. These are manufactured in high-strength 
(high tensile) steel alloys in all of the common 
threads and in a range of headed and headless 
types which equate with the normal types of 
screws, bolts and grub screws. These screws 
and bolts are provided with a hexagon-shaped 
socket in the head which takes a key or wrench 
which allows them to be easily inserted or 
removed. The three head types shown in Figure 
8.2 can be obtained readily, being described as 
socket-head cap screws, grub screws and coun¬ 
tersunk screws. 

A common method of using cap screws is to 
counterbore the top of the clearance hole to 
allow the head to be recessed below the surface. 
The head depth must suit the thickness of the 

Figure 8.2 Three types of Allen screw. 

item to be secured, but if reasonably thick items 
need to be fastened by screws, the result of 
counterboring and recessing produces a neat 
arrangement, as Figure 8.3 shows. 

These screws, known colloquially as Allen 
screws, are intended as a high-strength engi¬ 
neering fastener. They are consequently avail¬ 
able in a very wide range of sizes, modellers’ 
suppliers usually stocking BSF, BSW and BA 
sizes, in various lengths, from 3 /sin. diameter 
down to 8 BA. They are also available in a wide 
range of metric sizes from M20 to Ml.4. They 

Figure 8.3 Recessed Allen screws on a rear toolpost. 

are generally available having a black finish and 
are ideal for the assembly of tools or when their 
high strength is needed, as for big-end bolts for 
IC engines, for example. 

The hexagon wrenches are universally 
known as Allen keys. 

Providing location for bolted fastenings 

One important point about nut and bolt 
fastenings is that they are designed to clamp 
items together, not to locate them in some par¬ 
ticular alignment. The hole drilled to accept a 
bolt or screw is consequently called a clearance 
hole. This is deliberately made larger than the 
screw in order to allow for tolerances in screw 
size and to permit slight positional errors in the 
placing of fixing holes in the components to be 
joined. The fixings themselves cannot be relied 
upon for location, only for clamping. 

If positive location of the screw-clamped 
parts is a requirement, this must be arranged by 
separate means. For some components this is 
easily done by slight modification to the design, 
but for others, specially fitted pins or specially 
made bolts are required. Two examples are 
shown in Figure 8.4. The items shown comprise 
the upper and lower halves of a two-tool turret 
designed to hold commercial parting-off blades 
on a lathe. To provide location, the lower half is 



Figure 8.4 The two parts of the tool clamp on a rear toolpost 
showing the locating register and alignment pins. 

provided with a precisely turned spigot which 
locates in a boring in the toolpost body. 

To prevent the upper and lower halves of the 
clamp from becoming disoriented, two ground 
silver steel pins are pressed into the lower half. 
These locate in accurately sized (reamed) holes 
in the upper half, ensuring accurate location of 
the parts. 

This use of locating pins, or dowels, is very 
common, both as a means to ensure that no 
misalignment occurs when the clamping bolts 
are slackened off (as here) or to ensure that 
parts are fitted together only in the correct ori¬ 
entation. This latter use is simply arranged by 
placing dowels asymmetrically so that incorrect 
assembly is not possible. 

Hardened and ground dowels are available 
commercially and can be fitted into accurately 
reamed holes in the items to be joined. Alterna¬ 
tively, silver steel pins may be used, since this 
material is accurately ground to size. It can, of 
course, be hardened also but this is not always 
necessary unless frequent removal and replace¬ 
ment of the items is envisaged. Indeed, it may 
be desirable to fit a hardened steel bush to mate 
with the dowel if very frequent movement or 
dismantling is envisaged. 

The lower part of the toolpost turret, with 
its accurate register for location, exemplifies 
the usual approach to location of circular work. 
Provided that there is room, a register or spigot 
locating in a bore is frequently the best way to 
ensure accurate location. The rear cover for a 
steam locomotive cylinder is shown in Figure 
8.5 and will be seen to have a shallow spigot 
turned on the inside which locates in the cylin¬ 
der bore. This ensures concentricity between 
the bore and the central piston rod hole when 
the cover is assembled to the cylinder block, 
provided that both the central bore and the 
register, or spigot, are machined at the same 
setting in the lathe. 

Figure 8.5 Cylinder covers are usually provided with a 
shallow register which locates the cover in the bore. This is 
essential for covers which are bored for the piston rod. 

Bolts and fitted bolts 

If items having other than minimum thick¬ 
nesses are to be fastened using nuts and bolts, 
bolts should strictly be employed, rather than 
screws, the difference being that a bolt has an 
unthreaded portion of shank below the head. 
This portion is usually machined at the nominal 
diameter of the thread and is consequently 
liable to be closer to its nominal size than the 
threaded portion. A bolt is, therefore, capable 



Figure 8.6 Small- and large-headed 4BA screws. 

of providing somewhat better location than the 
‘all-thread’ screw and should be used whenever 
there is sufficient thickness to allow a length of 
plain shank to be employed. 

Four BA screws are shown in Figure 8.6, two 
of which have hexagon heads which are one 
size smaller than the standard. Such screws 
(and similar bolts) are frequently used in mod¬ 
elling work since they provide a better scale 
appearance than fastenings with normal-sized 
heads. The screws shown are 4BA. 

In some situations, fitted bolts may be used 
to allow the fastening to perform both the 
locating and clamping functions. This is useful 
whenever there is a possibility that the fric¬ 
tional clamp provided by the nuts and bolts will 
not be sufficient to resist the likely loads. An 
example is the model locomotive reversing 
lever stand shown in Figure 8.7. The stand is 
held to the frames by three bolts, but since it is 
subject to a turning motion (a torque) in use, 
large clearances between the bolts and the holes 
in the frame and stand may ultimately mean 
that the stand will overcome the frictional grip 
provided by the nutted bolts and a sloppy fit 
will result. 

Locking of the nuts helps to prevent the fric¬ 
tional grip slackening, but the problem may 
still not entirely be prevented unless a positive 
location for the stand is incorporated. This 

* V 

Figure 8.7 The reversing lever stand for a model locomotive. 

could be provided by dowels of the type shown 
in Figure 8.3, but since both the stand and the 
frame are only '/sin. (3mm) thick, there is little 
real ‘meat’ into which a dowel can be fitted 
(one component, at least, really needs to be rea¬ 
sonably deep in order to provide adequate grip 
for the dowel). 

An alternative is to clamp frame and stand 
together and to ream the fixing holes accurately 
to size by machining both items together. In the 
case illustrated, there was sufficient room for 
three such holes. Three, special, hexagon-head 
bolts were then made up having an unthreaded 
portion turned to a good (tight) fit in the 
reamed holes with a smaller screwed portion to 
be nutted and locked into position. The special 
bolts with precisely sized shanks thus provide 
both the clamping and the location (resistance 
to rotation) which are required, the nutted fas- 



Figure 8.8 A fitted bolt for the 
stand of Figure 8.7. 

tening overcoming the objection to the fitting 
of dowels in the thin plates. 

For obvious reasons, these fastenings are 
known as fitted bolts. The type used for the 
reversing lever stand is shown in Figure 8.8. 
These were turned up from hexagon mild steel 
and the parallel portions made to a good fit in 
the reamed holes. Given a good fit and prop¬ 
erly tightened (and locked) nuts, this type of 
fastening is suitable in situations where a hard¬ 
ened dowel cannot be used. 

Thread locking 

Lock nuts 

Whenever a fastening is liable to become loos¬ 
ened in service, for example, if subject to vibra¬ 
tion, a means of locking it is usually provided. 
To allow simple locking, a number of specially 
produced lock nuts are available - the most 
common of these are self-locking nuts. These 
comprise two basic types, each of which is 
effectively a standard nut having an extension 
piece machined on one side. The extension is 
either formed as a threaded portion of the nut, 
but machined (slit) in such a manner that this 
part of the thread can be distorted during 
manufacture so that it is a tight fit on the screw 
or bolt. The alternative type, and nowadays the 
more common, has a hard plastic insert fitted 
into the extension. This resembles a washer in 
that it is unthreaded, but the central hole is 

drilled or moulded to be a tight fit on the screw 
or bolt thread. The plastic originally used for 
these nuts was nylon and the trade nam eNyloc 
Nut has persisted as the colloquial description 
of these types. The alternative, all-metal type is 
normally known as a self-locking nut. Both 
types are illustrated in Figure 8.9. 

In the past, a much-used locking nut was the 
castle nut, or castellated nut. This is essentially a 
standard (full) nut machined with four or six 
radial slots across one face. These serve to allow 
a split (cotter) pin to pass through a diametral 
drilling in the screw or bolt and through two 
opposite slots in the nut. The cotter must be a 
good fit in both the drilling and the slots in 
order to ensure that no rotation of the nut is 
possible once the cotter is inserted. 

These are seldom used on models, except to 
represent a prominent feature on the proto¬ 
type, and other means are usually employed to 
retain nuts when this is done for reasons of 

Figure 8.9 A//tocand self-locking nuts. 

Full and half nuts 

The published thread specifications normally 
specify standard dimensions for all aspects of 
the threaded fastenings which they describe. 
These include head dimensions for bolts and 



screws, together with two thicknesses of nut, 
known as full and half or standard and thin nuts. 

Thin nuts are provided for use as lock nuts 
and are frequently described as such. For lock¬ 
ing, they may be used in pairs, one to provide 
the fastening and the second locked tightly 
against the first to prevent slackening. Where 
there is room, it is usual to fit a full nut for 
fastening and to use a half nut for locking. 

Liquid thread locking 

Paint and varnish have been used industrially 
for the locking of small threads for many years. 
A thread-locking varnish has been the most 
common material used, this usually being 
painted onto the protruding thread of the 
screw after tightening the nut. This is perfectly 
adequate as a locking method since the varnish 
fills the screw thread (and any clearances 
between the nut and the screw at the outer end) 
and therefore prevents the nut slackening until 
adequate force is applied to it. The method is 
cheap, causes no surface damage and does not 
add to the number of items to be assembled, yet 
allows removal when required. 

Specially prepared varnishes are not neces¬ 
sary since ordinary paints can be used, enamel 
or gloss being perhaps best since these have 
more ‘body’ than emulsion or undercoat. 
Rather than applying the paint or varnish exter¬ 
nally, it is perhaps best to place a drop on the 
screw thread before its insertion into the 
tapped hole or before placement of the nut. 
Clearances between the two threads are then 
filled with paint and removal requires applica¬ 
tion of adequate torque to break the paint film. 

Current techniques for metal joining have 
made available various grades of locking fluids 
having more precisely defined characteristics 
than a drop of whatever old paint happens to 
be available. Loctite is the most readily avail¬ 
able range which offers medium- and high- 

strength versions (Loctite 242 Nutlock and 
Loctite 270 Studlock, respectively) for place¬ 
ment on the threads before engagement, or 
Loctite 290, which is a penetrating adhesive for 
application to threads after assembly. 

These fluids are members of a group known 
as anaerobic compounds, the group including 
several other fluids which form permanent or 
semi-permanent bonds between metals. All of 
these fluids are described in Chapter 9. 

Shortening screws 

One frequently required operation is the short¬ 
ening of screws. The requirement arises be¬ 
cause of the non-availability of short screws in 
the smaller sizes, these often being produced in 
only a limited selection of lengths. Shortening a 
screw is, of course, not a problem, but creating 
a nicely square end which will readily engage 
with a nut does frequently cause aggravation. 
The way out of the difficulty is to utilise a 
device which was once one of the first items 
made by the instrument maker’s apprentice. 
This is the nut plate. 

At its simplest, the nut plate is a piece of mild 
steel plate having in one end a few tapped holes 
to suit the range of threads which you normally 
use. The holes can be quite closely spaced and 
the plate should be no thicker than the shortest 
screw you think you will need. For the smaller 
sizes, such as the BA threads, this might be '/sin. 

For shortening, a screw is inserted into a 
tapped hole and screwed in sufficiently to leave 
the unwanted thread clear of the plate. The 
plate should be sufficiently long that it can be 
held in the vice, with the end with the tapped 
holes upstanding sufficiently to use a hacksaw 
to cut off the excess thread and to use a file to 
smooth the ends. If the holes are tapped 
squarely in the ends of the plate, the end of the 
screw is automatically made square when it is 



filed off flush with the plate’s surface. Remov¬ 
ing the screw from the tapped hole removes 
most of the burrs that remain on the end, and a 
few strokes from a smooth file quickly produce 
a small chamfer. 

The nut plate may be as elaborate or as sim¬ 
ple as you please, from a piece of mild steel 
plate, as described above, to a stepped plate 
which allows specific screw lengths to be pro¬ 
duced every time. It might be hardened, if you 
wish, but this is by no means essential, and the 
choice therefore rests with the individual. My 
plate, made from a strip of black mild steel, is 
shown in Figure 8.10. 

Rivets and riveting 

General introduction 

Until the development of reliable, high-strength 
welded assemblies, larger engineering struc¬ 
tures were normally assembled by riveting 
together cut and shaped plates and suitable 
‘doubling’ or strengthening pieces. The result¬ 
ant structures possessed a degree of flexibility 
(which is beneficial) and enabled large assem¬ 
blies to be created from relatively small plates. 
Hot riveting was the method frequently 
employed, red hot round-headed rivets being 
pushed into pre-drilled holes and their plain 
ends formed into a round head while still red 
hot. The round-headed shape provides a good 
clamping area, and forming the rivet while still 
red hot ensures that it can be expanded to fill 
the holes in the plates, and the head can be 
formed fully. The rivet also increases the pres¬ 
sure on the plates as it contracts on cooling. 

In modelling, hot riveting is not necessary, 
but it is a requirement that the rivet fills the 
holes in the plates and therefore provides lateral 
location of the parts being joined. A good head 
also needs to be provided on the formed rivet. 

Figure 8.10 A simple nut plate. 

In modelling, this is usually desired principally 
on the grounds of appearance, but a good head 
shape, matching the commercially produced 
head, also allows the required strength to be 
achieved, if this is necessary. 

Rivets are produced in many forms - flat 
head, mushroom head, tubular, semi-tubular, 
bifurcated etc. - in a variety of materials and 
finishes. Those of interest to the modeller are 
likely to be countersunk or round-headed 
(usually called snaphead rivets) in steel, iron, 
brass, aluminium or copper. 

Rivet spacing 

In an engineering sense, a riveted joint needs to 
be correctly proportioned if it is to function 
correctly and realise the full strength potential 
of the materials forming the joint. When 
considering modelling, the basic engineering 
criteria seldom apply since the aim is to pro¬ 
duce a miniature of some device which closely 
resembles the device itself, usually called the 
prototype. Rivet spacing and head size are 
therefore frequently determined by the proto¬ 
type itself and the choice of size is removed. In 



many instances, the rivets on the model may be 
dummies masquerading as fastenings but in 
reality are only present to represent the appear¬ 
ance of the prototype. 

For joints which are to be joined by solder¬ 
ing or brazing, it is frequently necessary to 
provide temporary support and location during 
the joining operation. In these instances, any 
fastenings used do not contribute to the ulti¬ 
mate strength of the joint and they can be 
spaced out conveniently provided that they will 
ensure adequate closeness of the joint compo¬ 
nents during heating and cooling. If riveting is 
employed in these instances, design criteria 
again play no part. 

Other modelling uses for rivets are in situa¬ 
tions in which, traditionally, overscale rivets 
have been used, for example for attachment of 
angles to the buffer beams of model locomo¬ 
tives. Use of such overscale fastenings means 
that it is not appropriate to form the rivet head 
fully on the ‘visible’ side. It is common to ham¬ 
mer the rivet down into a countersink and then 
to file it off flush with the outer plate. To 
achieve adequate strength, overscale thick¬ 
nesses are often employed for the components, 
making rivet selection more a question of what 
can be squeezed in, rather than what is theoreti¬ 
cally desirable. 

However, there inevitably arise situations in 
which joints do need to be riveted and a simple 
treatment of the subject is therefore included 
here. First of all, rivet size. For general engi¬ 
neering structural work, it is normally accepted 
that, whatever the type of joint, the rivet diam¬ 
eter should be twice the thickness of the plates 
being joined. However, the formulae quoted do 
not provide a linear relationship between plate 
thickness and rivet diameter, a 2in. (50mm) 
rivet being recommended for 2in. thick plates 
and a I'/iin. (38mm) rivet for lin. (25mm) 
thick plates. 

Published data refer, for the most part, to the 
type of joints likely to be made in forming all- 

riveted, full-size boilers. Consequently, their 
applicability to the smaller and lighter assem¬ 
blies to be found in the modeller’s workshop is 
limited, but for practical purposes it may be 
assumed that rivet diameters between 1 Vi and 
twice the plate thickness should be employed. 

Riveted joints are classified according to 
whether they are single, double or triple riveted 
and whether they are butt or lap joints. Single 
and double riveted of both types are shown in 
Figure 8.11. Butt joints may be strengthened by 
the use of a second reinforcement plate, effec¬ 
tively sandwiching the main plates. For a dou¬ 
ble-riveted butt joint, the second plate is fre¬ 
quently arranged to span only the two central 
rows of rivets being itself only single-riveted to 
the other plates. 

Rivet spacing, and the relationship between 
the rivet holes and the edges of the plates are 
important in forming a sound joint. During the 
riveting process (described below) the plates 
are squeezed together and some distortion 
inevitably results. If the rivets are too close 
together, or too close to the edge of the plates, 
splitting can occur while the rivets heads are 
being formed. The rivet hole should never be 
closer to the edge of the plate than the rivet 
diameter and the rivet spacing should be Vh to 
3 times the rivet diameter. 

Figure 8.12 summarises the above simple 
rules for rivet size and spacing, but it should be 
emphasised that these figures must be used only 
as a guide. All-riveted construction is relatively 
rare in model work, the joint usually being rein¬ 
forced in some way, rather than the rivets being 
relied on entirely. Riveted water tanks are 
normally sealed by soft soldering and the rivets 
therefore become decorative rather than func¬ 
tioning as fastenings. In model boiler work, 
butt joints are frequently used in the construc¬ 
tion of boiler barrels and fireboxes but the 
joints are made sound and leakproof by silver 
soldering after riveting, and the principal 
concern during riveting is to ensure good clos- 





mi SBH p«mb 



Single riveted Double riveted 













Single riveted 

Double riveted 

Figure 8.11 Riveted joints. 




















, Vs2 

7 /m 


7 /lt 

J Mt« 









Vi 6 








For cither sizes, calculate as above 


























7 /t« 


! 'A2 

2 A2 

7 /lS 













1 A 








For other sizes, calculate as above 

Figure 8.12 Recommended rivet spacings. 



ing of the plates and butt strap(s) without 
distortion of either so that the silver solder can 
do its job of uniting the items to provide proper 
strength to the joint. 

Unless you are particularly skilled in the 
design of pressure vessels, it pays always to 
follow a particular boiler design meticulously, 
particularly regarding the arrangements speci¬ 
fied for any butt joints. 

Basic procedure 

One of the problems posed by the riveting 
operation is possible damage to the plates being 
joined. This is caused by the hammering which 
shapes the rivet head. This can have the effect 
of squashing the plates and hence stretching 
them. The effect is more noticeable if the plates 
are copper, soft brass or aluminium since these 
materials are very ductile and can readily be 
squeezed into alternative shapes. 

If the plates to be joined are substantial, the 
effect may not be noticeable. This is usually 
also the case where the length to be riveted is 
short and the rivets are not placed too close to 
the edges of the plates. If the items to be joined 
are long, with closely spaced rivets, there may 
be significant stretching and considerable 
increase in length, especially for an item such as 
the locomotive running board shown in Figure 

In the case of this assembly, the running 
board itself is a robust 16 swg mild steel sheet, 
but it will be riveted to a brass angle since this 
was the only correctly-sized material available. 
The mixture of materials is unfortunate since 
the soft angle will distort easily during riveting. 

Two points arise from the above. First, the 
rivets used should not be significantly harder 
than the plates to be joined. They are normally 
of the same material as the plates, or the softer 
of the two, if the metals are dissimilar. Second, 
for a long length such as the running board of 

Figure 8.13 A model locomotive running board. 

Figure 8.13, only one item should be drilled 
fully prior to starting the riveting operation. 

If both are fully drilled, holes in the two 
items rapidly become displaced as rivets are 
successively inserted and their heads formed. It 
is essential to drill only the one component, say 
the running board itself initially, and then to 
drill the angle progressively as riveting takes 

It is helpful to start at the centre of a long 
length and also beneficial to insert some rivets 
here and there, say every three or four inches 
(75mm or 100mm) along the plates, as an 
initial operation. This establishes the two parts 
in correct alignment and the effects of distor¬ 
tion are minimised. 

Forward planning of this sort is required, 
since the brass angle will curve into an arc as 
the rivets are inserted, as shown in Figure 8.14, 
and without some holes drilled, and held, at 
least by temporary screws, progressive distor¬ 
tion will render the task of positioning the 
angle for riveting more and more difficult. 

Distortion of the angle can, of course, be 
minimised by using soft rivets and by avoiding 
any excessive hammering beyond that which is 
absolutely required to form the head of the 
rivet. In this respect it is probably best if the 
commercial head is placed against the angle but 
since this brings the home-made head upper¬ 
most in this instance, it may be preferable to 
place the commercial head at the top and to 



Local stretching due to rivet dosure 

deal with the distortion as riveting progresses. 

For all riveting, it is important that the rivet 
is properly supported during the forming 
operation. This is especially important when 
riveting soft materials otherwise the force of 
the hammer blows may be taken by the plates 
and severe distortion will result. The riveting of 
soft copper, as in boilermaking, is frequently a 
problem in this respect and it is often preferable 
to use screws (but not brass or steel) for holding 
the plates together for silver soldering. 

Riveting dolly 

The operation of inserting and forming the 
head of a rivet is referred to as ‘rivet driving’, 
‘heading up’ or ‘dosing up’, this latter descrip¬ 
tion describing the action of the rivet in bring¬ 
ing the plates together. With few exceptions, 
one side of the formed and closed-up rivet will 
comprise a round or snaphead. On occasions, 
this can simply be the commercial rivet head 
since, if the rivet is to be countersunk on one 
side it is the countersink which can most simply 
be formed. This can be done using only a suit¬ 
able hammer, but a correctly shaped support 
for the commercial head will be required in 
order to allow rivet closure and preserve the 
head shape. 

This support is usually called a dolly and tra¬ 
ditionally takes the form shown in Figure 8.15. 

A sturdy bar is drilled at some convenient point 
to allow the dolly to be inserted. This is a short 
length of hardened and tempered silver steel 
machined to fit the drilling in the bar and hav¬ 
ing a machined recess in the upper end into 
which the rivet snaphead is a good fit. 

This form of support is not necessarily the 
best for the single-handed worker however, 
since the edge of the hard dolly will mark the 
work if it is not supported absolutely squarely 
in relation to the dolly during the closing up 
(hammering) process. It is frequently better to 
dispense with the dolly and to machine the 
recess for the rivet head directly in the bar 
itself. This allows the plates which are to be 
joined to lie directly on the bar’s surface and 
there is little likelihood that they will be 
marked during riveting. 

From descriptions in many books it would 
seem that the traditional method of creating 
the recess for the rivet head is to drill an appro- 

Figure8.l5 The traditional form of riveting dolly. 



priately sized ‘starter’ hole using an ordinary 
twist drill and then to hammer a suitable hard¬ 
ened steel ball into the drilling to create the 
required shape of recess. This doesn’t seem 
very scientific even assuming that the correct 
size of ball is available, and it inevitably results 
in the ball flying across the workshop to be lost 
in some far corner. 

A much better method is to make a cutter 
which cuts the correct profile directly, and 
avoids the need for hardened balls and ham¬ 
mering altogether. The most suitable form of 
cutter may be made from silver steel, turned to a 
suitable diameter and radiused on the end 
to suit the rivet head. This end radius, as 
rqachined, will resemble the rivet head (a 
spherical radius) and to create a cutting edge, 
roughly half of the end is machined or filed 
away to create a round-ended D-bit (see Chap¬ 
ter 15). The cutter should be hardened and tem¬ 
pered (pale straw) after which it may be used to 
cut the snaphead recess as a follow-up to drill¬ 
ing an initial hole using a twist drill. 

The dimensions of the recess can be 
obtained from Figure 8.16 which shows the 
proportions of the common snaphead rivets. In 
the USA, the snaphead rivet is known as the 
button-head rivet. 

/-0.9D Rad 


Figure 8.16 Snaphead rivet dimensions. 

Figure 8.17 Dimples for snaphead rivets, produced by 
drilling, in the end of a I in. (25mm) square bar. 

The purists will perhaps frown, but it is 
possible to grind the end of a twist drill to form 
the recess for the snaphead and those shown in 
Figure 8.17 were cut by this means. A set of 
three recesses is shown, two close to the 
corners of the bar and one placed equidistant 
from the sides and the end. This allows rivet 
heads to be supported close to obstructions on 
the workpiece (or other rivets) and allows the 
bar’s top surface to be used in different situa¬ 
tions to support the work. 

When forming the recess, it is best to test the 
shape and depth by using a commercial head 
but care must be exercised in selecting the 
standard against which the recess will be judged 
since some rivets in a batch always have slightly 
malformed heads, so select the best out of half- 
a-dozen or so. 

Closing up the rivet 

The essential stages in closing up a rivet can be 
summarised as follows: 

1. the drawing-together of the plates 

2. the squashing of the rivet so that it expands 
into the holes in the plates 



3. the forming of the head. 

These essential steps must be performed in the 
above order. If the plates are not drawn 
together into intimate contact before the proc¬ 
esses to swell and form the rivet are complete, 
the rivet will expand between the plates and the 
joint will not be sound. 

All of the processes concerned with the clos¬ 
ing up and forming of the rivet must be per¬ 
formed with the above objectives in mind and 
this influences the method and the design of 
punches used to form the head. In this respect it 
is essential that the rivet is properly supported 
during the closing operation so that hammer 
blows, or blows from a forming punch, truly 
affect the rivet and not the plates being joined. 
The more solid can be the support, the better 
the results are likely to be. 

The simplest type of rivet closing is that in 
which the unformed end of a snaphead rivet is 
hammered into a countersink and afterwards 
filed off flush with the surface of the plate. This 
method is frequently used in situations in 
which the strength of the riveted joint demands 
a larger rivet (and hence rivet head) than scale 
appearance would demand on a model, such as 
the attachment of frame fixing angles, or horn- 
blocks, to model locomotive frames, as shown 
in Figure 8.18. 

Figure 8.18 The rivets holding the small angles to this buffer 
beam have been hammered down into countersinks and filed off 

For such situations, the snaphead of the rivet 
is placed inside and the countersink formed on 
the outside of the frames. A plain, or ordinary, 
countersink is not ideal in these situations since 
it leaves a depression tapering away to nothing 
creating a very weak edge to the formed rivet. It 
is best to use an ordinary twist drill to form the 
countersink, creating a recess as shown in 
Figure 8.19 having a shallow, parallel-sided 
portion which provides increased strength for 
the formed head. 

Before bringing the two items together, each 
must be de-burred so that good contact is 
achievable. The holes should have been drilled 
so that the rivets are a tight fit, and the rivets 
cut off, leaving enough material to fill the coun¬ 
tersink. Cutting off must leave the rivet end 
reasonably smooth and burr-free since the first 
operation is to ensure that the head is seated 
correctly against the lower plate, and the two 
plates are in close contact. To do this, a brass or 
copper punch of the form shown in Figure 
8.20A is used to punch the two items down on 

Straight-sided countersink 


Figure 8.19 The recommended shape of the countersink has 
a short parallel portion which gives strength to the formed rivet. 
The method of supporting the head of the rivet on a dolly should 
be noted. 




b "I 

Diameter - D 

Make in brass 



Figure 8.20 A set of punches such as these is needed for 
forming a snaphead on the end of a rivet. 

to the rivet head which is supported in a dolly, 
as described above. This ensures that the work 
is ready for rivet closing; the rivet head is prop¬ 
erly seated in the dolly and in contact with the 
lower plate, and the plates are closely in con¬ 

For forming the end of the rivet to fill the 
countersink, an ordinary hammer is used, the 
first blow being directed straight down onto 
the rivet with the objective of swelling it within 
the holes. It should be tested at this stage to 
check that it is indeed firm in the hole. 

With the rivet head once more supported in 
the dolly, successive blows are now aimed at the 
end of the rivet to spread it into the straight¬ 
sided countersink. Once each rivet has been 
dealt with in this way, the ends are filed off 
flush, when they should be virtually undetect¬ 
able, as Figure 8.18 shows. 

Forming snaphead rivets 

If a snaphead is required on both sides of a riv¬ 
eted joint, the rivet must be closed up using a 
series of shaped punches. The object is still to 
ensure good contact between the plates, to swell 
the rivet in the hole and then to form the head. 
These processes cannot readily be achieved 
using a single, snaphead-shaped punch and the 
closing up must therefore be performed in sev¬ 
eral stages. A suitable set of four punches is 
shown in Figure 8.20. 

The recess shapes shown are designed to 
perform the closing functions progressively. 
Punch A should be made in something soft such 
as copper or brass since it is designed to be used 
directly against the upper plate to force both 
plates together and to press the rivet into firm 
contact with the lower plate. 

Once the rivet is cut off to length (see below) 
Punch B is used to spread the top of the rivet 
slightly, but more importantly, to compress it so 
that it expands in the hole. After use of this 
punch, it is wise to check that the rivet is firmly 
held in the plates and correctly seated. 

Following the initial compression, Punches 
C and D are used to form the head progres¬ 
sively into the required shape. 

The above description assumes that the rivet 
length standing clear of the upper plate pro¬ 
vides the correct amount of material to create a 
satisfactory head shape. For the snapheads of 
the types shown in Figure 8.16, an upstand of 
between 1.6 and 1.7 times the rivet diameter is 
generally satisfactory, although the amount 
required is dependent upon the relative 
hardness of the rivet and the materials being 
joined, and the fit of the rivet in the holes. It 
is beneficial to treat the first one or two rivets 
as the means of determining the length 
required, assuming that the drilling has been 
(or will be) carefully carried out so that a 
standard fit between rivet and holes will be 



For the common rivet sizes, the following 
upstands should be used as a guide: 

Rivet Diam. 



0.1 Oin. to 0.11 in. 

V) 2 in. 

0.15in. to 0.16in. 


0.20in. to 0.21in. 


0.25in. to 0.27in. 

J /i6in. 

0.30in. to 0.32in. 

As noted above, the cutting off of the rivet 
needs to leave it relatively burr-free and clean 
to allow Punch A of Figure 8.20 to be used to 
close the plates and seat the head correctly. 
Cutting off with side cutters is convenient but 
does not satisfy the basic requirement for a 
cleanly cut end. Cutting off by this means can¬ 
not, therefore, be performed before the rivet is 
inserted into the hole and if performed after¬ 
wards will frequently be found to eject the rivet 
from the plates, rendering the use of Punch A 
obligatory. With side cutters it is difficult to 
judge the cut correctly to leave the required 

The best compromise is to use a drilled plate 
(like the nut plate described above in relation to 
the shortening of screws) perhaps with a spacer 
washer below the head to ensure the correct 
length. Side cutters (or similar) may then be 
used to cut the rivet and a few strokes with a 
sharp file will leave the end sufficiently burr- 
free to use directly. If very many rivets are 
needed, a hardened plate and a sharp cold 
chisel will be found to expedite the rivet¬ 
cutting process. 

Press fits 


An extremely common way to assemble two 
cylindrical items in a semi-permanent way is to 

utilise a press fit. In this type of fit, a boring or 
housing is made just too small in diameter to 
accept the item which is to be fitted into it. The 
two must be pressed together with some force 
in order to allow the inserted item to enter the 
bore. To achieve this type of fit there must be 
‘interference’ between the two items, this just 
being another way of saying that the bore is 
slightly too small. 

In pressing the items together, there must be 
some expansion of the housing and some 
contraction of the inserted item. How much 
distortion occurs is dependent on the relative 
strengths of the two parts and on the amount of 
interference which has been allowed for in 
making them. The amount of interference also 
affects the difficulty, or otherwise, of removing 
the inserted item and whether removal may be 
required is one factor in determining the allow¬ 
ance for interference. 

Press fits are typically used for fitting bear¬ 
ings, either plain bearing bushes or ball- and 
roller-bearing types. For plain bearings, the 
bush is made or fitted so that it provides 
running clearance for the shaft without further 
boring or reaming after assembly. Commer¬ 
cially produced plain bushes are available 
having precisely sized bores and outside diam¬ 
eters, specially for this type of fitting, but it is 
clearly important that the interference fit used 
is just right in the sense that the bush must not 
be squeezed in by any significant amount when 

For the fitting of ball and roller bearings, 
both the housing and the shaft must be sized to 
provide the correct interference fit with the 
relevant part of the bearing. As for plain bear¬ 
ings, only a Might’ interference is allowed. This 
calls for accuracy in sizing both components so 
that the inner and outer of the bearing are fitted 
sufficiently tightly to prevent their rotation in 
the housing, or on the shaft, but the bearing is 
not distorted when assembled, hence taking up 
its internal clearances and damaging the balls 



or rollers, and the tracks. 

An alternative use of interference fits is the 
permanent pressing together of two items 
which will not need to be subsequently disas¬ 
sembled. In these cases, a greater amount of 
interference is allowed in order to ensure no 
relative movement between the two items in 
service. In modelling, or model engineering 
generally, common uses for this type of fit are 
for fitting locomotive wheels onto their axles, 
and for the ‘building up’ of crankshafts, in 
addition to the ordinary fitting of bearings, 
described above. 

The interference fit 

To allow precise control of sizes and tolerances 
in industry, ranges of standard fits are specified 
by the various national authorities. These allow 
for the manufacture of shafts and housings with 
a range of tolerances for both which provide for 
various fits from loose clearance, through run¬ 
ning and location to push and press fits. In some 
specifications, the description ‘force fit’ is used 
rather than press fit. For the fitting of bearings 
to their shafts and housings and permanent 
fitting together of cylindrical components, push 
and press fits are ordinarily required. 

Industrially, it is valuable to establish sepa¬ 
rate tolerances for shafts and housings since 
items can then be manufactured independently 
and yet still guarantee that a specified fit will 
result when the parts are fitted together. There¬ 
fore, the published tables are more concerned 
with tolerances of the individual components, 
and the guidance provided does not immedi¬ 
ately permit the recommended amount of in¬ 
terference to be deduced. Since both the hole 
and the shaft are allowed to vary from the 
nominal size, the resulting interference is a 
variable amount, depending upon the actual 
sizes achieved. The amount of interference 

allowed is also related to the diameter of the 
work and this must be borne in mind when 
making the parts. 

As noted above, bearings must be fitted 
using only small amounts of interference in or¬ 
der that distortion does not result. The lighter 
push fits from the British Standard designation 
are used in these cases since the frictional drag 
created by a properly lubricated bearing is low. 
Push Fits are also used in any situation in which 
one component is weak, for example in the 
form of a thin sleeve, like the commercially 
produced plain bearings. For these cases, an 
interference of .0005in. per inch of diameter 
(.005mm per 10mm of diameter) is normally 

For a press fit, an interference of between 
.OOlin. and .0015in. per inch of diameter is 
normally satisfactory for those cases in which 
the fitted length is not greater than twice the 
diameter and neither component is in the form 
of a thin sleeve. The figure of .OOlin. per inch 
of diameter is that normally suggested as the 
maximum for good quality cast iron compo¬ 
nents, but for fitting model locomotive wheels 
onto their axles, or similar situations in which 
the components are reasonably robust, most 
authorities recommend an interference of 
.002in. per inch of diameter on the basis that 
movement of a wheel on its axle is, to say the 
least, highly undesirable. The same interference 
is also recommended for the pressing together 
of built-up crankshafts. 

If you are used to working in the metric 
system, these figures translate to .01mm and 
.02mm per 10mm of diameter. 

The figures for interference fits given above 
are those recommended for ferrous parts. Such 
fits are less successful for non-ferrous parts due 
to the softer (more elastic) nature of the materi¬ 
als. Consequently, if these fits are required, the 
above allowances should be increased by a fac¬ 
tor of 2 in order to ensure a satisfactory joint. 



The pressing operation 

The assembly of press-fitted items can naturally 
be performed in a press. This usually comprises 
a one-piece casting or forging, consisting of 
base, column and head not unlike a bench drill 
in concept. The head carries a sliding sleeve 
which moves up and down at right angles to the 
base, driven through a lever or arm by a rack or 
screw arrangement providing considerable 
mechanical advantage. If a press is not avail¬ 
able, the bench vice may be used instead, but is 
not so convenient since the two parts of the 
workpiece, together with any spacers or load¬ 
spreading plates, need to be juggled into posi¬ 
tion between the jaws before closing the vice 
onto the assemblage while holding all squarely. 

Since the vice is not all that convenient a 
replacement for the press, it is sometimes better 
to use a hammer, together with suitable 
punches, to force the items together. Another 
alternative is to use a drawing-in bolt (or draw- 
bolt) and nut to pull a bush or bearing into its 
housing. This is only applicable to situations in 
which both items have a through bore. 

Whichever method of pressing is used, care 
must be taken that the force is exerted exactly 
at right angles to the axis of the work and in the 
case of fitting ball or roller bearings, that the 
load is taken on the correct part of the bearing. 
This generally requires the making up of drifts, 
spacers or bushes to assist the pressing opera¬ 
tion, to ensure, for example, that the inner 
member of the bearing is not used to press the 
outer into its housing, or vice versa. 

If an assembly containing several axles and 
their bearings needs to be put together, the 
assembly sequence must be chosen so that it is 
not necessary to fit bearing inners and outers at 
the same time otherwise the forces required 
will be transmitted through the balls or rollers, 
thereby damaging the races. If shafts or axles 
need to be driven into race inners, either a soft- 
faced hammer must be used or a brass or 

aluminium alloy bush should be used to avoid 
damage to the axle end. 

The fitting of plain bearing bushes or sleeves 
is naturally more straightforward since only 
one pressing or pulling operation is required 
and in many instances a simple bolt-and- 
washers arrangement will suffice. 

Additional security 

Properly executed, interference fits in ferrous 
components are perfectly suitable for built-up 
crankshafts, for holding locomotive wheels 
onto their axles or other similar, high-torque 
situations. However, if you lack faith, or have a 
particularly heavy-duty application in mind, a 
little additional security may be desirable. One 
way that this can be provided is by pinning the 
items together by use of a radial pin passing 
through both. An alternative is the use of pins 
or screws fitted partly into both items. 

An example of this method of fitting is 
shown in Figure 8.21 which is the driving pul¬ 
ley cluster for a small lathe. It comprises three 
parts; the steel driving disc shown, a cast centre 
which incorporates a drive gear, and a two-step 
pulley casting. The steel disc is provided with a 
hole in which the driving pin which passes 
through the bull wheel locates. The disc is held 
to the pulley cluster (which is fitted firmly to 
the cast iron centre) by four countersunk 
screws. To ensure that adequate torque is trans¬ 
mitted, the press fit of the disc on the centre is 
reinforced by two parallel pins pressed into 
holes drilled half in the disc and half in the 
centre piece. The pins were made a close fit for 
these holes and were driven (hammered) into 
position. They provide a positive location for 
the disc on the centre and reinforce the press fit 
of these two items which would otherwise be 
relatively weak because of the use of a thin disc 
due to lack of space. 



Alternatives to the press fit 

As described above, the allowance for press fits 
in ferrous components for ordinary duties is 
normally something between .001 in. and 
.0015in. per inch of diameter. This places quite 
tight tolerances on the associated machining 
processes and means that extremely careful 
work is required when making both shaft and 
housing. Very small departures from nominal 
sizes can easily render an interference fit impos¬ 
sible to press together or, at the other extreme, 
so loose as to have no interference at all. Fortu¬ 
nately, it is possible to assemble such items as the 
pulley cluster of Figure 8.21 by gluing, using one 
of several adhesives which are now available. 

The assembly shown has three components, 
as described, plus the small pins providing the 
positive location between the steel disc and the 
centre. Adhesives may be used for all of the 
joints; pulley cluster to centre, disc to centre, 
pins into drilled holes and even screws into 
pulley cluster. Since some adhesives provide 
adequate strength even when radial gaps are as 
large as .004in. or .005in. (.10mm to .13mm) 
the machining of the various diameters can be 
undertaken assuming wider tolerances, subject 
to there being sufficient accuracy to maintain 
the various items in reasonable alignment and 
concentric to one another. There are particular 
benefits with the two securing pins since the 
holes for these may simply be drilled and the 
pins pushed in after coating with adhesive. This 
is much easier than attempting to create short, 
blind holes of sufficient accuracy to allow the 
manufacture of pins having a carefully control¬ 
led amount of interference. 

A further method of assembly which might 
also be suitable in particular instances is the use 
of soft or hard solders. Although soldering is 
not especially suitable in the case of the assem¬ 
bly shown in Figure 8.21, it is a valuable means 
of metal joining. Since the use of both solders 
and adhesives requires the introduction of the 

Figure 8.21 The driving gear cluster for a small lathe, 
showing the axial pins which transfer the drive from the pulley. 

binding material into the joint, and both meth¬ 
ods generally require some clearance between 
the parts, the use of solders and adhesives is 
described in Chapter 9. 

Semi-permanent attachment 


It hardly seems to be covered by the chapter 
heading of ‘Metal joining’, but there is a fre¬ 
quent need for cylindrical parts to be attached 
to one another so that torque can be transmit¬ 
ted between one component and another, and 
there remains the possibility of simple separa¬ 
tion of the items. So, something having the 
assembled characteristics of a press fit is 
required, but without the need to apply great 
force when separating the mating components. 



Wheels, arms and sector plates fitted to cir¬ 
cular shafts immediately come to mind, where 
a firm fit is required, and good transfer of 
applied forces, but the fitting must incorporate 
the possibility of later removal and replace¬ 
ment without damage to the parts, or spoiling 
of the fit. In these instances, it is usual to ream 
or bore the hole for the shaft or axle accurately 
to size so that the fitted item has a push fit onto 
its shaft. Additional security is then provided by 
using screws, pins or keys to resist the antici¬ 
pated turning forces. 

The items used to provide the additional 
security may be fitted on the radius of a circular 
section through the two items and are then 
described as being ‘radial’. Alternatively, they 
may be fitted parallel to the axis of the axle or 
shaft, in which case the description ‘axial’ is 

Radial screws and pins 

The simplest form of radial fitting is the simple 
grub screw which is used to hold a bored sleeve 
onto a shaft. The sleeve is drilled and threaded 
to accept a screw which is inserted to bear on 
the shaft and lock the shaft and sleeve together. 
To provide additional security and allow more 
force to be transmitted, it is usual to drill a 
shallow dimple in the shaft, into which the end 
of the screw can locate, to provide a more posi¬ 
tive drive. 

This type of fixing is not only used for such 
simple things as control knobs, but is frequently 
the method by which steam engine eccentrics 
are fixed to their shafts. Eccentrics need to be 
set to the correct position after assembly to their 
shafts and the process of setting is a repetitive 
one. Consequently, the screws are not normally 
tightened fully until several adjustments have 
been made. 

It may be difficult to hold the items together 
sufficiently firmly to drill a dimple at the 

required location, without the possibility that 
the setting will be lost, and in such cases Allen 
grub screws having shaped ends are used, 
which provide additional security. Some types 
present a circular edge to the shaft, and since 
the screw is hardened, the end ‘bites’ into the 
shaft and creates its own positive location. This 
type of fastening is illustrated in Figure 8.22, 
together with a flat-ended type which is used if 
the shaft has a machined flat. 

A more secure attachment can be provided 
by using a radial pin to secure a boss or collar to 
a shaft. Three different types of pin are used for 
this duty: a plain, parallel pin, or a taper pin, 
both of which are fitted into a carefully sized 
hole which is produced by a process known as 
reaming (see Chapter 10) or a hollow, hard¬ 
ened pin which is fitted into a plain, drilled 

These latter pins, known as roll pins, are 
naturally the easiest to fit since they require 
only a drilled hole, which is quickly and easily 
made. Roll pins are rolled (from a type of 
spring steel) so that they are just larger than the 
nominal diameter. Being springy they can be 
driven into a suitable hole and just as easily be 
driven out, yet they are quite secure when in 
position in a hole. The amount of turning force 
(torque) which can be transmitted is limited by 
the fact that the pin is hollow, but if the load is 

Figure 8.22 Illustration of the end shapes of Allen grub 



Figure 8.23 A selection of roll pins. 

Figure 8.24 Two taper pin reamers and a selection of taper 

The type of pin illustrated in Figure 8.21 lies 
parallel to the axis of the assembly and is cor¬ 
rectly described as an axial pin, or key. For the 
placement of this type of key, the items are first 

not severe, the roll pin is ideal. Figure 8.23 
shows a range of roll pins. 

If greater torque needs to be transmitted, a 
solid pin is required. This might be just a length 
of silver steel or can be a taper pin. Both are 
accurately ground to size and can provide a 
secure and solid attachment provided they are 
fitted to a correctly sized and well-finished 
hole. In the case of a taper pin, the hole needs 
to be finished with a specially produced reamer 
whose taper matches that of the pin. This 
requires a hole to be drilled sufficiently large 
for the end of the reamer to enter the hole. The 
reamer is then put in the drilling machine in 
place of the drill and the hole enlarged until the 
length of the hole is tapered and the pin will 
enter to the required depth. 

The shallow taper on the pin and the hole is 
designed so that the pin jams itself into the hole 
when it is ‘seated’ by a sharp tap from a ham¬ 
mer. Any surplus pin projecting from the hole 
can be cut off with a hacksaw and the end of the 
pin rounded or flattened with a file, since the 

Figure 8.25 The fitting of an outside crank on a model steam 
locomotive. The ends of the taper pin have not yet been cut off. 

pins are not especially hard. Figure 8.25 shows 
an external crank on a model locomotive which 
is held in place by a taper pin. This pin has not 
yet been cut off at the ends. 

Two taper-pin reamers and some taper pins 
are illustrated in Figure 8.24. 

Similar considerations apply to the fitting of 
parallel pins as to taper pins. The pin must be 
nicely parallel and the hole must be correctly 
sized and have a good internal finish so that the 
pin fits properly, without shake or sloppiness. 
This means finishing the hole to size by 
reaming after drilling an initial hole just large 
enough for the reamer to enter. 

Obviously, it is necessary to drill both items 
together when fitting any sort of radial pin, and 
it might also be necessary to identify which way 
round the items should be fitted together in 
some instances, although a taper pin naturally 
only fits its tapered hole when entered from the 
correct side. 



assembled and the hole for the pin is drilled by 
starting on the junction between the two parts 
and drilling down in the usual way. If the pin is 
to be glued into position, it is adequate to use a 
simple drilled hole since a little clearance is re¬ 
quired for the adhesive and little other atten¬ 
tion needs to be paid apart from cleanliness. 

If the pin is to be held by friction, it must be 
a tight fit in the hole and this must be reamed or 
otherwise finished to the correct size, so that 
the pin can be driven in. Provided that a 
method of withdrawal is provided, there is no 
reason why the pin cannot be removed, or the 
two fitted items pulled apart, should it be nec¬ 
essary to separate the components of the 

Drilling the hole for the pin is straightfor¬ 
ward if the assembly can be positioned under 
the spindle of the drilling machine and the two 
components which are to be drilled are of simi¬ 
lar hardness. If the shaft is steel, say, and the fit¬ 
ted item is a soft material, such as aluminium, 
the drill is likely to ‘favour’ the aluminium and 
wander off the required line. It is effectively 
pushed away by the hard steel into the soft 
aluminium and the result is a disaster. 

In these instances, it is better to machine the 
two parts separately and the usual type of key is 
rectangular in cross-section. This allows a 
keyway to be machined in both parts, and the 
only dimension which is critical during these 

Figure 8.26 The leadscrew on my lathe is driven by a 
changewheel which engages a key set into the end of the 
leadscrew shaft. 

operations is the width of the slots since the key 
does not need to take up the full depth which is 

An example of a keyed fitting of this type is 
shown in Figure 8.26 which shows the driving 
end of the leadscrew on a lathe. The leadscrew 
can be driven by a gear fitted to the end, and is 
fitted with a half-moon key, known as a Wood¬ 
ruff key, which fits a machined recess in the end 
of the shaft. The gear has a corresponding 
square recess (a keyway) machined in it, which 
fits the upstanding part of the key. 

Further illustrations of this arrangement are 
shown in Chapter 16, together with examples 
of straight rectangular keys which serve a simi¬ 
lar purpose. 



Metal joining 2 - 
solders and adhesives 

Types of solder 

Solders are described as hard or soft. Soft sol¬ 
ders are alloys of tin and lead, usually around 
60 per cent tin and 40 per cent lead, which melt 
at about 185°C. Both constituents are soft, as 
are the resultant alloys. Colloquially, the tin- 
lead alloys are normally described simply as 
solder, whereas hard solders are referred to as 
silver solders. 

Hard solders are alloys of silver and other 
elements having melting points in the range 
from 610°C to roughly 760°C. 

Another commonly employed method of 
metal joining is to use a brass comprising 50 per 
cent each of copper and zinc as the solder. This 
process is known as brazing and the brass wire 
used for the purpose is known as brazing 
spelter. This is not a commonly used process in 
the amateur’s workshop, since brass becomes 
liquid only above 875°C and a great deal of 
heat is required. 

Brass is also not ideal for use as a solder, and 
alloys of silver, and other elements, with cop¬ 
per and zinc, create alloys which become liquid 
at lower temperatures than brass and which 

have better properties generally for this type of 

Due to the long use of the term brazing, 
this persists in modern parlance and the term 
silver brazing is used, as well as hard soldering 
and silver soldering, all to describe the same 

Solders are most commonly supplied as rods 
or wires. These are used with separately applied 
fluxes which may be in the form of liquids or 
pastes, although one very common form of soft 
solder is manufactured as a hollow wire which 
is filled with flux. 

Use of solders in modelling 

Soldering is a very useful technique for model¬ 
ling, since it permits assemblies which repre¬ 
sent forgings or castings on the prototype to be 
built up from smaller components, allowing 
complex shapes to be produced. Since the 
solder used for soft soldering is an alloy of 
weak metals, the resulting joints do not have 
great strength, but if an assembly is decorative, 



Figure 9.1 Soldered fittings inside a water tank filler lid. 

rather than functional, soft soldering may be 

Since a solder can be made to flow through¬ 
out a joint, it is a useful way to complete an 
assembly which needs to be sealed, and if the 
item is small, the solder might be used both for 
the purposes of assembling the parts and form¬ 
ing the seal. This might be the case for small 
water, petrol or oil tanks for models. 

The ability to make assemblies is exempli¬ 
fied by the water tank filler lid shown in Figure 
9.1. The lid has a locating hoop on the upper 
surface which locates the closing hinge strap, 
and this has been pushed through two drilled 
holes and then soldered into the lid, on the 
inside. The hinge strap is held into the lid by a 
headed brass pin which is secured into the lid 
by a brass washer soldered to its inside end. 
Although the lid is functional, no great strength 
is needed from these joints and they continue to 
serve the function after some years of use. 

The tank filler has been soldered into the 
tank top for this model, and the tank was sealed 
by solder, although its component parts were 
screwed together. 

On the same model, the rear lamp irons 
shown in Figure 9.2 have an integral rib on the 
left-hand side. On the model, this has been rep¬ 

resented by a short length of copper wire soft 
soldered into a shallow sawcut on the upstand¬ 
ing part of the iron. 

The examples of soldered items described 
above are not highly stressed joints, and they 
have been made by soft soldering. If a strong 
assembly is needed, it should be silver soldered, 
as noted above, since silver solders are much 
stronger than soft solders. 

The buffer stock of Figure 9.2 is one item 
which might be made by silver soldering 
together its component parts. To make the 
stock, a square of mild steel can be silver sol¬ 
dered to a short length of round steel bar, leav¬ 
ing just the minimum amount of material to be 
removed from the face of the square flange, 
and the outside diameter of the round bar. 
Making the complete stock from a length of 
square steel requires much material to be 
machined away, material which, after all, one 
has had to purchase, and the silver-soldered 
assembly is a more viable way to create these 

There are other tasks for which silver solder¬ 
ing is the natural choice, principally the manu¬ 
facture of model boilers and their fittings. The 
silver solders used have high melting points, 
making silver-soldered assemblies suitable for 

Figure 9.2 Details on the rear bunker of a 5-inch gauge 
pannier tank. The lamp irons have a bar soft-soldered to their 
sides and the buffer stock is the type of item which might be 
made by silver soldering a square plate to a short stub of round 
steel bar. 



use with high-pressure (and therefore high- 
temperature) boilers and steam circuits. 

The soldering process 

Soldering consists of melting a material (a 
solder) into a clamped-together joint. The sol¬ 
der forms an alloy with the metals in the joint 
binding them together. The intention is not 
usually to fill gaps in the joint, and the items to 
be joined must be in close contact. 

When making a soldered joint, the solder 
melts at a lower temperature than the metal 
items to be joined. To introduce the solder to 
the joint, it is heated until the solder melts 
when touched on it. The solder flows through 
the joint, forming a bond with the hot materials 
and uniting them when the heat is removed, 
allowing the joint to cool and the solder to 

To achieve an adequate bond between the 
solder and the metallic items in the joint, the 
metals must be clean. The components in the 
joint must be de-greased and all tarnish 
removed before the soldering operation com¬ 
mences, and freedom from formation of tar¬ 
nishing compounds must be maintained during 
soldering. Unless the metal is exceptionally 
dirty, simple mechanical cleaning by use of em¬ 
ery cloth is all that is required, but any deposits 
of oil or grease must be removed first by use of 
a suitable solvent. Cleaning must be performed 
immediately prior to soldering and once 
cleaned, the metal should not be touched. 

Cleanliness during the soldering operation is 
maintained by use of a flux. This is designed to 
be liquid at the soldering temperature so that it 
flows over the surfaces within the joint and 
prevents the formation of oxides and tarnishing 
compounds. Pre-cleaning, and the use of a suit¬ 
able flux, are vital to the production of sound 
soldered joints. 

Solders can be obtained which melt at 
temperatures lower than that of boiling water 
(100°C), or at around 200°C. There is also a 
group of solders which melt between 600 and 
800°C, and it is possible to use a type of brass as 
a solder. This melts at 875°C. The range of 
solders is very wide, and the requirements for 
heating the joint can be just an electrically oper¬ 
ated soldering iron, or may require the use of a 
propane or oxy-acetylene torch. 

Basic procedure 

Essential steps 

The process of soldering is quite straightfor¬ 
ward if carried out correctly, but to be success¬ 
ful, the process must be carried out as follows: 

1. The material in the joint must be cleaned, 
and must remain clean throughout the 
heating and soldering operations. This 
means adequate pre-cleaning and the use 
of an appropriate flux. 

2. The joint must be firmly held together dur¬ 
ing soldering and the gaps must be within 
the range recommended for the solder 
used. In practice, this generally means 
close fits. 

3. The metal in the joint must be hot enough 
(right through the joint) to melt the solder 
which is in contact with it. 

This last point is probably the most important 
of all. If the job is not sufficiently hot, solder 
cannot flow through the joint. 

The normal method of carrying out the sol¬ 
dering operation when using solder in stick 
form, is to heat the items to be joined, testing 
their temperature periodically by applying the 
solder stick to them, just outside the area being 
heated, until the solder melts and flows when 
touched on the surface. In this respect, it is 
helpful to heat the joint from one side, but to 



apply the solder stick to the other, on the basis 
that this ensures that the joint is hot right 

The close mechanical fit of the parts, and the 
cleanliness of the metal must be maintained 
during the whole operation. If the metal is 
dirty, solder cannot flow over the surface since 
the dirt acts as a barrier, preventing contact 
between the solder and the base metal which, in 
turn, prevents the formation of a sound joint. If 
the gap between the items to be joined is too 
large, the solder will not form a bond since it is 
not generally designed to bridge large gaps. 

Heating the work 

Soldered joints made by model engineers gen¬ 
erally require quite large amounts of heat to 
bring the work to the temperature at which the 
solder will melt when touched on the surface. A 
small silver-soldered joint requires the work to 
be brought to red heat, and a soft-soldered joint 
might be, for example, to seal the water tank 
for a large-scale tender or locomotive. 

A blowlamp (or blowtorch) is likely to be the 
heat source. The most convenient general type 
is the propane torch since a range of burners is 
available, capable of burning gas at different 
rates and of providing large or small amounts 
of heat. A description of a typical torch/burner 
system is given in Chapter 4. 

Small silver-soldered joints might be made 
on a small tin-box hearth, of the type illustrated 
in Figure 4.24, but the assembly of a large 
structure such as a locomotive or traction 
engine boiler will require a metal bench, or 
brazing hearth of the type shown in Figure 
4.23, in view of the large amount of heat which 
is required for the later stages of assembly. 

There is also a need for plenty of room 
around the hearth in view of the large amounts 
of radiated heat from the structure and the 
throw of the flame produced by the larger burn¬ 

ers. It goes without saying that there should be 
no combustible material adjacent to the hearth. 

Holding while heating 

Holding the pieces to be joined firmly together 
during the heating and cooling process is vital. 
Larger items are usually screwed or riveted 
together, which provides adequate security, but 
the components of smaller assemblies need to 
be located firmly together, otherwise move¬ 
ment can occur, especially as the water-mixed 
fluxes used for silver soldering dry out. The 
water is driven off as steam, and the flux melts 
as the temperature rises, and these events can 
easily displace small items which are not ad¬ 
equately located. 

For smaller assemblies which are to be sol¬ 
dered, the basic requirement is for some sim¬ 
ple, cheaply made clamps. If a blowlamp or 
torch is being used to heat the job, the clamps 
are subjected to exposure to the naked flame, 
and they may discolour. They are also likely to 
become contaminated with flux, and are rap¬ 
idly spoiled. A few clamps should be made up 
in the workshop and reserved especially for use 
when soldering. 

For soft soldering, wooden spring clothes 
pegs can be used. They are ideal for use as tem¬ 
porary clamps, especially once their ends have 
been cut away, but they are not suitable when 
using a naked flame and must be reserved for 
small jobs which can be heated by a soldering 

There are also some small clamps which go 
under the name of curl clips. They are ideal for 
holding smaller items, or for use where space is 
limited. One blade comprises two parallel arms 
but the other has a cross bar near the outer end. 
The clips are absolutely ideal as clamps and, 
although metallic, do not have too large a mass 
to prevent their use for soldering when an iron 
is used. Being made in aluminium, or an 



aluminium-based alloy, their use for silver sol¬ 
dering is restricted since a propane burner will 
quite easily melt these materials, but they are 
occasionally available in stainless steel and this 
type can be extremely useful. 

Soft soldering 

Common types of soft solder 

Soft solders are alloys of tin and lead, usually in 
the proportions of 60 per cent tin and 40 per 
cent lead. The actual proportions do differ 
between different grades of soft solder, confer¬ 
ring different characteristics on the alloys in the 
region of the melting point. 

Most solders (and this applies to silver sol¬ 
ders also) do not have a melting point in the 
sense of a single temperature above which they 
are liquid and below which they are solid, al¬ 
though it is possible to prepare alloys which do 
have a single, identifiable melting point. Such 
alloys are called eutectic alloys and a soft solder 
comprising 62 per cent tin and 38 per cent lead 
is eutectic and melts at 183°C. 

Alloys containing other proportions of tin 
and lead do not have a single melting point but 
are said to have a melting range i.e. two tem¬ 
peratures are specified. The lower of these is 
called the solidus and is effectively the freezing 
point since below this temperature the alloy is 
solid. The higher temperature, the top of the 
melting range, is called the liquidus since above 
this temperature the alloy is liquid. Between 
solidus and liquidus the alloy is in a ‘pasty’ state 
- neither fully liquid nor solid. 

For applications in which good spreading of 
the solder is required, that is, where the solder 
is required to flow readily through the joint, a 
short melting range is required. This is pro¬ 
duced by utilising proportions of the alloying 
elements close to the eutectic point, that is, 

close to 62 per cent tin. For example, 64 per 
cent tin, 36 per cent lead has a melting range 
from 183°C (solidus) to 185°C (liquidus) 
whereas 60 per cent tin and 40 per cent lead 
produces an alloy having a melting range from 
183°C to 188°C. 

Solders with a long melting range are very 
useful when gap filling or surface filling are re¬ 
quired and were used for car body filling before 
the days of glass-reinforced resin fillers and 
also for making ‘wiped’ joints in lead pipes for 
plumbing. Reducing the tin content to 50 per 
cent produces an alloy having a melting range 
from 183 to 212°C and an alloy containing 
only 30 per cent tin, with 70 per cent lead, has a 
melting range of 183 to 25S°C. 

As might be expected from the above 
descriptions of typical solder characteristics, a 
wide range of alloys is available having differ¬ 
ing properties. The two major classifications 
are tinmans’ solder, having a short melting 
range, and plumbers’ solder, intended for 
wiped joints and for use as a body filler, and 
therefore having a wide melting range. The 30 
per cent tin alloy referred to above is of this lat¬ 
ter type (183 to 255°C melting range). 

The most familiar form of tinmans’ solder is 
cored solder used for electrical work. This 
comprises a hollow wire into which a non- 
corrosive, rosin-based flux is loaded so that 
solder and flux are introduced into the work 
simultaneously. Soft solders are also available 
as solid sticks or bars. These are used with sepa¬ 
rately applied fluxes. 

High-temperature soft solders 

There is frequently a need to carry out two or 
more soldering operations on the same assem¬ 
bly and there is value in having solders available 
with quite different solidus temperatures. 

The addition of a small amount of silver to a 
tin-lead alloy significantly raises the solidus 



temperature, creating a high-temperature soft 
solder which is not particularly expensive. The 
alloy most frequently stocked by model engi¬ 
neers’ suppliers is Comsol (from Johnson 
Matthey Metals Ltd.) which has 1.5 per cent 
silver, the remainder being tin and lead. This 
alloy is eutectic, passing from the solid to the 
liquid state at 296°C. It is intended for the sol¬ 
dering of armature windings on electric motors 
but its availability makes possible two-stage 
soldering using tinmans’ solder as the second 

Comsol was also occasionally recommended 
for the final caulking of stays in copper boilers 
which have otherwise been assembled by use of 
silver solder, but this practice is not now very 

Low melting-point alloys 

By incorporating a quantity of cadmium or 
bismuth into a tin-lead alloy, the melting range 
(or indeed melting point) can be lowered by 
about 40°C. An alloy containing 50 per cent 
tin, 32 per cent lead and 18 per cent cadmium 
is eutectic and has a melting point of 145°C. An 
alloy of 49 per cent tin, 41 per cent lead and 10 
per cent bismuth has a melting range of 142 to 

Alloys of these types are not generally avail¬ 
able from local suppliers and modellers usually 
understand the phrase ‘low melting-point sol¬ 
der’ to mean an alloy used for soldering white 
metal castings which themselves have a melting 
point of around 200°C. Clearly, the use of nor¬ 
mal tin-lead solder on these alloys means there 
is a very real danger of melting the castings. 
Consequently, fusible alloys (their commercial 
description) are utilised for joining white metal 
components and are much used in small-scale 
railway modelling. 

Alloys are produced having different melt¬ 
ing ranges, utilising tin and lead in differing 

proportions together with bismuth and/or cad¬ 
mium. Many alloys bear the names of their 
inventors. One of the best known is Wood’s 
Metal which contains 50 per cent bismuth, 25 
per cent lead and 12.5 per cent each of cad¬ 
mium and tin. The most interesting feature of 
Wood’s Metal is its melting range which is 70 to 
72°C. It thus melts in not-quite-boiling water. 

Several of the fusible alloys melt at tempera¬ 
tures lower than 100°C and are used for filling 
pipes and tubes before bending due to the rela¬ 
tive ease with which they may be melted into 
place and afterwards removed. A commercial 
version of a low melting-point alloy, sold spe¬ 
cifically for pipe bending, is Cerrobend. 

These alloys are well adapted to the solder¬ 
ing of white metal and they are produced in 
stick form and sold as low melting-point sol¬ 
ders under various commercial names. Their 
use as solder demands a fairly strong flux since 
the molten alloy tends not to wet the base metal 
very well. It is usual to employ an acid flux (see 
below) when using bismuth-based alloys. Sol¬ 
der producers generally market a suitable liq¬ 
uid flux which is sold under a commercial 

Alternative solders 

Over the last decade or so, small-scale railway 
modelling has undergone something of a revo¬ 
lution and there is much emphasis nowadays 
on really fine scale models. This desire for a 
scale appearance in the smaller scales has 
resulted in the extensive development of kits 
for locomotives, wagons, buildings, signals and 
accessories of all sorts, the parts for which are 
etched in thin sheets of brass or nickel silver. 

The wide availability of etched kits, which 
are best soldered together, has resulted in a 
demand for a wider range of solders which can 
be used for kit assembly. There is naturally a 
need for solders with different characteristics 



and melting points (or melting ranges) and the 
increased demand for specialised products has 

To avoid the need for a metallurgical degree 
in order to determine the best solders to use, 
suppliers have developed simple means of iden¬ 
tification and generally publish cheap booklets 
which describe the solders and the related 
fluxes in detail. 

One supplier whose products are widely 
available markets the following soft solders. 
Number 70 described as ‘very low melting 
point’, Number 188 described as having ‘a 
short melting rang e',Number224 described as 
having ‘a wide melting range’ and Number 243 
described as ‘containing silver’. 

You need to consult the supplier’s booklet to 
determine exactly what the difference is 
between the four types, but it seems likely that 
the names actually identify the melting points, 
or allude to the liquidus temperature of the 

If you have need of these specialist solders, 
you are more likely to find them in a model 
railway stockist, rather than a model engineer¬ 
ing supplier. 

Forms of solder 

For most soldering work, a solder having good 
spreading characteristics and a short melting 
range is required. For soft soldering, this is sat¬ 
isfied by tinmans’ solder. The most useful form 
is a 60-40 tin-lead alloy in wire form with a 
number of cored longitudinal holes filled with 
rosin-based flux. This is known as cored solder. 
In use, flux and solder are introduced into the 
joint almost simultaneously, the flux arriving 
first since it flows at a slightly lower tempera¬ 
ture. The flux is corrosion free and the residues 
are readily removed using white spirit. 

Cored solder wire is available in different 
diameters, denoted by swg numbers, 18,22 and 

24 swg being readily available. It is best to use 
the thinnest wire which is available, preferably 
24 swg (.022in., 0.56mm). In any event, it is the 
cheapest per unit length and since it is easy to 
introduce too much solder to the joint, the 
small diameter is to be preferred. Cored solder 
should be bought in as large a quantity as can be 
afforded since it is expensive in small quantities. 

If a large job demands a much larger diam¬ 
eter of solder, a stick of uncored tinmans’ will 
probably serve, since this is available in rough¬ 
cast form about ’/win. (4.5mm) square, but it 
must, of course, be used with a separately ap¬ 
plied flux. For most modelling, a 16 swg cored 
solder will generally be adequate. 

Solders are also available in the form of 
paint i.e. a suspension of solder particles in a 
flux base. These are excellent for small jobs, 
providing, as they do, a means of introducing 
the solder before the joint is assembled. For 
larger joints, their disadvantage lies in the fact 
that much of the fluid is flux, and there may be 
insufficient solder present to effect a really 
sound joint. Paints are considered in more 
detail below, under the heading ‘Getting solder 
to the joint’. 


Tinning is a method of preparation which is 
used extensively for the leads of electronic 
components and printed circuit boards. 

Tinning simply means bonding a layer of 
solder to two pieces of metal before bringing 
them together to make the joint. The layer of 
solder is microscopically thin so it does not pre¬ 
vent the correct butting together of the two 
parts. Solder is, however, placed in the joint 
before starting. The advantage of tinning is that 
the solder does not readily tarnish and the 
process of making a subsequent joint between 
two tinned components becomes a quick and 
simple operation. 



The actual tinning is carried out simply by 
heating clean metal, possibly already pre¬ 
fluxed, melting solder and flux on to the surface 
and wiping off the excess solder whilst it is still 
molten. Rosin-cored solder is very convenient 
for this, but if a more active flux is required, or 
if it is desired to use a non-cored tinmans’ solder 
stick, the flux must be applied before heating 
commences. The wiping off of the excess solder 
can be done with almost any damp, clean cloth, 
but a piece of damp synthetic sponge, as used in 
soldering iron stands, is essentially lint-free and 
leaves no fibres behind to impede the subse¬ 
quent soldering. 

The advantage of treating the parts sepa¬ 
rately in this way is that the solder is applied, 
and the bond with the base metal is formed, 
while all is in sight. Cleaning of any flux 
residues from the tinned parts is also easy 
before the joint is assembled and a more vigor¬ 
ous flux can be used for the tinning operation. 
Since the tinned surface is not very susceptible 
to tarnish formation, the tinning also assists 
maintenance of a clean surface at the time of 
soldering, which is not the case for brass, for 
example, unless the cleaning is carried out 
immediately prior to soldering. 

The tinning operation produces a very thin 
layer of solder bonded to the base metal which 
for all practical purposes can be assumed to be 
non-existent (dimensionally, that is). When two 
items have been tinned in this way, it is only 
necessary to coat them with flux, press them 
together and heat up the pair until the solder 
melts, adding just a little additional solder to 
the joint. 

Fluxes for soft soldering 

Cleanliness during all soldering operations is 
maintained by use of a flux. This dissolves any 
residual tarnish from the metal (which must be 
well cleaned first) and also prevents formation 

of oxides on the surface of the molten solder. 
The selection of a flux depends principally on 
the metals to be joined. Fluxes are classified 
into different chemical groups and may be 
organic or inorganic, acids, salts, rosin-based 
and so on. Classifying each as a good or bad 
flux requires an estimate of its tarnish-removal 
capability, its corrosiveness and its stability and 
effectiveness over the range of temperatures at 
which it is expected to be effective. 

Acids tend to have good temperature 
stability, be good or very good at tarnish 
removal but have a high or moderate corrosive¬ 
ness. Removal of residues after soldering is very 

If no corrosion can be allowed, as for electri¬ 
cal work and much modelling, a rosin-based 
flux is normally used. Such fluxes have rela¬ 
tively poor temperature stability and have only 
fair tarnish-removal properties. Nevertheless 
they are used extensively for electrical and elec¬ 
tronic work where their non-corrosiveness is of 
supreme importance and where tinned surfaces 
are normally used. These are much less suscep¬ 
tible to the formation of oxides than the base 
metals and do not require the use of such an 
active flux. Tinning is an important way in 
which the strength of a soft-soldered joint can 
be improved and since it brings also a freedom 
from the formation of surface contaminants, it 
is a valuable technique to employ. 

Non-corrosiveness of the flux is also an 
important consideration for modelling, so 
rosin-cored solders and rosin-based fluxes are 
frequently used. Due to their relatively poor 
tarnish-removal properties, the base metals 
must be well cleaned, or tinned, before solder¬ 
ing begins. 

The main constituent of rosin-based flux is 
water-white rosin. This is distilled from pine 
sap and consists of 80 to 90 per cent abietic 
acid (also known as sylvic acid) which is a mild 
organic acid. It has only poor tarnish-removal 
properties and will not penetrate a heavy 



tarnish layer, nevertheless it works efficiently 
to protect tarnish-free surfaces during solder¬ 
ing and can be considered as corrosion-less. In 
a practical rosin-based flux, additional activa¬ 
tors are sometimes included to improve the 
tarnish-removal properties and this gives rise to 
the term ‘activated rosin flux’. These are the 
most widely used fluxes for electrical work and 
are used for forming the flux cores in cored 

If rosin is overheated, it turns a dark brown 
colour and loses most of its tarnish removal 
properties. If a problem is experienced in mak¬ 
ing a joint, the result is that the heating may 
continue for longer than usual as an attempt is 
made to make the solder ‘take’, causing the flux 
to overheat thereby rendering it useless from 
the point of view of tarnish removal. If the joint 
is allowed to cool and a second attempt made 
to solder it without first cleaning all of the over¬ 
heated flux from the base metal, the overheated 
(and useless) flux cannot perform its function 
of tarnish removal and a failure again results. 
As can be imagined, the effect can be cumula¬ 
tive. If a problem of this sort is experienced, it 
is best to separate the items, clean down to new 
metal completely and start again. Some rosin¬ 
like fluxes are now available which have a 
higher charring temperature and this allows 
some latitude if there is a likelihood of over¬ 

Residues from rosin-based fluxes should be 
removed by using white spirit or methylated 
spirit since they are not water-soluble as are the 
residues of some other fluxes. 

For soldering metals which readily form 
oxides, such as copper, brass and steel, a flux 
which is more active than rosin is desirable. If 
the work can be thoroughly cleaned after¬ 
wards, the flux known generally as ‘killed spir¬ 
its’ can be used. The active ingredient in this is 
zinc chloride (formed by dissolving zinc in 
hydrochloric acid) and although it is highly 
corrosive it can be neutralised by washing 

thoroughly with water. One commercial form 
of killed spirits is marketed as Baker’s Fluid. 

Provided that the joint is accessible and can 
be properly washed, a flux based on zinc chlo¬ 
ride is safe to use. If making closed assemblies, 
such as water tanks, it is best to use a really 
active flux (for copper or brass) only for an ini¬ 
tial tinning operation on the plates, afterwards 
washing thoroughly before soldering the as¬ 
sembly together using a less-corrosive flux. 

Another active flux is phosphoric acid which 
is used mainly for stainless steel. Various com¬ 
mercial fluxes for this material are based on this 
acid, but since it is not very efficient at wetting 
the surface it is sometimes combined with other 
chemicals and supplied in the form of a paste. 
Commercial fluxes have developed greatly over 
the last few years and are now much more 
widely available. They can be relied upon to 
give good results when used over the recom¬ 
mended temperature range and with the metals 
for which they are designed. Some are corro¬ 
sive, but many are water-soluble and corrosive 
residues are simply removed by washing. On 
no account should the washing or neutralising 
operation be omitted. 

Silver soldering and brazing 

Hard solders 

Hard soldering can be considered as two sepa¬ 
rate operations - brazing and silver soldering. 
Nowadays, brazing is not much practised but it 
is nevertheless a useful method of metal joining 
and it was at one time widely used for joining 
steel components. Brazing simply consists of 
using a 50-50 brass wire (brazing spelter) as a 
solder, melting it into a clamped-together joint, 
in the presence of a flux. Borax was tradition¬ 
ally used as the flux used for brazing. 

Brazing spelter has almost entirely given way 



to alternative hard solders, the most common 
of which are the silver-bearing alloys, generally 
known as silver solders. These are closely con¬ 
trolled alloys having precise characteristics 
which are available having different melting 
ranges, and spreading characteristics. The 
changes are brought about by changing the 
silver content of the solder, together with the 
proportions of other alloying elements, to 
change the melting range and the temperature 
below which the alloy is fully solid (the solidus). 

By choosing solders with different solidus 
temperatures it is possible to carry out two or 
three silver soldering operations on one item 
without any real danger that heating for the 
later stages of assembly will melt previously 
applied solders. This is very important when 
assembling a model boiler made in copper, for 
example, since by its nature, the assembly must 
be performed in several stages. 

Table 9.1 


Melting Range (°C) 
Solidus Liquidus 

BS 1845 

Easy-flo No. 2 








Silver-flo 16' 




Silver-flo 24' 




* Silver-flo 16 and Silver-flo 24 were previously known 
as B6 and C4 alloys, respectively 

Silver solders are most readily identified and 
defined by using the British Standard designa¬ 
tions (from BS 1845) but in practice these are 
not often used. This is due to the wide availabil¬ 
ity of solders from one manufacturer, Johnson- 
Matthey Metals Ltd. The practice has devel¬ 
oped of using this manufacturer’s designations 
rather than the BS references. The most com¬ 
monly available silver solders from this manu¬ 
facturer are shown in Table 9.1. 

Table 9.1 shows four silver solders having 
different compositions and exhibiting different 
melting ranges. Easy-flo No. 2 and Argo-flo are 

alloys of silver, copper, cadmium and zinc, the 
former having 42 per cent of silver and the 
latter 38 per cent. Argo-flo has the wider melt¬ 
ing range and is used when larger fillets are 
required or when wider gaps need to be filled, 
since it is suitable for gaps up to .OlOin. 
(0.25mm). Easy-flo No. 2 is suitable for gaps 
up to .006in. (0.15mm). 

The same manufacturer also supplies Easy- 
flo No. 1, corresponding to the BS 1845 desig¬ 
nation AG1. This alloy contains 50 per cent 
silver and is consequently more expensive than 
No. 2, but it is slightly more fluid when molten 
and produces finer fillets than the cheaper 
alloy. No. 1 has a melting range from 620 to 
630°C. The Easy-flo and Argo-flo alloys are 
recommended for all general work. 

For the first stage of any two-stage silver sol¬ 
dering which is required, Silver-flo 16 or Silver- 
flo 24 are usually specified. For these alloys, the 
silver content is indicated by the appended 
reference number. These alloys are, however, 

Cadmium-free silver solders 

General-purpose silver solders such as Easy-flo 
No. 1 or No. 2 and Argo-flo are alloys of silver, 
copper, cadmium and zinc. These materials 
developed out of the use of a simple brazing 
spelter which is a 50:50 brass comprising an 
alloy of copper and zinc. The addition of some 
silver to such an alloy reduces the liquidus tem¬ 
perature and creates an alloy that is signifi¬ 
cantly better than a plain brass for use as a 
solder. Further improvements in the alloy are 
brought about by the addition of some cad¬ 
mium which further reduces the melting point 
and also brings other benefits. These include an 
improvement in the ability of the molten solder 
to wet the joint surfaces and a reduction in the 
surface tension, giving the alloy a better capil¬ 
lary action so that it is more readily drawn into 



the narrow clearances in the joint. Cadmium is 
also cheaper than silver, thereby reducing the 
cost of the solder. 

The problem with cadmium, and to a lesser 
extent zinc, is that they are both elements 
which are characterised by low melting and 
boiling points. Cadmium, for example, melts at 
321°C and boils at 767°C, while zinc melts at 
419°C and boils at 907°C. Both elements are 
naturally molten during the soldering process 
and although the boiling point of neither ele¬ 
ment should ordinarily be reached during the 
process, some cadmium and zinc vapour is nor¬ 
mally given off. 

The vapours which are produced combine 
readily with oxygen to form cadmium and zinc 
oxides, both of which are toxic. By far the most 
dangerous of these is cadmium oxide and expo¬ 
sure to quite low levels of this compound can 
be fatal. All silver soldering must, therefore, be 
carried out in an area which has good ventila¬ 
tion. Heating should be rapid and not pro¬ 
longed more than is absolutely necessary to 
effect the joint. 

To avoid the major hazard posed by the pres¬ 
ence of cadmium, the higher melting point 
alloys such as Silver-flo 16 and Silver-flo 24 are 
formulated without the incorporation of cad¬ 
mium and are therefore described as cadmium- 
free. Even when using these alloys, adequate 
ventilation should be provided and the work 
should be brought up to soldering temperature 
as rapidly as possible and the joint completed 

The flux may be affected, and lose its active 
properties, if it is raised to too high a tempera¬ 
ture or even if it is held within its working 
temperature range for too long. Heating must 
be rapid (the temperature of the work must rise 
quickly through the melting range of the sol¬ 
der) particularly when using alloys having a 
long melting range, otherwise some separation 
of the elements forming the alloy may occur. 
This effect is known as liquation. 

Forms of silver solder 

The most commonly available types of silver sol¬ 
der are those listed in Table 9.1. Different forms 
of these alloys are available, normally compris¬ 
ing rods, wire and strips in various standard 
sizes. Industrially, thin foils are available in some 
grades but these are not generally to be found in 
modellers’ suppliers. Rods and wires are gener¬ 
ally the same thing, except that the description 
‘wire’ is applied to the smaller diameters below 
/$ in. ( 1.5mm). Wire is generally supplied in 
coils containing a few metres of one size. Rods, 
on the other hand, have larger diameters and are 
sufficiently rigid to be supplied in straight 24in. 
(600mm) lengths, hence the description ‘rod’. 

Strip in 600mm lengths is also available, 
being of rectangular cross-section, from 
about 1.5mm x 0.6mm to 5.0mm x 1.0mm. 
These are quite useful as an alternative to large- 
diameter rods where reasonable quantities of 
solder need to be introduced into the joint. 

Due to the relatively high cost of the alloys, 
the smaller cross-sections are usually preferred, 
but it is a mistake to believe that only a small- 
diameter wire is required since this is not very 
stiff and it is difficult to control when being 
applied to the joint. Due to the high tempera¬ 
tures needed for silver soldering, and the large 
sizes of some assemblies (boilers and the like) it 
is usually necessary to operate at some distance 
from the workpiece. A long, stiff rod is advan¬ 
tageous and allows precisely controlled dabs of 
solder to be applied economically. Naturally, if 
much small work is undertaken, a fine wire can 
also be kept in stock. 

If there is a likelihood that two or three 
different types of silver solder will be kept in 
stock, they should be kept separately, and iden¬ 
tified in some way, so that it is not possible to 
mistake one for another. The difference in the 
appearance of the different solders is hardly 
noticeable, and the bundles should be labelled 
in some way to prevent confusion. Figure 9.3 



Figure 9.3 Always label silver solders otherwise they are 
impossible to distinguish. 

shows labels made by winding masking tape 
around bundles of solder, folding it back upon 
itself, and identifying the type using a ballpoint 
pen. Notice that partly used sticks have been 
put back into the bundles. 

Silver solder can also be obtained in the 
form of paints and can be used to place the 
solder in the joint before assembly, as described 

Fluxes for silver soldering and brazing 

As for soft soldering, flux selection for hard 
soldering is determined by the characteristics of 
both the solder to be used and the metal to be 
joined. The melting range of the solder must be 
matched by the active temperature range for 
the flux and the activity of the flux must suit the 
metal to be joined, and its freedom, or other¬ 
wise, from oxidation at the soldering temp¬ 

With the exception of the old-fashioned 
brazing operation using 50:50 brass, for which 
borax was the standard flux, fluxes for hard 
soldering are usually purchased against particu¬ 
lar trade names, the products of Johnson 
Matthey Metals Ltd. once again being com¬ 
monly available. If two-stage hard soldering is 
required, as for model boiler work, two fluxes 

will be required, one which suits the Easy-flo 
melting range (around 600°C) and a second 
which is active at 800°C, corresponding with 
the melting ranges ofSilver-flo 16 and Silver- flo 
24. There are some fluxes which are suitable 
for use over the full temperature range which is 

Easy-flo flux, having an active range of 550 
to 800°C, complements the Easy-flo silver sol¬ 
ders and is available in a general-purpose type 
or one specially formulated for use on stainless 

Although Easy-flo flux remains active up to 
800°C, it is usually recommended that Tenacity 
Elux No. 4A is used with the Silver-flo alloys 
since it remains active up to 850°C. For 
stainless steel, Tenacity Flux No. 5 is recom¬ 
mended, which remains active between 600 
and 1000°C. 

Easy-flo and Tenacity fluxes are available in 
paste or powder forms, the latter being the 
more commonly available type from modellers’ 
suppliers. Before use, powder fluxes must be 
mixed with clean water to form a paste having 
the consistency of thick (not whipped) cream. 
The flux must be allowed to stand for a few 
minutes after mixing and should be given a 
thorough final stir before use. 

The flux is painted onto the cleaned surfaces 
of both plates prior to their assembly since it is 
too thick to be drawn into the small clearance 
gaps which must be employed, even when using 
the Silver-flo solders with their longer melting 
range. Any pre-soldering riveting or bolting 
together must be undertaken after the metal 
has been cleaned and coated with flux. A good 
fit of the parts, together with the minimum 
number of rivets or screws, is beneficial. A 
further coating of flux should be applied to the 
assembled joint crevices prior to heating up for 
the soldering operation. 

Figure 9.4 shows a model boiler being pre¬ 
pared for the backhead to be silver soldered 
into position. The backhead and the inside of 



Figure 9.4 The backhead of a locomotive boiler secured into 
the wrapper and generously fluxed. At this stage, the firehole 

ring has not been fluxed, and the nipples for the stay ends remain 
to be fitted. 

Figure 9.5 The central hole in the hearth permits the boiler 
to be rested on the front of the firebox for silver soldering the 

the outer wrapper have been cleaned and 

backhead into position. 

coated with flux, and the assembly screwed 

together using gunmetal screws. Flux has been 
generously applied all around the main joint 
and it remains to fit the two nipples for the 
upper stays and add flux to these and to the 
outside of the firehole ring tube. 

For this operation, the boiler barrel is posi¬ 
tioned in the central hole in the hearth, and the 
throatplate is supporting the assembly on top 
of the hearth. This is shown in Figure 9.5. The 
central hole in the floor of the hearth is also 
useful when the front of the boiler is being 
soldered, since the boiler can be supported 

below the hearth, allowing the smokebox end 
to project into the working area as shown in 
Figure 9.6. 

Figure 9.6 The smokebox tubeplate can be silver soldered by 
supporting the boiler on its firebox below the hearth and 
allowing the smokebox end to project through the central hole. 



In both of the above examples, the boiler is 
packed around with firebricks to provide insu¬ 
lation and concentrate the heat of the torch. 

When the heating commences, the lighter 
elements in the flux (the water or the liquid 
medium in which the flux particles are sus¬ 
pended) soon evaporate, leaving a coating of 
flux particles on the surface. As the tempera¬ 
ture is raised, the flux becomes molten and 
flows over the surface, cleaning it and prevent¬ 
ing oxidation by excluding the air. Once the 
plates are hot enough, silver solder applied to 
the surface melts and runs over the clean metal, 
displacing the molten flux. When using a solder 
with a short melting range, such as Easy-flo No. 
2, it will be seen to ‘flash’ round the hot joint, 
over quite long distances, almost instantane¬ 

Flux residues 

When the job cools, flux residues remain as a 
solid coating which must be removed before 
the joint can be examined. Different fluxes 
leave residues presenting different degrees of 
difficulty in their removal, some being soluble 
in water but others requiring a weak acid or 
alkali solution if chemical removal is desired. 

The Easy-flo fluxes form residues which are 
slowly soluble in hot water so that a simple im¬ 
mersion in the domestic sink, together with 
some mechanical assistance (scraping) will usu¬ 
ally suffice to remove the residues. Tenacity 
Flux No. 4A forms a residue which is virtually 
insoluble in water and a 10 per cent sulphuric 
acid solution (in water) is recommended as the 
best means of removal. This ensures complete 
removal without any mechanical assistance and 
the acid bath also cleans the base metal, dissolv¬ 
ing any scale (oxide) and producing bright, 
clean metal. The solder is also cleaned, giving a 
bright finish to the joint and permitting a 
proper examination to be made. 

Figure 9.7 The appearance of a correctly made silver- 
soldered joint. Solder has run through the joint, from the other 
side, around each tube to present a significant fillet of solder. 

Figure 9.7 shows the appearance of the loco¬ 
motive boiler of Figures 9.4 to 9.6 at an earlier 
stage in its completion. The inner and outer 
assemblies have just been united, but the illus¬ 
tration shows the results of the earlier soldering 
during which the tubes were soldered into the 
firebox tubeplate. For this operation, the solder 
was applied to the opposite side of the tube- 
plate, and the solder seen around each tube has 
flowed through during the soldering operation, 
to provide a bright ring on the inside, confirm¬ 
ing good penetration of the solder, and indicat¬ 
ing the soundness of the joint. 


The process of cleaning using a 10 per cent 
solution of sulphuric acid in water is generally 



known as pickling, the dilute acid being 
referred to as the pickle. A 10 per cent acid 
solution is easily prepared from commercial 
sulphuric acid in the ratio of nine volumes of 
clean water to one volume of commercial acid. 
ADDED TO ACID, since the first drops to 
touch the acid are likely to cause a chemical 
reaction not unlike a small explosion, thus 
spreading acid in all directions. ALWAYS ADD 
ACID TO WATER. This way, the procedure is 
quite safe, the only dangers really being due to 
the corrosive nature of the acid itself. 

The primary requirement is for a large, safe 
container in which the acid solution can be 
stored. Pickling in a ten per cent solution of 
sulphuric acid is normally reserved for use on 
brass or copper assemblies and is usually associ¬ 
ated with boilermaking. A reasonably large 
container is required, the minimum size being 
such as will allow a model boiler at least to 
stand on its back or front and be half immersed 
in the pickle. 

Figure 9.8 shows a large chemical vat which 
I use. This has a sealed lid which is held in place 
by a steel hoop which is tightened onto the top 
of the vat by a lever toggle. The vat is about 
one-third full of dilute acid and meets the 
requirement to immerse about half of the 

Figure 9.8 A safe container is essential if acid pickling is to be 

In many instances, a builder’s plastic bucket 
will serve for the actual pickling and a screw- 
top plastic barrel will be suitable for storage. 
Since the acid is corrosive, even in its dilute 
state, there is some advantage in having a stor¬ 
age container which is also large enough to use 
for the pickling operation since this avoids the 
minor splashes caused by the transfer between 

It is worth noting that, if clothing is splashed 
by the acid, a hole will result. This may be con¬ 
tained to some extent by washing immediately 
with water, but a hole will surely appear in due 

It goes without saying that any splashes on 
the skin must be washed off with plenty of 
water, but particular care must be exercised 
when handling the concentrated acid since this 
is extremely corrosive and should only be 
handled when wearing rubberised gloves and 
protective goggles. 

When the concentrated acid is poured from 
its container, which is likely to be a narrow¬ 
necked, plastic ‘Winchester’, care must be 
taken not to allow the acid to gulp in the neck 
of the bottle, as air is drawn in, since this causes 
an unsteady stream of acid to be poured, in turn 
causing droplets to be expelled in random 
directions. Pouring must be slow and steady, 
allowing air to enter the neck of the bottle 

There are also possible dangers in the 
pickling operation itself. Some of the earlier 
writings on the subject of pickling have given 
rise to the belief that it is necessary to place the 
job in the pickle while it is still very hot. Since 
chemical reactions are usually speeded up as 
the temperature increases, placing a very hot 
workpiece in the acid produces a spectacular 
cleaning of the metal and removal of the flux 
residues. Unfortunately, it also produces a spec¬ 
tacular fizzing and splashing, accompanied by 
the emission of fumes and this markedly 
increases the likelihood of damage to clothing. 



It is also difficult to avoid inhaling the fumes, 
and there is an overall danger to health. 

It is perfectly safe to place a warm workpiece 
in the acid solution, but the general advice to 
allow a large silver-soldered boiler to ‘cool to 
black’ and then pickle, often repeated over the 
years, should be regarded with circumspection 
since it is definitely not sufficient just to allow 
the redness to go out of a large copper assembly 
and then place it straight into the pickle bath. 
Let it cool at least to being fairly warm. The 
cleaning of the metal will not be so rapid, but it 
will be safe. 

After pickling, the work should be cleaned 
by a thorough washing in plenty of water and 
the joint crevices scoured with kitchen cleaning 
powder (e.g. Vint or whatever is available) and 
a stiff nailbrush, afterwards washing thor¬ 
oughly once more. This allows the joints to be 
inspected without fear that pockets of acid 
remain in, or on, the job, which might damage 
clothing, tools and so on, in the workshop. 

Thorough cleaning is also necessary to 
ensure complete removal of flux residues, since 
these are usually mildly corrosive. 

A silver-soldered joint 


The making of the joint is not essentially differ¬ 
ent from soft soldering, except that the flux is 
supplied as a powder which needs to be mixed 
with water before use, and the soldering opera¬ 
tion takes place at a much higher temperature 
than is required for soft soldering. 

The large size of some assemblies which 
need to be silver soldered means that there is a 
need to provide large amounts of heat, but if a 
range of burners is available, a suitable one can 
be chosen for the job in hand. 

After making the parts of the assembly and 

deciding the method of locating them together, 
immovably, during soldering, the first opera¬ 
tion is to ensure that the mating faces are really 
clean. Any burrs left by the cutting and shaping 
processes should be removed and any deposits 
of oil or dirt removed by wiping with a clean 
cloth or tissue. 

The next process is to clean the surfaces 
thoroughly using a medium grade emery cloth, 
afterwards avoiding contact with the cleaned 
surfaces with the fingers in order not to deposit 
any grease on them. 

Flux must be spread on the joint surfaces 
before assembly, so the next task is to mix up 
some flux. If the flux is available in the form of 
a ready-mixed paste, this can be used directly, 
but if a powder flux is being used, which is the 
most convenient form for the occasional user, 
the powder must be mixed with clean water 
until the right consistency is achieved. 

Flux is expensive, but it must be present 
throughout the joint before heating com¬ 
mences, otherwise the work will oxidise and a 
sound joint will be impossible to achieve. The 
mixed flux needs to be rich in flux, and not a 
thin, watery mixture. 

Mixing should commence by putting what 
you perceive as enough powder for the job in 
hand into a clean container. Water needs to be 
added to the powder, but less water, by volume, 
is required than there is powder, so water is 
added only in small quantities. 

For a small job, only a small amount of flux 
powder, say, of the order of half to one tea¬ 
spoonful, is needed. Add two drops of water 
(only) to this amount of powder and mix using 
a medium-sized artists’ paint brush. Add one or 
more drops of water if the mixture is not fluid 
enough, but on no account add more than a 
drop at a time, otherwise the mixture will cer¬ 
tainly become too wet. 

If this happens, more flux powder can be 
added, but it is amazing how much is needed to 
correct even a small over-addition of water, and 



it is much better to err most definitely on the 
side of too little water and add it drop by drop. 

When the work is clean, and the flux mixed 
to the right consistency, which is always 
described as ‘the thickness of cream’, the flux- 
and-water mixture can be painted onto the 
joint faces. The mixture must wet the work. If 
the pool of liquid contracts into a small pool 
when the brush is removed, the metal is dirty 
and the mixture is not wetting the surface. If 
wetting does occur properly, the mixture will 
stay where it is put on the metal. 

When both surfaces have been coated with 
flux mixture, the joint can be assembled. The 
fact that wet flux is now present means that 
assembly by riveting is not really a viable 
option. A screw or clamp is to be preferred, or a 
rivet (or rivets) may be used to provide location 
while relying on the mass of the parts to hold 
them together. However, positive location is 
vital, so that accidental disturbance cannot 
occur during heating and cooling. 

Once the joint components are located ad¬ 
equately, the work can be transferred to the 
hearth and firebricks placed around it to con¬ 
tain the heat yet allow access for the torch. In 
this respect, it is worth noting that conven¬ 
tional burners, which have air holes in the 
burner tube itself, will not burn correctly if 
pointed into a restricted corner. So, enough 
space needs to be allowed around the object to 
be heated. 

The torch must now be made ready for heat¬ 
ing. This means fitting the torch handle assem¬ 
bly to the hose and connecting the required 
neck tube and burner. With the outlet valve on 
the gas bottle turned off securely, the hose and 
hose failure valve, or regulator, is connected up 
(left-hand thread for propane cylinders) and 
with the valve on the handle assembly turned 
off securely, the valve on the cylinder is 
switched on and the hose failure valve (if in use) 
set to allow gas to pass to the handle and 

It is valuable to add a little more flux around 
the joint to be certain that there is an adequate 
supply present in the region(s) in which the 
silver solder will be applied. 

A reasonable length of solder should be 
taken from stock, bearing in mind that the 
work and the hearth will become hot and an 
ungloved hand will not be able to approach the 
work very closely. 

Making the joint 

After all of this preparation, the burner can be 
lit! The flame needs to be adjusted by setting 
the valve in the handle assembly so that the 
burner is operating correctly, after which the 
flame can be applied to the work. 

The initial aim is to raise the temperature of 
the joint generally so that the water is evapo¬ 
rated from the flux mixture. This causes the 
flux to regain its white appearance, but it 
coagulates and should envelope the joint if 
there is sufficient present. If the amount looks 
rather sparse, it may be wise to stop at this 
stage, let everything cool, clean up thoroughly 
and start again. 

With sufficient flux present, heating contin¬ 
ues until red heat is approached. The flux 
begins to melt, and takes on a clear, glass-like 
appearance, surrounding the joint. 

At this stage it is possible to see that the flux 
is doing its job in preventing oxidation, since 
the joint area is (or should be) bright and clean¬ 
looking, while the area outside the immediate 
region of the joint, where there is no flux, will 
be black, or obviously discoloured. 

When the work has reached red heat, the 
flame is moved away slightly and the stick of sil¬ 
ver solder touched onto the joint surface, typi¬ 
cally the intersection of the two overlapping 
parts. If the work is hot enough, the solder melts 
on contact with the joint. If a solder with a short 
melting range is in use, such asEasyflo No. 2, it 



will run along the joint instantaneously. The 
flame is moved along the joint, assuming that it 
is long, and the process repeated. When the 
whole joint, or all of the parts of the joint, have 
been soldered, the burner is turned off, and the 
job allowed to cool. 

If, when the solder is applied to the joint, it 
melts but does not run, the work is not quite 
hot enough and the flame must be brought back 
once more onto the joint until the solder bead 
which has been left on the work melts, and runs 
through the joint. 

The solder cannot be placed directly in the 
flame since it melts immediately and it must 
always be applied to the work with the flame 
removed. It then acts as a thermometer, con¬ 
firming, or not, that the work is hot enough to 
allow the joint to be made. 

A demonstration observed 

If all of the above seems very precise and fussy, 
it is because the adoption of this procedure will 
ensure the best chance of a sound joint. As an 
example of how quick and easy it can be, let me 
relate the story of a demonstration once seen in 
a nightschool classroom. 

A student had a small, forged spring which 
needed a temporary repair, and we were prom¬ 
ised that the course tutor would “see to it, all in 
good time”. 

On the evening in question, the tutor saun¬ 
tered down to our end of the room and sta¬ 
tioned himself at the end of the large confer¬ 
ence table around which we all sat for the 
evening’s work. 

“If anybody’s interested” he said, “I’m just 
going to do a spot of silver soldering” and he 
placed a piece of firebrick and a DIY blow¬ 
torch, with gas canister attached directly to the 
handle, on the tabletop in front of him. 

He asked the young lady for the pieces of her 
broken spring and then produced from a pocket 

a tiny piece of emery cloth or carborundum 
paper, about 1 in. (25mm) square. He cleaned 
up the broad ends of the two pieces of spring 
with the abrasive paper, then laid one piece 
carefully on the firebrick with the clean surface 

He carried a little spittle on one finger and 
placed it on the clean spring end and then 
equally carefully placed the other part of the 
spring on the firebrick, overlapping the first by 
a small amount. Further spittle was added to 
the overlap. 

Out of his shirt pocket, he took a small twist 
of paper which he opened very carefully to 
reveal a very small quantity of white powder, 
some of which he sprinkled carefully on to the 
spittle-coated overlap. 

Having twisted the scrap of paper up to seal 
the packet once more, he placed it back in his 
shirt pocket and withdrew a short length of 
silver solder. 

Fortunately, there was a smoker present, so 
the blowtorch was lit and adjusted, and the 
flame applied to the overlap of the spring. In no 
time, the small spring was red hot, and one deft 
touch with the silver solder stick caused the 
solder to melt and run through the joint. 

The job was done - no fuss, no bother, and 
all the time talking about general aspects of 
silver soldering, but letting the assembled 
students draw their own conclusions about the 
procedure, and the ease with which it might be 
done. Quite a difference between this demon¬ 
stration and my first attempts at silver soldering 
a 5-inch gauge locomotive boiler. 

Getting solder to the joint 


In modelling, the type of joint frequently 
required means that relatively large areas of 



metal are butted together and the resulting 
joint is soldered by heating the two parts and 
applying solder and flux at the edges of the 
joint. In applying the solder externally, a 
significant fillet is sometimes built up on the 
outside of the joint. This fillet is often not 
desired and it must be removed afterwards by 
use of a scraper or some other mechanical 

One way to reduce the size of this fillet (it 
may not be eliminated completely) is to intro¬ 
duce the solder into the centre of the joint 

The most obvious way to introduce the 
solder is to place it in position before the items 
in the joint are brought together. If the joint is 
to be soft soldered, tinning of the parts prior to 
assembly is the most convenient method, the 
technique for which is described above. 

Another technique is to apply solder and 
flux in the form of paint, prior to assembly of 
the joint. 

Direct access to the joint 

Alternatively, access holes can sometimes be 
provided to allow the solder to be introduced 
into the centre of the overlapping joint, 

All that is required is a small access hole 
drilled in one of the components of the joint, 
roughly central on the overlap if it is small. For 
a larger joint, several holes should be drilled. 
Once the two items have been brought up to 
temperature, solder can be introduced through 
the hole(s), right into the middle of the joint, 
from where it can spread outwards. This tends 
to eliminate any unsightly external fillet, al¬ 
though if the solder spreads as it should, some 
will reach the edge of the joint, and a small fillet 
will appear. Provided that too much solder is 
not introduced, the fillet is usually small and a 
much neater job results. 

Solder paints 

A common commercial method of placing 
solder in a joint before assembly is found in the 
plumbing trade for which joint fittings are 
made which already incorporate a ring of sol¬ 
der. To make a joint, a length of pipe is simply 
cleaned at one end, fluxed and inserted into the 
joint piece. Heating the pipe-and-joint assem¬ 
bly melts the solder, which flows through the 
cleaned and fluxed joint, thus sealing the union. 

This method is not exactly ready-made for 
modelling, but solder paints can be obtained 
which comprise solder particles suspended in a 
liquid flux. The well-mixed paint is spread onto 
the cleaned surfaces of the joint materials, the 
two parts of the joint clamped together and the 
whole joint heated. The solder particles melt 
and flow throughout the joint, solder normally 
appearing as a small fillet at the external 
boundary. Heat is then removed and the clamps 
taken off when the joint has cooled and the sol¬ 
der solidified. 

The method has the great merit that solder is 
introduced directly into the centre of the joint, 
where it can do most good, and only small 
fillets show at the edges of the joint. Paints are 
available containing a suspension of either 
silver solder or soft solder particles and the 
method can be used to make both soft- and 
hard- soldered joints. 

Since the paint is only partly solder, only a 
small quantity is introduced into the joint, 
hence the better appearance of the finished 
joint. However, the paucity of solder does 
mean that the method is not ideally suited to 
the sealing of joints and should, therefore, be 
used with circumspection on water, petrol or 
oil tanks, or pressurised joints. 

The solder particles in a paint rapidly settle 
to the bottom of the jar and a thorough stirring 
is essential before use. Although solder paints 
include gelling agents to help prevent rapid 
settlement of the particles to the bottom of the 



jar, it is best to stir regularly if a series of joints 
is to be made. 

The lighter elements of the suspension 
medium evaporate progressively from solder 
paint and stock needs to be reconstituted after a 
time. Provided that the paint is not completely 
dried out, tap water can be used, but if a crust 
has formed on soft solder paint, it should be 
removed and discarded since the solder parti¬ 
cles will be oxidised and useless for their pur¬ 
pose. Some suppliers of solders and fluxes 
recommend periodic addition of an appropri¬ 
ate flux from their range to prevent excessive 
drying out. 


Types of adhesive 

Three groups of adhesives are widely used in 
engineering for the joining of similar or 
dissimilar materials. These are Epoxy Resins, 
Anaerobic Compounds and Cyanoacrylate 
Adhesives. These different groups of com¬ 
pounds have different basic characteristics: the 
as-supplied and final (set) forms and appear¬ 
ance, the bond strength, setting or curing time, 
suitability for joining different groups of mate¬ 
rials and the environmental capability (tem¬ 
perature and relative humidity) which the 
adhesive is designed to withstand when set. 
Since each type of adhesive uses different com¬ 
pounds, and basically produces different 
results, each can be considered separately. 

However, one general point is worth making 
- plastic materials do present some problems in 
relation to gluing operations. There are several 
basic types, of which the Polyolefins, Poly¬ 
carbonates, Styrene-based and Nylon/Acetal 
types are the most common. There are also 
other materials, such as glass- or carbon-fibre 
reinforced materials which must be glued with 

Figure 9.9 This bunker side illustrates what might be done to 
arrange attachments. The upper part of the tee section beading 
around the top edge is glued into position, and the handrail is a 
steel hoop to which two brass collars have been silver soldered. 
This assembly is soft soldered to the bunker side. 

particular adhesives if satisfactory bonds are to 
be created. It is necessary to know reasonably 
precisely what type of material you are dealing 
with before attempting to glue plastic compo¬ 

An example of the use of adhesives is shown 
on the locomotive bunker in Figure 9.9. The 
top edge of the side platework is trimmed on 
the prototype with a tee-section beading. On 
the model, this was represented by riveting a 
brass strip around the top edge and then gluing 
the radiused top of the tee into position using a 
2-part liquid glue. 



As an aside, the handrail was made by silver 
soldering two collars on to the rail and then 
soft soldering these to the bunker side 

Epoxy resins 

Epoxy adhesives are supplied as two compo¬ 
nents, one part being the epoxy resin and the 
second being the hardener. These components 
are supplied as viscous (thick, or slow-flowing) 
pastes. To use the epoxy, a 1:1 mix of resin and 
hardener is prepared by stirring together the 
two parts and the mixed adhesive is then coated 
onto cleaned joint surfaces. The items to be 
joined are then clamped together and the joint 
set aside to allow time for the adhesive to 
harden, or cure. 

Hardening times may be as long as 24 hours 
at room temperature for standard epoxies, but 
rapid or quick-set variants are available in some 
ranges. These have a hardening time of roughly 
10 minutes at room temperature which is 
advantageous if the joint cannot be clamped 
together conveniently. The chemical reaction 
which causes hardening to occur commences 
immediately the two components are mixed 
together, thus establishing a working or han¬ 
dling time for the mix. For standard epoxies, 
this is 1 to 2 hours at room temperature, 
whereas for the rapid-setting types it may be as 
short as 5 minutes. 

The setting time of standard epoxies can be 
significantly reduced by curing at an elevated 
temperature, the curing times being typically 
24 hours at 20°C, 3 hours at 60°C but only 20 
minutes at 100°C. The heat generated by a 
reading lamp can be beneficial in this respect 
and two hours under a 60-watt reading lamp 
placed immediately above the work is usually 
sufficient to achieve an initial cure so that the 
job may be handled. Curing at room tempera¬ 
ture requires 3 to 5 hours before the joint may 
be handled. 

If an epoxy resin is heated to assist curing, 
this reduces its viscosity, allowing it to flow 
more readily, and more easily fill the joint, thus 
conferring added strength to the cured joint. 

Surface preparation of the joint materials 
prior to using epoxy adhesives is normally by a 
simple ‘abrade and clean’ process. That is, by 
use of an abrasive paper to produce a rough¬ 
ened surface, followed by a dusting off of the 
material loosened from the surfaces. Chemical 
cleaning is not usually required, but the materi¬ 
als to be joined must be clean and tarnish-free 
and this presents some difficulties in relation to 
brass and aluminium alloys which form tarnish 
coatings very rapidly in air. Difficulties may 
therefore be experienced with these metals. 

Epoxies are generally well suited to forming 
metal-to-metal joints or for joining metal to 
other materials such as wood, plastics and even 
glass. The difficulty with glass is the prerequi¬ 
site abrasion, without which the bond strength 
is relatively poor. Plastics also present other 
problems since different forms of epoxy resin 
do not form bonds of similar strength with all 
plastics and the manufacturers’ data sheets 
should be consulted for detailed information. 

Once fully cured, epoxies have the capability 
to sustain the bond strength exceedingly well at 
temperatures as low as -50°C and up to 100°C. 
The quick set or rapid versions of these adhe¬ 
sives do not usually show such good high- 
temperature performance and for some of 
these also, the strength achieved, particularly 
the impact strength, is significantly worse than 
the regular types. If good strength is required, it 
is best to use a standard epoxy resin adhesive 
and to design the joint so that clamping during 
the long curing time will be straightforward. 

Some epoxies are available in high strength 
variants, as also are some ‘loaded’ epoxies 
which incorporate a metallic filler to provide 
increased hardness in the cured state. Alterna¬ 
tively, epoxies may be loaded to provide electri¬ 
cal conductivity when set. Metal-loaded epox- 



ies are suitable for the repair or filling of cast¬ 
ings, but if they are intended for filling rather 
than for use as adhesives, the two components 
frequently take the consistency of putty. Mixing 
of these is by kneading the putty-like compo¬ 
nents together, again by taking equal quantities 
of the resin and the hardener. 

Anaerobic compounds 

Anaerobic compounds are adhesives which 
solidify in the absence of air, ‘anaerobic’ simply 
meaning ‘having no oxygen from the air’. Pro¬ 
vided that they are excluded from contact with 
air, these compounds will solidify and the nor¬ 
mal method of use is simply to introduce them 
into a close-fitting joint. This allows setting to 
commence in the centre of the joint and there is 
no further requirement for a setting agent or 

These types of adhesive comprise a wide 
range of materials which have different 
strengths and characteristics when set. This 
allows their use for the formation of permanent 
bonds or, at the other extreme, to provide a 
sticky, gap-filling compound used for locking 
ordinary screwed fastenings to prevent their 
slackening under vibration, yet allow normal 
disassembly. This method of thread locking is 
described in Chapter 8. 

Since a range of compounds has been devel¬ 
oped, different materials are provided for 
specific duties, the ranges normally being 
described as high strength or medium strength 
retaining compounds, and the low-strength, or 
screw-locking types. The high-strength materi¬ 
als are used for permanent bonds. These 
descriptions over-simplify the nomenclature 
however, since even at the low-strength end of 
the range (the thread-locking compounds) vari¬ 
ous strengths are available for different uses. 

Low-strength anaerobic compounds are 
intended for thread locking. Two basic types 

are available, the first being intended for appli¬ 
cation to the threads before assembly, while the 
second type has sufficiently good creep to be 
applied to the threads after assembly. The 
Loctite range of anaerobics is perhaps the most 
readily available and this provides a total of 
four thread-locking compounds. These are 
Loctite 221 (Screwlock), Loctite 241 (Nut- 
lock), Loctite 270 (Studlock) and Loctite 290 
(Penetrating Adhesive). Loctite 270 has the 
highest strength, being intended for the locking 
of studs which will not normally require 
removal, 241 (Nutlock) has about half the 
strength of Studlock and 221 (Screwlock) is 
half the strength again and is intended for low- 
strength locking of screws which may require 
adjustment or where easy disassembly is 
required. It is also suitable for use with fasten¬ 
ers made in soft metals. 

For permanent or semi-permanent bonding 
of cylindrical parts, anaerobic compounds 
provide a more convenient alternative to the 
press or interference fits which were previously 
used. As described in Chapter 8, these forms of 
fit require accurate machining of the compo¬ 
nents and careful pressing together in order 
to achieve correct results. The advantage of 
using an anaerobic adhesive is in the wider 
machining tolerance which the gap-filling 
capacity of the adhesive allows. The machining 
of the parts is less critical, and is therefore 
usually quicker. 

Retaining compounds, as the high-strength 
anaerobics are called, are formulated for two 
specific duties - the permanent fitting or 
assembly of cylindrical items which will not 
require disassembly, and the fitting of such items 
as ball and roller races which are subject to wear 
and may therefore eventually require removal. 
In the Loctite range, 601 (Retainer) is the high- 
strength (permanent) type and642 (Bearing Fit) 
is the lower-strength alternative. Loctite 641 
has a strength slightly less than Studlock while 
601 (Retainer) is about twice as strong. 



The gap-filling qualities of these two com¬ 
pounds allow the items to be slip fitted so there 
is no question of pressing parts together. This is 
especially beneficial when fitting ball or roller 
races since there is no possibility of squeezing in 
the races and thereby damaging the bearing. 
Loctite 641 (Bearing Fit) can be used with radial 
gaps up to .006in. (0.15mm) whereas 601 will 
accommodate a radial gap of .004in. (0.1 mm). 

Although described as being of use for per¬ 
manent fits, bonds made using Loctite 601 
(Retainer) can be disassembled should the need 
arise. The working temperature range of the 
cured adhesive extends to beyond 150°C but at 
this temperature the bond strength is between 
30 and 50 per cent of that at room tempera¬ 
ture. Warming up the joint will normally make 
it possible to break the joint bond by pressing 
the two items apart. Heating the cured material 
does not significantly affect the bond strength 
which is retained when the joint has cooled to 
ambient temperature once again and the press¬ 
ing operation must therefore be undertaken 
while the joint is still hot. 

Surface preparation of the joint prior to 
application of anaerobic compounds is depend¬ 
ent on the grade to be used. The thread-locking 
grades should be used on clean threads, but will 
cure effectively even if the parts are coated with 
a rust-preventive coating, provided that this is 
not excessive. In industrial applications this 
means that threaded fasteners are used in the 
as-received condition, without any special pre¬ 

For retaining compounds, a degree of clean¬ 
ing is desirable, the aim being to produce clean, 
bare metal. Cleaning should be performed 
using the solvent recommended by the manu¬ 
facturer. Most anaerobics are soluble (in the 
uncured state) in chlorinated solvents such as 
trichloroethylene and these materials are usu¬ 
ally recommended for use as surface cleaners 
prior to application of the adhesive. Some 
increase in ultimate strength is provided by 

roughing the surfaces and a treatment with 
medium grade abrasive cloth, prior to use of 
the solvent cleaner, helps to provide a key 
within the joint. 

Although the anaerobics harden naturally 
when deprived of oxygen, the speed at which 
curing takes place is affected by the material in 
the joint, curing occurring less rapidly on alu¬ 
minium, cadmium, zinc and stainless steel than 
on iron or mild steel. Primers are consequently 
sometimes available to accelerate the curing 
process. Even without a primer, curing still 
takes place on the less active metals, but the 
times to achieve handling and full strength 
(normally about 15 minutes and 3 to 5 hours, 
respectively) are increased. For the amateur 
worker, only stainless steel is likely to create 
problems in this respect. 

The anaerobic adhesives are designed for use 
in metal-to-metal joints and although they will 
bond to some plastics, there are many for 
which anaerobics are not suitable. If this type of 
application is required, the manufacturer’s 
literature should be consulted. 

The anaerobic compounds include a range of 
materials intended for use as replacements for 
paper, cork or fibre gaskets which are used for 
sealing pressurised joint faces. These can be ex¬ 
tremely useful for modellers since their use 
avoids the need to cut and punch small gaskets 
in conventional joint materials. Sealants are 
usually supplied in a tube or syringe with appli¬ 
cator nozzle built-in. A thin bead of sealant can 
consequently be readily applied around the 
joint face and the parts assembled. Once the 
joint is closed up, the sealant cures from the cen¬ 
tre in the normal way. Surface preparation prior 
to sealing is usually by use of a suitable solvent. 

Cyanoacrylate adhesives 

The most common form of cyanoacrylate 
adhesive is the familiarLoct/te Super Glue. Like 



most of the cyanoacrylates, this hardens in 
a few seconds on almost any material and no 
heat or long-term clamping is required. The 
hardening time can be as short as 10 seconds, 
hence the description instant glue. This feature 
brings with it the danger that accidental bond¬ 
ing of the skin may occur, so these adhesives 
must be used with great care. 

It is worth noting that, although hardening 
occurs in just a few seconds, full strength is nor¬ 
mally only achieved after 12 hours. 

Although suitable for use with many materi¬ 
als, some types of plastic are not readily joined 
by these adhesives and special variants are 
available to suit these more difficult applica¬ 
tions. Porous surfaces, such as wood and 
ceramics also present possible problems and an 
activator is frequently recommended in these 
instances. This acts as an accelerator for the 
hardening process. The activator is normally 
applied to one surface to be joined and the 
adhesive to the other but it is sometimes neces¬ 
sary to coat both surfaces with activator, 
depending on the materials to be joined. 

Some specialised applications do require 
specially formulated versions of the adhesive. 

particularly those which require the use of an 
activator. These are catered for by special packs, 
usually sold under descriptive names which 
make their intended use obvious. Among avail¬ 
able types are included those suitable for glass- 
to-metal joints, especially for use on toughened 
glass, as used for car windscreens, and some 
for use on rubber and plastics. These are 
provided with full instructions and a general 
specification of the materials for which they are 

Cleaning of the joint materials using a chlo¬ 
rinated solvent is sometimes a requirement 
prior to bonding but a clean and dust-free 
surface is frequently all that is required. Once 
fully hardened, cyanoacrylates are generally 
satisfactory at temperatures up to 80°C and if 
service use is likely to exceed this, a high-tem- 
perature variant should be chosen. These allow 
satisfactory use up to 120°C. Since the maxi¬ 
mum in-service temperature is relatively low, 
the bonds may be broken down by heating the 
joint above the maximum and peeling it apart. 
After heating, the glue residues are generally in 
the form of a tacky strip which may be removed 
by scraping. 



Hole production 


The production of holes, normally referred to 
simply as ‘drilling’ is an absolutely basic 
requirement of any engineering manufacturing 
activity. For smaller holes, the normal tools 
used are twist drills which are available in a 
wide range of closely spaced sizes. The range of 
sizes likely to be of interest will be from about 
.0135in. (0.35mm) to '/iin. (12.7mm) or so in 
diameter although the occasional use of larger 
drills may be required, depending on the facili¬ 
ties which are available and the type of work 
being carried out. 

Certainly, for holes above lin. (25mm) in 
diameter and perhaps from '/iin. upwards, 
alternative means of hole production are nor¬ 
mally employed. These will take the form of 
tools having single-point cutters, such as the 
trepanning tool, some of which require a ‘pilot’ 
hole to be drilled prior to their use to ensure 
positional accuracy. An adjustable boring head 
may be available for the production of larger 
holes, this being a single-point cutter mounted 
in an adjustable holder and having a fine adjust¬ 
ment of the diameter being cut and which 
consequently is very versatile in use. These 

types of tool definitely take us out of the drill¬ 
ing types of operation. 

The production of holes is not restricted to 
just plain drilling or boring but encompasses 
also the related operations of countersinking, 
counterboring and reaming, each of which is 
described separately below. 

Holes may, of course, be produced on the 
lathe as a drilling or boring operation, or per¬ 
haps on a vertical mill or mill/drill, but the most 
usual method is to use a drilling machine. 

As noted above, the amateur worker may 
have little interest in drills beyond Van. or so in 
diameter since the slow speeds required for 
larger drills are not normally available on the 
smaller drilling machines. The power on a 
smaller machine is also normally less than is 
required for the larger drills and the normal 
maximum chuck size is likely to be '/iin. or 
around 12.5 mm. 

The drilling machine 

Twist drills are nowadays such a universally 
available tool that there can be few people who 



have not used them for producing (drilling) 
holes. The usual method of use is to fit a drill 
into the chuck of a DIY electric drill, and drill 
away. If you have the luxury of a 2-speed, or 
variable-speed drill, you will be aware that 
larger drills need to run at a slower speed and, 
as an experienced ‘driller’, you will appreciate 
the general need for squareness when drilling 
holes. However, DIY and related drilling is fre¬ 
quently quite forgiving in terms of squareness, 
and the materials to be drilled are often rela¬ 
tively soft, apart from concrete and brick, for 
which special techniques are used. 

If you have tried to drill a large hole in a 
hard material, you will know that considerable 
force is needed to push the drill into the work 
and if much drilling of hard metals is required, 
an arrangement which gives some leverage 
(mechanical advantage) is advantageous. The 
need for squareness is also much more stringent 
in engineering and there needs to be a re-think 
of the drilling method. 

One of the simplest ways to effect the 
change is to use a bench drill adaptor for the 
DIY drill. Such adaptors mount the drill into a 
frame which slides on a vertical column 
mounted into a casting which forms the base, 
or pedestal, of the tool. The frame is usually bi¬ 
ased upwards by a spring which counterbal¬ 
ances the mass of the drill. 

A long lever is pivoted to the column, or an 
attachment to it, and is linked to the frame so 
that pulling on the handle presses frame and 
drill downwards and into the work, which is 
fixed to, or resting on, the base casting. The 
closeness of the lever pivot to its attachment at 
the drill frame means that the pull on the 
handle is magnified at the drill and consider¬ 
able pressure is applied to the drill point to 
force it through the work. The adaptor also 
satisfies the need for squareness by supporting 
the drill to the column and maintaining align¬ 
ment with the work throughout the whole 
range of movement. 

Figure 10.1 A bench-mounted drilling machine fitted with a 
'Ain. (12.5mm) capacity chuck. Modem drills are fitted with a 
transparent guard around the chuck. 

A more regular type of bench drill is shown 
in Figure 10.1. This is typical of the medium¬ 
sized machines which can be purchased from a 
wide variety of sources. The drill is built 
around a column of ground mild steel, 2in. 
(50mm) in diameter fitted to a substantial cast 
iron base which can be screwed or bolted to the 
bench. At the head of the column, a large cast¬ 
ing provides a rear mounting for the drive 
motor and at the front, provides a bearing for a 
sleeve which contains a spindle. 

The spindle is driven by the motor through a 
vee-section belt, both spindles being provided 
with a 4-step pulley, as shown in Figure 10.2. 
This machine thus has four speeds. The spindle 
is driven from its pulley cluster by a key which 
locates in a keyway running down the spindle, 



Figure 10.2 The drill spindle is driven through a 4-speed 
drive arranged between a pair of pulleys. 

which can therefore be allowed to slide up and 
down. The bottom of the spindle carries a drill 
chuck which can be both rotated and advanced 
downwards under control of a feed handle. 

To maintain the rigidity of the spindle loca¬ 
tion, it advances in one with a sleeve which 
provides support even when the spindle is 
extended downwards. To allow the sleeve to be 
advanced, it is cut with teeth, like a rack, which 
mesh with a small gear on a shaft which has a 
control lever attached to it so that it may be 
rotated. This is seen in Figure 10.3. Pulling on 
the lever rotates the shaft and pushes down the 
sleeve and drill spindle. In Figure 10.3, the 
sleeve and spindle are fully down, this machine 
being provided with just over 2in. (50mm) of 
‘down feed’. 

A return spring is fitted to counterbalance 
the mass of the chuck, spindle and sleeve so 
that the ‘at rest’ position of the chuck is ‘up’. 
On the machine illustrated, the return spring is 
contained within the circular housing enclosing 
the left-hand side of the down-feed shaft, and is 
seen in Figure 10.1. 

There is a frequent need to drill holes 
partway through a workpiece and the down- 
feed mechanism usually incorporates an adjust¬ 
able stop which can be set to prevent spindle 
and sleeve advancing farther than is required. 
On most machines, this takes the form shown 

Figure 10.3 The down feed handle routes through three- 
quarters of a turn to bring the chuck to the fully down position. 

in Figure 10.4, but the machine of Figure 10.1 
has the stop incorporated into the right-hand 
side of the down-feed gearshaft and is in this 
instance an adjustable collar surrounding the 

Since drills vary in length according to their 
diameters, and items to be drilled also vary in 
size, it is essential to have a movable table 
below the drill head. Drilling machines have an 
adjustable table, roughly equivalent in size to 
the base, which can be moved up and down the 
column and locked in any position. On a 
medium-sized machine such as the one illus¬ 
trated, the table is moved up and down by 



Figure 10.4 This conventional depth stop is fitted to a mill/ 
drill and is fitted with a scale engraved in mm. 

The table is drilled with a central hole, and 
can be set with the hole immediately below the 
axis of the chuck and spindle to allow a drill to 
pass completely through the work without 
damaging the surface of the table. The table can 
take any position around the column and can 
be positioned to one side if this provides better 
support for the work, but there is then the pos¬ 
sibility that the drill will mark the table if it 
pierces the work and something which can be 
‘sacrificed’ must be interposed to prevent this. 
Since the table can be rotated about the col¬ 
umn, it can be swung out of the way in order to 
place the work directly on the base. 

The main physical characteristics of the drill 
is the height beneath the chuck, the others 
being the down feed and the distance between 
the column and the spindle axis. Further con¬ 
siderations are the size of the chuck which is 
fitted and the spindle speeds. Both of these 
features should match one another since the 
lowest speed really determines the maximum 
size of hole which can be drilled or bored. 

Bench drills may be said to be made in small, 
medium and large sizes. The one illustrated 
here is of medium capacity, allowing work up 
to about 8 in. (200mm) to be positioned below 
a ‘/an. (12.5mm) drill, which is the capacity of 
the chuck, and allowing 4'/an. (115mm) 
between the column and the drill centre. The 
drill actually has 12in. (300mm) clearance 
between the chuck and the base. 

This drill has what might be described as a 
‘fixed’ chuck. That is, it is not designed to be 
removed from the spindle, but is attached per¬ 
manently to it. Larger machines naturally have 
larger spindles and they may be large enough to 
be bored with one of the standard Morse tapers 
at the lower end. In these cases, a drill chuck 
mounted to a tapered arbor can be used in the 
drill, or taper-shank drills used directly. 

The Morse tapers are one of a range of shal¬ 
low tapers which are described as ‘self-holding’ 
i.e. a tapered item such as a drill or drill chuck 
can be jammed into a matching tapered socket, 
where it holds itself in. Provided that the drill is 
not too large, the friction grip is adequate, but 
larger drills need a more positive drive and the 
end of the taper is machined with a rectangular 
tang which engages in a recess inside the spin¬ 
dle. The tapered ends of a drill chuck arbor and 
a large, taper-shank twist drill are shown in 
Figure 10.5. 

If the spindle is provided with a Morse taper 
fitting, it is also normally slotted just above the 
position occupied by the driving tang so that a 
tapered drift can be inserted for driving out the 
drill or drill chuck. Figure 10.6 shows a V-tin. 



Figure 10.5 Some drilling machines accept drills and chucks 
which have Morse taper arbors. These can also be used on a 

(19mm) drill with taper shank, mounted 
directly into the spindle of a large drilling 
machine. The slot which provides access for 
driving out the drill is visible at the top of the 

This large machine is a floor-standing model, 
a type generally called a pillar drill, built around 
a column 4Viin. (115 mm) in diameter, with the 
chuck roughly 4 feet (1220mm) above the base. 
The machine head incorporates a 2-speed gear¬ 
box and 4-step pulleys to drive the spindle and 
has some very low speeds, making it suitable for 
drilling with large drills. It can also be used for 
boring large holes, using a boring head. 

On a pillar drill, the movable table is usually 
sufficiently large that it requires some form of 
gearing to lift and lower it on the column and a 
rack is normally fitted to allow this, a pinion 
shaft with a large handle being fitted into the 
rear of the table for winding it up and down. 

If you envisage the need to do heavy work, 
and you have room, a pillar drill is certainly 

At the other extreme, if you envisage a great 
deal of work with small drills, a high-speed 
bench drill can be useful. A machine having a 
top speed of 2000 to 3000 rpm with a small 
chuck accepting drills up to s /32in. (4mm) in 
diameter is often described as a sensitive drill¬ 
ing machine since the down-feed handle does 

Figure 10.6 V4in. (19mm) drill directly mounted into the 
spindle of a large pillar drill. 

not provide a huge mechanical advantage but is 
designed to provide a good ‘feel’ when using 
small drills. 

These two machines, at the extremes of 
what is readily available, might suit individual 
modellers, but for most, a bench drill having 
a Vi-inch capacity chuck (12.5mm) will gener¬ 
ally satisfy their needs. It is, however, useful to 
have a machine with a Morse taper to the 
spindle which matches the lathe or other ma¬ 
chines in the workshop, since this allows taper- 
mounted drill chucks to be interchanged be¬ 
tween them. 



Drilling machine accessories 

When a drill cuts, it tends to rotate the work, a 
tendency which must be resisted. The force 
which the drill exerts can be very high, and 
positive means need to be taken to resist it. For 
this purpose, the work is held in a small vice, 
suitable for use on the drill table, in which the 
part may be held while drilling. A vice suitable 
for this task is known as a machine vice. 

A typical vice intended for use on a drill is 
shown in Figure 10.7. The vice is machined flat 
on the underside of its base and the upper sur¬ 
face of the base is machined parallel to it, main¬ 
taining parallelism with the drill table. The base 
casting incorporates an upstanding, fixed jaw, 
and a movable jaw slides in the central slot in 
the base, being retained by a plate which is 
recessed into the underside. The moving jaw is 
positioned and tightened by a knurled sleeve 
with an internal thread which engages a screw 
retained in the moving jaw. 

When the work is being drilled, the ten¬ 
dency for it to turn is resisted by holding the 
vice handle firmly to prevent rotation. The jaws 
of this vice are fitted with hardened face pieces 
which are machined with vee-shaped grooves 
that allow circular bars to be held securely, but 
it should be noted that the sharp corners of 
such grooves do mark circular work which is 

Figure 10.7 A machine vice for the drilling machine. 

large in relation to the groove width and this 
must be remembered if damage to already pre¬ 
pared surfaces is to be avoided. 

The vice shown has jaws 3'Ain. (83mm) 
wide and l!Ain. (32mm) deep, and provides a 
maximum opening of 2in. (50mm). It suits my 
‘/ 2 -inch (12.5mm) capacity bench drill which 
has a table 8in. (200mm) square. If the 
workpiece is very large, and the hole being 
drilled is small, for example, if '/win. (1.5mm) 
diameter rivet holes are being drilled in a loco¬ 
motive running board 30in. (750mm) long, it is 
sufficient to resist the tendency for the work to 
turn, when the drill is cutting, by hand pres¬ 
sure, since the ‘hand hold’ acts as a sufficient 
radius for the tendency for the work to turn 
(the torque) to be resisted easily. 

For large holes, say, from !/ 2 in. (12.5mm) 
diameter upwards, hand pressure may not be 
sufficient to resist the torque exerted by the 
drill, and there is then the need to bolt down 
the work, or the vice with the work in it, to the 
drill table. The base of a machine vice is pro¬ 
vided with slotted lugs so that it can be bolted 
down when required, the drill table being slot¬ 
ted to allow this. 

For a small drilling machine for which the 
table has through slots, plain nuts and bolts are 
sufficient for clamping items to the table, but 
the tables of larger drilling machines may have 
a special type of slot in which the table is not 
pierced through, preventing the use of ordinary 
fastenings for holding down the work or the 
vice. In these cases, the slots are machined to an 
inverted tee shape and special nuts, or bolts 
with specially shaped heads, are used for hold¬ 
ing down. These are known as T nuts or T 
bolts, and they have to be machined to a section 
which matches the slot. Tee slots are also com¬ 
monly used on lathes, and the requirements for 
these special attachments are described in 
Chapter 16. 

A further essential accessory for the drill is 
an accurate pin chuck. This is needed to hold 



Figure 10.8 A pin chuck having a turned spigot is used for 
holding small drills in larger chucks. 

small drills since a Vi-inch or Vi-inch capacity 
chuck (12.5 or 16mm) cannot be expected to 
hold the smallest sizes, and there are often 
occasions when these are used. 

Pin vices or pin chucks for use on the drilling 
machine have a smooth-turned finish to the 
shank so that they may be held truly in the drill 
chuck. A typical pin chuck is shown in Figure 
10.8. The body comprises a turned shank with 
a screwed-and-bored lower end, onto which a 
knurled sleeve can be screwed. The body has a 
hollow end which accepts one of three split col¬ 
lets, each of which has a head which is coned 
both ways so that screwing the sleeve on to the 
body closes up the collet, causing it to grip the 
drill which is inserted down the centre. 

The three collets, together, cover the range 
from virtually zero to about Vton. (2.4mm) 
diameter. The minimum size of drill that an 
ordinary '/ 2 -inch orViin. drill chuck will grip, is 
normally '/ 32 m. or 0.8mm, so a pin chuck 
extends the gripping capability of the main 
chuck down to the very small sizes. 

Figure 10.9 shows the pin chuck in use for 
spotting through the holes in an attaching angle 
into a frame plate on a model locomotive. In 
this case, the small size of the chuck has 
allowed the work to be approached, even 
though there is a large motion plate riveted to 
the angle. 

Twist drills 

The twist drill is the most widely used general 
type of drill. It is available in two basic forms - 
those with a plain (parallel) shank and those 
having a Morse taper shank. The parallel- or 
plain-shank type is the most widely used (and 
useful) type although there are occasions when 
the availability of taper-shank drills for direct 
mounting in the tailstock or mandrel tapers of 
the lathe is extremely useful. 

A selection of parallel-shank twist drills is 
shown in Figure 10.10. Two types are shown: 
some general purpose, or jobbers’ drills and the 
shorter stub drills, in the same range of sizes. 

Figure 10.9 A pin chuck also allows easier access to the work 
in those instances in which a large chuck would foul the work. 



Figure 10.10 Jobbers' drills are the normal ones used, but 
the shorter stub drills are sometimes used for drilling directly in 
the work, without centre punching the location of the hole. 

Jobbers’ drills are the ones most frequently 
used, and many workshops manage with only 
this type. 

A twist drill has two spiral flutes machined 
up the body and at the tip, the drill is ground to 
a cone shape so that two cutting edges are 
formed. The flutes are deep in order to create 
large cutting edges, and allow the cut material 
to be ejected easily away from the cutting lips. 
This means that the drill is not very strong and 
is easily bent if subjected to any side force, so 
that the drill is inclined to ‘wander’ away from 
the truly axial path and this sometimes leads to 
positional inaccuracy. If the hole is deep, drill 
wander can cause the hole to be drilled out of 
square with the work’s surface. 

A deep, out-of-square hole naturally bends 
the drill, and since it is also quite hard and rela¬ 
tively brittle, this means that breakages can 
occur if the side loads become large. The 
shorter stub drills are designed to minimise the 
bending, since the shorter an item is, the less it 
bends for a given force. Stub drills are benefi¬ 
cial when positional accuracy is required. 

Drills are made in an extremely wide range 
of sizes, from virtually nothing, to very large 
indeed, in closely spaced sizes. The preferred 

range of British Standard drill sizes is now 
based on a metric series although some frac¬ 
tional imperial drills are contained within the 
standard. If you are setting up a workshop from 
the beginning, metric drills will have to be pur¬ 
chased except for the few imperial sizes which 
are allowed by the standard and therefore 
remain in the catalogues. 

Within the previously used British Standard 
drill sizes, three reference systems were in 
normal use. These were the fractional sizes, 
rising in increments of '/Min. from '/32m., and 
the number and letter drill sizes, ranging from 
number 80 (.0135in. diameter) to number 1 
(0.228in. diameter) and from letter A 
(0.234in.) to letter Z (0.413in.). The size incre¬ 
ments within the number and letter ranges 
were quite small, the difference between 
numbers 59 and 60 (.040in.) being only .OOlin. 
and between letters Y and Z .009in. Tables 10.1 
to 10.3 list the previous letter and number drill 

To assist in selecting metric drills as equiva¬ 
lents of the previous standards, a recom¬ 
mended set of sizes is also given in the table. 
Since imperial fractional sizes are still among 
those recommended, the list of equivalents 
includes some of these where they provide a 
closer alternative to the previous sizes. 

In the range of metric drills, the increments 
are also small and sizes between 3 mm and 
10 mm diameter are available in increments of 
0.1mm. In addition, within this range, the 
intermediate sizes 0.25mm, 0.5mm and 
0.75mm are available. Between 1mm and 3mm 
diameter, the increment is .05mm. Above 
10 mm preferred and second preference sizes 
are recommended, these latter being somewhat 
more expensive than the preferred range which 
rises in the sequence 10.00, 10.20, 10.50, 
10.80,11.00 and so on. The second preference 
range provides the 0.10mm increments in addi¬ 
tion to 10.25 mm, 11.25 mm etc. 



Table 10.1 Drill Gauge (Number Series) Drills, Sizes 1 to 60. 










































3/3 2in. 









2.25 mm 





















































































































J /Min. 





































Note: Imperial sizes are those specified for the drill gauge. 

Table 10.2 

Drill Gauge (Letter Series) Drills. 






Recommended Size 

Metric (Drill 

Alternative Gauge) 
















15 /Min. 

































J /iin. 






























17 /Min. 













Note: Imperial sizes are those specified for the drill gauge. 



Table 10.3 Drill Gauge (Number Series) Drills, Sizes 61 to 80. 



















































0.45 mm 



























0.35 mm 







Note: Imperial sizes are those specified for the drill gauge 

Sharpening twist drills 


Twist drills are sharpened by grinding. For 
larger drills, this means using the bench 
grinder, but smaller drills below about '/sin. 
(3mm) in diameter can be ground by hand on a 
slip stone or oil stone. Accurate drill grinding is 
an art which many people find difficult to 
acquire and over the years drill grinding aids 
have been developed both for home manufac¬ 
ture and for commercial sale. Do not despair 
before you start - as noted elsewhere, the 
whole of model engineering requires skills 
which can only be gained by practice and drill 

grinding should just be regarded as one of the 
skills that you must acquire. 

Drill point angles 

The drill point takes the form shown in Figure 
10.11. It is ground to a flatfish cone shape in 
such a way that two cutting lips are formed, 
one at the base of each flute. The precise shape 
of the drill point is important in promoting 
good cutting and in helping the drill to cut at its 
correct size. Drill grinding is, therefore, one 
skill which must be practised until consistent 
results can be achieved. 

The angle of the cutting edges should be 59 

Figure 10.11 A detailed view of the point of a twist drill. 



degrees each side of the centre line. This angle 
is naturally a compromise, chosen to maximise 
the general usefulness of the drill. A sharp angle 
at the point helps to centre the drill in the start¬ 
ing hole better than a larger one, but making 
the included angle at the point smaller length¬ 
ens the cutting edges and requires more power 
to drive the drill into the material being cut. A 
shallow-angle point drills faster than a more 
acute point, even though better centring is 
obtained with the latter. 

Both lips of the drill must be ground to the 
same angle and must be of the same length, oth¬ 
erwise the point of the drill (in reality, a line 
where both flanks of the drill meet in the 
web) will not be in line with the centre of the 
drill body. This causes the point to describe a 
circle when the drill is rotated, throwing the 
workpiece around and making the drill difficult 
to start in a pilot hole, or on the centre punch 

If one flank of the drill is longer than the 
other (the point is not at the centre) the drill 
produces an oversize hole. The longer of the 
two cutting lips also absorbs more power and 
this unbalances the load on the drill. Since the 
body is deeply fluted, only a relatively small 
amount of material holds the drill together - 
this is the central core, or web. To provide as 
much strength as possible, the flutes become 
narrower towards the shank, but at the tip, the 
deeper flutes leave only a small web and if an 
uneven load is placed on the drill lips, a break¬ 
age can result. Any difference in lengths of the 
cutting lips on the two sides of the drill, 
whether due to unequal angles or to a mis¬ 
placed point, must be avoided. 

The cutting edge 

If the lips of the drill are to cut, it is vital that 
there is clearance behind them. This means that 
the coning of the point must be done so that the 

lips touch (and cut) the work but the drill point 
surface behind the lip does not. This then 
allows the lips to slice their way into the work 
and the material is drilled. This clearance angle 
is generally made between 12 and 15 degrees, 
as shown in the side view of the drill point, 
Figure 10.11. 

In creating the clearance angle, material is 
ground away immediately behind the cutting 
lip thereby removing material which provides 
strength in this important position behind the 
cutting edge. The clearance angle cannot there¬ 
fore be made excessive since removal of too 
much material seriously weakens the drill and 
also reduces the heat conduction capacity of 
the cutting region, causing heating of the drill 
point. Large clearance angles cannot be used on 
account of both of these effects. 

If clearance is too small, or non-existent, the 
point of the drill merely rubs on the work and 
overheating is the result. Carbon steel drills are 
less common now than was once the case, so 
overheating is less of a problem (it is possible to 
draw the temper of a carbon steel drill by over¬ 
heating it) but too low a clearance angle stops 
the drill cutting and turns the process of hole¬ 
making into one of heat production. Clearance 
angles must be maintained in the range 
between 12 and 15 degrees. 

Grinding smaller drills 

Commercially, the flanks of the drill point are 
ground as the flanks of a cone i.e. the flanks are 
slightly convex when viewed from the side. 
This convexity is quite noticeable on large 
drills, around Van. (12.5mm) in diameter, but 
if examined closely, the surface will actually be 
found to be not far removed from being flat. 
On smaller drills, the convexity is barely dis¬ 
cernible and it will be quite satisfactory to 
sharpen small drills with the flanks of the point 
ground flat. Point angle and clearance at the 



cutting lip must, of course, be maintained, but 
the use of flat flanks simplifies the achievement 
of the desired shape. 

For small drills even the fine stone of the 
grinder is too fierce and something gentler is 
required otherwise much of the drill will be 
removed when all that is required is a fine skim 
off the point. A slip stone, or general-purpose 
carborundum stone may be used and the drill 
sharpened by hand, it then only being necessary 
to hold the drill at the correct (compound) 
angle and to rub it gently backwards and for¬ 
wards on the stone. In order to maintain the 
required angle, a watchmaker’s glass should be 
used and the drill point examined after every 
few strokes since, even at the low speeds of 
hand grinding, only light touches are usually 
required unless the drill has been broken or 
damaged through contact with hardened vice 
jaws or some such object. 

For small drills, a gauge to check the angle of 
the point (as described below for the larger 
drills) is not very valuable unless it is made to 
fairly fine limits. The eye is remarkably good at 
detecting faults in shape however, especially if 
symmetry is really the only criterion, and it will 
not be found difficult to see when the drill 
point is not quite right and what needs to be 
done to achieve the desired shape. 

Grinder safety 

For sharpening drills, a double-ended bench 
grinder having coarse and fine wheels, of the 
type described in Chapter 14, is normally used. 
For drill grinding, only small amounts of mate¬ 
rial need to be removed and the fine wheel is 
entirely satisfactory. 

A grinding wheel is designed to be used on 
its periphery and it is strongest when loaded on 
the edge. It is relatively weak when subjected to 
a load on the side, and may shatter, with disas¬ 

trous consequences, if subjected to a large side 
load. Heavy work, such as grinding and sharp¬ 
ening lathe tools must therefore only be under¬ 
taken on the edge of the wheel. 

Although described as being made in high¬ 
speed steel, drills are less tough than lathe tools 
and only very light touches of the drill on the 
grinding wheel are required. In any event, there 
is a need to remove only small amounts of ma¬ 
terial from the drill tip and very short and light 
contact is all that is needed. 

The arrangement of the wheel guards on a 
bench grinder leaves the sides of the wheels ex¬ 
posed. Since it is normal to approach the 
grinder with the edge of the wheels presented 
to the operator, the most natural approach for 
grinding the end of the drill at an angle is by 
presenting the end of the drill to the side of the 
wheel. In this position, the hands are relaxed 
and comfortable, and are in the position most 
suited to control the drill accurately. Conse¬ 
quently, it was normal to teach apprentices to 
sharpen drills in this way and it still remains the 
natural method for those who were so taught, 
including the modeller who was invited to 
demonstrate drill grinding in my workshop, so 
that photographs could be taken. 

Note particularly, however, that this is defi¬ 
nitely not the preferred method of using the 
wheel, and act accordingly. Safety goggles 
should be worn whenever any grinding is per¬ 
formed and the wheel guards and safety guards 
must be in position at all times. In Figures 
10.12 and 10.13, the safety guards have been 
removed to allow the photographs to show 
clearly the movement of the drill that is re¬ 

It is very important that the tool rest is ad¬ 
justed so that there is only a very small gap be¬ 
tween it and the wheel, to ensure that nothing 
can be pulled down into the space between the 
wheel and the guard. If this does happen, the 
wheel is very likely to be broken by the subse¬ 
quent jam-up. 



Figure 10.12 The starting position for grinding a large 
drill on the side of the grinder wheel. 

Figure 10.13 The position of the hands and the drill 
after passing one lip over the surface of the grinding wheel. 

Grinding larger drills 

As noted elsewhere, in relation to the filing of 
curves and compound shapes, it pays to have a 
clear idea of what is required before commenc¬ 
ing. This can best be obtained by making up a 
template. Since drill grinding will be a continu¬ 
ing activity, a basic template of drill point angle 
should be made up in a suitable piece of steel 
sheet - 20 swg or 1mm is about right. A tem¬ 
plate should not be necessary for estimating the 
clearance angle since the actual value of this is 
not all that critical and may be estimated by eye. 

To grind the drill and create the cutting lips, 
a compound movement of the drill against the 
grinding wheel is required. 

The side of the wheel is much the most con¬ 
venient to use, and the movement of the drill 
which is required is shown in Figures 10.12 and 

10.13. Figure 10.12 shows the position as the 
cut starts, and Figure 10.13 the finishing posi¬ 
tion, indicating the change in approach angle 
which the drill must make in order to create the 

Initially, the drill is held at 31 degrees to the 
line of the grinding wheel, with the cutting lip 
horizontal. To grind the flank of the point, the 
drill must be rotated and, at the same time, the 
shank lowered, in order to create the clearance 
angle behind the cutting lip. 

The drill is held loosely between the thumb 
and middle and index fingers of the left hand 
with the thumb located in one flute, and the 
hand supported on the rest on the grinder. The 
drill is controlled by the right hand, being 
pushed forward and rotated, at the same time, 
lowering the right hand to create the clearance 



For drill grinding, very little material nor¬ 
mally needs to be removed and the fine wheel 
of the grinder is consequently satisfactory. 

Grinding and sharpening aids 

Small drills can quite adequately be sharpened 
by hand on a small slip stone, grinding off the 
lips of the drill to a flat profile rather than the 
commercially produced cone shape. To grind 
the cone-shaped facets on a large drill on a 
grinding wheel, a compound action of the 
hands is required, as described above. This 
action can be mimicked by a relatively simple 
attachment for an ordinary grinder, and these 
attachments are available commercially or as 
parts and drawings for home manufacture. 

The principle of operation of a drill grinding 
jig (the usual name) is illustrated in Figure 
10.14. The drill is placed in a carrier or holder 
which has a pivot rod which can rotate in a 
bearing in the top of a stand. The carrier can be 
swung to pass the end of the drill across the side 
face of the grinding wheel. The carrier is angled 
so that the drill lies at 30 degrees below the 
horizontal and the face of the lip presented to 

Figure 10.14 Arrangement of a drill grinding attachment for 
the bench grinder. 

the wheel is ground at the required angle. The 
feed of the drill towards the wheel is controlled 
by a feedscrew attached to the carrier, against 
which the end of the drill rests. 

To generate the 12 to 15 degrees relief angle 
behind the cutting edge, the bearing for the 
carrier is not vertical but is inclined towards the 
wheel. The bearing has a short, spiral slot 
machined in it, and a cross pin in the pivot shaft 
engages with this slot so that the carrier is lifted 
out of the stand as it rotates. This means that 
the carrier and the drill advance towards the 
wheel as they rotate, grinding progressively 
more material off the point and automatically 
backing off the cutting edge. 

To ensure that the cutting edge is correctly 
ground, there needs to be a stop at the end of 
the carrier, against which the flute of the drill 
bears, so that the cutting edge is vertical at the 
start of the cut, otherwise the backing-off cone 
is not correctly aligned on the end of the drill. 

The attachment ensures that the two princi¬ 
pal angles on the drill point are automatically 
maintained during grinding and the operator 
need only adjust the cut on the two lips to 
ensure that they are of equal length. 

Drilling speeds 

Recommended cutting speeds for drilling are 
substantially the same as those used when turn¬ 
ing and drill speeds may therefore be inferred 
from the information given in Chapter 14 
(Tables 14.2 and 14.3). When drilling, the cut¬ 
ting speed is that calculated for the drill diam¬ 
eter and the speeds to use may be read directly 
from Table 14.3. However, although there is 
seldom a need to turn items much smaller than 
'/sin. diameter, there are frequent occasions on 
which very small holes are required and it is 
useful to reproduce Table 14.3 here (as Table 
10.4), confining it to the range of sizes below 



Vim. (12.5mm) and extending it downwards to 
show the speeds required for drills down to the 
smallest normally required, the No. 80, or 

Notice particularly from Table 10.4 the very 
high rotational speeds which are required in or¬ 
der to keep small drills cutting at the correct 
speed. In practice these very high speeds are sel¬ 
dom available, the limit for a normal bench 
drill usually being of the order of 3500 rpm. It 
can, however, be useful to adopt higher speeds 
on some occasions and if very much drilling is 
likely to be undertaken in the smaller sizes, the 
construction or acquisition of what is usually 
called a sensitive drilling machine might be 
contemplated. This type of drill is specifically 
designed for use with small drills and it pro¬ 
vides a very sensitive feed arrangement and uti¬ 
lises only a small chuck in conjunction with 
high rotational speed. 

Table 10.4 Drilling Speeds, Diameter and rpm. 


(in.) (mm) 


Cutting Speed ft/min 
100 85 70 


No. 80 







No. 70 










































The following cutting speeds should be used: 
brass, nickel silver, aluminium alloys, 300 ft/min 
BMS, copper, gunmetal, phosphor bronze, 85 to 100 ft/min 
cast iron, 70 ft/min; stainless steel, silver steel, 60 ft/min. 

Starting the drill 

Centre marking 

An ordinary jobber’s drill cannot be expected 
to make its own centre, and for two reasons: 
first, it has only a thin web holding it together, 

and it is therefore relatively flexible, and 
second, the rather flat point of standard drills 
(118 degrees included angle) makes the drill 
poor in starting the drilling operation. 

Short stub drills, with their much shorter 
fluted length than jobbers’ drills, will start a 
hole immediately below their axis, and if the 
need is just for a hole, or if some other means is 
available to determine its position, there is no 
need for the marking-out and centre-punching 

For these reasons, and to ensure positional 
accuracy, the centre position of a drilled hole is 
marked by use of a centre punch. This produces 
a small, shallow, vee-shaped dimple which not 
only allows the operator to see where the hole 
should be, but also permits a small twist drill to 
find the position when drilling commences. 
This initial drilling must, however, be with a 
small-diameter drill. A larger drill, with its 
thicker web and larger point, is difficult to start 
directly in a centre punch dimple. 

A small drill more readily finds the punched 
centre and will even bend significantly to com¬ 
mence drilling in the centre if the operator has 
not located the dimple immediately below the 
drill’s axis. Obviously, any significant bending 
is likely to break the drill if it is excessive or is 
allowed to persist during any reasonable depth 
of penetration into the work and so must be 
corrected immediately by adjusting the posi¬ 
tion of the job in relation to the drill’s axis. 

If a drill does find the centre, but remains 
bent, it is not operating in its designed condi¬ 
tion and cannot be expected to produce an 
accurately positioned hole which is square to 
the face of the workpiece. If the tip of a small 
drill is brought into contact with the work and 
cannot find a centre, it bends and describes a 
wide arc on the work and will break or bend if 
pressure is applied. 

For drilling small holes, it is usually not nec¬ 
essary to clamp the workpiece to the drill table, 
although it is desirable that the job should be 



mounted in a machine vice if its size and shape 
allow. With loose mounting of the work and 
the vice on the drill table, the operator can 
adjust the position of the work to prevent the 
drill bending thereby avoiding the risk of drill 
breakage and ensuring that the hole is correctly 
positioned. The initial drilling operation is 
normally by use of a small drill which can 
be expected to find the small centre punch 
dimple to create a good centre (or initial hole) 
should a larger drill eventually need to be put 

Sometimes, really accurate location of drilled 
holes is required. In these instances, it may be 
that two or more holes are required in some 
accurately defined relationship, in which case 
consideration should be given to co-ordinate 
marking out or a centre drilling operation which 
might perhaps use the lathe’s feedscrews and 
dials as the means of obtaining the required 
accuracy. If position is required accurate to what 
may be discerned by eye, it is usual to use a small 
drill simply to enlarge the centre initially, with¬ 
out attempting to drill to any depth. The posi¬ 
tion of this enlarged centre may then be viewed 
with a watchmaker’s glass (loupe) or magnifier 
relative to the marking-out lines to judge its 
positional accuracy. Figure 10.15 shows a small 
brass plate in which the centres of holes near the 
corners have been enlarged using a small drill. 

Figure 10.1S Centre punch dimples enlarged using a small 
drill as a preliminary to drilling at the required size. 

Correcting a misplaced centre 

If an enlarged centre does not correspond with 
the desired (marked) position, it must be 
‘drawn over’ in the required direction. Various 
dodges can be employed for this operation, and 
the method to use depends upon the magnitude 
of the error and the relative size of the finished 
hole in relation to the existing enlarged centre. 
Workpiece thickness also affects the choice of 

Assuming that the error is small, the simplest 
way is to set up the workpiece at an angle below 
the drill so that further drilling causes the 
centre to move in the desired direction. Once 
sufficiently deep, the central position can be 
expected to have moved by a discernible 
amount and its position may then be examined 
as before to determine whether the amount of 
movement has brought the centre into line. It is 
then necessary to put in a larger, stiffer drill 
with the work supported directly to the drill 
table to establish a centre or hole which is 
square to the base of the work. 

If the initial centre enlargement has revealed 
a gross error in position, it is probably wise to 
re-establish the correct position by drilling 
through from the other side (if a through hole is 
required) since this produces a good hole. To 
assist the location of the hole, it is generally 
best to drill through any associated holes, if this 
is possible, to provide a reference for the hole’s 

If drilling from the reverse is not possible, or 
the hole is blind, an approach must be made 
from the marked-out side. It is usually best in 
these cases to take out the drilled depression by 
chipping away the adjacent material to produce 
a flat surface on which to start again. A milling 
operation (refer to Chapter 16) may be used to 
machine a small flat, if this is necessary. 

If a misplaced hole position has been estab¬ 
lished in thin material, but the drill used for 
centring is small in relation to the final size, it is 



Scribed lines defining Misplaced pilot hole 

Figure 10.16 If a hole is drilled in the wrong position in thin 
material, it can sometimes be corrected by filing the hole to bring 
its centre to the correct location. 

best to drill through and pull the hole over by 
filing, assuming that a small file is available to 
enter the hole. This is illustrated in Figure 
10.16. The hole is filed out in the required 
direction so that the effective centre is moved 
by the necessary amount. Enlargement is 
required in both directions in order to allow a 
larger drill to centre itself on the required axis 
to drill a pilot hole in the required position. 

As can be inferred from all of the above, it is 
generally beneficial to drill the initial pilot hole 
using a small drill, small, that is, in relation to 
the final diameter. Provided that the initial cen¬ 
tre punch dimple has been correctly positioned 
in relation to the marking-out lines, ordinary 
observation of a small drill as it finds the centre 
allows adjustment of the workpiece position, as 
drilling starts, to achieve a correctly positioned 

The foundation of good positional accuracy 
for holes lies in accurate marking out and 
centre punching of the position. Careful atten¬ 
tion needs to be paid to the centre punching, to 
ensure that a clean dimple is created, exactly at 
the right position. If the dimple has been 
punched several times in order to attempt to 

locate it correctly, it will almost certainly not be 
ideally shaped. 

Using a centre drill 

In some situations, marking out and centre 
punching of hole positions is not desirable on 
the basis that it cannot be performed suffi¬ 
ciently accurately. However, if the work can be 
accurately positioned below the axis of the 
drill, a short, stiff drill may be used to drill its 
own centre, without prior marking out being 
necessary. The suitability of stub drills for this is 
described above, but another type of short drill 
is also suitable, and may be preferred. This is 
the centre drill. Its primary use is for creating 
centres in the ends of work which is to be 
mounted between centres on the lathe. Its use 
for this purpose, and for drilling a centre prior 
to the use of a twist drill in the lathe, is 
described in Chapters 13 and 15. 

The centre drill, being short and stiff, is 
ideally suited to the creation of centres on the 
drilling machine, given some means of estab¬ 
lishing the workpiece in the correct position. 
One such situation is shown in Figure 10.17. A 

Figure 10.17 A simple indexing head bolted to the 
table of the drilling machine so that the work can be 
indexed round to spot the fixing hole positions. 



circular workpiece, still mounted in the lathe 
chuck, is fitted to a simple indexing spindle, the 
mounting base of which is bolted to the drill 

With a centre drill fitted into the drill chuck, 
the work is indexed to the required number of 
positions and the centre drill used to drill the 
centres directly, the work and indexer having 
been set so that the holes are drilled at the cor¬ 
rect diameter. The centred holes are afterwards 
opened up to the required diameter using an 
ordinary twist drill, without difficulty. 

If the lathe can be adapted so that the 
workpiece can be mounted on the cross-slide, a 
similar arrangement is possible for the centring 
of holes having linear rather than angular 
displacements, the cross-slide feedscrew being 
used to provide accurate positioning. Should a 
vertical slide be available, calibrated movement 
in two dimensions is possible. Similar methods 
are, of course, used on vertical milling 
machines and mill/drills. 


Clearing the drill of swarf 

Once the drill has commenced drilling, steady, 
but not excessive, pressure must be applied to 
keep the drill cutting. When drilling mild steel 
or aluminium alloys, and also some bronzes, 
the cut material comes away in long curls 
which rotate with the drill and represent a haz¬ 
ard. Pressure on the drill needs to be released 
from time to time to cease cutting temporarily 
and cause the curls to come away in short 
lengths, otherwise they may become caught up 
on the operator’s hands or around the drill and 

When drilling brass, cast iron or gunmetal, 
the cut material is generally created in the form 
of chips. In a deep hole (deep in relation to the 

diameter of the drill) the chips progress only 
slowly up the flutes of the drill, under pressure 
from the newly formed chips at the cutting 
edges, and the flutes can become blocked due 
to chips not being ejected sufficiently quickly. 
The drill must be withdrawn completely from 
the hole quite frequently to allow material to 
fall away and clear the flutes. Without this, 
small drills are likely to break due to build up of 
material in the flutes. 

Breaking through 

If a through hole is being drilled, the point 
at which the drill breaks through must be 
approached with some caution. At the instant 
of breaking through, the hole is not quite com¬ 
plete. The drill point has penetrated the work, 
but on the lower surface the hole has not yet 
reached the full diameter of the drill. 

The flutes of the drill may be regarded as a 
very quick screw thread. Given a chance, the 
drill is inclined to screw itself into the work and 
will do this if, for example, one flute becomes 
jammed against some obstruction in, or on the 
edge of, the hole. 

As the drill breaks through, some material is 
usually pushed out of the hole ahead of it, 
assuming that there is space below the 
workpiece. As the point breaks through pro¬ 
gressively, this ‘pushed ahead’ material forms a 
ring around the end of the hole, as shown in 
Figure 10.18. If the drill is finally forced 
through this ring of material so that it cuts 
through in less than one half turn, the ring of 
material is not removed completely, leaving the 
material in the condition shown in Figure 

The drill point has been forced through the 
hole and the cutting edges cannot remove 
material since they are outside the hole. The 
edges of the flutes, which are not designed for 
cutting, are now caught on the unremoved 



Figure 10.18 A ring of material is pushed out of the hole 
ahead of the drill. Its size is dependent on the material, the 
sharpness of the drill, and the pressure which is applied. Such a 
large burr as this should not ordinarily be produced. 


Figure 10.19 As the drill breaks through, it can get caught up 
in the ring of material which it is pushing ahead of it. If the 
material is tough, and the drill is forced through under great 
pressure, the drill can become jammed in the work, with nasty 

material and the drill screws itself into the hole 
very rapidly. 

When drilling under power in the drilling 
machine, three possibilities exist: either the job 
rotates with the drill, if not held down suffi¬ 
ciently well, or the job remains stationary and 
the drill screws itself rapidly into the hole, or 
the drill stalls in the hole and the chuck keeps 
turning, spoiling the shank of the drill as the 
hard jaws rotate against it. 

Backing up the work 

All of the above situations are hazardous and 
are to be avoided. Preventing breaking through 
is perhaps the most sensible precaution to take 
since it avoids the problem entirely. This can be 
achieved by backing up the workpiece i.e. by 
clamping the work firmly to a piece of scrap 
material, preferably of the same type as that 
being drilled. The backing piece needs to be 
reasonably thick, otherwise the drill may break 
through it, with the same consequences. 

The use of any sort of backing is generally 
beneficial. Offcuts of soft materials are fre¬ 
quently used for this purpose, aluminium alloys 
being good in this respect, but any reasonably 
parallel block can be used. Composite materials 
such as Tufnol are useful but a dense chipboard 
will be found satisfactory and is readily avail¬ 
able, as scrap, in small pieces. Figure 10.20 
shows a block of chipboard in use for backing- 
up a small brass plate which is being drilled 
near its corners. 

Use of backing also removes the need to 
ensure that the drill will pass correctly through 
the clearance hole in the drill table. This allows 
the table to be positioned to support the work 
to best advantage, which is beneficial, for 

Figure 10.20 To assist the drill to break through the work 
cleanly, it is helpful to position some backing below the work. An 
offcut of flooring grade chipboard has been pressed into service 



example, when drilling on the edge or end of a 
large workpiece. 

Size and finish 

Having established a suitably sized centre, the 
actual operation of drilling a hole by use of a 
twist drill is straightforward, but there are 
several points which need to be borne in mind. 
First of all, a drill must not be expected to drill 
a hole exactly at its nominal size. Even if the 
drill is perfectly ground, it will cut holes of dif¬ 
ferent sizes in different materials. For example, 
a sharp Van. (12.5mm) drill might cut a hole 
.004in. to .006in. (0.1mm to 0.15mm) larger 
than its nominal diameter in steel but within 
.OOlin. or .002in. (0.025mm or 0.05mm) of its 
own size in copper. A newly sharpened drill 
also tends to cut bigger than a drill which has 
been dulled at the corners. 

The second point to bear in mind is that a 
drill does not necessarily provide a good 
surface finish inside the hole, particularly if the 
drill is not newly sharpened. The rough finish 
results from the fact that the drill is designed to 
cut on its end and material is torn out of the 
sides of the hole rather than cut in a clean fash¬ 
ion, the effect usually being more obvious if the 
drill is not really sharp, or the material is 
particularly tough. 

If a correctly sized hole having a good finish 
is required, an alternative procedure known as 
reaming is utilised, the cutter (a reamer) being 
designed to cut on its outside diameter in order 
to produce the required surface finish and size 
in the hole. Its action is quite different from a 
drill. Reamers and reaming are described below. 

Progressive opening out 

The technique of starting the drilling operation 
with the work held in the machine vice, but not 

otherwise held down, is described above. This 
allows easy positioning of the work to bring the 
centre ‘pop’ immediately below the axis of the 
drill by observing that the drill does not bend 
when finding the centre. For small holes, say, up 
to '/iin. or s /i«in. (9mm) in diameter, a hand-held 
machine vice will adequately resist the torque 
produced by drilling. For larger drills, the 
torque exerted by the drilling process becomes 
significant and it is necessary to consider clamp¬ 
ing the machine vice, or the work itself, to the 
drill table. Should clamping be considered 
appropriate from the outset, a tapered spindle 
held in the drill chuck provides the means to 
locate the hole centre accurately below the drill 
spindle axis. This is similar to the method of 
centring the four-jaw chuck with reference to 
the tailstock, described in Chapter 13. 

Once the work is clamped into position, 
successively larger drills may be put through 
progressively to bring the hole to its final size, 
the clamping preventing any displacement of 
the hole during drilling. Enlarging the hole 
progressively in this way also helps to avoid an 
oversize hole since the fact that all except the 
first drill are cutting only on their outer edges 
tends to avoid errors caused by faulty grinding 
of the drill point. It also reduces the power 
required to drill the hole and may improve the 
finish inside the hole. Usually, progressively 
doubling the diameter produces good results, 
so a Van., 'Ain., Van. (or 3mm, 6mm, 12mm) 
sequence is satisfactory. 

Large drills are not very well suited to find¬ 
ing a small centre, due to their thicker web 
(larger point). Also, a large drill requires a 
slower speed, which is again not conducive to 
good centring. There are benefits all round in 
the progressive approach to the final size. Hav¬ 
ing the work clamped below the drill axis also 
helps to avoid difficulties when attempting to 
open out a hole with a large drill since it guar¬ 
antees that the drill, if correctly ground, cuts 
equally on both edges. If the work is not 



clamped down, but positioned by eye, one lip 
of a large drill sometimes catches in the pre¬ 
drilled pilot hole and throws the workpiece 
(and machine vice) around, chipping uneven 
lumps out of the work as it does so. 

A similar effect can also be obtained when 
attempting to ream a hole to final size with the 
reamer mounted in the drill chuck, and it is ad¬ 
visable to clamp the work securely during both 
drilling and reaming otherwise the desired 
good finish in the bore may not be achieved. 

If a drilled hole is needed close to a particu¬ 
lar diameter, then that size of drill is naturally 
used to create it. A drill generally produces a 
hole nearest to its nominal size if it removes 
only a small amount of material from an exist¬ 
ing hole, and the enlarging sequence should, 
therefore, allow for the final drill to remove 
only .020in. (0.5mm) of material, and the pre¬ 
vious drill chosen with this in mind. This 
method of sizing a hole is recommended when 
an accurate size is required, but a fine finish to 
the hole is not necessary, for example to create 
a close-fitting hole when riveting. 

Removing burrs 

Most machining operations do not usually take 
place absolutely cleanly, since some material is 
not cut away, but is deflected out of the way. 
When drilling, some material is pushed out¬ 
wards on the entry side and some is also pushed 
out, ahead of the drill, on the exit side. This 
means that burrs form on both sides of the 
work. The size of the burr is a function of sev¬ 
eral factors, but the most important is the 
bluntness of the drill, burrs becoming larger as 
the drill becomes blunter. So, the first essential 
is to keep drills sharp. Nevertheless, burrs do 
form and must be removed, otherwise the fit 
and the appearance both suffer. 

The traditional way to remove burrs is to use 
a slightly larger drill to put a small chamfer on 

the edge of the hole. It is usually sufficient to do 
this by hand, particularly if the drill is small, 
since it can just be twirled in the fingers, and the 
burr is easily removed. For models which are 
miniatures of some real object, you may feel 
that the method is inappropriate, since even a 
small chamfer at Vi 2 scale, represents something 
huge on the prototype. There may also be 
situations where, even on real items such as 
tools, the chamfer formed may appear 
unsightly. In these cases, it is often better to 
remove the burr by rubbing the surface of the 
work with a smooth file. If the burr is small, this 
usually removes it adequately, but it may turn 
the burr over into the hole. In these cases, the 
drill should be put into the hole again, by hand, 
and twirled in the fingers, when it will remove 
the vestiges of the burr, and leave a nice, sharp 

For large holes, it is best to use a scraper, or 
one of the specialised de-burring tools. 

Triangular holes 

A problem which is sometimes encountered, 
particularly when opening out holes in thin 
sheet materials, is that of the ‘not-round’ hole. 
The habit of the cutting edges of a large drill 
catching on the rim of a small pilot hole is 
described above. This principally occurs if the 
drill axis and the hole are not aligned when 
contact first occurs. 

Since the drill is running eccentrically to the 
pilot hole, it tends to throw the workpiece 
around in a circular motion. However, the drill 
may still be cutting the work somehow, and the 
combination of the two motions causes the 
cutting edge (or that part of the edge which is 
cutting) to describe a triangular motion relative 
to the work. The result is a hole with a triangu¬ 
lar form, as shown by the example of Figure 
10.21. Since the problem is caused by the work 
moving about the drill axis, the solution is to 



Figure 10.21 The start of two triangular holes in a steel 
plate. These have been caused by the larger drill not being 
correctly aligned with the pilot hole. 

bolt the work to the drill table (under the drill 
axis) and thus prevent movement. 

The problem is likely to occur also, when¬ 
ever the drill approaches the work without 
proper guidance from its point or its outside 
diameter. It is likely to be worse, therefore, for 
larger holes drilled in thin sheet materials, or if 
a pilot hole is being opened up, as described 
above. A palliative may be to back up the work, 
since this gives the drill point something to cut, 
and as it forms, the dimple guides the drill 
point and provides a steadying action. 

The problem is heightened when drilling 
softer materials such as aluminium alloys, since 
the drill is then able to cut even though it is not 
in the ideal situation. Thin alloy sheets are the 
worst offenders in these respects. 

Drilling and punching thin sheets 


Small holes in sheet materials are usually 
among the most straightforward drilling opera¬ 

tions. The holes are frequently required only 
for fitting attaching parts (rivets or screws) and 
position is often not critical since what matters 
is that hole positions should match those in the 
mating item, through, or into which the 
fastenings pass. There are naturally situations 
in which hole positions are important, in clock 
plates for example, but whenever the hole is 
small in relation to the material thickness, there 
is usually no difficulty in actually producing 
(drilling) the hole. 

As drill size increases, the depth of the drill 
point, with its standard, 118-degree included 
angle, naturally also increases. If the material is 
thin, this means that the drill point breaks 
through the work before the drill is cutting at 
its full diameter. This leads to the situation in 
which the cutting edges of the drill are liable to 
catch in the edges of the hole, in the same 
manner that this can occur when opening out 
an existing hole with a larger drill, as described 

Under these circumstances, there is little 
guidance for the drill in the material, since the 
point has broken through, and the drill is not 
yet cutting at its full diameter. In any event, if 
the material is thin, there is little material to act 
as a guide, and it becomes very difficult to 
achieve positional accuracy. 

There is also another problem. If the mate¬ 
rial is thin, it has little strength, and when the 
drill starts to cut, material tends to ‘climb’ up 
the drill rather than the drill cutting into the 
material and the result is often disaster. 

One way to mitigate these effects is to clamp 
the work securely between two pieces of scrap, 
preferably of the same material as the work- 
piece. All three items must be drilled at the 
same time, and their combined thickness must 
suit the diameter of the drill being used. The 
top packing (at least) must be sufficiently thick 
and strong to withstand any tendency for the 
thin workpiece to climb up the drill. 



Modifying the drill point 

Another way to approach the problem is to 
re-grind the end of the drill so that the point is 
flatter. There is a limit to this, however, since a 
flat-ended drill naturally has no point to pro¬ 
vide guidance, and as the included angle of the 
point increases the force needed to push the 
drill through the work also increases. A com¬ 
promise can be adopted by using an included 
angle of 145 degrees or so, 72 degrees each side 
of the drill’s axis as frequently recommended in 
some workshop practice books. 

Using this drill point angle means that a 
s Aim. drill has a point ‘depth’ of .025in. and a 
Vitin. drill, a depth of.OlOin. so, if thin sheets 
do need to be drilled, the hole must be quite 
small for the drill to be cutting at its full diam¬ 
eter before breaking through. In metric terms, 
the above figures equate with a 4mm diameter 
drill, and 0.64mm thick material, or a 1.6mm 
diameter drill, and 0.25mm material, all of 
these figures relating to an included angle at the 
drill point of 145 degrees. 

Pre-drilled to full depth 
of drill point 

Wooden block damped 
to drill table 

Thin sheet drilled with pilot 
hole and placed on wooden block 

Pressure from drill 

Figure 10.22 One method of drilling a thin sheet. 

Deforming the workpiece 

A further way to cope with thin materials is to 
make the material thicker (effectively) by 
deforming it into a more suitable shape. This 
can be done by placing a block of hard wood 
below the drill point, either bolted to the drill 
table, or held in a machine vice which is itself 
bolted down. The drill can then be used to drill 
a shallow depression in the block, just deep 
enough to cut to the drill’s full diameter. 

The situation when drilling the thin sheet 
material is shown in Figure 10.22. The work is 
pre-drilled with a small pilot hole and this is 
placed immediately below the drill point. The 
rotating drill is brought down firmly onto the 
work, guided by the pilot hole and the pressure 
forces the thin metal into the depression before 

the drill starts to cut. The deformed material, 
which has the same diameter as the drill, acts as 
a guide during the drilling operation, and the 
drill always cuts at its full diameter. 

Punching holes 

If very thin sheets need to be pierced, it may be 
best to abandon the idea of drilling and instead 
use a punch. These are available commercially 
but they can be made up fairly easily. A case- 
hardened mild steel block can be used in asso¬ 
ciation with a punch made from silver steel 
which has been hardened and then tempered to 
dark purple. If only a few holes are needed, it is 
not essential to harden the mild steel block 
which acts as the die. 



Drilling brass and gunmetal 

Reference to Chapter 14 will show that the cut¬ 
ting point of a lathe tool is created by grinding 
clearance angles on the tool. These are known 
as top rake, side rake and front clearance. For 
turning brass or gunmetal, top rake is normally 
ground at zero degrees, or may even be nega¬ 
tive, the tool being ground downwards towards 
the point. 

For a conventionally sharpened drill, the 
equivalent of top rake is determined by the 
helix angle (spiral) of the drill, as shown in 
Figure 10.23. Drills are available having differ¬ 
ent helix angles, normally described as having a 
‘standard’, ‘quick’ or ‘slow’ helix, but unless 
the flute is straight, there must always be a posi¬ 
tive rake at the cutting edge. 

Front of cutting edge / , - 

stoned away x S', 

Top rake 

Helix angle, Effectively = top rake 

Figure 10.23 A close-up view of the cutting edge on a drill, 
showing how a small flat can be put on with a slip stone before 
using the drill in gunmetal or cast brass. 

The tendency for the drill to screw itself into 
the work when breaking through is described 
above. A cutting tool with positive rake shows a 
great propensity for digging in when cutting 
brass or gunmetal. That is, there is a tendency 
for the cutting edge to get under the surface and 
force itself into the work. On the lathe, this is 
not usually a problem since the tool feeds are 
ordinarily through screws which can resist the 
pulling action which the work has on the tool. 

When using the drilling machine, the situa¬ 
tion is quite different. The down feed is via a 
rack, which can just as easily be pulled from 
below as pressed down by the handle. In addi¬ 
tion, considerable force is normally required 
when drilling and there is consequently consid¬ 
erable pressure on the feed handle. Should the 
drill dig in, and be pulled into the cut, it will 
‘walk away’ rapidly, screwing itself into the 
workpiece and most probably out of the other 
side, totally out of control. During this process, 
the drill is ripping material out of the hole, 
screwing itself in at a rate determined by the 
helix angle, virtually cutting itself a thread in 
the manner of a tap. The finish in the hole is 
therefore not good, size is not likely to be cor¬ 
rect and if you wanted a blind hole, it may now 
be too late. Positional accuracy may also suffer 
and the work is possibly spoiled. 

There are, of course, dangers posed by the 
dig in. The drill is caught in the work, which is 
trying to rotate with the drill and considerable 
torque is necessary to resist this. If the holding 
force is not sufficient, the job rotates and the 
fingers, or whatever is in the way, may be 
damaged. The work itself, or even the drilling 
machine may also be damaged. 

Since prevention is better than cure, the 
remedy is to modify the cutting edges of the 
drill so that they have zero rake. At one time, 
straight flute drills were produced especially 
for use in brass, but these are no longer avail¬ 
able. Fortunately, however, the drill’s cutting 
edge may easily be modified to produce the 
desired rake. 



The enlarged view of the drill’s cutting edge 
shown in Figure 10.23 shows that the edge has 
a top rake equal to the helix angle. If the edge is 
modified, by removing the shaded portion, the 
top rake is reduced to zero degrees. To achieve 
this change, only a small amount of material 
needs to be removed, creating a flat just a few 
thou wide. This can be achieved quite simply by 
rubbing a fine slip stone down the cutting edges 
a few times, parallel with the axis of the drill. 
Once modified in this way, the drill will cut 
normally in brass and gunmetal and will exhibit 
none of the grabbing and digging in which will 
otherwise occur. 

After modification of the edge, the drill needs 
re-sharpening before use in other materials. 

All drills for use in brass do not need to be 
modified. When drilling ordinary through 
holes in thin sheets, drills up to 3 /i«in. (5 mm) or 
so in diameter are quite satisfactory if conven¬ 
tionally ground, especially when drilling half- 
hard sheet (the normal temper). However, 
brass or gunmetal castings are a different 
proposition, due to their different composi¬ 
tion, and great care must be exercised when 
drilling even small holes in these materials if the 
drill point has not been modified. 

Cross drilling 

The cross drilling of circular rods is a frequent 
requirement. It can be carried out by eye but 
this doesn’t always produce good results and 
some form of drilling jig is frequently 
employed. However, if the need is only an 
occasional one, it may not be worthwhile mak¬ 
ing a jig and the ‘by eye’ approach may serve. 

Assuming that the item must be machined 
flat on the end, it must at some time be in posi¬ 
tion in the lathe. The tool point is set at centre 
height for facing the end and once the turning 
operation is complete it may be used to scribe a 

line across the diameter on the end. Without 
rotating the chuck, the tool can also be used to 
scribe a line along the rod exactly correspond¬ 
ing with the marked diameter thereby provid¬ 
ing a reference for the hole position. 

Sometimes, a turning tool does not produce 
a very fine line when used for scribing the refer¬ 
ence lines (it depends on the tool and material, 
which way the tool moves against the station¬ 
ary work and how much pressure is applied). It 
may be preferable to position a rectangular bar 
in the lathe toolpost, adjusting its height with 
reference to the centre of rotation which is 
visible on the turned end. Once set, the bar may 
be used to scribe the diameter and longitudinal 

Marking out for the hole position along the 
length allows a centre punch dimple to be put 
in and the work can then be mounted in the 
machine vice ready for drilling. The work 
needs to be positioned so that the diametral line 
across the end stands vertical with respect to 
the drill table. It might be possible to check this 
with a square standing on the table, but if the 
workpiece is short, or the base of the machine 
vice has integral lugs, like the one illustrated in 
Figure 10.7, this may not be possible. 

A machine vice without such lugs is a useful 
alternative for small work, since it allows a 

Figure 10.24 A circular workpiece with a centre line scribed 
on its end, can be aligned for cross drilling using a square. 



Figure 10.25 A small drill might be used for aligning the work 
for cross drilling. 

closer approach to the work, as illustrated in 
Figure 10.24. 

If a square cannot be used to align the work, 
it is best to align the rod’s diameter with the 
drill axis by placing a small drill in the chuck 
and adjusting the rod’s position in the vice so 
that the point moves parallel to the scribed line 
as the chuck is lowered. This is shown in Figure 

10.25. You may find it is easier to do this using 
the point of a pocket scriber, since the finer 
point makes proper alignment easier to detect. 

Figure 10.26 A pair of vee blocks is the preferred method for 
holding round work. 

An alternative method of holding small 
work is by use of a pair of vee blocks. These are 
rectangular blocks which are machined with 
one or more vee-shaped grooves in which a cir¬ 
cular bar can be rested. Some types have rectan¬ 
gular grooves machined along both sides and 
are supplied with a stirrup clamp which can be 
used to hold the work, as shown in Figure 

10.26. A long workpiece can be clamped to one 
block and be supported by the other, providing 
a rigid arrangement and allowing the drill and 
chuck to approach the work. 

A vee block is again not helpful if the work is 
short, and another method must be sought. A 
simple alternative to a full-blown cross-drilling 
jig can be created if a bush is made up having 
the same outside diameter as the rod to be 
drilled and itself having a true axial hole down 
its centre. Workpiece and bush can then be set 
up in the machine vice, as shown in Figure 

10.27, and the bush used to guide a small pilot 

Should a jig be thought desirable, it can best 
be arranged as a vee block incorporating a 
workpiece clamping arrangement but also hav¬ 
ing a bridge piece which can accept a guide bush 
centrally over the vee. Something specially 
made is therefore required, provided with inter- 

Figure 10.27 Cross drilling can be simplified by using a drilled 
bush as the guide, but bush and workpiece need to be exactly 
the same diameter and the machine vice must have parallel jaws 
so that it grips both the bush and the work equally. 



changeable bushes to act as guides for differ¬ 
ently sized drills. Numerous designs have been 
published over the years, and making one of 
these can be an interesting and useful exercise. 

Reamers and reaming 

A twist drill, even if correctly ground, is not 
capable of drilling a hole to an exact dimension 
and will, in general, cut larger than its nominal 
size. One way of testing whether the drill is 
cutting near its nominal size is to drill a hole in 
some piece of scrap material and then test the 
fit of the shank in the drilled hole. This should 
reveal any gross oversize in the drilled hole and 
allow any fault to be rectified. This test, 
although widely practised, should be regarded 
only as an informal check, since the shank of a 
drill is always finished to a smaller diameter 
than the drill itself. 

The method of drilling a hole slightly 
smaller than that required, and following up 
with a drill of the required size, to produce an 
accurate-sized drilled hole, is described above. 
However, since a drill is designed to cut on its 
end, the finish imparted to the bore is, in gen¬ 
eral, rather rough. A drilled hole is not suitable 
for use as a bearing or whenever an accurately 
sized smooth bore is required, and reaming is 

Figure 10.28 A group of reamers, up to Yan. ( 16 mm) in 

employed whenever a good finish and/or a 
specified size are required. 

Figure 10.28 shows a group of reamers up to 
s /8in. (16mm) in diameter. The cutting edges of 
a reamer are formed on the outside diameter by 
cutting a number of flutes along the length. 
These usually progress in a slow spiral from tip 
to shank, but some reamers have straight flutes. 
There are from four to ten flutes cut on the 
reamer, depending on its size. 

The reamers illustrated in Figure 10.28 are 
designed to produce parallel-sided holes and 
are passed completely through a pre-drilled 
hole to clean up its internal surface. Since par¬ 
allel-sided holes are produced, these cutters are 
known as parallel reamers. 

Although described as a parallel reamer, a 
small amount of taper is generally incorporated 
in order that it may enter a drilled hole slightly 
smaller than the nominal diameter. Since the 
reamer cannot readily be sharpened while still 
remaining at the same size, it is advantageous if 
it removes only small amounts of material, and 
the initial drilled hole is, therefore, only slightly 
smaller than the final hole size. 

There are essentially two types of parallel 
reamer - hand reamers and machine reamers. 
Either type can have a parallel shank, with 
squared end, so that a tap wrench can be used 
to turn and drive the reamer, or may have a 
tapered shank, making it suitable for holding in 
a machine tool. 

The essential difference between hand and 
machine reamers is that hand reamers are 
significantly tapered in their length. A hand 
reamer is used by drilling a hole slightly smaller 
than the finished size, but large enough for the 
small end of the reamer to enter. The reamer is 
then fitted with a tap wrench, entered into the 
drilled hole and turned by hand so that the 
drilled hole is enlarged to the size cut by the 

Machine reamers have very little taper, but 
are sharpened to cut on the end, in addition to 



the sides, and can therefore be forced into a 
hole which is smaller than the reamer’s end 
diameter. Accurate reaming in this manner 
requires that the reamer is accurately in line 
with the pre-drilled hole. This means that the 
work must be clamped below the drilling 
machine spindle (or in another machine tool) 
so that drill and reamer follow the same path 
successively, and there is no tendency to force 
the reamer to one side and cause uneven cutting 
in the drilled hole. 

There is quite a wide variation in the amount 
of taper which is ground on a reamer. Among 
my stock of reamers, extending from s /gin. 
(16mm) down to '/i«in. (1.6mm), some 20 
reamers in all, five have tapers between .015in. 
and .022in. (0.4mm to 0.6mm), five have no 
taper and the remainder have tapers between 
.003in. (0.1mm) and .007 in. (0.2mm). The 
tapers on the hand reamers are remarkably 
consistent throughout the range of sizes, for 
example the s /sin. being tapered .017in., the 
3 /«in. by .022in. and the s /hin. by .015in. 

All of these reamers have been bought just as 
reamers, from various sources over many years, 
and are perhaps not typical of current produc¬ 
tion. By way of contrast, a new 12mm ( is /j 2 in.) 
reamer from a well-known British manu¬ 
facturer, identified on its protective box as a 
hand reamer, has a tip diameter of 11.5mm, 
making the taper the equivalent of .020in. The 
maximum diameter of this reamer is 12.01mm 
and the plain, squared shank is ground to 

When using a reamer, the initial drilling 
prior to reaming should be arranged so that the 
reamer removes only a small amount of mate¬ 
rial. If the reamer is tapered, the normal prac¬ 
tice is to measure its tip and select a drill which 
just allows the reamer to enter the hole. The 
reamer is then put into the drilled hole and 
turned under pressure to enlarge the hole pro¬ 

gressively to the reamer’s size. 

The cutting edge of the reamer flute is 
ground with a parallel portion, which may be 
from .005in. to .020in. (.13mm to .50mm) 
wide, immediately behind the cutting edge. 
There is considerable rubbing action during the 
cutting process and, as a consequence, reamers 
are inclined to chatter. The reamer must be pre¬ 
vented from merely rubbing and must, there¬ 
fore, be pressed steadily into the work. 

There are real benefits in clamping the work 
rigidly to the drill table and performing both 
the initial drilling and final reaming without 
disturbing the clamps. The reamer must then be 
in line with the drilled hole and square to the 
work, and steady pressure can be exerted from 
the feed handle. One possible problem is that 
the flutes of the reamer, which can be very 
shallow on small hand reamers, can become 
clogged with swarf. This causes the reamer to 
jam in the hole, always with unfortunate conse¬ 
quences. This is one reason for drilling the ini¬ 
tial hole close to the final size, so that there is 
little material to remove. 

Ideally, the reamer should pass once into, 
and once out of, the hole, so it is unfortunate if 
it has to be withdrawn partway through, just to 
clear the flutes. 

When reaming under power in the drill (or 
lathe) speed should be reduced to roughly a 
quarter of that used for drilling at the same size. 

Reamers should never be turned backwards 
as this tends to turn the cutting edge over and 
rapidly blunts the tool. 

Since new reamers are ground with a slightly 
oversize maximum diameter, they generally cut 
slightly oversize when new. If reaming is being 
carried out to create the bore of a bearing, this 
is usually of no consequence, but if holes are 
being reamed to accept items for press or inter¬ 
ference fits, this must be borne in mind when 
making the corresponding parts. 



Counterboring, countersinking and 
spot facing 


Associated with drilling there are several 
frequently performed operations. These are the 
counterboring and countersinking of the 
mouth of a drilled hole and the local facing of 
the workpiece around a drilled hole, referred to 
as spot facing. This operation may precede the 
drilling of the hole but can also often be per¬ 
formed afterwards. 

These operations may be performed using 
commercially available cutters but are fre¬ 
quently undertaken using home-made tools 
since commercial products tend to be rather ex¬ 
pensive and the process of manufacture is easily 
carried out using high-carbon steel (silver steel). 


Countersinking is perhaps the most common of 
the three operations. This is often carried out 
using a rose countersink, as shown in Figure 
10.29. This is a specially made cutter designed 
for creating the 90-degree cone in the mouth of 
an existing hole to accept a countersunk-head 
screw. For the modeller, the rose bit has some 
in-built disadvantages, the principal of which is 
the large number of cutting edges which is nor¬ 
mally provided. 

The depth of the flutes reduces towards the 
point and in this region there is little space in 
the flutes to accommodate the cut material. 
When used in small holes, the flutes quickly be¬ 
come blocked and the bit ceases cutting since 
the edges are masked by the jammed-in mate¬ 
rial. Frequent clearing of the flutes is therefore 
necessary. The problem is emphasised by the 
presence of an 8 BA screw in Figure 10.29. This 
thread is .086in. (2.2mm) in diameter, and is 
certainly not small by modelling standards. 

Figure 10.29 A rose countersink compared with an 8BA 

Figure 10.30 A pair of 2-flute countersinks with pilots. 



A more useful form of cutter, particularly 
for small holes, has fewer flutes and a two-flute 
cutter, not unlike a centre drill, but having a 90- 
degree cone, is a much more useful form of the 
countersink. Figure 10.30 shows a pair of com¬ 
mercial cutters of this sort, both of which are 
provided with a plain pilot portion, to steady 
the cutter in the hole, made to suit a particular 
screw size. Each suits just the one size and is 
described as a ‘2BA countersink’ or whatever. 
The use of only two (or sometimes four) 
cutting edges normally prevents build-up of cut 
material on the edges, except when cutting 
aluminium alloys, which are almost always a 
problem in this respect. 

The use of a pilot on the countersink tends 
to reduce chatter since it acts to steady the cut¬ 
ter’s position at the mouth of the hole and gives 
a smoother finish to the cut. The ordinary rose 
bit has a great tendency to chatter if firm pres¬ 
sure is not maintained, and the rough finish can 
be difficult to eradicate once it has occurred, 
although a change of speed can be beneficial in 
this respect since it changes the rate at which 
the cutting edges contact the chatter marks. 

A useful form of countersink cutter can be 
made as shown in Figure 10.31. This is made by 

turning the end of a piece of silver steel to a 90- 
degree cone and then filing or machining away 
a short length of the coned end to half the 
diameter of the rod, less .002in. or .003in. 
(.05mm to .07mm). Hardening and tempering 
to light straw produces a tool which is chatter- 
free when cutting, but does need firm pressure. 
If, instead of having a plain shank, the cutter is 
given a knurled handle, it can be turned by 
hand and used for de-burring holes. 


Counterboring is normally carried out for the 
same reason that countersinking is employed; 
to recess the heads of screws or bolts below the 
surface of the workpiece. The requirement here 
is to produce a flat-bottomed enlargement of 
the mouth of a clearing-size hole, for a particu¬ 
lar size of screw, to create a recessed seating for 
the head. 

A flat-ended cutter having a parallel pilot is 
essential for this operation. Once again, cutters 
of this type are available commercially and nor¬ 
mally take the forms shown in Figure 10.32. A 
flat-ended cutter with the correct size pilot is 

Figure 10.31 A countersink made from silver steel. 



Figure 10.32 Two commercially made counterbores. 

Figure 10.33 A counterbore made in silver steel. 

ground to provide two or four cutting edges and 
can be used as a second-stage cutter to create the 
counterbore at the mouth of an existing hole. 

Such cutters can readily be made in the home 
workshop, using silver steel, which is then hard¬ 
ened and tempered to pale straw, as described in 
Chapter 3. Figure 10.33 shows a close-up of a 
home-made counterbore. The main diameters 
are turned in the lathe, following which the cut¬ 
ter is filed down on both sides to match the 
width of the pilot. To create the cutting edges, 
the resultant flats on each side of the pilot are 
backed off, again by filing, to provide clearance 
behind the cutting edge. This needs to be car¬ 
ried out carefully so that, on both sides, a 
minute portion of the faced shoulder is left as a 
witness. To provide the equivalent of a lathe 
tool’s top rake, a shallow groove is filed across 
the cutting face, using a small round or half- 
round Swiss file. This also needs to be done 
carefully so that the cutting edge is sharp and 
the witness on the face remains. The groove 
cannot be too deep otherwise the cutting edge is 
weakened and the tool’s possible life is reduced. 

Some backing-off of the outside diameter, 
just to provide clearance, may also be benefi¬ 
cial, since if the outside of the cutter remains at 
the full diameter it rubs on the sides of the 
counterbored hole and rapidly heats up, caus¬ 
ing further jamming as it expands. Once again, 
the filing must be done with some care to avoid 
spoiling the cutting edge and leave a witness at 
the full cutter diameter. 

Once filing is completed, the cutting edge 
must be polished (but not rounded or blunted) 
and the tool can then be hardened and tem¬ 
pered as described in Chapter 3. 

Spot facing 

Spot facing is carried out whenever the 
workpiece surface is not truly square to the axis 
of a drilled hole and it is required to fit a bolt or 



screw so that it abuts the surface adjacent to the 
hole. This can happen if the drilled hole is not 
at right angles to the surface or if the surface is 
in the ‘as cast’ condition and has not yet been 
machined. This is likely to be the case where a 
casting is provided with bosses at the positions 
of attaching-bolt holes and needs machining 
only across the tops of these bosses. This might 
be done by a milling operation but it is 
frequently simpler to use a counterbore which 
will provide a local seating for the nut and/or 
washer, the seating being cut following the 
drilling operation so that the tool may incorpo¬ 
rate a pilot to centre the faced area around the 
hole. No great depth of penetration is required, 
it being necessary to clean up the surface to 
provide the seating. A cutter exactly like a 
counterbore is required, but slightly larger, in 
order to accommodate the washer and allow 
clearance for the nut or bolt head. 

Boring larger holes 

Once the required hole size becomes larger than 
the capacity of the drilling machine’s chuck, or 
beyond the capacity of the spindle taper to hold 
a taper-shank drill, the drilling operation 
becomes that of boring. Various types of adjust¬ 
able cutter are used for boring (or cutting) 
larger holes, ranging from variants of the old- 
fashioned, carpenters’ brace-mounted tank 
cutter, through fly cutters to the more precise 
boring heads. A boring head is simply a body 
which provides a mounting for a cutting tool 
holder which can be offset from the axis of rota¬ 
tion under precise control. A home-made 
boring head is shown in Figure 10.34. The tool 
is angled in the end of a cylindrical holder which 
is mounted in a block which can move along a 
slideway formed in the body. The block is 
tapped to engage an adjusting screw which is 
free to turn in the body endplate, but is retained 

by an inserted key. Rotation of the screw allows 
the block and tool holder to be positioned in the 
slide, thus adjusting the radius at which the cut 
is taken. For this boring head, the screw is 3 /sin. 
BSF (20 tpi) providing an increase in diameter 
of 0.1 in. per turn. 

The block can be locked in the slide through 
the action of an Allen grub screw which presses 
a brass pad onto the block. This is visible in Fig¬ 
ure 10.34. 

An alternative form of boring head is shown 
in Figure 10.35. This has a circular body which 
provides rather less variation in diameter than 
the one shown in Figure 10.34, but it has a 40 
tpi adjusting screw and may be set more pre¬ 
cisely than its ‘big brother’. In this instance, the 
cutter takes the form of a forged-and-ground 
boring tool, not unlike the type used on a lathe. 
This provides the means to bore a deep hole, or 

Figure 10.34 A large adjustable boring head fitted to a No. 2 
Morse taper arbor. 

Figure 10.35 A small adjustable boring head on a No. I 
Morse taper arbor. 



a hole deeply recessed inside the workpiece. 
There is obviously a limit to how long the cut¬ 
ter might be, since additional length decreases 
the rigidity, but some form of extension is fre¬ 
quently necessary. However, as with all 
machining operations, overhang should always 
be reduced as much as possible. 

The boring heads illustrated are mounted on 
Morse taper arbors, and are suitable for mount¬ 
ing directly into the spindle socket of the drill¬ 
ing machine, if the spindle can accommodate it, 
but this needs to have a sufficiently low speed 
to enable larger holes to be cut. Cutting a 4-in. 
(100mm) diameter hole requires a spindle 
speed of 80 rpm in mild steel (85 ft/min (26m ] 
min) cutting speed). 

A boring head mounted on a Morse taper 
arbor may also be mounted in the mandrel 
taper in the lathe and can be used for boring 
items which are secured to the cross-slide. 
However, the drilling machine spindle taper 
ordinarily incorporates a recess to engage the 
driving tang of a standard taper but this is not 
usually provided in the lathe mandrel. An alter¬ 
native type of Morse taper arbor is more useful 
if use in the lathe is envisaged. This has a tapped 
hole at the small end so that a threaded drawbar 
may engage the arbor to pull and hold it into 
the mandrel taper. Without such assistance, the 
taper cannot be expected to hold except for 
very light cuts, and even these may be too much 
at larger diameters. An arbor of this type is fit¬ 
ted to the boring head of Figure 10.34 and its 
threaded end is shown in Figure 10.36. 

Various other forms of tool with an adjust¬ 
ment of the cutting diameter are available, less 
sophisticated (and therefore less expensive) 
than a boring head. They can easily be made up 
in the home workshop. Figure 10.37 shows one 
which I made up, and Figure 10.38 provides a 
dimensioned sketch. 

The cutter (a Vsin. diameter, high-speed tool 
bit) is carried in a l 3 /&in. diameter carrier, Vuin. 
thick, the cutter bore located 3 /sin. off the axis 

Figure 10.36 Tapped hole for a drawbar in the end of a No. 
2 Morse taper arbor. 

Figure 10.37 An adjustable hole and washer cutter. 

of the carrier. A 4BA grub screw holds the cut¬ 
ter in place and the carrier is split and recessed 
so that two 4BA Allen screws may be fitted to 
allow the carrier to be clamped to an arbor. 

The arbor itself has a 3 /sin. diameter shaft, 
V/mn. long, and a two-step location for the 
carrier which is offset from the shaft by '/sin. A 
'/sin. diameter hole for a pilot is drilled on the 
axis of the shaft. The 5 /sin. diameter flange is 
made 'Am. long leaving the V4in. diameter to be 
s /isin. long. 

The arbor should be made first and the 
carrier bored to be a nice fit for both diameters, 
afterwards being split and tapped for the 
clamps and reamed for the cutter bit. 




Tap 2 holes 4BA 
and drill this side 4BA clear 

Figure 10.38 A drawing for the hole and washer cutter. 

The dimensions suggested here provide a 
cutter which will cut holes from s /8in. to 1 '/sin. 
diameter, but if you need different sizes, then 
change the dimensions to suit. 

The original cutter was made to cut holes of 
Va n. and lin. diameter in '/sin. thick steel. To 
cut a lin. diameter hole, a speed of about 325 
rpm is required in mild steel, and the drilling 
machine must be capable of rotating the cutter 
at a sufficiently slow speed, as shown in Table 
6.3. Do bear this in mind before building a cut¬ 
ter for 4-inch holes! 

Although the hole in the shaft for the pilot is 
only '/sin. diameter, it is better to use something 
larger in practice, so a Van. pilot with one end 
turned down can be secured with Loctite. There 
is quite a side thrust on the pilot due to the out- 
of-balance mass which will wear if not hard¬ 
ened. A case-hardened mild steel pilot, or one 
made from hardened silver steel, is preferred. 

The cutter is shown in a configuration in 
which it is suitable for cutting holes, but if the 
cutting tool is ground appropriately, it may 
equally well be used for making washers. 



Screw threads 


Everyone must be familiar with the use of nuts 
and bolts for fastening things together, and 
with the basic concept that nut and bolt must be 
made to fit one another. Colloquially, nut and 
bolt are said to be ‘the same size’. However, this 
is not literally true since the nut must be a free 
fit on the bolt and clearances have to be 
allowed for when making both items. Before 
cutting one’s own threads it is important to 
understand the basic concepts behind the 
manufacture of mating nuts and bolts (for want 
of a better phrase). 

When referring to threads, it is usual to refer 
simply to the nominal outside diameter of 
the bolt or screw. In reality, four distinct diam¬ 
eters are involved - the maximum, or major 
diameter, and the minimum, or minor diameter 
- for both the bolt and the nut. These are 
shown in Figure 11.1. The minor diameter is 
also referred to as the core diameter. 

The most common forms of screw thread 
are those having a simple vee form of the type 
illustrated in Figure 11.1. Since it is essential to 
ensure that the major diameter of the nut is 



Figure 11.1 Screw thread terminology. 

larger than that of the bolt, and the minor 
diameter of the bolt is smaller than that of the 
nut, the thread crests and roots are not usually 
formed as sharp vees but are flattened or 
radiused to create the necessary clearances. 

Figure 11.2 shows a nut and bolt mated 
together. The major diameter of the bolt is 
reduced by flattening the crests of the thread 
and the minor diameter of the nut increased by 
the same means. Nut and bolt therefore mate 



Figure 11.2 The fit between a nut and a bolt. 

correctly and there is no interference between 
the major and minor diameters of the two 
parts. Flattening or radiusing of the crests is the 
normal means by which the clearances are cre¬ 
ated and a vee-form thread does not normally 
show the full shape of the vee at the crest or 
root. A small reduction in the theoretical 
thread height is usually specified. 

The major and minor diameters for nut and 
bolt are defined in a relevant standard for the 
particular thread series. Together, they natu¬ 
rally define the height or depth of thread which 
is a useful measure when threads are to be cut 
using a single-point tool on the lathe. The 
height (depth) of the thread is also used to 
define other characteristics of the shape. 

It goes without saying that nut and bolt must 
have the same pitch of thread (the distance 
between adjacent crests or roots) and the nomi¬ 
nal bolt or screw diameter, together with the 
pitch and shape of the vee form, define the 
basic characteristics of the thread. The full 
height (full vee) of the thread is defined by the 
vee angle and the pitch. Rounding or flattening 
(truncating) at the crests and roots is usually 
specified as a fraction or proportion of the full 
vee-form height. 

Every thread system (Whitworth, Metric, 
British Association etc.) has a form which is 
defined by an appropriate specification. In the 
UK, such specifications are issued by the British 
Standards Institute. Relevant extracts from the 
standards are widely published in engineers’ 
reference books which should be referred to for 

fuller information about thread forms. Full 
details about particular threads are only likely 
to be needed when screwcutting in the lathe, 
but an appreciation of the general characteris¬ 
tics of thread form is also helpful when using 
the more convenient taps and dies for thread 
production. Details of the more common 
thread series are given below, followed by a 
description of the use of taps and dies. 

For modelling activities, the smaller and 
finer pitch threads such as British Association 
and the fine pitch Whitworth form are likely to 
be of most interest, but since machine tools and 
accessories utilise some larger threads, the more 
common of these are also described below. 

Thread series 

Whitworth and BSF threads 

In the early days of the Industrial Revolution, 
no standards existed for screw threads. Such 
‘ironmongery’ as was required was locally 
made, each machine builder using his own 
standards. In 1841, Joseph Whitworth deter¬ 
mined to establish a standard for screw threads 
and decided to measure samples of nuts and 
bolts from various sources. From the results 
obtained, average values for diameter, pitch, 
and nut and bolt head sizes were calculated. 

This, the first attempt at standardisation, 
resulted in specifications being generated for all 
of the major dimensions for a complete thread 
system. This provided a range of nominal diam¬ 
eters, and corresponding pitches, definitions of 
bolt head and nut dimensions, the shape of the 
vee form and the tolerances on major and minor 
diameters to ensure interchangeability of 
threaded items from different makers. 

The pitches chosen by Whitworth, although 
based on average pitches in use in his day, even¬ 
tually came to be considered as rather coarse 




Figure 11.3 Whitworth and BSF thread form. 

and with the improvement in techniques gener¬ 
ally, a finer series of pitches seemed desirable. 
The original Whitworth scheme had been 
formally adopted by the British Standards Insti¬ 
tute and became known as British Standard 
Whitworth (BSW) and the fine pitch series was 
issued as British Standard Fine, or BSF. 

Both BSW and BSF thread series use a vee form 
having a 55-degree included angle, as shown in 
Figure 11.3. The form has radiused crests and 
roots, the radius being related to the pitch of the 
thread (P). Since the vee-form angle is constant, 
the depth of thread is related to the pitch. For 
the Whitworth thread, having a 55-degree vee 
form, and the standard rounding of the crests, 
the depth of thread is numerically equal to 
0.640 X P. In most thread series (although there 
are exceptions) it is normal for the pitch to vary 
with the diameter so that larger diameter screws 
use coarser threads. In the scheme finally devel¬ 
oped from Whitworth’s proposals, nominal 
diameters from'/Urn. to 6in. were standardised, 
having pitches from .0167in. or 60 threads per 
inch (tpi) to 0.4in. (2 Vi tpi). 

As specified, the thread has one standard 
form, with one major and one minor diameter 
specified. For a nut, the major diameter is 
increased by a factor equal to 0.16P and for a 
screw, the minor diameter is decreased by the 
same amount. A nut therefore has radiused 
crests at its minor diameter while a screw is 
radiused at its major diameter. 

As an alternative to the radii shown in Figure 
11.3, the standard allows a flat-topped, or 
truncated, thread form. The amount of trunca¬ 

tion is related to the pitch but is effectively a 
flattening of the crests corresponding with 
removal of the alternative radius to the point 
where it just meets the straight vee form. 

As for all standards, tolerances are applied to 
ensure that there is always interchangeability 
between nuts and bolts, to allow the produc¬ 
tion tooling a degree of wear. The tolerances 
which are provided broadly allow the major 
diameter of a nut to be greater than, but not less 
than, the standard size and a similar tolerance is 
applied to the minor diameter of a screw. 

As for some other thread forms, the pub¬ 
lished standards for BSW and BSF allow for both 
nuts and bolts to be manufactured to ‘close’, 
‘medium’ and ‘free’ limits. These different fits 
are represented in the standard tables by pro¬ 
gressively wider tolerances applied to the major 
and minor diameters. Most ordinary nuts and 
bolts are manufactured using the medium or 
free fit limits since these allow wider tolerances 
and hence greater tool wear before out-of- 
tolerance items are produced. These different 
fits are unlikely to be of interest except to mak¬ 
ers of ‘production’ quantities of screwed items. 

Table 11.1 Common Imperial Threads 




55-degree vee 

BS Fine Coarse 

tpi tpi 

60-degree vee 


tpi tpi 






s Ai 





7 Ai 


















7 /i« 




















} /4 





7 A 












Table 11.1 shows the standard pitches adopted 
for both Whitworth and BSF threads. As noted 
above, both thread series utilise a 55-degree vee 
form and adopt the standard shapes shown in 
Figure 11.3. The BSF series does not extend to 
sizes below 3 /i6in. in diameter since the BSW 
sizes below this utilise acceptably fine pitches. 

Unified and American Standard threads 

The Unified thread series was established 
jointly by the USA, Canada and Great Britain 
with the objective of creating a standard which 
would allow interchangeability of threaded 
items between the three countries. The result¬ 
ant Unified Series does not quite cover the same 
range of sizes as the American Standard Series, 
but all of the unified threads are included in the 
American Standard. 

Two separate series, Unified National Coarse 
and Unified National Fine (UNC and UNF) are 
provided for, roughly equating with BSW and 
BSF, but utilising a 60-degree vee form rather 
than the 55-degrees of the British Standard. 
Neither series caters for fractional diameters 
below 'Ain. although there are some smaller 
diameters designated by number references (see 
below). Table 11.1 shows the adopted pitches 
from 'Ain. to lin. diameter. For the Unified Fine 
Series the maximum specified diameter is 1 'Ain. 
at 12 tpi, but the Coarse Series extends to 4in. 
diameter (4 tpi). These larger sizes are not likely 
to be of interest to the modeller. 

Both UNC and UNF thread series include a 
range of fine-pitched threads extending down¬ 
wards below 'Ain. in diameter, both series 
allowing a smallest basic diameter just below 
‘/i6in. These sizes are shown in Table 11.2. 

In order to allow a distinction between these 
numbered threads and others such as the BA se¬ 
ries and the whole number metric sizes, these 
threads are almost always referred to by size 
and pitch. Thus, the references are 2-56 UNC 

Table 11.2 UNC And UNF Small-diameter Threads 




Turns per inch 


























and 2-64 UNF etc. the coarse or fine pitch 
series usually being identified also. 

The thread forms specified for the Unified 
Series are very similar to the Whitworth and 
BSF form shown in Figure 11.3 (except that the 
vee form is 60 degrees) with truncation of the 
crests to create the required clearances. 

Model Engineer threads 

When considering the simple case of nuts and 
bolts, it is usually the case that larger diameters 
utilise coarser pitches. The coarser (and deeper) 
threads, being basically stronger than the finer 
(and shallower) ones, naturally suiting the 
larger diameter bolts and screws. For the same 
reason, coarse pitch is normally used if alu¬ 
minium alloy components need to be threaded 
since this provides more strength due to the 
thicker root of the thread. 

In cutting screw threads for models, condi¬ 
tions are frequently quite different from those 
in the ‘real’ world. There are particular 
problems to solve such as cutting threads on 
thin-walled tubes, and arranging fastenings for 
miniature items. Fine-pitch threads are 
extremely useful in both cases (the depth of 
thread is small) although rather larger diam¬ 
eters with fine pitches are usually needed. 

There is also a frequent need to utilise items 
which are threaded both internally and exter¬ 
nally, to be screwed onto (into) both threads 



simultaneously. This requires threads of differ¬ 
ent diameters, but the same pitch. There is 
therefore value in having threads with finer 
pitches relative to their diameters, and also 
thread series having the same pitch throughout 
a range of different diameters, usually called a 
constant-pitch series. 

Fine- or special-pitch threads are allowed by 
the BSW specification which contains recom¬ 
mendations for standard depths of thread for 
all pitches from 4 tpi to 20 tpi (rising in steps of 
2 tpi) and also for 24,26,28,32,36 and 40 tpi. 
From these recommendations there have devel¬ 
oped the special, fine-pitch, Model Engineer 
threads which provide two distinct, constant- 
pitch series having pitches of 32 and 40 tpi. The 
standard allows any size of bolt to be threaded 
with any of the special pitches, but what has de¬ 
veloped is a range of taps and dies for cutting 
threads from '/sin. in diameter to '/ 2 m. in incre¬ 
ments of either '/win. or '/isin. as shown in Table 

For the standard form, the depth of thread 
may be calculated from 0.64 divided by the 
turns per inch. Thus the depth of thread for 32 
tpi is .020in. and for 40 tpi is .016in. Compari¬ 
son between Tables 11.1 and 11.3 will show 
that '/sin. x 40 tpi and 5 /32in. x 32 tpi are stand¬ 
ard BSW threads while 3 /i«in. x 32 tpi is from 
the BSF range. The 40 tpi range is most interest¬ 
ing since the thread depth is .016in. making the 
core diameter just about '/j 2 in. smaller than the 

Table 11.3 Whitworth-form Model Engineer Threads 


Minor (Core) Diameter 


40 tpi 

32 tpi 




























nominal size. It is thus possible to tap a 7 /32in. 
diameter hole with a Vain, x 40 tpi thread, al¬ 
though this does not mean that 7 /32in. is the cor¬ 
rect size hole to drill before cutting a '/tin. x 40 
tpi thread. See below. 

British Brass thread 

Although not covered by a British Standard, 
there is a well-established thread which has been 
used for many years for gas fittings, brass tubes 
and general work. The origin of the thread 
seems to have been lost, but it is still widely used 
for gas fittings in sizes above '/«in. and is also 
used in Germany for the imperial sizes 'A, Vis, 
Vs, 7 /i6, Vs, 3 /4, 7 /s and 1 inch diameter. 

For all diameters, the Brass thread uses a 
Whitworth-form (55-degree) of 26 tpi. It there¬ 
fore complements the 32- and 40-tpi Model 
Engineer threads and provides a further fine- 
pitch series. The minor or core diameters for 
the range of Brass threads are shown in Table 
11.4, the depth of thread for the Whitworth 26 
tpi thread being .025in. 

Table 11.4 British Brass Thread 
(55-degree Whitworth Form, 26 tpi) 


Minor Diameter 















7 A 




Metric threads 

The International Metric Thread System, usu¬ 
ally known by the abbreviation SI (from the 
French Systeme Internationale) was established 



following an International Congress held in 
Zurich in 1898. Metric threads have since been 
adopted widely, the principal non-participant 
being the USA where the Unified Thread System 
is still widely in use. 

The current standard is that known as the 
ISO Metric Thread Series (from the Interna¬ 
tional Organisation for Standardisation) repre¬ 
sented in the UK by British Standard 3643. This 
specifies a range of standard diameter-and- 
pitch combinations ranging through basic 
major diameters from 1mm to 300mm. The 
thread takes a 60-degree vee form with flatten¬ 
ing of the crests to provide the basic clearances, 
although rounding of the crests of bolts is 
allowable provided that the standard truncated 
outline is not exceeded. 

A coarse- and fine-pitch series of threads is 
specified in BS3643 in a range of diameters 
which rise in small increments from 1mm. The 
allowable diameters are classified into first-, 
second- and third-choice recommendations, 
providing very close spacing of screw sizes. The 
preferred, first-choice sizes are shown in Table 

The second- and third-choice sizes provide 
threads with major diameters between the sizes 
shown in Table 11.5, making an extremely 
wide range of sizes available, however, not all 
of these are produced commercially. 

To distinguish the Metric series of threads 
from others which use simple number refer¬ 
ences, the Metric series is usually referred to as 
M2, M2.5 etc. This practice avoids confusion 
with other series such as BA and the smaller, 
number-referenced, Unified threads, although 
the BA designator is almost always used. 

As noted elsewhere, imperial size materials, 
while still available in some sizes, are slowly 
being replaced by metric alternatives. It is likely 
that some items, such as copper tubes, will 
remain available for the foreseeable future, but 
the gradual change to metric sizes will eventu¬ 
ally mean that imperial threads will become less 

Table 11.5 ISO Metric Pitches (mm) 

First Choice 


Coarse Series 

Fine Series 














































useful, and therefore less widely used. 

There will still be a need for a constant- 
pitch, fine thread series for the ‘metric model¬ 
ler’ and three ranges of diameters having 
pitches of 0.5 mm, 0.75mm and 1.0mm, as 
shown in Table 11.6, have been recommended. 
These constant-pitch series are not yet widely 

Pipe threads 

To allow for the threading of pipes and cou¬ 
plings, standard threads have been developed 
allowing both tapered and parallel (normal) 
threads to be employed. In the UK, the British 
Standard Pipe Thread (BSP) utilising the 55- 
degree Whitworth form is still widely used. In 
the USA, the Society of Automotive Engineers 
(SAE) standard is typical of a range of such 
threads which utilise the 60-degree vee form 
adopted for other American threads. 

One problem posed by these threads is the 
size description which is applied, this being the 
nominal bore of the tube for which the thread 
is intended. Thus, Vt BSP has a major diameter 



Table 11.6 Recommended Constant-pitch Metric 
Threads For Model Engineers 



0.5 mm Pitch 

Core Diameters 
0.75mm Pitch 

1.0mm Pitch 

























































of 0.383in. and similarly the SAE Dryseal Taper 
Thread specified as '/sin. is designed for a pipe 
of outside diameter equal to 0.405in. Only the 
smaller sizes are likely to be of interest to model 
engineers since they are occasionally specified 
for boiler fittings, especially the BSP sizes. These 
small sizes have the characteristics shown in 
Table 11.7. 

Table 11.7 British Standard (Parallel) Pipe Threads 


(Nominal Bore 
of Tube) 




Depth of 


28 tpi 




19 tpi 

0.518 in. 


3 /iin. 

19 tpi 




14 tpi 



The sizes listed in Table 11.7 are taken from the 
definition of the parallel form of the BSP thread 
which is intended for general engineering use. 
The more common form of the BSP thread is the 
tapered one, for which the standard taper is 1 
in 16. The specifications contain recommenda¬ 
tions for the length of ‘useful thread’ and the 
size of the smaller diameter end which is the 

measuring or gauging point. This is the same as 
the major diameter given in Table 11.7 for the 
parallel thread of the same nominal size. If 
tapered threads do need to be cut, reference 
should be made to an engineer’s reference book 
which will provide full information. Usually it 
will merely be a question of cutting a thread to 
fit a commercial plug or coupling and trial-and- 
error methods may therefore suffice. 

British Association threads 

One of the most useful thread forms for the 
modeller is the British Association (BA) Series. 
Although British by name, this series is based 
on a metric standard, the basic size being a 
screw of 6mm nominal major diameter having 
a 1mm pitch. This is size number 0, known sim¬ 
ply as OBA. This is the largest size. 

The other BA sizes are given simple number 
references, from 1 upwards, each succeeding 
smaller size having a nominal diameter which is 
a fixed proportion of the size above. The pitch 
of thread becomes progressively finer as the 
nominal diameter decreases, the increase in 
pitch also being a fixed percentage of the previ¬ 
ous size. Table 11.8 shows the BA series of 
threads, decreasing in size from OBA, giving the 
nominal metric major diameters for the screw 
of each size. The BA thread series uses a vee- 
form having a 47.5-degree included angle. 

Although based on a metric standard, the BA 
series of threads actually includes one or two 
which are extremely useful to the imperial 
modeller. Table 11.8 includes the imperial 
equivalents of the major diameters of the stand¬ 
ard-sized screws. As can be seen, 5 BA is 
extremely useful since its major diameter is just 
.OOlin. greater than '/sin. It is also possible to 
achieve a reasonable 7BA thread on a 3 /32in. 
(.094in.) rod and 2BA can be formed on a 3 /i6in. 
(0.188in.) diameter. 

As originally conceived, the BA series 



Table 11.8 British Association (BA) Thread Series 




Major Diameter 

Nominal Nut Hexagon 
Across Flats (AF)* 
















































































* = 1.75 x major diameter 

Thickness of ordinary nuts = 0.90 x major diameter for 
0, 1 and 2BA; = 0.95 x major diameter for smaller sizes 
Head thickness for hexagon head bolts and screws = 
0.75 X major diameter 

included 26 threads from 0BA to 25 BA, the 
very small sizes being intended for watch¬ 
making and similar applications. The most 
commonly used sizes are the even-numbered 
threads from 0BA to 16BA, although this latter 
size, close to '/bin. in diameter is a relative 
rarity. Sizes below 16 BA have been replaced by 
alternative horological threads and are no 
longer available. 

As noted on the previous page, 5 and 7BA 
threads are useful to imperial modellers due to 
their close approximation to standard rod sizes, 
and equipment to cut these threads can be valu¬ 
able. The odd BA threads do, however, have ap¬ 
plications to modelling generally since bolts and 
nuts having one size smaller head than standard 
are available from modellers’ suppliers. 

When modelling, you ideally need nuts and 
bolts which are miniatures of the prototype 
items. Unfortunately, small threads such as the 
BA series were not designed with this in mind 

and the hexagons used are consequently out of 
proportion (for model work) with the screw 
and bolt diameters. An improved appearance 
can be achieved by using a nut or bolt head one 
size smaller than normal, for example, using a 
7BA head on a 6BA bolt, or a 5BA head on a 
4BA bolt. Nuts are also available with the 
smaller hexagon. Table 11.8 includes the across 
flats (AF) dimensions of standard nuts or bolt 
heads to allow selection of stock should it be 
necessary to manufacture these in the home 
workshop. The definition of thickness of ordi¬ 
nary (not thin) nuts is also given together with 
the method of determining the head thickness 
of hexagon-head bolts and screws. Some mod¬ 
ellers’ suppliers do stock hexagon bar in the BA 
nut sizes. 

Identification of threads 

Naturally, if a thread needs to be identified, 
with a view to making a matching screwed 
item, it will frequently be possible to fit a nut to 
a bolt and thereby mate one with another. 
However, even if this is possible, the process 
does not actually reveal which thread you are 
dealing with since it only confirms that you 
have a nut and bolt which can be screwed 

Figure 11.4 Thread pitch gauges. 



To identify a thread absolutely, it is necessary 
to confirm at least one diameter, to measure the 
pitch and confirm the angle of the vee form. 
These latter two features are best identified by 
use of thread pitch gauges. These are available 
for all of the standard thread series and three 
are illustrated in Figure 11.4. These are for the 
Whitworth, BA and Metric (SI) series. Each 
gauge comprises a set of steel blades, each one 
cut with a different pitch, at the appropriate 
thread form, covering all the pitches of the 
thread series, or some convenient range. The 
gauges shown cover the following ranges: 

• Whitworth 4 tpi to 62 tpi 

• BA OBA to 10BA (pitches not specified) 

• Metric 3.0mm to 0.25mm pitches 
Unless a thread gauge is available, the accurate 
determination of pitch is the major problem, 
the best alternative being the use of a tap as a 
comparison. Determination of the vee form 
may also present difficulties and a gauge which 
can be used for comparison is the most conven¬ 
ient means to do this. The thread gauge pro¬ 
vides for ready identification of both of these 
features of a thread. 

Figure 11.5 shows a thread gauge in use. At 
A, the correct 12 tpi Whitworth gauge is shown 
positioned in the vee of a Viin. diameter 
Whitworth screw. The thread gauge matches 
perfectly. The 10 tpi gauge is shown adjacent to 
the screw in Figure 11.5B. The mismatch is 
obvious - the gauge will not enter the vee of the 
thread, thus confirming that this is not the 
correct pitch. 

Naturally, in order to select a thread gauge 
for comparison, it is necessary to decide 
whether the thread in question is imperial or 
metric, but a measurement of the outside diam¬ 
eter (major diameter) of a screw usually reveals 
the basic type. For the screw illustrated in 
Figure 11.5, the outside diameter measured 
0.493in. or 12.52mm, using a dual-system 
vernier gauge, making it the minimum speci¬ 
fied diameter (0.4937in.) for Vi BSW and 

Figure 11.5 12 tpi and 10 tpi BSW thread gauges compared 
with a A BSW screw. 

therefore not likely to be metric. It may be a 
Unified thread, but if so it would be 13 tpi (if 
UNC) or 20 tpi (if UNF). If it were BSF, it would 
be 18 tpi and there is little doubt in this case 
that the size is % BSW 

Provided that the size allows, identification 
of an internal thread follows the same proce¬ 
dure except that the minor diameter is meas¬ 
ured to provide the essential information 
concerning the size. Naturally it is not possible 
to observe the fit of the thread gauge in the 
thread, so the trial must be made by touch. 
However, the fit is so good when the correct 
pitch gauge is used that there is usually little 
doubt concerning the pitch. 

If the internal thread is too small for conven¬ 
ient measurement of the minor diameter, trial- 



and-error methods must be used to find a screw 
which fits, and by inference to determine the 

Cutting internal threads 


Thread production is most conveniently car¬ 
ried out using hand tools known as taps and 
dies. Taps are used to cut threads in pre-drilled 
holes, or for the making of nuts. Dies are used 
for cutting threads on circular rods, or for mak¬ 
ing screws or bolts. Both taps and dies are avail¬ 
able in different forms. 

Screw threads may also be cut on the lathe 
and for the larger or more unusual threads this 
is frequently the method employed. Adaptation 
of a lathe for screwcutting is described in Chap¬ 
ter 16. 


Put simply, a tap is a hardened screw, cut to the 
correct vee form, pitch and diameter and 
machined with cutting edges so that it may be 
screwed into a suitable hole to cut a thread of 
the required type. Three types of tap are nor¬ 
mally employed: 

• a taper (or first) tap which starts the cutting 

• a second tap 

• and a bottoming (or plug) tap which com¬ 
pletes the cutting. 

Two or three taps are generally necessary 
simply because excessive turning force (torque) 
is required to cut a thread to its full depth using 
a single tool, and a specially made tap is used 
for starting the thread-cutting process. 

A set of three taps (for 2BA) is shown in Fig¬ 
ure 11.6. Each comprises a precisely shaped 

Figure 11.6 A set of taper, second and plug taps. 

thread of the required vee form and size, car¬ 
ried on a fluted body, providing three cutting 
faces. On a taper tap, the thread is ground away 
progressively towards the tip, at a shallow 
angle so that the end of the tap is small enough 
to enter a hole whose size is related to the 
minor diameter of the thread to be cut. This is 
the starting tap. The taper tap does, however, 
incorporate a useful length of full thread and 
provided that the workpiece allows the tap to 
pass through the hole completely, a taper tap 
may cut the thread adequately. 

If a blind hole must be threaded, the taper 
tap can only be used as the first stage, since a 
full thread extending to the bottom is normally 
required. The initial thread cut by a taper tap 
must be cut further to create a full-form thread. 
A second tap is ground to a much steeper taper 
than a taper tap, and for a much shorter length, 
and it is used to open out the thread formed by 
the first operation. 

Bottoming, or plug, taps have no taper as 
such but are ground with a small ‘lead’ to assist 
them to be eased into the cut. A plug tap is used 
after a second to bottom out and complete a 
thread in a blind hole. 

Hole sizes for tapping 

Before a tapping operation can begin, a hole 
appropriate to the thread to be cut must be 
drilled in the workpiece. At first sight, it might 
appear that this should be the minor diameter 



of the thread, but for several reasons this is not 
the case. First of all, the cutting action of the 
tap has to be considered since its action is not 
simply that of cutting away and removing 
metal. All metals are ductile to some extent i.e. 
they can be squeezed or drawn (extruded) into 
different shapes. Ductility naturally varies 
between different materials. Copper for exam¬ 
ple is soft and very ductile but cast iron has low 
ductility. Steel has intermediate ductility while 
aluminium and its alloys tend to be like copper. 

When a tap forms its thread, some metal is 
cut and removed, but there is also an action 
which tends to push the metal out of the way of 
the tap and squeeze or extrude it into a thread¬ 
like shape. If the drilled hole is larger than the 
minor diameter of the thread, the effect of the 
tap is to cut at the major diameter but to 
squeeze metal down into the minor diameter of 
the tap so that a full-form thread is produced 
even though the initial hole (the tapping hole) 
is larger than is theoretically necessary. Since 
ductility varies, there is an argument to support 
the view that the tapping drill size should vary 
according to the material being tapped. 

The second question to consider is whether 
a full-form thread is actually needed. The 
answer to this is “No”, and for several reasons. 
The concept of close-fit, medium-fit and free- 
fit nuts and bolts is described above. The dif¬ 
ferent fits are specified by allowing greater 
tolerances on the major and minor diameters, 
allowing differently dimensioned items to 
be produced. The strength of the resulting 
screwed components is virtually unaffected by 
these dimensional changes since the strength of 
nut and bolt together is not changed signifi¬ 
cantly until about one-half of the engaged 
thread is removed. 

Within limits, the actual depth of thread in a 
nut can vary without materially affecting the 
strength. The tapping size hole, and hence the 
amount of thread which is formed, may there¬ 
fore be selected for other reasons. 

The amount of thread which is present is 
usually expressed as a percentage of the speci¬ 
fied full thread, allowing for rounding or trun¬ 
cation. It is on the basis of the required depth of 
thread, and the material to be tapped, that the 
tapping size hole is sometimes determined. 

Due to the extrusion effect, which varies for 
different materials, it is usually recommended 
that tapping drills should be selected especially 
for the material to be tapped. The selection is 
made by choosing a drill which cuts a hole that 
leaves sufficient material only for a given 
percentage of the full thread depth. The per¬ 
centage depth of thread which is aimed for 
depends upon the ductility of the material 
being cut (the ease, or otherwise, with which it 
may be squeezed into a new shape) a typical 
range of recommendations being: 

• Copper and aluminium 70 per cent 

• Gunmetal 75 per cent 

• Steel and brass 80 per cent 

• Cast iron 90 per cent 

The force required to turn a tap when cutting 
the thread depends on the amount of metal 
being removed and is related to the percentage 
depth of thread which is to be achieved. For a 
large and relatively strong tap, it may be possi¬ 
ble to aim for, and achieve, a 90 per cent depth 
of thread but for a small BA thread an attempt 
to produce as full a form as this almost certainly 
results in a broken tap. A large percentage 
depth of thread should not be attempted, espe¬ 
cially in the smaller sizes. 

Many published tables of tapping drill sizes 
appear to have been drawn up totally without 
regard to the percentage depth of thread. 
Certainly, the percentage is frequently not 
stated and such tables should be regarded with 
some circumspection. Lists which do specify 
percentage depth of thread generally recom¬ 
mend figures between 70 and 85 per cent. 
These percentages are most definitely on the 
high side, especially for smaller sizes, and a 
depth of thread of greater than 75 per cent is 



Table 11.9 Recommended Tapping Drills (Mm); Whitworth-form Threads (55 Degrees) 




tpi Drill 


tpi Drill 



26 tpi 

Model Engineer 

32 tpi 40 tpi 

/ 6 














J /l6 







7 Al 















Vi 6 





































S A 












7 A 












Note: 65 per cent depth of thread for '/« diameter and below, 75 per cent depth of thread for other diameters 

quite unnecessary, and very likely to lead to 
broken taps. 

To keep things simple, it is adequate to uti¬ 
lise only two different values for percentage 
depth of thread, 75 and 65 per cent, using the 
lower percentage for smaller threads, say, be¬ 
low 'Ain. (6mm) in diameter and the higher 
percentage for all other sizes, irrespective of the 
material being tapped. Tables 11.9 and 11.10 
show the recommended tapping sizes for the 
Whitworth-form threads and also for the BA 
series based on the above percentage depths of 
thread. These sizes are calculated without any 
allowance for extrusion of the material being 
cut. Only drills from the current British Stand¬ 
ard recommendation are listed since these are 
now the preferred sizes. 

When deliberately cutting threads having 
only 65 per cent depth of thread, it must be 
borne in mind that the tapping hole must not be 
oversize - the drills used must be correctly 
ground so that oversize holes are not produced. 
Sharpening and testing of drills is described in 
Chapter 10. 

If alternative depths of thread are required, 
the relevant tapping drill diameter can be calcu¬ 
lated from: 

Drill diameter = nominal diameter - 
(2 X thread height x (E + 100)) 

where E is the percentage depth of thread. 

Thread height is defined by: 

• 0.64 + tpi for Whitworth-form threads 

• 0.60 x Pitch for British Association 

Table 11.10 Recommended Tapping Drills; British 
Association Threads; 65 Per Cent Depth Of Thread 


Tapping Drill 


Tapping Drill 


























Vt 2 in. 







For other threads, a standard engineers’ 
reference handbook (such as The Model Engi¬ 
neer’s Handbook by Tubal Cain, published by 
Nexus Special Interests) will provide the rel¬ 
evant details. 

Tapping operations 

With the tapping-size hole drilled, the thread¬ 
ing operation may be performed. This is simply 
carried out by screwing the taps progressively 
into the hole. To do this, a tap wrench is fitted 
to the squared end of the tap thus allowing the 
driving torque to be applied. 

As well as needing to be rotated, the first tap 
to enter the unscrewed hole (normally a taper 
tap, except for the finer threads) needs to be 
steered into the hole in order to achieve and 
maintain squareness. The tap wrench must fit 
the tap firmly to allow gentle guidance to be 
given to the tap. Figure 11.7 shows a selection 
of tap wrenches ranging from a small tee 
wrench suitable for the BA sizes to a large tradi¬ 
tional type that suits taps up to 3 /sin. or so in 
diameter. Personal choice, and the size of the 
tap, dictate the type to use and a range of sizes 
such as that shown will be required. 

For the smaller BA sizes, even the smaller tee 
wrench shown is too gross and a pin chuck may 
be preferred when sizes are very small. Some¬ 
thing quite lightweight and flimsy is in order 
for taps smaller than, say, 8 BA since this 

Figure 11.7 Tap wrenches of the two principal types. 

matches the strength of the tap and provides a 

better feel during the cutting operation. 

When starting the first tap in the plain hole, 
gentle downward pressure combined with 
about a quarter turn of the tap will jam it into 
the hole as it just starts to cut. At this stage, the 
tap must be checked for squareness to the sur¬ 
face of the workpiece, in two directions at right 
angles to one another. If your eye is good, no 
aid may be necessary, but if in doubt, stand a 
square on the job and check that the tap stands 

If all is not well, turn the tap another quarter 
turn, putting gentle pressure on the tap wrench 
(very gentle pressure if the tap is small) in the 
required direction and check once more for 
squareness. It is best not to turn the tap further 
into the work until squareness has been 
achieved and if the first two quarter turns have 
not achieved this, it is best to remove the tap 
and start again, before much metal has actually 
been removed. Although there is a limit to the 
number of times that this can be done, it is im¬ 
portant that the tap should start squarely as it is 
impossible to correct misalignment once a sig¬ 
nificant amount of thread has been cut. 

Taper taps do generally tend to lead them¬ 
selves into the hole quite nicely provided that a 
reasonably square start is made. Experience 
will bring confidence in persuading the tap to 
align itself squarely with the hole axis. If tap¬ 
ping is a new operation to you, start by cutting 
something reasonably large, say 2BA, Vit BSF 
or M5, in Brass about 'Ain. (6mm) thick, and 
steer well clear of blind holes and very small 
sizes until confidence has been built up. 

If the tapped hole is a ‘through’ hole, a taper 
tap is adequate to cut a full thread, particularly 
in thin work. For longer holes, or blind holes, 
second and plug taps do need to be used to bot¬ 
tom out the thread. There is usually no diffi¬ 
culty in detecting when the tap bottoms in the 
hole but this position must be approached with 
some care or the tap may be too highly stressed. 



Tap breakages do occur and are principally 
due to attempts to tap holes with too large a 
percentage depth of thread rather than through 
hitting the bottom of a blind hole. Indeed, 
experiments to determine the torque required 
to cut different thread depths suggest that to at¬ 
tempt to cut threads having 100 per cent depth 
results in 100 per cent of tap breakages! The 
use of 65 and 75 per cent depth of thread 
avoids this problem for the most part but if 
small threads are required in stainless steel, 50 
per cent depth may be a more sensible aim. The 
mating screws do need to be cut accurately to 
size however, and it is vital that the tapping drill 
cuts absolutely the correct size. 

Removing a broken tap is usually a problem 
since they nearly always break off flush with, or 
just inside, the hole. You may be lucky and find 
that the broken tap can be screwed out using 
pliers, but if not it must be broken up or driven 
out using a small punch. Driving out is only 
possible if the hole is not blind. Either way, the 
hole is unlikely to be suitable for tapping at the 
required size and it must either be opened out 
to suit the next larger size or opened out signifi¬ 
cantly so that a plug can be used to fill the hole 
and allow another attempt to drill and tap the 
hole correctly. 

Blunt taps are one cause of breakages and 
the workshop’s stock should be examined from 
time to time so that problems can be averted by 
purchasing replacements before a breakage 

Tapping holes for fine-pitch threads 

For threads in the series having 32 tpi or 40 tpi, 
there is a need only for second and plug taps. If 
you examine a taper tap for a fine-pitch thread 
such as’/sin. x 32 tpi, you will see that the tap is 
tapered and backed off to such an extent that 
there is virtually no thread at the end. As such a 
tap starts to cut, there is little thread being 

formed initially and therefore no strength 
available to hold the tap squarely in the hole. 
The slightest side pressure tends to ‘strip’ the 
small amount of partly formed thread and the 
result is usually not a tapped hole but a hole 
which is reamed out to some intermediate size 
by the front end of the tap. 

As a consequence, it is best to start the tap¬ 
ping operation for large-diameter fine threads 
using a second tap so a taper tap is not required 
for threads larger than !4in. diameter in 40 tpi, 
Vitin. in 32 tpi and ’/sin. in 26 tpi. 

Avoiding burrs at the mouth of the hole 

If a tapping-size hole is drilled and the hole 
tapped, the operation produces satisfactory 
results except at the mouth of the hole. The 
extrusion (metal flowing) effect which tends to 
cause the tap to push metal out of the way 
rather than cutting it, can occur more readily at 
the mouth of the hole since there is open space 
into which the metal can be readily pushed. 
This produces an unsightly hump around the 
hole which destroys the flatness of the surface 
locally and impairs the fit of whatever will be 
attached to the tapped workpiece. Figure 
11.8 A shows the problem. 

Various makeshift ways are sometimes 
adopted to remove the hump, such as filing the 
surface or de-burring the mouth of the hole 
using a drill. Both methods destroy part of the 
thread form however, and it is then necessary 
to put the tap in again to clean up the thread. 
All of this can be avoided by drilling the initial 
hole as shown in Figure 11.8B. After drilling at 

A B 

Figure 11.8 Counterboring a hole to avoid a burr at the 
mouth when tapping. 



Figure 11.9 Counterbored tapped holes In a boring head 

the tapping size, the appropriate clearing size 
drill is entered into the hole, not very deeply, 
but just enough to equate with two pitches of 
thread, say. This provides a small internal space 
in which the tap can create its burr and leaves 
the surface of the workpiece unblemished by 
the tapping operation. Figure 11.9 shows a 
mild steel block tapped in this way. 

Cutting external threads 

Threading dies 

Although there are solid dies, the most com¬ 
mon form is the split type, a range of which is 
shown in Figure 11.10. These are made in a 
range of standard outside diameters to suit the 
thread size for which the die is made, the 
standard nominal diameters being 'VUin., lin., 
I s /i6in. and 2in. this range covering dies for 
threads up to lin. (25mm) in diameter. The 
next standard die size above this is 2!4in. diam¬ 
eter but this is definitely outside the model 
engineering range of sizes. 

A die consists of a hardened steel disc having 
a central threaded hole of the appropriate size 
and form. To provide the cutting edges, 3,4 or 

5 holes are machined through the die around 
the central, threaded hole, creating not only the 
cutting edges but also the clearance necessary 
to allow the cut material to be ejected from the 
centre of the die. 

The dies shown in Figure 11.10 are of the 
split type i.e. they are slit into one of the 
clearance holes and the slit is machined with a 
vee-shaped chamfer on its outer edge. A die is 
hardened and tempered, and retains a little 
spring and a split die may, therefore, be opened 
out or closed in a little thus adjusting the size of 
thread which it cuts. To allow adjustment of the 
die, a standard holder incorporates three 
screws which engage the die, one in the slit and 
the two others, one on each side of the slit. 

Dies are provided with two machined 
recesses into which the two side adjusting 
screws of the die holder can locate. This is illus¬ 
trated in Figure 11.11. UK and American prac¬ 
tice is different in respect of the positioning of 
the side recesses, American preference being 
for positions 90 degrees each side of the slit, 
but UK practice positioning them at 30 degrees 
on each side. 

The central screw in the die holder engages 

Figure 11.10 Split dies are supplied in a range of standard 
outside diameters. 



Figure 11.11 A split die and die holder, showing the adjusting 

the chamfered slit and when screwed in, its 
pointed end opens up the die causing it to cut a 
larger thread. Alternatively, with the central 
screw withdrawn, the side screws may be used 
to squeeze in the die and hence cut a smaller 
thread. There is the possibility of producing 
grossly undersized or oversized threads, and 
the result of the threading operation must be 
checked by cutting a thread on a piece of scrap 
material and testing the result. 

If the thread being cut is to mate with a 
tapped hole, then the best test is to tap a hole in 
a piece of scrap material and make the screw to 
fit it. If the die is opened out too much, it cuts a 
shallow thread, making the minor diameter too 
large and preventing the screw from entering 
the test hole. In this case, the die must be closed 
up (a little!) and run down the thread again, the 
process being repeated until a proper fit is 

If the thread being cut is to mate with a com¬ 
mercially produced nut, then the nut, or prefer¬ 
ably several from the same batch, should be 
used for the test. For the smaller threads which 
might be produced, it is very important that the 
dies are set correctly, and do not cut undersized 

Cutting the thread 

Starting a die on the end of the workpiece is 
generally difficult - at least starting it squarely 
is. It is essential to check the die holder’s posi¬ 

tion relative to the rod, particularly by looking 
across the holder. Achieving squareness is quite 
easy when comparing the left-hand arm against 
the right-hand arm and pressure can be brought 
to bear to correct any lack of squareness here, 
but not at right angles to the line of the handles. 
The die must be started, squareness achieved 
and then the holder turned through a further 
90 degrees into the cut and squareness checked 
again, once more using the line of the handles 
as a guide. 

To assist starting, the die is ground with a 
significant chamfer to provide a generous lead 
to the cutting edges. This chamfer is usually 
ground on the side of the die carrying the iden¬ 
tification of its size and thread form and is not 
always carried out on the back, a practice 
which can have useful advantages. Starting is 
also assisted by opening up the die and by 
providing a good chamfer on the item to be 

Once the cutting has been started squarely, 
the die holder is simply screwed into the cut. 
The material which is removed from the 
workpiece is squeezed into the clearance holes 
in the die (which create the cutting edges) and is 
produced in continuous curls if the die contin¬ 
ues to be screwed into the cut. To break up the 
curls, the die should periodically be reversed 
for a short rotation and the action should be 
one of cutting for one or one-and-a-half turns 
and then reversing the die for a quarter of a 
turn, or perhaps a little more. The breaking of 
the curl can easily be felt and this type of action 
considerably eases the job of cutting the thread. 
For short or small threads, the breaking of the 
curl is not so important. 

A more satisfactory way to cut external 
threads is to perform the operation on the lathe 
using a tailstock-mounted die holder. Such a 
holder maintains axial alignment with the work 
when starting and cutting the thread and is 
more likely to produce a correctly aligned 
thread. A holder which will accept ,3 /isin. and 



lin. diameter dies will cater for threads up to 
3 /8in. (10mm) diameter and most definitely 
gives better results than a manual operation. 
Use of the tailstock die holder is described in 
Chapter 15. 

The chamfer cut into the leading edge of the 
die means that it will not cut a full-form thread 
up to a shoulder on the work. If the face of the 
shoulder is required to butt against the mating 
part, one or other of the items must be 
machined to allow full engagement of the two 
parts. This may be done in two ways: either by 
cutting away the oversize, incomplete thread 
left by the die, or by counterboring the end of 
the screwed hole to provide clearance. Counter¬ 
boring to prevent burrs at the mouth of a tapped 
hole is described above, and illustrated in Figure 
11.9. The drilling of a somewhat deeper clear¬ 
ing-size hole at the mouth, perhaps with a small 
countersink or chamfer, will provide clearance 
for the unthreaded portion of the screw. 

On some dies, the reverse side is not cham¬ 
fered to any marked degree and it is sometimes 
possible to form the full thread a little nearer a 
shoulder by finally putting the die on to the 
formed thread the wrong way round. 

Alternatively the unthreaded portion of the 
screw should be removed as part of the opera¬ 
tion to turn the screw blank diameter and form 
the shoulder, as shown in Figure 11.12. This 
operation, known as undercutting, is per¬ 
formed on the lathe. See Chapter 15. 

Lubrication during thread cutting 

Most cutting processes benefit from the pres¬ 
ence of a lubricating and cutting fluid at the 

Figure 11.12 A thread undercut at the shoulder. 

point at which cutting takes place. Fluid is per¬ 
haps most useful when high rates of metal 
removal are desired since the flow extracts heat 
from both the tool and the workpiece. If the 
fluid has an oily base, it also acts as a lubricant 
and can improve the sliding contact which 
occurs between the work and the tool thereby 
significantly improving the finish. 

When thread cutting using taps and dies, 
high rates of metal removal are not normal in 
the amateur’s workshop, even if using the 
tailstock die holder with the lathe powered, 
since the threads produced are usually quite 
short. Nevertheless, there is benefit in provid¬ 
ing lubrication for the threading process. The 
traditional lubricant for threading was tallow 
but there are now available various commercial 
preparations specifically designed for slow- 
speed cutting operations. They are valuable in 
promoting smooth cutting and preventing the 
tap or die tearing the material and help to pro¬ 
duce the desired smooth surface finish. Model 
engineers’ suppliers generally stock these prod¬ 
ucts which are known by the general name of 
tapping compounds. 



The lathe - essential machine tool 

What the lathe can do 

Provided that it is of adequate size, the lathe is 
capable of carrying out all of the machining 
operations which are essential for the amateur 
worker. Apart from the drilling machine, it is 
likely to be the only machine for cutting metal 
which the amateur has at his disposal and it can 
be developed into a very versatile machine tool. 

Basically, the lathe provides for a bar or rod 
of material to be rotated and a fixed cutting 
tool to be brought into contact with it. Since 

Figure 12.1 A turning operation which is reducing the outside 
diameter of a steel bar. 

the tool is harder than the workpiece, material 
is removed from the work as long as the tool 
continues to be fed into the rotating material. 
This basic operation is shown in progress in 
Figure 12.1. 

The workpiece may require material to be 
removed from its end face. In this case the tool 
is turned through 90 degrees in relation to its 
position when surfacing the outside of a bar, 
and the metal cutting operation performed on 
the face, as shown in Figure 12.2. This opera¬ 
tion is known as facing. 

Figure 12.2 A facing operation machines the end surface of 
work which is rotating in the lathe. 



The above operations are described as turn- mandrel, mounted in a robust casting at the 

ing; the workpiece is rotating and the cutter is a left-hand end of the bed of the machine. This 

single-point tool that is fed into the work to end of the machine is known as the head and 

remove material. Turning is essentially what a the assembly that supports the spindle, or man- 

lathe is designed to do, but by suitable adapta- drel, is known as the headstock. 

tion it may be arranged to perform the func- The foundation of the lathe is its bed. This is 
tions of other machines. Some adaptations of usually a heavy iron casting which is machined 

the lathe are described in Chapter 16. at the left-hand end to accept and locate the 

headstock and machined down the remainder 
of its length, on the top surface and on the 
edges, in order to locate a carriage, or saddle, 
General description which can be moved along the bed. 

The machine is strictly known as a centre 
A simple outline drawing showing the essential lathe since its earliest use was for turning long, 
parts of a lathe is shown in Figure 12.3. The thin work which was mounted between centres 
machine comprises a motor-driven spindle, or for machining. This method of use is still con- 

Figure 12.3 The main features of a lathe. 



venient for many jobs. The centres are merely 
spigots with pointed ends which engage in 
cone-shaped holes in the ends of the work- 
piece. One centre is mounted in (or on) the end 
of the mandrel and the other is mounted in (or 
on) a supplementary casting which is fitted 
towards the right-hand end of the bed. This end 
is known as the tail of the machine and the 
supplementary casting is the tailstock. 

The tailstock supports the centre at the same 
height above the top surface of the bed as the 
axis of the mandrel. This dimension, known as 
the centre height, is used to describe the basic 
capacity of the lathe, although the diameter 
that can be accommodated by the machine is 
sometimes specified. 

For driving (rotating) work which is 
mounted between centres, a lathe carrier is 
fitted to the job at the mandrel end. This is usu¬ 
ally a casting which fits over the workpiece and 
is provided with a set screw so that it may be 
clamped to it. The end of the carrier acts as a 
radial extension to the workpiece and is driven 
by a peg fitted to a plate attached to the end of 
the mandrel. This is called the catchplate. 
Figure 12.4 shows work mounted between cen¬ 
tres and Figure 12.5 the driving arrangement. 
The procedure adopted for cutting the female 
centres in the ends of the workpiece so that it 
may be supported and driven is described in 
Chapter 15. 

Figure 12.4 Work mounted between centres on a lathe. 

Figure 12.5 A carrier used to drive work which is mounted 
between centres. 

For removing material progressively from 
the outside of a bar mounted between centres, 
it is necessary to move a tool in and out in rela¬ 
tion to the work. The saddle is provided with a 
machined upper surface so that a supplemen¬ 
tary slide may be mounted on it. This is called 
the cross-slide since it moves across the axis of 
the lathe bed. The cross-slide is provided with a 
feedscrew and handle so that controlled and 
precise movement of a tool mounted on the 
cross-slide may be achieved, allowing precise 
control of the cut and hence the diameter of the 

With a tool mounted on the saddle and 
cross-slide, controlled movement in two direc¬ 
tions is possible, both along and across the end 
of the workpiece. When work is fitted to the 
mandrel only, and not supported at the tail- 
stock centre i.e. fixed somehow to the mandrel 
and not mounted between centres, two cuts 
may be applied to the work, either along its 
length (a surfacing cut) or across its end (a 
facing cut). 

The above paragraphs describe the basic and 
essential features of a plain lathe. However, 
even a modest practical lathe has additional 
features which render it more versatile than 
the simple machine described, and the follow¬ 
ing paragraphs consider the practical require¬ 
ments for the headstock, saddle and tailstock 



The headstock 

Mandrel speeds 

In order to be suitable for a wide range of work, 
the mandrel must be capable of rotating at dif¬ 
ferent speeds. To be really versatile, the speeds 
need to range from a few tens of rpm to a few 
thousand rpm, and to accommodate this speed 
differential some form of gearing is usually 
incorporated into the speed-change arrange¬ 

Figure 12.6 The original headstock of a small lathe which was 
driven by a flat leather belt. 

Many modern lathes incorporate an en¬ 
closed, all-geared headstock drive but lathes of 
older design frequently use belt drive. Figure 
12.6 shows the driving arrangement of a small 
lathe designed some years ago for the amateur 
market. The mandrel is supported by two up¬ 
ward projections in the headstock casting and 
the space between is occupied by the drive com¬ 
ponents. These comprise a three-step pulley and 
two gears. The mandrel drive is here seen in its 
original form, with flat-belt pulley, before its 
conversion to vee-belt drive. 

Figure 12.7 shows the full drive, with the 
three-step, vee-belt pulley fitted. A counter¬ 
shaft is fitted above the mandrel and slightly to 
the rear, and carries a matching three-step 
pulley. The countershaft carrier is adjustable so 

Figure 12.7 The complete drive arrangement on a small 
lathe, showing the countershaft and the driving belts. 

that the belt can be correctly tensioned and has 
a quick-release lever to assist rapid belt reposi¬ 
tioning, although this is not visible in the figure. 
Drive to the countershaft is provided from the 
motor, mounted at the rear, with pulley sizes 
chosen to give a countershaft speed of about 
400 rpm. 

The completed drive to the mandrel is 
shown in Figure 12.8. With the pulley cluster 
and gears occupying virtually all of the space 
between the headstock bearing housings, there 
is no room for more than the three-step pulley, 
and this would appear to limit the range of 
mandrel speeds to just three. However, the two 
spur gears fitted to the mandrel provide the 
means to extend the speed range. The smaller, 
left-hand gear is mounted in an assembly with 



Figure 12.8 The rebuilt mandrel drive arrangement which 
now uses a vee belt. 

the three-step pulley, the whole assembly being 
free to rotate on the mandrel itself. The larger 
gear on the right-hand side, known as the bull 
wheel, is keyed or otherwise locked to the man¬ 
drel and provides the actual drive. 

Lying behind the mandrel is a second shaft 
which has two gears locked to it. The shaft has 
two alternative positions, being latched into 
each by a simple ball detent. As shown in the 
figure, the gears on the rear shaft (known as the 
back gear shaft) are engaged with the two man¬ 
drel gears and drive to the mandrel is from the 
pulley cluster gear to the back gear shaft and 
forward once more to the bull wheel. There is, 
therefore, a two-stage reduction in speed from 
the pulley cluster to the mandrel and the lathe 
is said to be in back gear. 

Direct drive is arranged by disengaging the 
back gear and driving the bull wheel instead 
from the pulley cluster. For this purpose, the 
bull wheel is provided with a captive pin which 
engages with a hole in the pulley cluster but 
which can be withdrawn when back gear drive 
is required. The head of this pin can be seen 
projecting from the right-hand side of the bull 
wheel in Figure 12.8. The arrangement pro¬ 
vides for six mandrel speeds to be available, in 
two ranges of three, a typical set of speeds for a 
lathe of this type being: 

• In direct drive 800 400 200 rpm 

• In back gear 160 80 40 rpm 

The actual range of speeds, particularly the 
maximum speed, is very much dependent on 
the design of the mandrel and its bearings. 
More speeds can be provided, by fitting in a 
four-step pulley for example, or by arranging a 
two-speed drive between the motor and the 

The back gears themselves need not be 
arranged at the back of the mandrel but can 
instead be arranged as a close-coupled pair of 
gears brought into engagement by a lever- 
operated eccentric mounting. Such an arrange¬ 
ment is shown in Figure 12.9. In this lathe, a 
sliding block attached to the bull wheel allows 
for engagement of the drive from the three-step 
pulley cluster when direct drive is used. 

The provision of a back gear is important 
since it allows large-diameter workpieces to be 
rotated slowly enough to maintain the periph¬ 
eral speed of the work past the tool (the cutting 
speed) within the normal range for the material 
being turned. 

It is obvious that, if the back gear is engaged 
without disconnecting the direct drive to the 
bull wheel, the mandrel will be locked. Since 
many accessories are mounted onto a screw 
thread on the end of the mandrel, the ability to 
lock it when removing or replacing these is very 

Figure 12.9 A back gear cluster mounted below the mandrel. 



convenient. This thread may be seen in Figure 
12.6 while Figure 12.5 shows the catchplate 
fitted to this thread and a male centre pressed 
into the tapered socket in the front of the 

The mandrel 

As noted above, the nose of the mandrel nor¬ 
mally carries a screw thread to allow accesso¬ 
ries to be mounted on it. Since, by their nature, 
screw threads do not provide accurate location, 
a simple threaded nose is not normally used. 
Instead, the mandrel incorporates a plain, par¬ 
allel register, as shown in Figure 12.10. The 
register is concentric with the mandrel axis and 
behind the register is an abutment face. Attach¬ 
ments screwed on to the mandrel have an accu¬ 
rately sized bore which fits the register, but 
have a relatively sloppy thread so that the regis¬ 
ter and the abutment actually locate the acces¬ 
sory and hold it aligned with the mandrel axis. 
The accessories are most commonly work¬ 
holding devices such as chucks, faceplates and 
collets, although they may also be used to hold 
such things as milling cutters. Workholding 
devices are described in Chapter 13. 

Figure 12.10 This mandrel nose has a thread for attachment 
of workholding devices, but is also provided with a parallel 
register which locates items mounted on it. 

Immediately behind the abutment face which 
locates accessories on the mandrel nose, the 
mandrel is supported in the headstock front 
bearing. To resist the axial pressure exerted on 
the mandrel by the turning process, the front 
bearing incorporates a thrust bearing. This may 
take the form of a thrust washer or a ball bearing 
specially designed to resist thrust, or thrust may 
be resisted by making the front bearing in the 
form of a taper thereby providing support and 
resisting the end thrust. 

The mandrel is also supported in the head- 
stock at its left-hand end but this bearing nor¬ 
mally takes a cylindrical form which does not 
contribute to the resistance to end thrust. A 
means to adjust the longitudinal clearance (end 
float) of the mandrel in the headstock bearings 
is usually provided so that there is also a 
restraint to prevent the mandrel being drawn 
out of the headstock in the direction of the 
tailstock. This would otherwise happen, for 
example, when drilling an axial hole in work 
that is supported only at the mandrel end (and 
this is in practice the most common method of 
operating the machine). In any event, the driv¬ 
ing pulleys may exert some side thrust on the 
mandrel and its positive location in the head- 
stock bearings is an essential part of the design. 

To allow turning to be carried out on the end 
of relatively long bars, the mandrel is drilled 
right through. The size of this hole is dependent 
on the outside diameter of the mandrel shaft, 
which is, in turn, determined by the general 
dimensions of the headstock assembly. The 
larger the lathe, the larger is this hole and the 
more versatile will the lathe be in dealing with 
large diameter bars. It is a fact of life that one’s 
own lathe does not have a large enough through 
bore for all of the jobs which one wishes to 
undertake. There are fortunately other ways of 
dealing with longer lengths of large diameter 

Not only is the mandrel drilled through to 
provide clearance for long bars, but it is also 



normally bored to a shallow taper at the nose to 
allow centres to be fitted into the bore. Stand¬ 
ard tapers for these sockets have long been 
agreed, the most common range for small 
machines being those known as Morse tapers. 

The shallow taper which is used (an approxi¬ 
mate 3-degree included angle for Morse tapers) 
means that the inserted centre jams itself into 
the socket, ensuring that it rotates with the 
mandrel. Other accessories may also be fitted 
in the same way, but if any great torque needs 
to be transmitted, positive steps have to be 
taken to ensure that the accessory is held into 
the taper because without some assistance the 
friction fit may not hold. When a centre rotates 
with the mandrel, this is not a problem, but a 
workholding device (or cutter) held in the taper 
socket does need some assistance to maintain 
the grip. However, once a tapered accessory 
has been inserted into the socket, it does need 
to be driven out fairly firmly. 

Many items are available having taper 
shanks. They include centres for use in the 
lathe, drills, drill chucks and accessories such as 
boring heads. For one popular lathe, collets are 
available to fit the mandrel taper, together with 
an adaptor to close them up on to the work- 
piece. These are described in Chapter 13. 

Headstock for screwcutting 

A plain lathe is one on which it is not possible 
to perform screwcutting. That is, it is not possi¬ 
ble to drive the saddle, and hence the tool, 
along the bed, automatically, at a rate that is 
related to the rotation of the workpiece. In 
truth, the design of the whole lathe is deter¬ 
mined by the requirement to make it a screw¬ 
cutting machine, but since the headstock needs 
to incorporate the required drive facility, the 
subject can be introduced here. 

The basic requirement is to lead the saddle 
along the bed in sympathy with the rotation of 

the workpiece or mandrel. A vee-shaped tool is 
then fitted to the cross-slide and saddle and the 
tool cuts a vee-shaped spiral (a screw thread) in 
the work, as it rotates. 

To lead the saddle along, a leadscrew is fitted 
along the length of the bed, supported in bear¬ 
ings at each end. At the headstock end, a key is 
fitted to the leadscrew to allow it to be driven by 
gearing from the mandrel. The end of the man¬ 
drel is permanently fitted with a drive gear so 
that further gears may be interposed between it 
and the leadscrew. The actual gearing is varied 
to suit the pitch of the screw thread to be cut. If a 
thread of 24 turns per inch (tpi) is to be cut, the 
gearing is arranged so that the saddle is driven 
along the bed by one inch as the workpiece 
makes 24 rotations, thus cutting the required 

The pitch of the leadscrew naturally affects 
the choice of gear ratio for a given pitch of 
thread, but since leadscrew pitches tend to be 
standardised it is possible to consult published 
tables for the arrangement of gears to be 
adopted for cutting common pitches of thread. 

The gears which link the mandrel and lead¬ 
screw gears are known as changewheels and a 
screwcutting lathe might be supplied with a set 
ranging from 20 teeth to about 75 teeth, in 
steps of five teeth. This range allows all normal 
pitches to be cut but to allow flexibility, larger 
gears are also usually manufactured. For impe¬ 
rial lathes, a ‘metric translator’ gear is some¬ 
times available which allows a close approxi¬ 
mation to true metric pitches to be cut. 

The changewheels are mounted on studs fit¬ 
ted into slots in a ‘banjo’ casting which clamps 
to the leadscrew bearing. Stud positions, and 
banjo position on the bearing housing, are 
adjustable to permit the full range of gear ratios 
to be set up and yet still allow correct meshing 
of the gears. 

It is usual to arrange for simple reversal of 
the direction of rotation of the gear train by fit¬ 
ting a ‘tumbler reverse’ to the headstock. This is 



Figure 12.11 Tumbler reverse, leadscrew drive and gearing 
at the headstock end of the lathe. 

shown in Figure 12.11. A three-position 
arrangement is provided which allows disen¬ 
gagement of the drive to the leadscrew, or 
selection of rotation in either direction. 

If a lathe is of the screwcutting type, the 
gearing between the mandrel and the leadscrew 
may also be used to drive the saddle along the 
bed when doing plain turning. In this case the 
lathe automatically feeds the tool into the work 
and the rate at which this occurs is termed the 
feed per revolution. This use of the headstock 
gearing is generally described as the self act 
since the machine is ‘looking after’ itself, 
although the operator has to be present to dis¬ 
engage the drive to the saddle at the appropri¬ 
ate point, unless the machine is an automatic. 

If the headstock is provided with both a back 
gear and an adaptation for screwcutting, it is 

normal to describe it as a back-geared, screw¬ 
cutting lathe, this description frequently being 
abbreviated to BGSC. 

The provision of a back gear is important for 
screwcutting since this usually cannot be done 
at a rush, especially for coarse-pitch threads, 
due to the need to disengage the drive at the 
point where the thread on the work is to end. 
The drive to the saddle is described below. 

The gearing to implement the screwcutting 
facility may be contained within a separately 
mounted ‘quick-change’ gearbox which offers 
lever-operated selection of the gear ratio be¬ 
tween the spindle and the carriage on which the 
screwcutting tool is mounted. Manual selection 
of the ratio by changewheels is a slower and 
dirtier process than the simple movement of an 
external lever on a quick-change gearbox, but 
changing this ratio is not undertaken too 
frequently. Lathes fitted with a quick-change 
gearbox are significantly more expensive than 
the changewheel type. 

The saddle and its fittings 


The saddle of a small (3'/2in., 89mm) lathe is 
shown in Figure 12.12. The saddle can be seen 
to have a cross-slide mounted on it. This is pro¬ 
vided with a feedscrew so that it may be tra¬ 
versed in a direction at right angles to the lathe 
axis (in and out relative to the workpiece). The 
feedscrew carries an engraved dial so that cali¬ 
brated movement of the slide is possible, the 
dial being marked every .001 in. or every 
.020mm or .025mm if the lathe is built to met¬ 
ric standards. On some machines the dial is 
loose on the feedscrew shaft, but driven by a 
friction-grip arrangement, so that it may be 
preset to some convenient reading to define, 
for example, the beginning or end of a cut. 



Figure 12.12 The saddle spans the bed of the lathe and has 
the cross-slide mounted on it. 


The front of the saddle carries an apron that 
allows a mounting for the saddle traverse han¬ 
dle to be incorporated and also the mechanism 
that allows the saddle to be driven along the 
bed. Figure 12.13 shows the apron associated 
with the saddle and cross-slide of Figure 12.12. 
The large handwheel on the right is rotated to 
drive the saddle along the bed manually. This 
handwheel engages with a rack which is bolted 
to the bed of the lathe, immediately below the 
upper machined surface, suitable gearing being 
interposed between handwheel and rack to 

Figure 12.13 The front of the saddle is closed by an apron 
which has the saddle traverse handwheel mounted at the right- 
hand side and the control for the half nuts on the left. 

allow the handwheel to be used for ‘putting on’ 
the cut when bar turning is required. The gear 
ratio is a compromise, however, since it must 
be slow enough to permit a cut to be taken by 
use of the handwheel, yet fast enough to permit 
rapid movement of the saddle out of the way, 
for example when the tool must be withdrawn 
for the next cut, or to allow measuring or drill¬ 
ing operations to be performed. 

Locking and driving the saddle 

If the lathe is to be used for taking a facing cut 
(across the end of the workpiece) the saddle 
must be locked to the bed and the cut taken by 
use of the cross-slide traverse. Locking the 
saddle directly to the bed is not particularly 
convenient and since locking it to the leadscrew 
is required in any event, for screwcutting, this is 
usually the method adopted for the small lathe. 
Reliance is then placed on the thrust bearings at 
each end of the bed to locate the leadscrew 
without endfloat, and a definite arrangement 
for adjusting this should be incorporated into 
the machine. 

The saddle is locked to the leadscrew by a 
pair of half nuts, or clasp nuts, mounted in 
slides inside the apron, and capable of engage¬ 
ment with, or disengagement from, the lead¬ 
screw. In Figure 12.13 the clasp nut lever is seen 
at the left-hand side of the apron. The lever 
rotates anti-clockwise to engage the nuts and 
clockwise to disengage them, these actions 
rotating the cup which engages with both half 
nuts. Anti-clockwise rotation lifts the lower 
half nut and pushes the upper one down so that 
the two halves engage the leadscrew, locking 
the saddle to it. 

When screwcutting, the leadscrew is driven 
by the changewheel train at the headstock end 
and this leads the saddle along the bed. An 
alternative method of driving the saddle is to 
rotate the leadscrew manually and this is one 



Figure 12.14 The leadscrew is sometimes fitted with a dial 
and handwheel so that the saddle can be moved by a known 

way of applying the cut when facing. It is also a 
valuable method of positioning the saddle 
when performing turning (or milling) opera¬ 
tions and for this reason the leadscrew may be 
provided with its own handle and have an 
engraved dial. The dial-and-handwheel on my 
lathe is shown in Figure 12.14. 


It is perfectly feasible to mount the tool directly 
to the cross-slide and use the lathe for turning 
as described above. This can make for a very 
rigid, no-nonsense arrangement, but it is com¬ 
mon practice to provide a further slide having 
screw-controlled movement and to mount this 
on the cross-slide. The resulting assembly is 
known as the topslide and it is this which nor¬ 
mally carries the toolpost, as shown in Figure 

The need for the topslide is made more 
apparent by considering that, under some cir¬ 
cumstances, the maximum diameter of work 
which can be accommodated on the machine is 
determined by the distance between the top 
surface of the cross-slide and the axis of rota¬ 
tion. It is desirable that this be made as large as 
possible, consistent with the other dimensions 

of the machine, but in maximising this distance 
the designer places the support for the tool well 
below centre height and something quite sub¬ 
stantial is required on which to mount the tool. 

Put another way, this means that there is 
room for a two-part assembly; a base which 
bolts to the cross-slide and an upper slide, pro¬ 
vided with a feedscrew and calibrated dial, so 
that controlled and precise movement is possi¬ 
ble for taking a cut. To allow flexibility in use, 
the topslide is arranged to be readily removable 
so that the saddle and cross-slide may be used 
for mounting work when the lathe is not to be 
used for turning. The attachment of the topslide 
is arranged for rapid removal and replacement. 
A positive location for the topslide is frequently 
provided but there is not normally a register to 
guarantee that the topslide is at right angles to 
the saddle. Indeed, it is valuable to be able to set 
the topslide so that movement of the tool is not 
parallel to the axis of the mandrel. 

When the tool is mounted to the topslide, 
the path taken by the tool is reproduced on the 
workpiece and unless the topslide is set so that 
it moves precisely parallel to the axis of the 
work, a tapered workpiece is produced. This 
ability to produce tapers is one advantage of the 
adjustable topslide. Tapered work is required 

Figure 12.15 The saddle can have a topslide mounted on it. 
This has its own calibrated feedscrew and a mounting post for 
the tool holding arrangement. 



fairly frequently, and rotation of the topslide to 
the required angle is one way to produce such 
work. The topslide base may be provided with 
a graduated angular scale to allow an approxi¬ 
mately correct angle to be set relatively easily. 
Such graduations do only permit an approxi¬ 
mate setting to be achieved and if a taper is 
required to fit an existing taper exactly, careful 
alignment or trial-and-error methods must be 
adopted to ensure a proper fit. 

Tool post 

The toolpost is the name given to whatever 
device is provided for clamping a tool onto the 
machine. For most lathes this is mounted on the 
upper surface of the topslide. It may be a simple 
clamp arrangement for a single tool or may 
comprise an adaptor for mounting a number of 
tools. Some lathes are provided with an adap¬ 
tor which accepts standard tool holders, per¬ 
mitting rapid interchange of tools while still 
guaranteeing that the top of the tool’s cutting 
point is placed at centre height once the holder 
has been set up (an essential requirement, see 
Chapter 14). 

The most common type of tool holder is the 
clamp-on-post type shown in Figure 12.16. 
This comprises a central pillar, threaded at the 
top for a clamp nut or ball-ended lever. Below 
the lever, a clamping plate is a loose fit on the 
pillar and is provided with a jacking screw fit¬ 
ted into a threaded hole in one end. Downward 
pressure from the clamp nut or lever, together 
with upward pressure from the jacking screw, 
provide clamping for the tool, which is nor¬ 
mally packed up to centre height by strips of 
material placed below it. 

The above type of clamp is perfectly satisfac¬ 
tory for much of the work which the lathe is 
required to perform. The tool may readily be 
mounted in different plan positions to ‘attack’ 
the work from the required direction and can 

Figure 12.16 The normal tool clamp is used to clamp a single 
tool to the topslide. This clamp has a ball-ended clamping lever 
and a jacking screw which are adjusted in conjunction with one 
another to clamp a tool firmly into position. 

be set to centre height readily as long as strips 
of material of various thicknesses are available. 
However, it frequently occurs that more than 
one tool is required in order to complete the 
turning operations on a particular item and the 
means to have several tools readily available is 
occasionally desirable. One way to arrange this 
is to purchase an interchangeable tool holder 
system, but this is a relatively expensive solu¬ 
tion to the problem and a holder capable of 
taking more than one tool is a less-costly alter¬ 

Figure 12.17 shows a four-way toolpost 
mounted on the topslide of Figures 12.15 and 
12.16. This fits onto the pillar and is provided 
with an indexing mechanism in its base that 
allows tools to be brought to eight positions, at 
45-degree intervals, with reasonable repeat¬ 
ability. Tools may be mounted up for turning, 
chamfering, screwcutting and parting off, for 
example, and these operations performed in 
quick succession once the tools have been set 
up. This is especially valuable when several 
identical components are required, and a rou¬ 
tine which uses several different tools needs to 
be repeated. 



Figure 12.17 The alternative tool mounting is by use of a 4- 
tool turret. This one can be indexed round into eight positions to 
bring the tools to the working position. 

The only serious disadvantages of the four¬ 
way turret are that the out-of-use tools do 
sometimes get in the way and they are also a 
potential safety hazard since their sharp ends 
are close to the operator when not in use. Nev¬ 
ertheless, it is a valuable means to satisfy the 
occasional multi-tool requirement. 

A further method of making more than one 
tool available is to employ a second toolpost 
mounted directly on the cross-slide. This type is 

Figure 12.18 A rear toolpost has some advantages for 
mounting tools for parting off work from the parent stock. This 
one has a 2-position turret. 

most commonly employed for parting-off 
operations and Figure 12.18 shows a two-way 
toolpost which is used for this purpose. This 
indexes into two positions, 180 degrees apart, 
to bring one of the two tools into use. Parting 
off is described in Chapter 14 where a consid¬ 
eration of the advantages of using a rear tool- 
post for this operation will also be found. 

The tailstock 

The use of the tailstock for supporting work 
mounted between centres is described at the 
beginning of this chapter. For such service, a 
simple casting capable of being locked to the 
bed and having some means to mount the cen¬ 
tre on it, is all that is required. However, other 
facilities are required, such as the capability to 
drill holes down the rotational axis of the 
work, and the tailstock has therefore developed 
into a more useful accessory. 

A typical tailstock is shown in Figure 12.19. 
The main casting is bolted to a base which lo¬ 
cates between the machined vertical faces of 
the inner U-section of the lathe bed. The base is 
held in alignment with the axis of the machine 
but the main (upper) casting may be moved 
relative to the base across the lathe axis. This 

Figure 12.19 The tailstock is movable along the bed, but can 
be clamped to it in any position. It is used to hold drilling chucks 
and other accessories. 



adjustment allows the axis of a tailstock- 
mounted centre to be moved relative to the 
mandrel axis so that it may be nearer to the 
operator, or farther away. This feature is a use¬ 
ful adjunct to the tailstock’s other facilities, but 
for the moment it is convenient to consider that 
the tailstock centre is in perfect alignment with 
the mandrel axis. 

The tailstock’s main casting is bored 
through at centre height so that a substantial 
sleeve may be fitted. The sleeve is a good sliding 
fit in the main casting bore but is prevented 
from rotating by a screwed pin which locates in 
a longitudinal groove cut in the outside of the 
sleeve. The sleeve is threaded at its right-hand 
end (farthest from the headstock) and this 
thread engages the thread in a bored-and- 
screwcut handwheel which is captive, but free 
to rotate, in the end of the main casting. Rota¬ 
tion of the handwheel drives the sleeve in or 
out of the tailstock barrel. 

The sleeve is itself sufficiently large to be 
bored through and is normally bored to one of 
the standard tapers, usually, but not always, 
matching that in the front of the mandrel. 
Accessories may be mounted into the sleeve 
provided that they are fitted with standard 
taper adaptors. Since adaptors are readily avail¬ 
able, drill chucks, die and tap holders and 
multi-tool turrets may be mounted into the 
sleeve, in addition to the standard centres, 
allowing axial holes to be drilled and threaded 
‘from the tailstock’, or work to be threaded 
externally by use of a die holder mounted in the 
tailstock. These operations are described in 
Chapter 15. 

To permit these operations to be performed, 
the tailstock can be locked to the bed. This 
allows the thrust to be taken when the tailstock 
handwheel is used to advance a drill into the 
mandrel-mounted workpiece. The tailstock 
sleeve can be locked in the barrel, since this is 
required when supporting work between 

Mandrel and tailstock tapers 

In considering the mandrel and tailstock bores, 
the normal practice is to bore these to a stand¬ 
ard, shallow taper. The tapers used are 
described as self-holding tapers i.e. a tapered 
accessory such as a lathe centre may be jammed 
into the mating tapered socket where it will 
hold itself in. Provided that an excessive turn¬ 
ing force (torque) is not applied, the accessory 
will be held in the bore until it is knocked out 
from the small end. The torque which can be 
resisted by the taper is certainly sufficient to 
resist the forces imposed when drilling a hole 
and the standard method of mounting a drill 
chuck in the lathe is by utilising a chuck with a 
tapered arbor and mounting it directly in the 
tailstock sleeve. A description of hole produc¬ 
tion methods on the lathe, and the use of the 
tailstock, is given in Chapter 15. 

The most common tapers in use for the man¬ 
drel and tailstock of small lathes are the series 
known as Morse tapers. These range from No. 
0 which is approximately 'Ain. (6mm)diameter 
at the small end, to the massive No. 7 which is 
2%in. (70mm) diameter at the small end and 
some lOin. long. The most common sizes of ta¬ 
per socket for small lathes are the No. 1 and 
No. 2 Morse tapers, being roughly 3 /«in. 
(9.5mm) and’/uin. (14mm) in diameter at their 
smaller ends. Some larger machines utilise the 
No. 3. It is convenient if both mandrel bore and 
tailstock sleeve are bored to the same taper so 
this is usually the case. 

Figure 12.20 shows the form of a standard 
Morse taper. The size of the small end of the 


c H 

Plug depth 





1 'A degrees 


Depth of hole 

Figure 12.20 Standard forms of Morse taper sockets and arbors. 



hole is specified by the gauge diameter, or plug 
diameter, A, and the length of usable taper by 
the plug depth, C. The hole depth, B, is usually 
the plug depth, C, plus Vttin. to allow for a 
reamer to be used to cut the taper and achieve 
the required plug depth. The standard dimen¬ 
sions for the four smallest sizes of Morse taper 
are shown in Table 12.1 below. 

Table 12.1 Morse Tapers 

No. of 












on Diam. 

























3 J /l6 



Notice, from the above table that the taper is 
not constant through the range of sizes but 
remains approximately .05in. per inch giving 
an included angle for the taper of about three 
degrees, or 1 Zi degrees per side, as shown in 
Figure 12.20. For the mandrel and tailstock 
bores, the Morse taper sockets take the form 
shown in Figure 12.20 except that the bores are 
not blind but continue as through bores, as 
described above. 

Figure 12.21 A lathe centre, a drill chuck and some drills 
having Morse taper shanks or arbors. 

Tapered arbors for lathe accessories are 
frequently of the plain type such as the centres 
shown in Figure 12.21, but it will be noticed 
that the drills and the arbor of the drill chuck in 
the figure are provided with a modified end in 
the form of a tang. This is designed to provide 
the drive for the drilling operation so that the 
taper provides the location and centring in the 
sleeve or spindle, but is not required to resist 
the torque which is exerted when drilling large 
holes. Provision for utilising the driving tang is 
not usually provided on a small lathe but is 
incorporated into larger drilling machines. 

Clearances and adjustments 

It will be obvious that the fits of the saddle to 
the bed, the cross-slide to the saddle and the 
upper part of the topslide to its base need to be 
maintained at what might be described as ‘nice’ 
clearances. That is, they should be free to slide 
under control from the associated feedscrews 
but should, as far as possible, be entirely free 
from shake or backlash. This means that the fit 
of the feedscrews should be good and they too 
need to be adjusted so that there is as little free 
play as possible. The different sliding members 
need to be maintained in correct alignment 
also, and so are arranged to move on slides 
which have adjustable clearances. 

The mounting of the slides and the saddle 
usually takes the form of a pair of dovetail 
slides, as shown in the cross-section of Figure 
12.22. Machined horizontal surfaces support 
the moving slide but it is located laterally by an 
upward projection with angled sides. The slide 
itself is machined with a mating cross-section 
but this is made wider than that on the lower 
member so that a filler strip , or gib strip, may 
be used to pack out the upper dovetail slot to 
adjust clearances. The gib is locked to the upper 
slide by screws which allow adjustment of the 



Movable slide 

Gib strip 

Adjusting screw 
and locknut 

Slide support 

Figure 12.22 A cross-section of the type of mounting for the 
slides on a small lathe. 

clearances. These screws locate in drillings or 
dimples in the gib ensuring that it moves with 
the slide. Adjustment of the screws removes any 
shake or play between slide and base and so 
allows the fit to be optimised. This adjustment 
should preferably be carried out with the 
feedscrew removed so that a proper feel of the 
fit can be obtained and the locking nuts (if pro¬ 
vided) should be tightened and the fit finally 
assessed before the feedscrew is remounted. 

Figure 12.23 shows the gib strip and adjust¬ 
ing screws, with lock nuts, on a slide of this 
type. This slide is also provided with two addi¬ 
tional screws which can be used to lock it in a 
fixed position whenever movement is not 
required and the opportunity can be taken to 
provide the additional rigidity that the locking 
of the slide affords. 

Figure 12.23 This underside view of a vertical slide shows 
the gib strip and the adjusting and locking screws. 

The freedom, or otherwise, of the slides is 
one indicator of the condition of a secondhand 
machine. Any sensible owner in the habit of 
using his machine will have the slides adjusted 
as tightly as possible consistent with there being 
no actual binding. On a worn machine, this will 
mean that a good adjustment for places where 
there has been little or no wear will be sloppy at 
the most-worn spots. This effect will be noticed 
in the fit between the saddle and the bed, where 
sloppiness will generally be found at the head- 
stock end, where most wear takes place, or 
binding at the tailstock end, according to the 
way the saddle clearances are adjusted. 

It pays always to keep the slides clean and 
lubricated with a thin grease or thickish oil and 
they should be cleaned and lubricated regularly. 
It should be noted however, that cast iron is 
very dusty stuff to machine and it is best to keep 
the lathe as dry as possible while this metal is 
being machined, afterwards thoroughly clean¬ 
ing the slides and re-lubricating them. In this 
respect, it is useful if the slides can be lubricated 
via an oil or grease nipple as introduction of the 
lubricant from the inside tends to push any 
debris out of the slideways. 

Correct adjustment of the gib strips ensures 
that the slide fits without shake or looseness in 
one direction, but the attachment of the slide to 
its feedscrew, and the fit of the screw in its nut 
can still give rise to backlash or free play. The 
cause of this is explained by reference to Figure 
12.24, which shows a section through a typical 
cross-slide, and the arrangement of the feed¬ 
screw and its nut. 

The nut is naturally part of the saddle or is 
bolted to it, and the feedscrew engages with it. 
The feedscrew usually has a square-form thread 
(strictly, an Acme thread, which has a flat crest 
and root and tapered flanks) which engages the 
nut. At the outer end, the feedscrew is provided 
with a thrust collar and a reduced-diameter 
portion which passes through the endplate, 
which is attached by screws to the cross-slide. 




Collar Feed 

Figure 12.24 A longitudinal section of the cross-slide showing 
the means of adjusting the free play. 

The outer end of the feedscrew is threaded 
with a conventional vee-form thread and the 
collar dial screws on to this and can be posi¬ 
tioned so that it and the thrust collar embrace 
the endplate closely. The feed handle is also 
threaded to match the end of the feedscrew and 
it is screwed on tightly so that it acts as the lock 
nut for the collar dial. 

There are two sources of free play in this ar¬ 
rangement. There must be clearance between 
the feedscrew and its nut, and there must be 
some clearance between the thrust collar and 

collar dial and the endplate. These clearances 
show themselves as lost motion or backlash 
when the rotation of the feedscrew is reversed. 
At each reversal, initial rotation of the 
feedscrew simply takes up the clearances. 

For a given position of the cross-slide, there 
are two readings of the collar dial - one indica¬ 
tion when driving the slide inwards and an¬ 
other when pulling the slide outwards. So, if 
the collar dial readings are to be used for posi¬ 
tioning the slide, the slide’s motion must always 
be in the same direction. 

Careful adjustment of the components on 
the outer end of the feedscrew, and attention to 
any wear on the endplate will result in only 
small clearances at this point. There must natu¬ 
rally be clearance between the feedscrew and 
its nut, but even this can be adjusted on some 
machines which are provided with a split nut 
provided with a setscrew which can be used to 
squeeze the two halves together. On other ma¬ 
chines, the nut is a low-cost, bolt-on item 
which can readily be replaced. 



Holding work in the lathe 


Chapter 12 provides a description of the fea¬ 
tures of a typical small lathe, the mandrel of 
which is provided with a register and screw 
thread so that accessories may be mounted on it 
and remain aligned with the axis of rotation. 
Most of the items mounted on the mandrel 
nose are used for holding the workpiece. This is 
certainly true when the lathe is used for its 
intended purpose of turning, i.e. rotating the 
workpiece and bringing a cutting tool into con¬ 
tact with it by use of the saddle and cross-slide 

The faceplate 

The manner of holding a shaft between centres 
and driving it by a catchplate-and-carrier 
arrangement is described in Chapter 12. This 
requires no accessories other than the simple 
ones described and is the manner in which early 
lathes were designed to be used, for example, 
for turning chair spindles and legs. 

If the work to be turned is large in diameter 
but relatively short, a between-centres method 
of holding and driving is not appropriate, and a 
mounting which more resembles the shape of 
the workpiece is required. This is the faceplate. 
In essence, this is simply a catchplate without 
its driving peg, but when such a device is 
designed to be used as a faceplate it is normally 
provided with slots so that clamps to secure the 
work may be easily bolted on. 

The faceplate is normally an iron casting 
bored and screwcut to suit the mandrel nose 
and turned on its front face and on the periph¬ 
ery, so that it runs true when mounted on the 
mandrel. The slots are usually cast-in and left 
unmachined, as is the rear surface. Figure 13.1 
shows a pair of faceplates for a 3'/2in. (89mm) 
lathe, the smaller of which will be seen to have 
some extra holes in its face. These have been 
drilled and tapped to allow relatively accurate 
location of work on the face when several iden¬ 
tical parts have been required and the faceplate 
has been used for holding the work. 

Figure 13.2 shows a partly turned locomo¬ 
tive wheel bolted to the faceplate for the final 
facing back to thickness of the crank boss. A 



Figure 13.1 A pair of faceplates. 

Figure 13.2 A wheel casting secured to the small faceplate. 

subsidiary mandrel, turned to be a good fit in 
the bore of the wheel, is fitted into the mandrel 
taper and locates the wheel relative to the cen¬ 
tre of rotation. This alignment is not vital to 
this final operation but was used previously to 
ensure that the tread and flange were turned to 
be concentric with the bore. 

The faceplate is adaptable to many turning 
operations, particularly when the work to be 
turned is large in diameter or is awkwardly 
shaped and unsuitable for holding by other 
means. One further advantage of the faceplate 
is provided by the gap in the bed which is fre¬ 

quently found on small lathes. This provides 
the capability to swing much larger diameters, 
just a few inches or centimetres long, at the 
headstock end of the lathe. In order to utilise 
this gap to the maximum effect, it is essential to 
use as slim a workholding device as possible 
therefore, generally, the faceplate will not be 
found to extend significantly beyond the end of 
the mandrel. 

The larger of the two faceplates of Figure 
13.1 is 9in. (229mm) in diameter yet it swings 
in the gap provided in the bed of the 3'/2in. 
(89mm) lathe for which it is an optional extra, 
as shown by Figure 13.3. 

Figure 13.3 The large faceplate occupying part of the gap in 
the bed. 



Figure 13.4 Faceplate dogs used to hold a steel plate to the 

The method used for holding work on the 
faceplate is very much dependent on the shape 
of the item and the machining which is 
required. Figure 13.4 shows the use of slotted 
clamps for workholding, called faceplate dogs. 
They are used for clamping the work by use of 
bolts passing through the clamps (and face¬ 
plate), the toe of the clamp bearing against the 
work and the heel being supported on packing 
raised to a suitable height. 

An alternative clamp is a section of angle, 
drilled for a clamp bolt and used in much the 
same way as a faceplate dog. Due to its shape it 
does not necessarily need packing at the heel 
and is considerably more convenient under 

Figure 13.5 L-shaped clamps holding a boiler backhead to 
the faceplate. 

some circumstances. Clamps of this type are 
shown in Figure 13.5. Both types of clamp are 
easily made and will repay the effort of their 
construction, being applicable also to the 
clamping of work to the cross-slide or vertical 
slide for boring or milling. 

The disadvantage of the clamping methods 
described above is that the clamps themselves 
obscure the outer face of the work and prevent 
the whole of this surface being turned. These 
methods are most suitable where the work is 
provided with lugs which allow clamps to be 
applied outside the area to be machined, or 
where the work only requires holes to be 
drilled or bored. 

A variation of the mounting method may be 
used when the complete face of the work needs 
to be machined. This utilises an angle plate 
bolted to the faceplate, the work being bolted 
to the angle plate rather than directly to the 
faceplate. This arrangement is shown in Figure 
13.6 which also demonstrates an alternative 
method of securing the work by use of a long 
bolt through the angle plate, in this instance, in 
conjunction with a strap of mild steel, which 
forms the clamp. A thick strap of mild steel can 
be used, as shown here, but a length of steel 
angle adds some stiffness to the clamp, and 
might also be used, depending on the situation. 

It frequently happens that work must be 
mounted out-of-balance on the faceplate (Fig¬ 
ure 13.6, for example) and quite large out-of- 
balance forces can be generated when the man¬ 
drel starts rotating. This is potentially danger¬ 
ous in that the friction clamps that are normally 
used may give way as the job tends to throw 
itself into orbit, but, in any event, it places an 
unnecessary strain on the mandrel bearings, 
particularly on the front bearing cap. Any out- 
of-balance mass naturally throws the mandrel 
about and since it has clearance inside its bear¬ 
ings, accuracy of the work will be affected and 
so a degree of balancing must be attempted. This 
is simply arranged, as shown in Figure 13.7. 



Figure 13.6 If an angle plate is mounted to the faceplate it can 
provide the means to hold a cylinder block casting so that the 
portface can be turned. Lacking an angle plate, a length of square 
angle can be used in its stead. 

Figure 13.7 Wark may be mounted off-centre and need to 
be counterbalanced. A suitable source of weights is the lathe 
changewheel set but care must be taken that the tool does not 
touch the changewheel 

Anything can be pressed into service for use 
as a balancing mass, but the basic requirement 
is satisfied by a few discs with holes through 
their centres so that they may be bolted into 
position. Lathe changewheels are ideal in this 
respect, but care must be taken not to allow 
them to strike the bed or cross-slide or to come 
into contact with the cutting tool. 


The four-jaw independent chuck 

A four-jaw independent chuck is the work¬ 
holding device which you must have - you can¬ 
not possibly manage without it. Not only is it 
capable of holding rectangular work but the 
independent movement of its jaws allows work 
to be accurately centred so that turned surfaces 
may be generated with reference to already 
turned diameters, or to reference points marked 
on the job. It is also suitable for turning circular- 
section bars, making it the most versatile of the 
available chucks. 

This chuck developed from an adaptation of 
the faceplate which seems to have been in use in 
the early part of this century. As noted above, 
there is some disadvantage in the use of the 
faceplate since all work may not be suitable for 
mounting by clamps, or dogs, pressing onto the 
outer face of the work. Indeed, for most items, 
gripping by the outside edge is preferable since 
this naturally leaves the face clear for machin¬ 
ing. Consequently, machine designers devel¬ 
oped small, bolt-on assemblies comprising 
hollow rectangular channels in which a screw- 
controlled sliding jaw was located. These could 
be mounted into the faceplate slots so that the 
jaws could be positioned radially to clamp to 
the outside of the work. The four-jaw inde¬ 
pendent chuck is a development of this idea. 

Figure 13.8 shows a 6in. (150mm) four-jaw 
chuck. It has a ground, cast iron body with four 
machined slides in which the jaws fit, each jaw 
being screwcut on the reverse to mate with a 
captive screw mounted in the body below each 
slot. The screws have square holes in their 
outer ends, and can be turned using a square- 
ended key. 

The jaws and slots are matched during 
manufacture and each is numbered accord¬ 
ingly; jaws normally being hardened and 
ground and the slots also ground to size. Jaws 
should not be interchanged between slots. The 
body of the chuck carries a stamped or en- 


Chuck body 

Figure 13.9 The normal mandrel fitting includes a screw 
thread and some form of location for the chucks etc, which will 
be mounted on it. On larger lathes, the chuck may be located by 
a short taper and be bolted into position. 

shows one type of backplate-to-mandrel fitting. 

All four jaws are reversible in their slides and 
are ground with two steps so providing flexibil¬ 
ity for mounting different types of work. Fig¬ 
ures 13.10 and 13.11 show work mounted in a 
four-jaw chuck. 

One possible disadvantage of the four-jaw is 
that it is not so easy to cope with out-of-balance 
workpieces, but this can usually be resolved by 
use of the faceplate, as described above. 

Figure 13.8 This four-jaw independent chuck is bolted to a 
backplate which is machined to fit the end of the mandrel. 

graved serial number and each jaw is numbered 
to match the body. It goes without saying that 
jaws should not be interchanged between 

Figure 13.9 shows that this particular chuck 
is mounted to a backplate which is bored and 
screwcut to fit the mandrel of the lathe. The 
chuck body is bored at the rear so that the 
backplate may have an integral, large-diameter 
spigot which locates the chuck accurately on it. 
There are two registers to hold the chuck body 
concentrically on the lathe mandrel. Figure 13.9 

Figure 13.10 A casting for the rear toolpost mounted in the 
four-jaw chuck. 



Figure 13.11 An awkward casting held in the four-jaw chuck 
for machining the outer face. 

Self-centring chucks 

Although relatively rare, self-centring four- 
jaw chucks are available and are naturally 
extremely convenient for holding both square 
and round work. The more normal self¬ 
centring chuck is the three-jaw, which is used 
for holding round and hexagonal workpieces. 

In a self-centring chuck, all jaws are driven 
in or out together, coupled to a plate within the 
chuck body on which a spiral thread, or scroll, 

Figure 13.12 A three-jaw chuck with one jaw removed and 
showing the internal scroll. 

is cut. The jaws carry mating sets of teeth for 
the scroll on their inner surfaces so that, when 
the scroll rotates, the jaws move in or out. Fig¬ 
ure 13.12 shows a 4-in. (100mm) three-jaw 
chuck with jaw 3 removed and reversed so that 
the jaw and scroll machining may be seen. Note 
that the jaws and jaw slides are all numbered to 
match one another and the jaws also carry the 
serial number of the body to which they have 
been fitted. 

The chuck illustrated in Figure 13.12 is of 
the type known as a geared scroll chuck since 
the scroll is driven through gears having a 
square socket in which the chuck key engages. 
An alternative type of three-jaw has the scroll 
fitted externally, at the rear of the chuck. In this 
type, the scroll is drilled radially to accept a 
tommy bar so that it may be levered round to 
open or close the jaws. This is called a lever 
scroll chuck and is normally less expensive than 
the geared scroll type. 

Unlike the four-jaw of Figure 13.8, the 
three-jaw chuck has its body machined for fit¬ 
ting directly to the mandrel nose, and it is de¬ 
scribed as a recess fitting chuck. The aim here is 
to reduce the overhang and maintain the 
workpiece much closer to the mandrel than is 
possible if the chuck is fitted to a separate 
backplate, thus providing better support during 

Figure 13.13 The chucking piece on a casting held in the 
three-jaw chuck. 



Figure 13.14 A round steel bar in the three-jaw chuck. 

Figure 13.1S The alternative pair of jaws holding a cast ring. 

Due to the shape of the scroll, the jaws can¬ 
not be reversed as can those of the four-jaw 
chuck shown in Figure 13.8 so two sets of jaws 
are provided. These are normally described as 
inside and outside jaws. Together, the two sets 
provide for versatility in use and Figures 13.13 
to 13.15 show different types of workpiece in 
the chuck for machining. 

Accuracy and concentricity 

In theory, a three-jaw chuck, if perfectly made 
and mounted, would hold round work axially 

in line with the mandrel and would do this for 
all diameters of work. It would be possible to 
mount a circular bar into the chuck, to reduce 
part of the rod to a smaller diameter and pro¬ 
duce two diameters which were perfectly con¬ 
centric with one another. 

In practice, this is not possible. No self¬ 
centring chuck will hold all workpieces abso¬ 
lutely true, and a new chuck will have a nomi¬ 
nal specified accuracy of .003in. or .004 in. 
(0.08mm or 0.10mm) total eccentricity. That 
is, if a perfectly circular rod were put into the 
chuck, the jaws tightened, and the mandrel 
rotated while measuring the position of a point 
on the surface of the bar relative to some fixed 
part of the machine, a total deflection of 
.003in. to .004 in. (.08mm to 0.10mm) will be 
measured. This eccentricity is called Total Indi¬ 
cated Run-out, orTIR. 

It is likely that a new chuck, even of ordinary 
grade will actually be capable of better per¬ 
formance than the permissible tolerance, due to 
selective assembly of the parts by the manufac¬ 
turer, but the three-jaw does not guarantee 
concentricity between the bar stock surfaces 
and newly turned diameters. If concentricity is 
required between the outside diameter of a 
round bar and further turned surfaces, either 
the bar must be set to the required degree of 
accuracy in the four-jaw independent chuck, or 
a larger stock bar must be used and both diam¬ 
eters must be turned at the same setting (as part 
of the same machining operation) without 
removing the work from the chuck. 

The dial test indicator 

A Dial Test Indicator, or DTI, is the instrument 
that permits the run-out of a circular bar 
mounted in a chuck to be measured. A DTI 
consists of a ball-ended lever or plunger con¬ 
nected to a pointer which moves over a scale as 



Figure 13.16 A plunger-type dial test indicator. 

the lever or plunger moves relative to the case 
of the instrument. 

One type of DTI is illustrated in Figure 
13.16. It has a circular case about 2in. (50mm) 
in diameter and is fitted with a plunger which 
passes through the case, along a diameter. The 
plunger is cut with a rack which engages a small 
gear set on a shaft, which has the pointer, or 
hand, on the front. As the plunger moves, the 
hand rotates over a printed scale, providing an 
indication of plunger movement. In order that 
the plunger can remain in contact with the 
object which is being tested, it is lightly spring- 
loaded outwards. For this DTI, the total plunger 
movement is Vim. (12.5mm). 

The scale has major divisions of .OOlin. (one 
thou) and has minor marks showing the 
.0005in. positions. The gearing between the 
plunger and pointer is such that the hand 
rotates once for a plunger movement of .050in. 

(1.27mm), and the number of rotations of the 
hand needs to be counted if large movements 
are to be measured. Some DTIs incorporate 
another hand for this purpose. 

This is not a normal use for a DTI however, 
and to emphasise this, the scale is engraved 
with a zero and marked to indicate plus and 
minus .025in. about this position. The printed 
scale is attached to the bezel of the case and 
both can be rotated to bring the zero line to 
some convenient position when making a 
measurement of plunger movement. 

DTIs having total plunger (or lever) move¬ 
ments of Vim. or lin. are the normal types, 
usually having dials engraved in .OOlin. incre¬ 
ments, but types with a smaller total movement 
are available, having engravings at .0005in. 
intervals. The corresponding main types of 
metric DTI provide 12.5mm or 25mm plunger 
or lever movement and have .01mm dial 

The disadvantage of the type of DTI illus¬ 
trated is that it cannot be used inside a hole, 
unless the hole is large enough to accept the 
complete instrument. A more versatile DTI is 
the lever type, an example of which is shown in 
Figure 13.17. In this DTI, a ball-ended lever is 
pivoted at one end of the case and is linked to 
the pointer such that .030in. movement of the 
ball end produces one rotation of the hand. The 
scale is marked with a zero line and shows plus 
and minus .015in. marked in .0005in. (half a 
thou) divisions. 

The lever on this DTI is loaded by a double¬ 
acting spring to a central initial position. It 
needs to be brought up to the work and pressed 
against it to produce an initial deflection in or¬ 
der that it remains in contact with the surface 
which is being checked. The dial and bezel can 
again be rotated on this DTI to bring the zero 
indication to a convenient position. The move¬ 
ment of the ball is restricted to plus and minus 
.045in. from its central position. 

Since the ball on the end of the lever is only 



Figure 13.17 A lever-type dial test indicator. 

'/i6in. (1.5mm) in diameter, it will enter a small 
bore, which the plunger type will not. In addi¬ 
tion, the lever is held to the internal levers by a 
stiff friction grip and can be positioned within 
an arc greater than 200 degrees, adding addi¬ 
tional versatility. 

Setting work in the four-jaw to run truly 


Initial turning operations in the four-jaw chuck 
may not actually require accurate setting up of 

the workpiece since the objective may simply 
be to clean up one face (of a casting, for exam¬ 
ple) to allow marking out and subsequent 
machining operations to be performed. As long 
as the job is securely held and reasonably bal¬ 
anced, this will suffice. 

Frequently, however, the workpiece requires 
turning, or needs holes drilling or boring, with 
respect to existing turned surfaces or marked- 
out reference locations. Different situations 
can be envisaged, which require different 
approaches, and these are considered sepa¬ 
rately below. 

Aligning to a cast-in feature 

When dealing with cast parts which are circu¬ 
lar, it frequently occurs that some surfaces 
cannot or will not be machined, but those that 
are turned should be reasonably true with the 
as-cast surfaces. The situation which comes to 
mind is when machining locomotive wheels or 
perhaps the spoked flywheel of a model trac¬ 
tion or stationary engine. The inside of the fly¬ 
wheel rim may possibly not be machined since 
the prototype was not, and it is desired to fol¬ 
low prototype practice on the model. Similarly 
for a locomotive wheel. The heel of the crank 
boss cannot normally be machined and yet it 
looks so much better if the heel radius is rea¬ 
sonably concentric with the axle mounting 
hole. In both of these cases reasonably concen¬ 
tric is the description used and as long as the job 
is concentric by eye that is all that is required. 

In these instances a ‘sticky pin’ is called for. 
This is nothing more elaborate than a dress¬ 
making pin stuck into a piece of modelling clay 
pressed onto the end of a tool mounted in the 
toolpost. This provides a precise enough fixed 
point against which to judge any run-out of the 
cast surface and the job may be moved over by 
adjusting the jaws in pairs until the desired 
degree of truth is obtained. Figure 13.18 shows 



Figure 13.18 A sticky pin in use when aligning the cast inner 
rim of a locomotive wheel. 

a sticky pin in use to align the rim of a tender 
wheel casting. 

Accurate alignment to a reference diameter 

The situation in which a component must be 
remounted in the lathe and turned or bored to 
be concentric with an already finished diameter 
calls for the most accurate setting up. This can 
only be achieved by making an accurate meas¬ 
urement of the run out of the reference diam¬ 
eter using a DTI. 

The technique simply requires an accurate 
DTI mounted in the toolpost or on a mounting 
base which can be stood on the bed or the 

cross-slide, so that run-out is measured directly 
on the dial. The DTI mounting must be rigid 
and free from shake, otherwise accurate setting 
is difficult or impossible to achieve. Figure 
13.19 shows a very simple lever-type DTI on a 
stand with a magnetic base, in use for this 

The DTI is attached securely to the bed of the 
lathe by the stand’s magnetic base with its lever 
resting on the parallel, circular part of the 
chimney which is mounted in a chuck having 
four independently adjustable jaws. The work 
can be moved in any direction relative to the 
centre of rotation of the chuck. The tubular 
centre part of the chimney has the base casting 
silver soldered to it, and the inside of the base 
has to be turned to mate with the inside diam¬ 
eter of the tube bore. 

If the tube (assumed to be truly circular) is 
aligned with the axis of rotation, the DTI shows 
the same reading throughout a complete rota¬ 
tion of the work. If this is not the case, the DTI 
shows the relative movement of the surface of 
the tube as it rotates, showing that the centre of 
the work does not correspond with the axis of 
rotation. The DTI is said to show a run-out as 
the work is rotated. Jaws and work must be 
repositioned to bring the outer surface of the 
tube to the true-running condition. 

Figure 13.19 A simple lever-type dial test indicator in use to 
set the run-out of the work to zero. 



The way to do this is to rotate the work until 
one of the jaws is aligned with the DTI lever (or 
plunger, if the DTI is of this type) and note the 
reading, or set the dial to read zero. Rotating 
the chuck (and work) through 180 degrees then 
brings the opposite jaw into alignment with the 
DTI and the difference in the position of the two 
places on the work, aligned with the two jaws, 
can be calculated. It should be decided which 
jaw is ‘up’ and which ‘down’ and the pair of 
jaws moved in the required direction, ensuring 
that they still grip the work. If the work is 
already close to the true-running position, the 
jaws should be moved sufficiently to change the 
DTI reading by half the difference between the 
two readings. 

If the work is still some way from the true 
condition, the DTI lever or plunger may lose 
contact with the work on one side. In these 
cases, again decide which jaw is up, and which 
down, and make a small adjustment to the two 
jaws in the required direction. The initial aim is 
to reach the condition in which the readings at 
the two positions are both on the scale of the 
DTI, after which, attention can be paid to the 
other pair of jaws, which are dealt with in 
exactly the same way. 

Once the DTI reading is on the scale for the 
full revolution of the work, final adjustments 
can be made to both pairs of jaws, and a final 
check made through a full revolution. 

If it is not possible to achieve satisfactory 
results, either the work is not truly circular, or 
the jaws are being tightened with variable pres¬ 
sure each time around, disturbing the readings, 
(or even squeezing the tube, in the example 

The particular example shown in Figure 
13.19 is clearly not one in which great accuracy 
is required, but a DTI, even the simple one 
shown here, can be used to bring an already 
machined diameter or bore very accurately to a 
true-running condition before further machin¬ 
ing is performed. 

Boring or turning to a marked-out position 

When marking out prior to machining, it is 
usual to mark the centre of any turned diam¬ 
eters or holes so that accurate location is possi¬ 
ble. For holes to be drilled on the bench drill, 
centre punching is always used to allow the 
drill to find the centre. A centre punch mark is 
also convenient when marking out work which 
is to be turned on the lathe, since it is readily 
identifiable and easy to locate accurately on the 
job. The centre punch dimple also allows a dial 
test indicator and a simple accessory to be used 
to set the job so that the dimple runs truly. 

Figure 13.20 shows an alternative way in 
which work can be set up in the four-jaw. A rec¬ 
tangular bar which is too long to hold between 
the jaws in the normal way has been set up 
cross-wise between the jaws. Notice also that 
two parallel blocks of mild steel have been 
placed behind the bar so that when it is tapped 
down to seat against them, it is parallel to the 
face of the chuck but its outer surface is clear of 
the jaws. 

The bar has a section of mild steel soldered to 
one end and this is to be turned to a curved pro¬ 
file relative to a centre punch dimple roughly in 
the centre of the bar. The bar is aligned approxi¬ 
mately before the chuck is mounted to the lathe 
mandrel (it is a lot easier to see when the chuck 

Figure 13.20 A block mounted cross-wise in the four-jaw 
chuck. The frictional grip is quite adequate for safe use. 



is horizontal) to minimise the time spent in 
achieving correct alignment. 

Once the chuck is mounted on the mandrel, 
the tailstock sleeve is fitted with a centre. This is 
brought up towards the work and another cen¬ 
tre interposed between it and the centre punch 
dimple in the bar as shown in Figure 13.21. If 
the chuck is rotated, the pointed end of the sup¬ 
plementary centre remains located in the dim¬ 
ple and if is this is not on the lathe axis, it natu¬ 
rally moves in a circle, taking the end of the sup¬ 
plementary centre with it. A DTI attached to a 
stand mounted on the lathe bed (or some other 
fixed part of the machine) can be used to meas- 

Figure 13.21 A centre mounted between a dimple in the 
work and a centre mounted in the tailstock. 

ure the movement of the supplementary centre, 
and the position of the work adjusted in the 
chuck until the dimple runs truly. This is illus¬ 
trated in Figure 13.22. 

The DTI was more conveniently used with its 
dial facing the chuck, but the principle is never¬ 
theless illustrated. Adjusting the work’s posi¬ 
tion when it is mounted in this cross-wise man¬ 
ner is a little more difficult than for a conven¬ 
tional mounting, and it must be tapped through 
the jaws to adjust in one direction, and all four 
jaws adjusted together to move in the direction 
at right angles. It is nevertheless a quite legiti¬ 
mate way to hold the work, even though the 
grip relies solely on friction, rather than grip¬ 
ping on an abutment on the chuck jaws. 

In cases where absolute accuracy of align¬ 
ment is not necessary, the centralised position 
of the centre punch dimple may be found by 
simply bringing up the tailstock centre and 
adjusting the position of the work to bring the 
dimple into coincidence with the centre’s 
point. Figure 13.23 shows how a half centre 
allows good visibility for this operation, dem¬ 
onstrated on the completed job illustrated ear¬ 
lier. This method of approximate alignment 
can, of course, be used as a preliminary to more 
accurate setting using the DTI. 

Figure 13.22 A dial test indicator can be used to monitor the Figure 13.23 A dimple on the work can be aligned to a 
loose centre when bringing the dimple in the work to the true- centre in the tailstock if a setting made by eye is adequate, 

running condition. 



Figure 13.24 A hole in the work can similarly be aligned to a 
short stub of round material held in the tailstock chuck. 

If the work has already been drilled to define 
a centre, alignment can be achieved by fitting a 
stub of suitable material into the tailstock 
chuck and adjusting the workpiece position to 
align the pilot hole with it. A tapered bar such 
as a scriber point often proves convenient but a 
parallel bar can be used, as shown in Figure 
13.24, where a flanged copper boiler back- 
head, on the faceplate, is being aligned for bor¬ 
ing the firehole. 


One way to guarantee concentricity between 
the axis of rotation of the mandrel and the 
workpiece would be to machine the mandrel 
with a parallel bore (on the lathe itself) exactly 
the correct size to accept whatever (round) bar 
stock one wished to machine down to a smaller 
diameter. If the bar stock was perfectly circular, 
concentricity would then be automatically 
achieved. This is not exactly a practical solu¬ 
tion but something very close to this idea is. 

If the mandrel is bored concentrically at its 
nose, it can provide a mounting for a number of 
identically sized sleeves each of which is accu¬ 
rately (and concentrically) bored to a specific 
size. Each sleeve then provides accurate align¬ 

ment of material of the particular size in the 
mandrel bore. A means must still be found to 
grip the material, but if the mandrel and sleeves 
are made as shown in Figure 13.25, and the 
sleeves are split symmetrically, a sleeve closes 
up on to the material when it is drawn or 
pressed into the mandrel bore, and therefore 
provides a grip. Correctly made, and kept 
clean, such sleeves offer the most accurate 
automatic alignment of round material with 
the mandrel axis. 

/- Spindle or mandrel 


Figure 13.25 A typical fitting of a collet into the mandrel 

The sleeves are called collet chucks, or more 
simply collets. To be effective, a positive means 
must be provided to press or draw the collet 
into the mandrel bore. This may be a draw bar 
which engages a thread on the inside end of the 
collet and draws it into the mandrel bore, or 
may be an adaptor for the mandrel nose which 
presses the collet in. Figure 13.26 shows two 
collets, the smaller one being of the draw-in 
type, while the larger is of the press-in type. 

This latter type is unusual in that it is 
designed to fit the No. 2 Morse taper bore in 
the mandrel of the lathe and is made having the 
relatively shallow 3-degree taper rather than 
the more usual 40-degree included angle of the 
draw-in type shown. 

A further advantage of the collet is the very 
small overhang which it allows, permitting 
turning to be undertaken very close to the man¬ 
drel nose where the work is well supported. 
The set-up is therefore rigid and coupled with 
the accuracy which is afforded, is conducive to 
good work. Collets are, however, relatively ex¬ 



Figure 13.26 Draw-in and press-in collets and the larger 
example fitted with its nosepiece ready for installation into the 
mandrel bore. 

Awkward shapes and second operations 


Some jobs do not really lend themselves to 
being mounted to the lathe mandrel by use of 
faceplate, chuck or collet and alternative means 
have to be found for holding the work. There is 
also the problem posed by the job which must 
be machined at both ends, often requiring a 
second setting up in the lathe, or a second 

The second-operation setting up may present 
difficulties if the chuck jaws grip a finished 

(already machined) surface since, being hard, 
the chuck jaws bruise the work even when tight¬ 
ened to normal pressures. Gripping already 
machined surfaces directly in the jaws is not 
appropriate, and packing needs to be inserted 
between the jaws and the partly finished job. 

Collets are especially useful for holding 
already machined diameters since they grip the 
work over larger areas and do not bruise the job 
as do chuck jaws. The only requirement is for 
the turned diameter to be correct for the avail¬ 
able collet. If collets are not available, a home¬ 
made variety known as a split bush can be made 
up, as described below. 

Some of the workholding methods which 
are appropriate to second-operation set-ups are 
also adaptable to awkwardly shaped work- 
pieces so both problems can be considered 

Problem pieces fall into several categories, 
including long work which is too large in diam¬ 
eter to pass through the mandrel, awkwardly 
shaped items and those second-operation set¬ 
ups which do not lend themselves to the use of 
chucks or collets for mounting the work. Some 
examples of each of these general types are 
considered below, but since the split bush is 
such a useful device, this is described first. 

The split bush 

Many small rods and pivots are made from 
materials such as silver steel and ground mild 
steel, which are already accurately sized on their 
outside diameters. If the lathe you have only has 
an old, worn three-jaw chuck which will not 
hold such material so that its outside diameter 
runs reasonably truly, you will find it a nuisance 
when the need is simply to reduce the material to 
a smaller diameter at one end, while maintaining 
the two diameters reasonably concentric. 

In the absence of a collet to hold the work 
truly, the four-jaw chuck could be used, but 



there is then the nuisance of having to set the 
work running truly, and the possibility of bruis¬ 
ing the work. As an alternative, a sort of tempo¬ 
rary collet, known as a split bush can be made 
up. A stub of mild steel is cut off, placed in the 
chuck, faced off and chamfered at both ends. A 
piece about the same length as the chuck jaws is 
satisfactory and the diameter should be chosen 
to give a wall thickness of about '/sin. (3mm) 
when the blank is bored to suit the workpiece. 
The position of number 1 jaw is marked on the 
outside diameter (centre punch mark) with the 
blank just protruding from the jaws and a bore 
of suitable size is made by centring, drilling and 
reaming or boring, this latter method produc¬ 
ing the truest hole. 

The marking of the position occupied by jaw 
1 when the bush is made, means that it can 
always be replaced in the same orientation, and 
the bored hole should run truly each time the 
bush is used. 

When it has been marked and bored, the 
bush is removed from the chuck and split by a 
single lengthways saw cut in a position which is 
central between two chuck jaws (the position 
opposite jaw 1 is convenient, since this has been 
identified). Once the burrs have been removed 
from the inside of the bore, a collet-like sleeve 
has been produced, but with only one slit. The 
bush can be placed in the three-jaw and will 
grip a circular bar (of the size for which it is 
made) when the chuck jaws are tightened. 

Figure 13.27 shows two split bushes which 
demonstrate the versatility of such devices. 
One has been reamed to grip !Ain. (6mm) diam¬ 
eter material, and the other has been turned 
and bored to hold short lengths of 7 /sin. 
(22mm) diameter steel. 

The bushes need to be carefully made, and 
should have only thin walls, otherwise they 
need large pressures from the chuck jaws in 
order to close up on the work. This bruises the 
bush and its potential accuracy is lost. Since the 
bush grips the work over a large area, as does a 

Figure 13.27 Two split bushes for mounting in the three-jaw 

collet, it naturally offers the advantage of not 
marking the work. The larger of the two bushes 
shown has rather a thick wall, which has been 
weakened by external saw cuts. 

Long, large-diameter bars 

Long, solid bars which are too large to pass 
through the mandrel bore can be held at one 
end in a three- or four-jaw chuck but must be 
supported at the other end in some way when 
being turned, otherwise a wobble may develop, 
and the end flay about, which is exceedingly 

If the end of the work can have a centre 
drilled in it, it can be supported on a centre fit¬ 
ted in the tailstock sleeve, the headstock end 
still being held in a chuck. If diameters at both 
ends of the bar need to be concentric, centres 
must be put in both ends, the bar supported 
between centres, and driven by a catchplate 
and driving dog. 

The difficulty comes in putting in the centres 
in the first place, since the normal method of 
facing and centring the end of a bar (described 
in Chapter 15) is not appropriate if the bar will 
not pass through the mandrel. Attempting to 
turn or centre when the job has a large over¬ 
hang from the chuck is a recipe for disaster. 



However, if the bar is reasonably stiff, which 
means not too long in relation to its diameter, a 
centre may be carefully put in to the unsup¬ 
ported end. Since the centre drill has two cut¬ 
ting edges, it presents a symmetrical load on 
each side of the axis of rotation of the work 
and, if put in with care, will provide the sim¬ 
plest means of cutting a centre. 

It is vital to consider the stiffness of the 
workpiece in relation to the overhang however, 
and no attempt should be made to face or turn 
the outer end until the tailstock centre can be 
used to provide support. A lin. (25mm) diam¬ 
eter bar with up to 6in. (150mm) overhang 
outside the chuck jaws can safely be centred 
without support, and even a Van. (12.5 mm) 
bar may be centred with this overhang, pro¬ 
vided that the end is running reasonably truly. 
The way to achieve this is described below. 

If the overhang of the bar is too long for a 
centre to be put in with safety, the end can be 
held in a special support, called a fixed steady, 
attached to the bed. Figure 13.28 shows a 12- 
inch (300mm) long bar held in a three-jaw 
chuck, with its end adjacent to a steady. This has 
a base which is clamped to the bed, and which 
has two adjustable fingers which can be brought 
up to touch the bar and then locked in position. 
The steady has a hinged upper half which can be 
clamped to the base, and which has a single 
adjustable finger that can be brought into con¬ 
tact with the bar and locked in position. 

The first thing, however, is to make the end 
of the bar run truly. To assist this, three turns of 
paper are wound around the chuck end of the 
bar, and the chuck jaws are not tightened fully, 
so that there is a little resilience in the grip pro¬ 
vided by the jaws. A DTI is set up with its stylus 
on the surface of the bar, near the end, and the 
chuck turned slowly by hand to locate the point 
at which the end of the bar is highest. When 
this point is located, the stylus of the DTI is 
lifted carefully away from the bar and the end 
tapped downwards with a soft-faced hammer. 

Figure 13.28 A 12in. (300mm) long bar held in the chuck, 
showing the steady which will be used to support the outer end. 

Figure 13.29 Paper wrapped around the bar at the 
headstock end. 

Figure 13.30 Tapping the end of the bar to bring it true. 



The set-up is illustrated in Figures 13.29 and 

The process is repeated, tightening the 
chuck jaws progressively, until the grip is secure 
and the bar end running truly. The lower fin¬ 
gers of the steady are brought into contact with 
the bar and locked, while the DTI is still moni¬ 
toring its position, so that any disturbance can 
be detected. After this, the steady is closed and 
the upper finger adjusted and locked so that the 
end of the bar is properly supported. The fin¬ 
gers are lubricated with a little oil and a centre 
drill used in the tailstock chuck to cut the cen¬ 
tre, as shown in Figure 13.31. 

It is naturally helpful if the end of the bar is 
reasonably square, but it is adequate if the bar is 
cut off using a saw and then filed. If the end 
needs to be faced, this can be done by support¬ 
ing the work on a half centre, and facing the 
end as described below. 

If a steady is not available, and the work is 
too long and flimsy for an unsupported cen¬ 
tring operation, the centres must be put in off 
the lathe. The workpiece should be sawn off 
squarely at both ends and filed so that it is rea¬ 
sonably flat and free from burrs. If the bar is cut 
off about ‘/isin. (1.5mm) longer than required, 
according to how squarely you can cut the 
ends, this allows for the ends to be faced to 
length later. 

Figure 13.31 The true-running bar. supported by the steady, 
being centre drilled. 

Figure 13.32 A bell punch. 

To centre the ends of the cut-off bar, the 
drilling machine can be utilised, but to allow 
the centre drill to start, a centre punch dimple 
must be put into each end. If nothing else is 
available, the end of the bar may be marked out 
using Jenny calipers or odd legs and the esti¬ 
mated centre of the bar marked with a centre 
punch. The ‘posh’ way to find the centre is to 
use a centre square, but either method is 
adequate as long as the bar is somewhat larger 
than the nominal finished size, to allow for any 

The correct tool for centre punching the end 
of the bar is a specially adapted punch known 
as a bell punch. This comprises a bell-shaped 
holder bored to a sliding fit for a cylindrical 
centre punch, as shown in Figure 13.32. If the 
bell is placed squarely over the end of a circular 
bar, the punch automatically aligns itself with 
the bar’s centre and a sharp hammer blow is all 
that is required to mark the bar. Naturally, if 
the bar is not faced off precisely, the centre 
punch dot may have a small positional error, 
but this can be accommodated by choosing a 
slightly oversize bar, as noted above. 

Hopefully, the bar will not be so long that the 
drilling machine cannot be used for centring the 
ends and it should be possible to arrange to hold 
it in the machine vice, or clamp it to an angle 



Figure 13.33 Centring a long bar on the drilling machine. 

plate so that there is adequate support. This 
operation is illustrated in Figure 13.33. The 
centres should be put in a little deeper than 
required so that the later facing of the ends will 
leave the bar with adequate centres. 

With the ends of the bar centre-drilled, it 
may now be set up between centres for the 
turning operation. If the tailstock is of the type 
which can be set over for taper turning, it must 
be set up first, as described in Chapter 15. 
Thereafter, the actual turning is straightfor¬ 
ward. There remains however, the facing of the 
ends. Naturally this must be carried out at the 
tailstock end but cannot be accomplished if a 
normal, fully-coned centre is used. A cut-away 
centre, known as a half centre is therefore used 
in the tailstock, as shown in Figure 13.34. This 
has almost one half of its thickness machined or 
ground away at the pointed end, leaving just a 
small, fully coned tip to support the work but 
still allowing the tool to face the end. 

Once both ends of the bar have been faced to 
length, a normal, full centre should be mounted 
in the tailstock since this provides proper 

Figure 13.34 A dose-up view of a half centre. 


Large tubes 

Large-diameter tubes may be dealt with in 
much the same way as large bars if the ends are 


first plugged so that they may be centre drilled. 
Since concentricity may not be so important 
(tubes frequently only require their ends to be 
faced off to length) it is sometimes possible to 
use wooden plugs for providing the centres, but 
the mandrel end of the tube can usually be held 
in, or on, the chuck jaws, thus simplifying the 
driving arrangement and requiring only one 
end to be plugged and centred. Mounting to the 
chuck also simplifies the method of driving the 

If several tubes need to be turned, it is 
worthwhile making up a simple holding fixture 
which can take the form shown in Figure 
13.35. This comprises a length of studding, 
with centres drilled in both ends, on which are 
mounted two wooden discs, turned to a suit¬ 
able diameter to be a hand press fit in the tube. 
The discs are secured to the studding by nuts 
and washers on each side and a lathe carrier can 
be mounted on the studding at the mandrel end 
to provide the drive. The friction grip between 
the discs and the tube is generally sufficient to 
allow facing to length without problems, but, 
as noted above, it may be possible to hold the 
mandrel end of the tube in the chuck and this 
then provides the drive. 

Figure 13.3S A length of studding and some wooden discs 
used for mounting large tubes between centres. 

The wooden discs can be roughly cut out, 
drilled through their approximate centres and 
then mounted on the studding for turning to 
size. The lathe’s top speed can be used for this 
operation and the turning performed quite 
adequately using an ordinary knife tool. The 
only problem is clearing away the sawdust! 

Thin discs 

Discs cut from sheet materials are sometimes 
needed and can present difficulties. If relatively 
small discs are required, having a central hole, 
they can be produced on the drilling machine 
using a trepanning tool or hole-and-washer 
cutter. For larger diameters, the drilling ma¬ 
chine may not have a sufficiently slow bottom 
speed to keep cutting rates within comfortable 
bounds but the lower speeds of the lathe’s back 
gear range do allow larger diameters to be 
turned, and the only problem which remains is 
how to hold the piece of thin sheet to the man¬ 
drel during turning. 

If a disc is required having a central hole, this 
should be drilled, reamed or bored first so that 
it may be used for holding the blank for turning 
the outside diameter. If the chuck jaws will 
enter the hole, and bruising of the work will not 
be a problem, the hole may be used directly for 
holding the work on the chuck. 

If the central hole is too small for this, a sim¬ 
ple stub mandrel can be made up. This can take 
one of the forms shown in Figure 13.36, the 
mandrel being quickly turned up from a stub of 
suitable material. If the disc is large in compari¬ 
son with its bore, two large washers should be 
used to increase the area used to grip the work, 
and if the friction grip provided by these is not 
sufficient, paper discs should be interposed 
between the washers and the job. In extreme 
cases, the use of emery paper discs will increase 
the grip still further. 

For stub mandrels and other chuck-mounted 



Figure 13.36 A selection of stub mandrels for mounting in 
the three-jaw chuck. 

adaptors, it is useful to mark the position taken 
by jaw no. 1 when the mandrel is turned to 
diameter so that it may always be placed in the 
chuck the same way, thus helping to maintain 
the truth of the turned diameter. 

If a disc is required which has no central hole, 
or if the hole is impossibly small in relation to 
the outside diameter which is to be turned, a 
chucking piece must be mounted on the blank 
prior to turning. This is a technique which is fre¬ 
quently adapted for castings and some jobs that 
serve to illustrate the method are shown in Fig¬ 
ure 13.37. These are all castings for a steam 
locomotive - the smokebox door and the front 
and rear cylinder covers. Although each is to be 
machined to produce a broadly disc-like part. 

Figure 13.37 Castings with chucking pieces. 

each casting is provided with a chucking piece 
which is ultimately sawn or machined off, but 
which serves to hold the item for the initial 
machining operations. The smokebox door 
serves to show how the chucking piece is used 
and also demonstrates one way in which a solid 
disc, or disc with very small central hole, may be 
dealt with. 

If the casting is reasonably circular, it is nec¬ 
essary only to clean up the chucking piece with 
a file (to remove any minor surface imperfec¬ 
tions) and to mount it in the three-jaw. This al¬ 
lows one side to be faced and the outside diam¬ 
eter to be cleaned up to size. If the casting 
shows too much run-out when set up in the 
three-jaw in this simple way, the four-jaw 
should be utilised instead, to allow any eccen¬ 
tricity to be corrected. If there is a large amount 
of wobble of the face of the disc, this may have 
to be removed before turning can begin in ear¬ 

If the thickness and diameter of the casting 
allow it to be held in the chuck with the chuck¬ 
ing spigot outwards, the spigot can be straight¬ 
ened by taking a cleaning-up cut on its outside 
diameter. If this is not possible, the spigot must 
be corrected by hand filing until the job runs as 
true as necessary to face the back and turn the 
diameter. Figure 13.38 shows these operations 
in progress. 

The smokebox door requires a small central 
hole to be drilled so this is put in from the 
tailstock after facing off a small diameter 
(about the same size as the chucking spigot) to 
provide a good surface for the centre drill to 
start. With the hole drilled, the casting is 
removed from the chuck and the chucking 
spigot sawn off. 

The central hole and the faced area on the 
inside allow another chucking piece to be made 
up and soldered into position so that the cast¬ 
ing can be held in the chuck for the shaping 
of the outside surface. The chucking piece can 
be made up from anything that is available, 



Figure 13.38 Turning a smokebox door held by its chucking 

provided that it can be soldered - a stub of 
brass is convenient. Since the door is drilled 
centrally, a short spigot should be turned on the 
end of the stub to locate in the hole and the out¬ 
side diameter of the chucking piece cleaned up 
so that it is concentric with the short spigot. 
The casting should then run reasonably truly 
for the final turning on the outside. Figure 
13.39 shows the door and chucking piece (the 
end of a length of stock brass rod) ready for sol¬ 
dering together. Soft soldering is adequate to 
attach the chucking piece to the inside of the 
casting and it need not be any greater than '/an. 
(12.5mm) in diameter for a 3'/2in. (90mm) 

Figure 13.39 A smokebox door and its temporary chucking 

door casting, although naturally, huge cuts at 
high speed are not possible. 

As noted above, if discs without a central 
hole are required, the soldering-on of a chuck¬ 
ing piece allows the outside diameter to be 
turned, and if soft solder is used the chucking 
piece can be melted off once turning is com¬ 
plete, and the disc cleaned up. 

Bored work 

The chimney top and base castings shown in 
Figure 13.40 both need to be machined all 
over. Both have awkward shapes, to say the 
least, and the bores have a rough, as-cast finish 
and usually a good draw or taper to ease the 
mouldmaker’s task at the foundry. With a little 
care, even these awkward shapes can usually be 
held in one of the chucks and Figure 13.41 
shows each of the castings mounted up for 
machining the bore. 

Figure 13.40 Chimney top and base castings. 

Once bored to size, a casting may be 
mounted on a stub arbor for machining the 
outside, as shown in Figure 13.42. If the arbor 
is turned to be a firm fit for the bore, the casting 
can be pushed on, especially if the arbor is 
given a slight taper at its outer end. The friction 
grip is again normally satisfactory provided 
that things are taken gently, but if slippage 
occurs, a scrap of paper can be used as packing 



to improve the grip. Failing this, one of the 
modern metal glues can be used to secure the 
item. If an adhesive is chosen which breaks 
down at elevated temperatures, the job can be 
released from the arbor by gently heating the 
pair until the bond breaks down. 

For turning the curves on items such as the 
chimney top, it is usually best to use a normal 
approach to establish the major diameters of 
the item, but to turn the curves by use of hand 
tools. Using these it is possible to produce such 
shapes relatively easily provided that a tem¬ 
plate is made to act as a guide as the shape 
develops, but it will frequently be found benefi¬ 
cial to remove the bulk of the metal, and estab¬ 
lish reasonably true surfaces using the topslide 
and cross-slide motions together to approxi¬ 
mate the required shape before using hand 

Screwed work 

Workpieces having screw threads on one end 
can often be held in a screwed bush for any sec¬ 
ond-operation work which is required. Male or 
female threads on the work can be utilised by 
making a suitable adaptor for mounting in the 
chuck and simply screwing the work into (or 
on) the adaptor. 

This method of mounting does not necessar¬ 
ily produce concentricity between the turned 
surfaces at the two ends of the job, since the fit 
of the screw threads on the adaptor needs to be 
just right to achieve this. The method is suitable 
if the end faces are required to be parallel or 
whenever the actual alignment is unimportant. 

Figure 13.43 shows an extension tube which 
may be interposed between a lens and a camera 
body as a means of altering the focusing range. 
The abutment faces at each end must be parallel 
or the image will not be correctly focused at all 
positions on the film. The male thread was, 
therefore, machined first and then a screwed 

Figure 13.41 Chimney castings held in the chucks for the first 
stage of turning. 

Figure 13.42 The chimney top mounted on a stub arbor. 



Figure 13.43 A screwed cylinder and the ring used to hold it in the chuck. 

bush was made, being bored, screwcut and faced 
off in the lathe ensuring, when making both 
threads, that the partly machined tube would 
enter the screwed bush right up to the abutment 
face. This is generally arranged by making the 
threads a sloppy fit and undercutting the male 
thread at the shoulder (see Chapter 15). 

Once mounted in the screwed bush, the tube 
was faced off at the second end (to bring it par¬ 
allel with the abutment face already machined) 
and then bored and screwcut itself to accept the 
lens thread. 

In this instance, the tube was made from alu¬ 
minium alloy and the bush was turned up from 
a slice of scrap brass. As noted elsewhere, it 
pays always to mark adaptors such as these 
with the position occupied by number 1 jaw 
when the device is made, as this helps to ensure 
accuracy should the bush be needed again. It 
also helps if the steps of the chuck jaws can be 
used to locate the adaptor, since without such 
location, a large, short cylinder is difficult to 
mount in the chuck without wobble. 



Principles of turning 

Tool materials 

Carbon steel 

The basic requirements for a lathe tool are that it 
should be hard enough to cut the work yet 
strong enough to withstand the forces imposed 
upon it during cutting. A high-carbon steel such 
as silver steel or gauge plate is suitable since it 
may be hardened and tempered to obtain a rea¬ 
sonable compromise between hardness and 
strength. These materials have the added advan¬ 
tage that they may be annealed and can be 
worked in the soft condition to form the tool 
shapes which are required. 

However, carbon steel cutting tools have the 
serious disadvantage that they are tempered at 
pale yellow (210°C). The cutting action of the 
tool is such that chips bear down on the top and 
slide along the surface. This causes a frictional 
load on the tool and both it and the workpiece 
are heated up in the process. Indeed, chips or 
turnings may be removed from the work in the 
blue temperature region (300°C) and the tool 
point can, therefore, heat up considerably. If 
heavy cuts are taken continuously for a long 

enough period, the tool point can quite easily 
heat up to a temperature at which the temper of 
a carbon steel tool will be ‘drawn’ and it will be 
rendered too soft to be used as a cutter without 
annealing, rehardening and tempering once 

Carbon steel tools are only suitable when 
light cuts will be taken and the tool will not be 
overheated and rendered soft. Many of the jobs 
which will be undertaken are, however, well 
within the capability of carbon steel tools - 
indeed they were the only type available at one 
time. It was only the desire to achieve faster 
rates of metal removal and the introduction of 
much harder grades of material which required 
machining, that caused the abandonment of 
such tools. 

One other situation in which the tool may 
become overheated is during resharpening. 
This is normally carried out on a grinder and 
friction between the fast-rotating abrasive 
wheel and the tool causes considerable heating 
and this may similarly draw the temper and 
render the tool useless. 

Although case-hardened tools are seldom 
used, it is perfectly possible to do so and is 



worth remembering should a cutter be inad¬ 
vertently made from a non-carbon steel, by 
mistake. The same arguments regarding the 
drawing of the temper also apply to the case- 
hardened tool, and once again, it should not be 

Alloy and high-speed steels 

Alloy tool steels were developed to provide 
resistance to what is called heat softening. In 
these materials, various alloying elements are 
added to high-carbon steels to produce 
increased ‘hot hardness’. Sometimes, a single 
alloying element is added to the steel, some¬ 
times two or three elements. Chromium, sili¬ 
con, tungsten and manganese are among the el¬ 
ements used, various proportions conferring 
different tool hardness and different upper 
working temperatures. 

The resulting alloy steels provide the capa¬ 
bility to remain hard at 350 to 400°C, but 
another group of related materials known as 
high-speed steels remains hard up to 600 or 
650°C. The name high-speed steel (HSS) is used 
as a general name when describing tool materi¬ 
als other than carbon steel and is sold generally 
under the HSS description which may be taken 
to describe a material with good hot hardness. 

All high-speed steels undergo the appropri¬ 
ate heat treatment during manufacture and are 
supplied in the hardened condition. They do 
not require any further treatment other than 
grinding to the required form, and will virtually 
last until resharpened to such a small size as to 
be no longer useful. 

Although extremely hard, they are neverthe¬ 
less resilient and have good strength. They may 
also be ground on ordinary grinding wheels, 
are about half the cost of similar tipped tools 
(see below), and will perform all of the turning 
which is required. 

High-speed steel is available in short lengths 

made especially for use as cutting tools, these 
normally being described as HSS tool bits. 
Square and circular sections are available, 
both having a ground finish, in increments of 
'/isin. (1.5mm) from Vuin. to '/iin. (4.8mm to 
12.5mm). Figure 14.1 shows a group of tool 
bits from which it will be seen that they are 
ground-off squarely at the ends, or roughly 
cropped to length. Both types therefore need 
grinding to the appropriate shape before use. 

Square bits are most commonly used for 
direct mounting in the lathe’s toolpost, al¬ 
though they may equally be used in holders such 
as boring bars and fly cutters. For home-made 
versions of such tools it is usually easiest to uti¬ 
lise circular tool bits since the mounting hole 
may then simply be drilled and reamed to the 
appropriate size. 

Figure 14.1 A selection of HSS tool bits. 

Hard-alloy materials 

In the search for even greater rates of metal 
removal, and the consequent need for materials 
with increased hot hardness, metallurgists were 
drawn to examine the properties of various 
carbides, based on the knowledge that it is iron 
carbide (cementite) which confers its hardness 
on steel. The carbides of tungsten, titanium, 



tantalum and cobalt were found to provide the 
desired improvement in hot hardness and vari¬ 
ous alloys of these carbides are now available, 
providing cutting materials which will operate 
at temperatures up to 1000°C, or well into the 
red hot range. The earliest of these alloys to 
come into general use was tungsten carbide and 
this name is used colloquially to describe tools 
having cutting points made in these alloys. 

The materials themselves, known strictly as 
cemented carbides, are extremely brittle and 
are not suitable for use as lathe tools unless they 
are supported by a backing of ordinary steel to 
provide the strength. Cemented carbides are 
costly to produce and solid carbide tools 
would, in any case, be prohibitively expensive. 
Consequently it is normal for the carbide to be 
supplied as a small tip on a steel bar, and the 
name tipped tool is used to describe this type. 

Two ways of providing the support are used. 
The first is to silver solder (or braze) a carbide 
tip onto a steel shank, thereby making what is 
commonly called a tipped tool. This is then 
ground to the required cutting angles and a 
strong, hard tool results. Since the carbide tip 
cannot be ground on an ordinary carborundum 
grinding wheel, commercially available tipped 
tools are supplied already ground to the 
required shape. They cost 20 or 30 per cent 
more than the same size HSS tool bit (which is 
not ground to shape, as purchased) but can be 
expected to give a long life between regrind¬ 
ings, especially for the amateur worker. Figure 
14.2 shows tools with brazed-on carbide tips. 

The second method of providing support for 
the cutter is to arrange that the carbide tip is 
mechanically clamped to a standard steel 
holder. The tips are manufactured to tight tol¬ 
erances, in square, diamond and triangular 
shapes, and a variety of holders is available, 
allowing selection of facing, bar turning, boring 
or screwcutting tool shapes. The cutter tips are 
usually drilled centrally for the clamping or 
locating spigot and rapid removal and replace- 

Figure 14.2 Tools with brazed-on tips. 

ment of the tip is possible through the action of 
a wedging or locking screw. One form of holder 
and tip is shown in Figure 14.3. 

The initial outlay for a holder and tip is rela¬ 
tively high, but replacement tips are approxi¬ 
mately the same cost as a Van. (6mm) circular 
or square HSS tool bit. Considering that no 
resharpening is required, but the tool is never¬ 
theless always sharp, the expenditure may be 
considered not too high a price to pay for the 

The disadvantage of tipped tools, of what¬ 
ever type, is the extreme brittleness of the tip 
itself. Any accidental heavy contact with the 
job, or a minor contact with the chuck jaws, is 
certain to knock a lump off the tip, thereby 
ruining it and requiring scrapping, or regrind¬ 
ing, if the tool is of the brazed-on type. 

Figure 14.3 A tool fitted with a replaceable clamp-on tip. 



For clamp-on tips, each corner of the tip is 
usable in turn, and accidental chipping, or final 
blunting of the tip, is rectified by bringing a 
new point into use. For the more common 
brazed-on tipped tools, a special grinding 
wheel is required, sufficiently hard to grind the 
cemented carbide tip. This is known as a green- 
grit wheel. These cost about 25 per cent more 
than normal grinding wheels, but once avail¬ 
able, will allow brazed-on tipped tools to be 
reground without problem. 

Industrially, the ability of tipped tools to 
operate at high tip temperatures is extremely 
important. In spite of the fact that such tem¬ 
peratures are unlikely to be reached in the 
amateur’s workshop, the tools are nevertheless 
very useful. The cutting tip is hard and has a 
highly polished top surface. The chippings or 
turnings slide over the surface very easily, away 
from the cutting zone, and the hardness makes 
the tools useful for difficult situations such as 
the initial cuts on iron castings, or even those in 
gunmetal, which sometimes have a hard skin. 

Theory of cutting 

Cutting action 

The basic process of cutting on the lathe is one 
of forcing a broadly wedge-shaped tool into the 
rotating workpiece. The action of cutting is one 
of tearing a ‘chip’ or ‘turning’ of metal out of 
the work. The chip bears down heavily on the 
tool, just behind the cutting point, and is 
severely compressed in the process. This is 
illustrated in Figure 14.4 which shows a section 
through the tool, looking straight on to the 
work. The process of removing metal continues 
so long as the workpiece is rotating and the tool 
continues to be fed into the cut. 

The finish left on the work by the initial tear¬ 
ing process is extremely rough and it is left to 

Figure 14.4 The basic cutting action. 

the sharpened point of the tool to remove tiny 
chips from this surface and leave a smooth fin¬ 
ish on the work. 

The cutting action demands that the tool 
should have adequate strength and toughness, 
which in practical terms means that it should be 
as large as possible (particularly in the vertical 
direction) and it should be well supported in 
the toolpost. 

Tool angles 

The best slicing action for the tool of Figure 
14.4 is provided by using as sharp a cutting 
angle as possible, but as Figure 14.5 shows, a 
narrow wedge angle results in a weak tool, 
making it unsuitable for cutting strong materi¬ 
als. Since materials vary in their composition 
and strength, the way in which the chip forms 
varies considerably as does the amount of 
power required to form a chip of a given size. 
Technically, different materials are said to have 

Strong edge Weak edge 

Figure 14.5 Weakness in a narrow-pointed tool. 



different ‘machinability’. Since the strength of 
the tool also needs to be considered, different 
tool materials can use different angles, even for 
cutting the same type of metal. 

In practical terms, this means choosing the 
wedge angle of the tool to suit the material 
being turned. A quite different tool shape is 
used for brass from that used for aluminium, 
these being the two extremes. 

Some compromises must be made in choos¬ 
ing the tool angles and it is possible to select for 
particular characteristics. For example, a tool 
may be ground to maximise metal removal 
without regard to surface finish, or to produce 
a good finish but not having the capacity to 
remove metal very rapidly. These compromises 
give rise to the descriptions roughing and 
finishing in relation to lathe tools and a two- 
stage process is sometimes adopted, using dif¬ 
ferent types of tool, first roughing out and then 

Another consideration also applies to selec¬ 
tion of tool angles - the cutting point must be 
the only part of the tool that touches the work, 
and angles must be ground on the tool to 
ensure that this occurs. Below the tool point, 
these angles are called clearance angles, and in 
the plan view, they are known as relief angles, 
as shown in Figure 14.6. 

clearance clearance 

Figure 14.6 Clearance, relief and rake angles. 

Clearance angles 

Front and side clearance angles are essential to 
ensure that the part of the tool below the cut¬ 
ting point does not rub on the work. If these 
angles are made too large, the support for the 
point is weakened, as shown by Figure 14.5, 
and it is best to standardise on a value of 10 
degrees for the clearance angles. If really tough 
materials need to be turned (stainless steel, or 
tool steels, for example) clearance angles of 5 
degrees can be helpful in strengthening the tool 

Relief angles 

If the side and end relief angles are made too 
large, a very sharp point is presented to the 
work. This weakens the tool and also makes it 
difficult to achieve a good finish, so large relief 
angles are also best avoided, and 10 degrees is 
the usual value adopted. Side relief is provided 
by grinding the end of the tool at an angle of 20 
degrees, in the plan view, and skewing the tool 
in the toolpost. 

Right-hand and left-hand tools 

Figure 14.7 shows plan views of the ways in 
which the tool can approach the work. At A, a 
tool is shown which performs a surfacing cut 
when fed towards the headstock. This tool is 
called a right-hand tool and requires side rake 
ground on the tool in the same direction as that 
shown in Figure 14.4. Figure 14.7B shows a 
left-hand tool (one which cuts when fed away 
from the headstock). This naturally requires 
side rake to be ground in the opposites direc¬ 
tion to that required on a right-hand tool. 

Notice from Figure 14.7C that, theoreti¬ 
cally, a left-hand tool is required in order to 
make a facing cut with the tool fed towards the 




Figure 14.7 Right-hand and left-hand tools. 

The accompanying Table 14.1 presents a list 
of side rake angles which can be used for 
turning common materials. As with many 
things, the angles selected represent a compro¬ 
mise between providing a sharp wedge angle 
and yet leaving a sufficiently strong point 
to withstand the cutting forces. Since the com¬ 
mon tool material is high-speed steel, Table 
14.1 has been compiled for tools made in this 

Table 14.1 Recommended Cutting Angles For High¬ 

speed Steel Knife Tools 

Side Rake 

Material (degrees) 

Free-cutting mild steel 20 to 25 

BMS 15 

Carbon steel (silver steel) 8 to 10 

Cast iron 10 to 12 

Hard brass and FC bronze 0 to 3 

70/30 ductile brass 5 

Copper, phosphor bronze and 

aluminiumbronze 20 to 25 

Monel metal 10 to 14 

Nickel silver 20 to 30 

Aluminium and alloys 25 to 40* 

* Aluminium alloys vary tremendously 

Tools for bar work 

centre of the workpiece. In practice, this is sel¬ 
dom considered necessary, and it is common to 
use a right-hand tool for this purpose. 

Rake for different materials 

The clearance and relief angles, which exist 
largely to ensure that only the point of the tool 
touches the work, can be standardised at 10 
degrees. With the side clearance, the side rake 
creates the wedge angle for the tool, which can 
have different values, depending upon the 
material being cut. 


The basic shape of a right-hand tool for turning 
external surfaces is shown by the views of 
Figure 14.6. For want of a better phrase, this is 
described as bar work although it might also be 
described as outside turning. The principles of 
turning inside a hole in the workpiece 
(normally known as boring) remain the same 
but since there are several means of hole 
production available, as well as boring, these 
are all described together in Chapter 15 and 
this chapter is restricted to a description of bar 



The knife tool 

A tool which is provided with rake, clearance 
and relief angles shown in Figure 14.6 is known 
as a knife tool. It is by far the easiest tool shape 
to create from a standard tool blank and is 
extremely easy to resharpen, when this is 
required. A right-hand knife tool is illustrated 
in Figure 14.8. 

The knife tool is satisfactory for both surfac¬ 
ing and facing cuts and will face the end of the 
work and cut up to and form a square shoulder 
on the work. If the end of the tool is ground 
back about 20 degrees in the plan view, this 
provides suitable relief angles, in the plan view, 
if the tool is skewed 10 degrees in the toolpost. 

Figure 14.8 A right-hand knife tod. 

Figure 14.9 A right-hand knife tool in use. 

Slightly modified tool shapes especially 
designed for the finishing operation are 
described below. These vary little from the ones 
already described and for most normal turning, 
knife tools having the side rake angles listed in 
Table 14.1, and front and side relief of 10 
degrees, as shown in Figure 14.6, are the nor¬ 
mal ones employed. Side and front clearance 
angles can be ground at 10 degrees, but if really 
tough materials need to be turned clearance 
angles of 5 degrees might be helpful. 

Alternative tool shapes 

Two modifications can be made to the tool in 
order to change its characteristics. The first of 
these is to grind the cutting point to a small 



radius, in the plan view. This has the double 
benefit of strengthening the point and of bring¬ 
ing a larger part of the tip into use for cutting, 
the finish is improved and so is the life of the 

Figure 14.10 Chatter on a locomotive wheel casting. 

The radius must not be too large, however, 
since a larger contact between the tool and the 
work imposes a greater load on both and 
greater deflections occur. As a general guide, a 
radius of about '/)2 in. (1mm) should be sup¬ 
portable on a medium-size lathe, and is suitable 
for a tool ground from high-speed steel. Any¬ 
thing larger may cause problems, but it does 
depend on the situation. 

As an example, the round-nose tool shown 
in Figure 14.10 has produced a good finish on 
the wheel tread, but it has also produced the 
uneven finish on the locomotive wheel flange. 
The tool has been used to form the side of the 
wheel flange. In this situation, the tool was cut¬ 
ting over a very large part of the tip. The work 
has consequently pressed down heavily on it, 
and it has deflected out of the way, reducing the 
load and causing the tool to deflect back up 

This has induced a vibration of the tool 
which caused a varying cut to be applied and 
this spoiled the finish on the work. This type of 
rippled finish is called chatter, and it is, need¬ 
less to say, unwanted. In this case the chatter 
was caused by too large a part of the tool being 
in contact with the work, but there are other 
causes, and these, and their cures, are consid¬ 
ered in Chapter 15. 

The large-radius, round-nose shape does 
have the advantage that it produces a very strong 
point, and it is frequently used to strengthen the 
end of a tool with a tungsten carbide tip, as is the 
example shown. 

Tool shape for good finish 

When turning, the tool must be traversed into 
the cut along the surface or face of the 
workpiece. The finish imparted to the work is a 
function of: 

(1) The finish on the tool. 

(2) The rate of traverse of the tool (if the tool 



is traversed too quickly, the result will be a 
cut like a screw thread). 

(3) The shape of the end (cutting point) of the 

The shape of the end of the tool is important in 
providing an overlap between cuts, or more 
correctly, in providing an overlap between suc¬ 
cessive turns in the spiral which the tool makes 
on the work. If a certain rate of feed is envis¬ 
aged for a finishing cut, say .010 in. per rev 
(0.25mm per rev), an overlap between cuts 
(spirals) can be arranged by grinding a .020in. 
(0.5mm) flat on the end of the tool as shown in 
Figure 14.11. Such a flat is most easily put on 
by hand using a slip stone and can be done at 
the same time that the tool is being honed. 

The flat must not be made too large how¬ 
ever, since it represents a departure from the 
principle that only the point of the tool should 
touch the work and it may induce chatter, 
which is not what fine finish is about! To this 
end a proper clearance angle must be ensured 
below the flat, which therefore needs to be put 
on with care. 

The best finish is obtained by using the ‘self 
act’ of the machine to provide the feed since this 
guarantees a steady traverse of the tool into the 
cut. However, it is a mistake to use too fine a 
rate of feed for the final cut in the belief that this 
will produce the best results. It does produce 
fine finish (initially), but it also blunts the tool, 
so if there is any significant length to turn, the 

tool may not be cutting to the same extent at the 
end of the cut and a size problem may result. 

When considering the length of cut, it is nec¬ 
essary to consider the length of the spiral which 
the tool traverses and not just the straight 
length of the work. If the feed is set to .OOlin. 
per rev (,025mm per rev) the work will make 
1000 turns for the tool to traverse a length of 
lin. (25.4mm). On a workpiece l.Oin. in diam¬ 
eter, this means the tool travels 260ft (80m) 
along a l.Oin. length. Feeds of the order of 
.OlOin. per rev (0.25mm) will therefore be 
found much more satisfactory for finishing, 
due to the shorter distance travelled by the tool. 
This is especially true if the material is relatively 
hard on the tool and likely to cause some wear 
on one pass. Stainless steel, cast iron and per¬ 
haps carbon steel might come into this cat¬ 

There are other reasons for keeping the tool 
cutting, such as the avoidance of chatter, and a 
reasonable rate of feed and good depth of cut 
will generally be found preferable. Certainly, 
feeds less than .005in. per rev (0.125mm) 
should not be used. 

Maximum-efficiency tool 

The usual form of this tool associated with the 
list of rake and clearance angles (Table 14.2, 
p.304) is that illustrated in Figure 14.12. This 
tool provides the most efficient shape to maxim¬ 
ise the rate of metal removal and is normally rec¬ 
ommended where this is the prime considera¬ 
tion. It is provided with both side and top rake 
i.e. the rake is ground at a compound angle. 

The tool is quite different from the conven¬ 
tional type. It is not provided with side clear¬ 
ance but is designed to cut along a significant 
portion of the side, this being arranged by 
grinding back the side to produce negative 
clearance. Consequently, the terminology relat¬ 
ing to the tool point changes, and the angles in 



the plan view are called the side and end cutting 
edge angles. The nose of the tool is radiused, 
essentially to strengthen the tip, since the tool is 
specifically designed for taking large cuts and a 
weaker, sharp point cannot be used. 

Figure 14.13 shows the reason for the adop¬ 
tion of this tool shape. Two tools are shown, 
one of which has zero side clearance while the 
other has a side cutting edge angle of 60 
degrees. A certain depth of cut is envisaged, 
together with a certain feed per rev, and the 
material which is removed from the workpiece 
is shown for each tool. Angling the side of the 
tool produces two effects - the chip which is 
being removed from the work is thinner, and a 
greater length of the tool is being used to 
remove the chip. Both effects naturally increase 
the life of the tool, hence its adoption when 
material removal is the prime consideration. 

The tool is essentially for removing material 
quickly and although it will not form a square 
shoulder, it is ideal for roughing out as a first 
stage when turning a shaft with large integral 
flanges, when much metal must be removed. 
Afterwards, the shaft can be turned to size, and 
a fine finish, as a second operation. 

In considering tool shapes and rake angles, it 
should be remembered that industrial processes 
are concerned with maximum economy. Con¬ 
sequently, tool angles are chosen to achieve a 
compromise between speed of metal removal, 
surface finish and tool life before regrinding 
becomes necessary. There is also much more 
power available on a professional lathe and this 
also influences the choice of speed and cutting 

The amateur is not usually so interested in 
maximising metal removal and will frequently 
not even want the bother of changing from a 
roughing tool to a finishing tool. Consequently, 
the preferred tool shape is the knife tool which 
is shown in Figure 14.8. Although designed for 
surfacing, it does cut well for facing the end of 
the work, or the abutment to a shoulder, and no 
other general-purpose tool is required for bar 

Compound-rake tools 

The knife tool is easy to grind consistently to 
the required angles and is equally easy to 
resharpen. It is consequently recommended for 
all external turning, or bar work. It is however, 
a simplified version of the generally accepted 
industrial form of tool which is described in 
most textbooks. 

It is generally argued that the tool shown in 
Figure 14.6 is in reality cutting on the front and 
the side, and it consequently needs to be 
ground with both side and top rake. The rake 
angles are thus visible when the tool is viewed 
from the side and from the front. 



Compound rake leads to better efficiency, 
both in the amount of material which can be 
removed, and in the life of the tool. So, com¬ 
pound rake can be beneficial and Table 14.2 
shows the range of side and top rake angles 
which are generally adopted. A range of values 
for both rake angles is given, since the angles 
adopted depend upon how one defines maxi¬ 
mum efficiency, and also since tool life, rate of 
metal removal and surface finish on the work, 
all need to be considered. Figure 14.14 shows 
how the top surface of the tool is ground slop¬ 
ing down away from the cutting point, for 
right- and left-hand tools with positive rake. 

Parting tools 

Once the initial turning operations have been 
completed, the workpiece will require parting 
off from the parent bar. This simply means 

Top rake 

Figure 14.14 Compound rake angles. 

Table 14.2 Recommended Cutting Angles For Com¬ 
pound Rake High-speed Steel Tools 



Top Rake 

Side Rake 

Free-cutting mild 


5 to 10 




20 to 25 




5 to 10 


Carbon steel 


4 to 6 

8 to 10 

Cast iron 



10 to 12 


6 to 10 


Hard brass and 



0 to 3 

FC bronze 




70/30 ductile brass 





2 to 6 

2 to 6 

Copper, phosphor 


8 to 20 

15 to 25 

bronze, aluminium 


10 to 20 

15 to 25 

Monel metal 


4 to 8 

10 to 14 


15 to 20 


Nickel silver 


10 to 20 

20 to 30 

Aluminium and 



25 to 40* 

* Aluminium alloys vary tremendously and it is difficult 
to give specific recommendations. 

using a narrow tool that cuts on the front and 
plunging it into the bar, removing a slice of 
material and severing the partly made compo¬ 
nent from the bar. Figure 14.15 shows an 
embryo bolt being parted from the parent 

It is clear that the parting tool cuts on the 
end rather than on the side, and the wedge 
angle must, therefore, be ground on the end 
(usually called the front). The tool is then said 
to have top rake, as shown in Figure 14.16, 
which represents the normal form of parting 
tool. The sides are ground to give a small clear¬ 
ance of 1 degree on each side to reduce the fric¬ 
tion and the tendency to jam. Front clearance is 
required but should not be excessive since this 
weakens the tool point. Zero top rake is some¬ 
times used to give good strength to the point 
and this is also beneficial when it comes to 
regrinding as only the front needs to be ground, 



Figure 14.15 Parting off a small bolt. 

I deg. each side \ 


5-10 deg. top rake 

I deg. each side 

Figure 14.16 The parting-off tool. 

but the material being worked may cut better if 
some top rake is given, but more than 10 
degrees will weaken the point. Five degrees 
represents a good working compromise 

To assist the workpiece to be severed with¬ 
out breaking off from the parent bar, and con¬ 
sequently still having a ‘pip’ attached to it, the 
front of the tool is sometimes ground at an 
angle so that the right-hand side leads the left 
and the job falls away just before the tool com¬ 
pletes the removal of material from the bar. 

Parting off is an operation which causes a 
great many problems, one reason being that the 
requirements for the tool are in conflict. Firstly, 
it is desirable that the tool is as thin as possible, 
since the metal it removes is just waste, but it is 
designed to cut over the full width of the face 
and this puts a strain on a narrow tool. To 
preserve strength, the tool has parallel or near- 
parallel sides so it tends to rub on the sides of the 
groove it is cutting. If the tool is not absolutely 
square to the work, it tends to bend and may jam 
in the groove and/or break off. The problems 
are magnified if a large diameter is involved 
since this requires a long tool with consequent 
aggravation of the strength problem. 

Grinding a parting tool from a standard 
high-speed steel blank is somewhat tedious, 
due to the amount of metal which must be 
removed. Figure 14.16 shows generally what is 
required and it is obvious that quite large 
amounts of metal need to be ground away. To 
overcome this nuisance, special parting-off 
blades are available commercially. These take 
the form shown in Figure 14.17. 

A narrow, hollow-ground blade, Van. 
(12.5mm) deep and 4in. (100mm) long is 
ground at top and bottom to fit a dovetail slot. 
The blade is ground to this cross-section 
throughout its length and may be resharpened 
by grinding only at the end. Provided with a 
suitable holder, which may be home-made or 
purchased from a commercial source, a blade 
will last a lifetime. If the holder can be held in a 



Figure 14.17 The end view of a commercial parting-off 
blade, mounted inverted in a reartoolpost. 

rear-mounted toolpost, the best of both worlds 
is available. 

Figure 14.18 shows an indexing, two-way 
tool holder specially constructed to hold two 
blades of different widths, in the inverted (rear) 
position on the cross-slide. This was made from 
a set of castings which is available commer¬ 
cially and remains on the cross-slide more or 
less permanently. 

Figure 14.18 A reartoolpost fitted with two parting-off 

The blades described above do not seem to 
be available commercially in widths less than 
*/i6in. (1.5mm). If many small components are 
to be made in relatively expensive materials, 
the waste represented by the use of such a wide 
parting-off tool may be unacceptable. This 
problem can be solved by the adoption of a dis¬ 
carded hacksaw blade as a parting-off tool for 
small jobs. Where the diameters are small and 
the tool extension consequently short, a blade 
mounted in a simple holder works very well 
and probably represents one of the narrowest 
parting-off tools which can be used. 

Of course, if all else fails, the component can 
always be cut off the bar using the hacksaw, and 
the job remounted in the lathe for facing off as 
a second operation. But please don’t saw it off 
in the lathe! 

Grinding and sharpening lathe tools 

The bench grinder 

If there is one general piece of advice which can 
be given, it is that one should never attempt to 
cut anything with a blunt tool. Some cutting 
tools cannot easily be resharpened without spe¬ 
cialised equipment, but lathe tools and drills 
(probably the cutters used most frequently) 
may readily be reground to create new cutting 
edges, provided that a bench grinder is avail¬ 
able. A grinder is also required to shape stand¬ 
ard tool bits into usable lathe tools, and it is 
thus an essential in the workshop. 

The normal type of grinder is illustrated in 
Figure 14.19. The one shown has Sin. 
(127mm) diameter wheels, and this size, or one 
with 6in. (150mm) wheels is necessary for the 
heavier work of grinding up lathe tools from 
standard blanks. Since this operation requires 
large amounts of material to be ground away, 
one of the wheels should have a coarse, open 



Figure 14.19 A 5 inch bench grinder. 

structure to allow material to be removed with¬ 
out generating too much heat. Nevertheless, 
heat will certainly be generated, and a bowl of 
cold water in which to cool the tool is an essen¬ 
tial accessory when grinding up lathe tools. 

For grinding the final cutting edge on a lathe 
tool, or sharpening drills, a wheel having a fine 
structure is required so that a fine, ridge-free 
finish is imparted to the tool. The grinder is 
normally double-ended and fitted with two dif¬ 
ferently graded wheels. 

Considerable power is absorbed in removing 
material rapidly from the hard steel of a lathe 
tool blank, and the machine may stall when 
pressure is applied if the motor is not ‘man 
enough’ for the task. A grinder with 5-inch 
wheels thus needs a motor rating of !A hp, or 
about 200 watts, and 6-inch wheels require 
something larger, say Vi hp or 350 to 400 watts. 
Anything less than these power ratings may 
make bulk removal of material a slow business. 

The wheels supplied with a new grinder will 
be suitable for grinding high-speed steel lathe 
tools and drills. If tipped tools need to be 
ground, a green-grit wheel is required. If one of 
these types of wheel is purchased, or indeed, 
any other wheel, it must match the machine for 
which it is intended. This means that its diam¬ 
eter and thickness should be correct, and its 
mounting hole diameter, but above all, it must 

be suitable for the rotational speed of the 

Grinding wheels are designed to be used 
below a specified rotational speed which must 
not be exceeded or the material will be 
overstressed and the wheel will burst apart. EX¬ 
DANGEROUS. All wheels must be marked 
with their maximum permissible speed, and 
any wheel which is not marked, or has lost its 
label, should be thrown away. 

Wheel diameter is important since the rests 
on the machine must always be set close to the 
wheel to minimise the likelihood of anything 
becoming jammed between the rest and the 
wheel, which could also cause a wheel break¬ 
age. Therefore, smaller wheels cannot usually 
be fitted, and neither can larger ones, due to the 
presence of the wheel guard. 

On the subject of tool rests, these are some¬ 
times too flimsy to be of much use, or are 
attached to flimsy guards, and improvements 
can usually be made to improve the rigidity and 
the utility of the rests. 

When creating a lathe tool from a high¬ 
speed steel blank, considerable amounts of ma¬ 
terial need to be ground away, and the blank 
must be pressed against the wheel with firm 
pressure. This pressure can only be resisted 
safely if the tool blank is pressed into the edge 
of the wheel, and the side of the wheel should 
never be used for the task of sharpening or 
grinding a lathe tool. 

Wheels are made by a moulding process, 
and the periphery is then ‘dressed’ to render it 
smooth, with the abrasive particles projecting 
from the surface of the matrix in which they are 
embedded. Due to the fundamental weakness 
of the wheel, the sides are not dressed, but are 
left in the as-moulded state and may be quite 
rough, even on a fine-grit wheel, and are dou¬ 
bly unsuitable for grinding lathe tools. 

Grinding wheels do wear in service, and the 
surface becomes clogged with abraded particles 



and the wheel ceases to cut cleanly. They occa¬ 
sionally need dressing to restore the surface to 
flatness and dislodge the abraded particles. 
This is usually done by using an industrial dia¬ 
mond which is sufficiently hard to cut the back¬ 
ing matrix of the wheel and dislodge the abra¬ 
sive particles. The diamond is cemented into a 
holder which acts as a handle, and can be sup¬ 
ported on the grinding rest. The dressing of 
a wheel naturally produces a large amount of 
highly abrasive dust and the workshop’s 
machinery must be covered completely before 
the operation commences if the grinder cannot 
be moved to a less sensitive location. A face 
mask and eye protection are essential wear dur¬ 
ing wheel dressing and during tool grinding. 

No attempt should be made to dress the side 
of a wheel since it is not strong enough to with¬ 
stand the required pressure. 

The wheels ordinarily fitted to grinders are 
intended to be used on hard materials such as 
those used for lathe tools and drills. Grinding 
soft materials such as mild steel, unhardened 
silver steel, and others, even softer, rapidly clogs 
the surface and renders the wheel useless. If you 
must grind soft materials, and it is strongly 
NOT recommended, fit an open-structured 
wheel such as one which uses silicon carbide as 
the abrasive. 

Tool grinding 

Given that the wedge angle is important in pro¬ 
moting good, clean cutting, it is clearly neces¬ 
sary to maintain the point angles on the tool 
within the normal range for the material being 
machined, during the grinding and resharpen¬ 
ing processes. Referring to Table 14.1 it 
appears that it is necessary to grind specific 
angles (accurate to the nearest degree) in creat¬ 
ing the tool point. As noted above however, the 
angles listed refer generally to the maximum 
efficiency type of tool and the recommenda¬ 
tions may therefore be varied in practice. As 

noted elsewhere, the chief concern is finally 
whether the tool cuts and produces the 
required finish, and sharpness and lack of rub¬ 
bing are the vital factors, rather than any strict 
adherence to published values for the various 
angles. It is, however, helpful to use a broad 
point to the tool for cutting tough materials, 
reserving tools with sharper points for lighter 
work, otherwise the point may break down 
very frequently 

If a lathe tool is to be ground from a high¬ 
speed steel tool blank, the coarse wheel of the 
grinder is used to remove the bulk of the mate¬ 
rial on the three faces, following which, the 
ground faces are rendered smooth by grinding 
them lightly on the fine wheel. The periphery 
of the wheel must be used for the severe task of 
grinding high-speed steel. Even the fine wheel 
may not produce the finish on the tool which is 
desirable, and the faces should afterwards be 
honed by hand, using a fine carborundum slip 
stone to produce a polished surface on the cut¬ 
ting edges. Honing must be carried out care¬ 
fully so that the angles ground on the tool are 
maintained and the edges do not become 

Establishing the basic ground angles initially 
is not easy, and practice will be required if the 
periphery of the wheel on an ordinary bench 
grinder is used for tool grinding. If a tool blank 
is applied to the wheel by hand and its position 
is controlled by eye, the angle at which the face 
is ground is not precisely determined. During 
the grinding process, the tool, or tool blank, 
must be withdrawn from the wheel repeatedly 
to determine how the grinding is proceeding. 
The repeated applications of the tool to the 
wheel may mean that the surface may be 
ground as a series of disorganised flats which 
leave the surface immediately adjacent to the 
cutting edge - and that is where it matters - not 
at the required angle. This means that the cut¬ 
ting edge is not presented to the work in the 
ideal condition, and a clean cut and good sur¬ 
face finish may not be achieved. In extreme cir- 



cumstances, the face of the tool adjacent to the 
cutting edge may be ground back too far, re¬ 
sulting in the tool simply rubbing and generat¬ 
ing heat. However, many workers find little dif¬ 
ficulty in developing the necessary skill and 
find the bench grinder entirely satisfactory for 
the preparation of lathe tools. 

Without a proper guide for the tool, it is 
only the skill of the operator that determines 
the shape of the tool point. What is preferable 
is a wheel with a flat surface, which is provided 
with an adjustable rest so that the tool, or tool 
blank, can be presented to the wheel at the cor¬ 
rect angle, repeatedly, thus ensuring that the 
ground faces are flat and lying at the required 
angles. The best form of wheel is the shape 
called a cup wheel, which is illustrated in Figure 
14.20. A cup wheel is broadly cylindrical in 
form, but may be formed as part of a truncated 
(cut off) cone. It is moulded with one end 
closed and pierced for a mounting hole. The 
other end is open, and the edge of the cylinder 
is dressed squarely. Typically, a cup wheel 
presents an edge ‘/iin. (12.5mm) wide which 
is intended to be used for grinding. This surface 
is flat, and can be maintained so by dressing, 
and will thus grind a flat surface on a tool 

If an adjustable rest is provided which allows 
the tool to be presented easily at a fixed angle to 
the wheel, the surfaces can readily be ground to 
create the tool angles. The range of angles 
which is listed in Tables 14.1 and 14.2 appears 


at first sight to be very wide, but it will be found 
that just a few values (only) can be used to pro¬ 
duce the whole range of tools likely to be 
required. These angles are 5,10,15,20 and 25 
degrees, and if means are available to set the 
adjustable rest accurately at these angles, the 
faces of any tool can readily be ground. 

An adjustable rest is suggested in the view of 
a cup wheel shown in Figure 14.20. Such a rest 
can be set readily if a series of templates is 
made, say about lVixlin. (40 x 25mm) in 20 
swg (1mm) material each having two corners 
sawn and filed to one of the angles. With the 
rest loosely held by its clamp, a template can be 
stood on the rest and the rest adjusted so that 
the template determines the angle which the 
rest makes with the wheel’s grinding face. The 
rest can then be clamped to its stand. 

Figure 14.21 illustrates how an adjustable 
rest is used to grind a knife tool from a high¬ 
speed tool bit. In Figure 14.21A, the rest has 
been set using a 10-degree template and the cup 
wheel is used to grind the front clearance and 
end relief angles for a right-hand tool. Front 
clearance is determined by the setting of the 
rest, and the relief angles are created by push¬ 
ing the tool towards the wheel at an angle of 
about 20 degrees. This is shown in the plan 
view of Figure 14.21 A. 

Note that the direction of wheel rotation is 
downwards past the rest, and the wheel is 
grinding the tool downwards from the cutting 
edge. This ensures that any burr left by the 
grinding process is on the unused (trailing) side 
of the tool. Cutting edges should always be 
ground in this way. 

Once the end of the tool blank has been 
ground, the side clearance is ground to the same 
10-degree angle, as shown in Figure 14.21B. 
Note, from the plan view in Figure 14.21B that 
the tool is maintained in line with the grinding 
edge of the cup during this operation. 

Finally, the rest is reset and the tool is turned 
to the position shown in Figure 14.21C and its 
associated plan, for grinding the side rake. 



A Grinding front clearance and side and front relief 

A View in direction of arrow A 

Cup wheel | Cup wheel 

Adjustable rest 

B Grinding side clearance 

A View in direction of arrow A 

to machine more readily if negative top rake is 
given. That is, the top of the tool is ground 
away so that the point of the tool is lower than 
the tool body. Up to 10 degrees of negative rake 
is sometimes used. 

Surface finish on the tool is important in 
promoting good finish on the work and 
although a coarse grinding wheel may be used 
to create the basic tool shape from an HSS tool 
bit, the final process must aim to produce as 
good a finish as possible on the tool. This must 
at least be performed on a fine wheel, but even 
this may not be entirely satisfactory in some 
instances. Honing of the surface of the tool by 
hand, using a fine slip stone is the only satisfac¬ 
tory means to create the required finish. This 
process should not destroy the carefully ground 
rake and clearance angles and a jig is frequently 
recommended for hand use to ensure that cor¬ 
rect angles are maintained and sharp edges do 
not become rounded. 

C Grinding side rake 

A View in direction of arrow A 

Figure 14.21 The procedure for grinding a tool using the 
adjustable rest. 

When grinding up tools, note particularly 
that brass requires little or no side rake and the 
tool can therefore be flat-topped, it being nec¬ 
essary to grind only the clearance angles and 
ensure that there is side and front relief (of 
about 15 degrees). Some brasses may be found 


In use, wear occurs on the top of the tool, due 
to abrasion in the area on which the com¬ 
pressed chips bear down. It also occurs on the 
front and side of the tool, in the areas just 
below the point, as shown in Figure 14.22. 
Sharpening of a knife tool is quickly carried out 
by grinding back the front face of the tool to 
remove the worn point (on all three faces 
simultaneously) and thus create a new cutting 
point without grinding the top surface. If the 
top surface is truly flat (has zero top rake) there 
is no change of tool height after resharpening 
and the tool may readily be mounted back in 
the toolpost and turning continued. 

If the tool has top rake, there is naturally a 
small loss of height at the point and this must be 
allowed for when remounting the tool. Never¬ 
theless, resharpening should always be done on 
the end, since it is in its length that the tool has 
its longest life. It is also best performed little 



and often since there is then little wear to take 
out. The job is then quick and easy and the tool 
is never allowed to become really blunt. 

If tungsten carbide tools with brazed-on tips 
need to be reground, this must be done on a 
green-grit wheel. The need for regrinding 
means that the tip is chipped and it is usually on 
the top surface that this is evident. Since the 
remainder of the top surface is smooth and 
highly polished, and already established at the 
correct rake angles, such tools should also only 
be ground on the front and the side to establish 
a good cutting edge. 

Height of the tool 

Since the rake angles are important in deter¬ 
mining how well the tool cuts, and also what 
quality of finish is achieved, it is important that 
the correct angle is created and maintained. In 
order to maintain the rake angles as ground on 
the tool, it is essential that it should be mounted 
at exactly the centre height of the machine. 

Figure 14.23A shows a tool with both side 
and top rake in the correct position in relation 
to the workpiece. The point is at centre height 
and this establishes the top rake angle correctly. 
Correct front clearance is also achieved by this 

If the tool is too high (Figure 14.23B) the 
top rake is increased, but more seriously, the 
front clearance angle is reduced and the tool 

point may not be touching the job and the tool 
will then merely rub on the work. 

As Figure 14.23C shows, incorrectly setting 
the tool too low seriously reduces the rake an¬ 
gle and although the front clearance is in¬ 
creased, and the tool will not rub, poor per¬ 
formance and poor surface finish may result. 
Additionally, when taking a facing cut, the tool 
will not reach the centre of rotation of the work 
and so will leave a pip in the centre. 

Attempting to turn with the tool too low is a 
recipe for disaster. Top rake is seriously 
affected, so the ability of the tool to cut the 
work is reduced and it may therefore dig in 
rather than dislodge a chip. If this happens, the 

Figure 14.23 The effect of incorrect tool height. 



workpiece will not necessarily stop rotating (it 
has a 'A horse power motor behind it) and it 
therefore climbs up over the tool. If it is small 
in diameter, it will bend and then jam up. If it is 
relatively strong (large in diameter) the jam¬ 
ming may occur instantly, but whichever way it 
is, the lathe is now stalled with the job pushed 
up above the tool and great strain is therefore 
placed on the chuck jaws and chuck body, the 
mandrel and front bearing cap and also on the 
tool, toolpost and topslide. Dig-ins should al¬ 
ways be avoided. 

Taking a facing cut is quite often the means by 
which the operator determines whether the tool 
is at centre height, the too-high or too-low con¬ 
dition being decided by determining whether 
the tool does reach the centre of the work, or 
passes below or above. This cannot be done if 
the work is hollow or is mounted between cen¬ 
tres, and a simple height gauge may be made to 
suit the centre height of the machine. 

It is frequently recommended that the tool 
should be set slightly high to compensate for 
the relative movement between tool and work- 
piece when cutting takes place as there is 
always bound to be some deflection of both 
items. The amount of recommended offset is 
small, however, the usual figure being between 
1 and 4 per cent of the diameter of the 
workpiece. It is doubtful if any real benefit is to 
be gained from this type of setting. The final 
test is always an empirical one, “Does the tool 
cut and produce the required finish?”. In any 
event, there are other factors which are prob¬ 
ably more important and, in any case, proper 
support should always be given to both the tool 
and the work in order to minimise any deflec¬ 
tions which occur. 

When surfacing other than purely cylindri¬ 
cal work i.e. when turning tapers or using form 
tools, or when screwcutting, tool height affects 
the final shape produced and correct tool 
height is vital for such operations. 

Cutting speeds 

The importance of maintaining the correct 
wedge angle on the tool to ensure clean cutting 
and a good finish is emphasised above. A fur¬ 
ther factor affecting the cutting action is the 
speed of the workpiece past the tool. 

It will be appreciated that some materials are 
naturally easier to machine than others, the 
softer or more brittle materials being easier on 
the tool which is therefore capable of removing 
material more rapidly. Different materials are 
said to have different machinability, four or five 
groupings being used by way of classification. 

Speed is important because different materi¬ 
als cut best at different speeds and unless cut¬ 
ting takes place at approximately the correct 
rate, the surface finish may be much poorer 
than could be achieved and the power required 
to remove material may increase. This is some¬ 
times noticeable if facing a large diameter, 
when it may be found that cutting is good at 
larger diameters but as the centre is approached 
the cut becomes ragged and uneven and mark¬ 
edly poorer results are achieved. This is because 
the cutting speed towards the centre is too low. 

Cutting speed is the peripheral speed of the 
workpiece at the diameter being cut, measured 
in ft/min. i.e. the circumference, in feet, multi¬ 
plied by the rotational speed (rpm). If you are 
used to working with metric units, the working 
circumference must be calculated in metres to 
yield a cutting speed in metres/min. 

The problem with attempting to recommend 
cutting speeds is that several factors are 
involved. These include the physical and mech¬ 
anical properties of the workpiece material, the 
rate of feed of the tool into the cut, the depth of 
cut, the tool size and point angles, and the tool 
wear which can be allowed before regrinding is 
required. Nevertheless, as described above 
there is a good cutting speed for each material 
and it is therefore useful to provide a guide. Due 
to the number of variables, different authorities 
tend to list different speeds depending upon the 



characteristics considered to be of prime impor¬ 
tance. However, the cutting speeds shown in 
Table 14.3 are either within the range of speeds 
recommended by the references consulted, or 
are very close to the figures, in cases where a 
very narrow range of speeds is specified. 

Although as amateurs we are not that inter¬ 
ested in maximising tool life or production 
rates, it is wise to adopt cutting speeds which 
are within the reasonable capacity of the tool 
since in any event, time spent resharpening is 
also non-productive for us. 

The table contains recommended speeds for 
turning when using high-speed steel (HSS) tools. 
If tipped (tungsten carbide) tools are used, 
speeds may generally be increased, a factor of 
two or three times being adopted commercially. 
This is possible due to the harder cutting point 
which the tip possesses and to the different 
wedge angle ground on the tool. 

Table 14.3 should also be taken to refer to 
plain turning or boring. When parting off (or 
using form tools) speeds must be reduced due 
to the width of cut being used, a reduction to 
50 per cent or even 25 per cent of normal cut¬ 
ting speed being desirable, depending upon the 
material and the actual width of cut. 

As an indication of the mandrel speeds 
(rpm) which might be used in practice, a simple 
table can be constructed to allow selection of 
basic speeds. This is shown in Table 14.4. 

Table 14.3 Recommended Cutting Speeds (HSS Tools) 


Cutting Speed 

Bright mild steel 

85 to 100 ft/min 



Phosphor bronze 

(26 to 30 m/min) 

Stainless steel 

60 ft/min 

Silver steel 

(18 m/min) 

Cast iron 

70 ft/min 
(21 m/min) 


Nickel silver (90 m/min) 
Aluminium alloys 

300 ft/min 

Table 14.4 Cutting Speed, Diameter and rpm 



Cutting Speed ft/min 
100 85 70 


!Ain. (3mm) 






'/•■in. (6mm) 






! /fcin. (9mm) 






'/jin. (12mm) 






5 Mn. (16mm) 






'/<in. (19mm) 






1 in. (25mm) 






2 in. (50mm) 






4 in. (100mm) 






6 in. (150mm) 






Cutting fluids 


When material is cut, there is naturally relative 
movement between the tool and the work. The 
resultant friction causes both the tool and the 
work to heat up, the heating effect increasing as 
the rate of metal removal increases. The tem¬ 
perature rise causes expansion of the work and 
the cutter, and this may lead to dimensional 
inaccuracy in situations in which the tool is cut¬ 
ting continuously. In an industrial production 
process, where the process is automated and 
continuous, the machine settings may need to 
be changed during any warm-up period. 

Using modern tools, there may be no other 
effect which needs consideration, but in the 
days when tools were made from carbon steel, 
significant heating could not be allowed since it 
could cause the tool material to be tempered to 
a softer condition. Rapid tool wear occurred 
and machines had to be stopped frequently for 
the tool to be changed. This serious disadvan¬ 
tage of carbon steel tools naturally limited the 
rates of metal removal which could be achieved, 
but the problem was mitigated by spraying a liq¬ 
uid coolant onto the work continuously. 

In spite of the adoption of the present-day 
types of high-speed steel tools, given this name 



simply because they do not lose their hardness 
at high temperatures, the practice of spraying a 
coolant onto the tool has developed, rather 
than diminished because cutting fluids lubri¬ 
cate, as well as cool, and this results in a better 
finish, longer tool life, lower temperatures, bet¬ 
ter accuracy, and so on. 

If all this talk of industrial processes seems 
somewhat esoteric, put a piece of s /sin. (16mm) 
silver steel into the lathe next time you have a 
spare moment, and set the lathe to run at 300 
rpm. With a sharp knife tool in the toolpost, 
take a cut of .030in. (0.75mm) and force the 
tool into the cut firmly using the saddle 
handwheel. Watch the colour of the turning 
develop as the cut progresses, and you should 
see it turn gradually from silver to straw and 
finally to blue as the tool and the work heat up. 
An experiment such as this should convince 
you that coolants were a necessity in the days of 
the widespread use of carbon steel tools. 

To be really effective, the cutting fluid needs 
to be present in large quantities, right at the 
point of cutting. This means that a pumped 
supply is necessary, and the machine has to be 
provided with a drip tray so that the overflow 
can be contained. Ideally the used fluid is col¬ 
lected, filtered and returned to the reservoir so 
that it may be reused. Thus, specialised equip¬ 
ment and/or adaptation of the machinery is 
necessary if the full benefits of using a cutting 
fluid are to be enjoyed. Some shielding of the 
machine is also necessary, since fluid can be 
thrown around in large quantities. 

The use of cutting fluid in the sort of quantity 
necessary for effective cooling needs special 
provision, but in the amateur’s workshop, 
continuous ‘flat-out’ cutting is unlikely to be 
required and so the need is not so much for 
cooling as for the improvement in surface finish 
which is achievable. It is not necessary to flood 
the work with coolant to achieve this, and some 
judiciously applied lubricant can be helpful, 
either applied as a mist spray, or as an occasional 
squirt from an oilcan, or applied by brush. 

Types of fluid 

Over the years, the development of cutting 
fluids has led to the use of a wide range of 
substances, from ordinary paraffin and other 
mineral oils, mineral and fatty oil mixtures and 
a range known as sulphurised oils. These latter 
fluids themselves comprise a wide selection 
with different uses, which vary according to the 
amount of sulphur they contain. 

Due to its wide use for machining steel, the 
best-known cutting fluid is soluble oil. For use, 
this is mixed with water to produce a white 
fluid which is known universally as ‘suds’. The 
oil content provides the lubrication and the 
water acts as the coolant, thereby satisfying 
both needs. The lubricating effect is naturally 
dependent upon the dilution, the traditional 
strength being determined by the relative costs 
of the two constituents - that is, mostly water! 

The objections to soluble oil as a cutting fluid 
are its smell, which is distinctive, to say the least, 
and the fact that the water content tends to 
promote corrosion. There is also the problem 
that large quantities are required for effective 
cooling, and something a little more oily is pref¬ 
erable when the need is most likely for lubrica¬ 
tion. There are, consequently, straight (not solu¬ 
ble) cutting oils which do not have an objection¬ 
able smell and which provide effective lubrica¬ 
tion even when present in small quantities, and 
these are to be preferred if a full pump-and- 
recovery arrangement is not provided on the 
machine. The pumping equipment is generally 
known as a suds pump. 

Choice of fluid 

Since the purpose of a cutting fluid is to assist 
removal of material from the work, it follows 
that the way in which the chip forms influences 
the type of fluid which is used. 

Some materials, such as cast iron and brass, 
machine very cleanly without the benefit of 



coolants. Most brasses cut very cleanly, and a 
good finish is produced, even at high speed. 
Cast iron provides a good finish due to the 
presence of graphite, which acts as a lubricant 
during machining. 

Industrially, special types of soluble oil are 
used when machining cast iron, but they have 
to be present in large quantities to wash away 
all traces of swarf, which is highly abrasive and 
must not be allowed to enter the bearings or 
slideways of the machine since it will there do 
great damage. 

As large quantities of coolant are not nor¬ 
mally available in the amateur’s workshop, and 
also since cast iron is only one of several mate¬ 
rials which are dealt with, it is normally ma¬ 
chined without the use of any coolant. Because 
the swarf is highly abrasive, it is usual to wipe 
any lubricant off the machine’s slideways and 
keep the swarf away as far as possible, in order 
that the oil and cast iron dust do not mix to 
form a grinding compound. It is also useful to 
arrange a shield to collect and direct the turn¬ 
ings to where they can be better dealt with and 
a stiffish piece of plastic sheet can be taped to 
the cross-slide to achieve this. 

Steels are very widely used and much 
machining of them is therefore undertaken. 
Steels are also tough and harder on the tool, 
tending to blunt it more rapidly and also heat it 
up. Industrially, a coolant is invariably used 
when machining steel, the most common type 
being soluble oil. If your machine is equipped 
to deliver the oil in sufficient quantities, and 
the smell poses no problem, then soluble oil 
should be used. If you can manage to cope with 
small quantities of the oil, then a drip feed or a 
mist spray can be beneficial in providing lubri¬ 

Aluminium and its alloys, although soft and 
easy to machine, have an annoying tendency to 
cause their rather ‘stringy’ turnings (swarf) to 
weld themselves to the tool’s cutting edge or 

tip. This naturally masks the point and impairs 
cutting and hence spoils the surface finish. For 
this reason, paraffin is normally used as a lubri¬ 
cant when machining aluminium alloys. 

When to use a coolant 

Given that the workshop may not be equipped 
with suds equipment, coolants are unlikely to 
be used habitually. When they are, it is their 
lubricating properties which are usually desir¬ 
able, and it is normally possible to apply suffi¬ 
cient fluid either by brush, oil can or drip feed. 

Ordinary turning can be performed at 
speeds that are sufficiently low to provide the 
surface finish which is required. Paraffin may 
be used when turning aluminium alloys to help 
prevent the swarf building up on the tip of the 
tool. Here, the tool needs lubricating and the 
paraffin can be applied to it, rather than the 
work, otherwise the high speeds which are used 
ensure that there is a good distribution of the 
coolant around the workshop and this is obvi¬ 
ously not desirable. Build-up also occurs on 
drill points and paraffin can also be used when 
drilling aluminium. 

Another operation which benefits from the 
presence of a lubricant is the drilling of deep 
holes in phosphor bronze. This material is diffi¬ 
cult to machine and a great deal of force must be 
applied before a chip will break away. There is 
much frictional heating as chips are deformed 
by the cutting edges and the resultant expansion 
can cause the drill to bind in the hole. A cutting 
fluid is useful both for its cooling and lubricat¬ 
ing effect. 

During turning, it is parting-off large- 
diameter work which leads to deep penetration 
of the tool into the work. Since the tool has 
very little clearance in the groove it is cutting, 
there is much friction and the operation once 
again benefits from the use of a cutting fluid. 



Basic lathe practice 



The basics of what a lathe can do are described 
at the beginning of Chapter 12, together with a 
description of the major parts of a lathe and 
their principal features. The lathe is capable of 
carrying out both turning and milling opera¬ 
tions although it is for the process of turning 

Figure 15.1 The tool used for facing this gunmetal casting is 
set too low in the toolpost and has left a small pip as it has passed 
across the work below centre height. 

that it is actually designed. Turning processes 
include plain turning, threading or screw¬ 
cutting and the production of form or taper 
surfaces. This chapter provides an introduction 
to those basic techniques of lathe utilisation 
which are common to all turning operations 
and the technique of screwcutting is among the 
topics described in Chapter 16. 

Setting up and supporting the tool 

The basic turning operations of facing and 
surfacing are described and illustrated at the 
beginning of Chapter 14, which also describes 
the basis for the grinding and sharpening of 
lathe tools to carry out these basic operations. 

Before a tool can cut correctly, it must be 
mounted in the toolpost so that its cutting point 
is exactly at centre height i.e. it must be possible 
for the point to pass through the rotational 
centre of the work. To achieve this, the tool is 
supported in the toolpost on suitable packing 
which brings the tip to correct height. Several 
strips of differing thickness are used to adjust 
the tool’s position, and a stock of tool-size 



Figure IS.2 The packing below the tool must be arranged to 
give proper support, with all of the loose strips aligned carefully. 

pieces in different thicknesses needs to be kept 
so that tools can be packed up correctly. 

A quick way to set the tool is to mount it up 
in the toolpost, approximately at the correct 
height, and then to take a facing cut across the 
work. Figure 15.1 shows a square block which 
has been faced off using a tool set so that the 
point is just below centre height. The result is a 
small pip of unmachined material at the centre, 
indicating that the tool has passed across the 
face of the work just below centre height. 

Figure 15.2 shows a correctly supported 
tool. Several metal strips have been used to 

Figure 15.3 The load on tool and work, when cutting, causes 
the two to move away from one another, causing the cut to vary. 

pack up the tool to the correct height. Their 
ends have been carefully aligned and the stack 
is positioned so that the full support offered by 
the toolpost base is utilised in supporting the 
tool. Provided that suitable strips are available, 
the packing is easy to arrange, but care must be 
taken to ensure that the tool is correctly 

Deflections of tool and work when cutting 

The situation when turning is shown in Figure 
15.3. When the tool is cutting, the uncut part of 
the work presses on the tool, causing it to bend 
down away from the work. However, from the 
work’s point of view, the tool is pressing against 
it, and it, too, bends, but this time away from 
the tool. This means that in the initial stages of 
the cut, or when facing, both the tool and the 
work are being forced apart and experience a 
‘load’ on one end. Both are supported only at 
the other end, and the amount of bending 
which occurs is related to the unsupported 
length, in such a way that the longer the over¬ 
hang, the greater is the deflection. Put another 
way, this means that the least deflection occurs 
when the unsupported lengths are short. In 
practical terms, this means that the workpiece 
must not extend very far from the workholding 
device (chuck etc.) and the tool must be prop¬ 
erly supported and not extend from the 
toolpost any further than is absolutely neces¬ 
sary. In other words, a short, stiff arrangement 
is required. 

Due to the deflections which occur, the tool 
and work are pushed away from one another, 
when the cut is actually taking place, and a 
slightly smaller cut is taken than would be the 
case were there to be no deflection. If the tool is 
put into the job twice at the same setting, it 
removes more metal the second time around, 
the actual amount being dependent on the size 
of the initial cut, the sharpness of the tool, the 



Figure 15.4 The deflections of tool and work may be 
sufficient to reduce the amount of material which is removed by 
the tool passing down the work. In these cases, a second pass of 
the tool along the work can result in a further cut being taken 
without advancing the tool any farther. 

material being machined and also the rigidity of 
the set up. 

The effects of tool deflection are illustrated 
in Figure 15.4 which shows the end of a long 
bar which is being turned, supported on the 
tailstock centre. Two effects are visible here, 
both resulting from passing the tool twice over 
the bar at the same setting, that is, without 
adjusting the position of the tool in relation to 
the work. 

The end V4in. (12.5mm) of the work has 
been turned twice, once as part of a cut taken 
along the length of the bar, and again, by slowly 
traversing the tool over the end portion. The 
polish imparted by this second cut is evident, 
and so is a slow left-hand spiral caused by 
winding back the tool to the tailstock end with¬ 
out withdrawing it from the work. 

The second cut taken by the tool is quite 
evident and this means that once the tool has 
traversed the turned surface in the correct 
direction, it must be retracted to clear the work 
before being returned to the start of the cut, 
when further feed may be applied and another 

cut taken. This is the reason that feedscrew 
dials are so valuable - without them it is only 
guesswork that allows a cut to be put on, 
whereas their use means that precisely set cuts 
can be taken every time. 

The load on tool and work is obviously 
dependent on the amount of material which is 
being cut, since a greater area of the work is 
pressing down on the tool when a greater cut is 
being taken. This area depends on both the 
depth of cut and the rate of feed of the tool into 
the work. The depth of cut is set by advancing 
one of the feedscrews by a known amount, and 
the rate of feed is determined by how fast the 
other feedscrew handle is turned. If the feed 
rate varies during the cut, the load and deflec¬ 
tions also vary, and so, therefore, does the 
amount of material which is removed. An 
uneven feed rate results in a ridged and rough 
surface, so if a smooth finish is required a 
steady rate of feed must be used. 

If the tool suffers a deflection when making 
the ‘going in’ cut, it is still deflected, and press¬ 
ing in on the work, when it reaches the end of 
the cut. If it is held at this position for a 
moment or two, it cuts itself a groove as it tries 
to regain its natural shape. As the cut dimin¬ 
ishes, the conditions for causing chatter are 
produced and if the tool is held too long at this 
position, the characteristic singing noise is 
likely to be heard, and the finish is spoiled. The 
problem of chatter is described below. 

If the handwheel is used to traverse the tool 
from the end of the cut back to the beginning 
without retracting it from the work, it takes 
another cut ‘coming out’, as illustrated in 
Figure 15.4. If the work is reaching the finished 
size, this second cut may be a severe embarrass¬ 
ment, to say the least. The surface finish on the 
work may well leave a little to be desired since 
the tool is not cutting at its most efficient when 
coming out. The tool should, therefore, gener¬ 
ally be retracted from the work before it is 
wound back to the beginning. 



If graduated dials are fitted to the feed¬ 
screws, it is a simple matter to withdraw the 
tool before winding back to the start, and then 
to advance the tool precisely by a known 
amount for the next cut. 

Retracting the tool before returning to the 
start of the cut may not always be absolutely 
necessary, since a roughing-out stage, often pre¬ 
cedes the finishing stage, and it clearly doesn’t 
matter if further material is removed when 
winding back - except, of course, that present¬ 
ing the tool to the work going the ‘wrong’ way, 
might well blunt it more than taking a normal 
cut. If finishing is being carried out, it obviously 
does matter, since a good finish and an 
approach to the final size have been (or should 
have been) planned, and it is important that 
more or less the same tool deflection occurs on 
every cut. 

Graduated dials are also helpful in allowing 
a definite approach to the finished size to be 
made without stopping to measure too often, 
since if the micrometer shows that the work is 
.050in. (1.26mm) over the required size, it is 
known immediately that the final cut will be 
taken with the tool advanced a further .025in. 
(0.63mm) into the work, give or take any dif¬ 
ferences in the tool deflections. 

Depth of cut and feed rate 

The rate at which material can be removed 
from the workpiece is dependent on the mate¬ 
rial which is being machined (some materials 
are softer, and easier to cut, than others) but the 
maximum rate is limited by the amount of 
power which is available to drive the work into 
the cut. If a deep cut is ‘put on’ and a high rate 
of feed is applied, the lathe mandrel may stall, 
since the power from the motor, transferred to 
the mandrel, is insufficient to remove material 
at the rate which the operator is trying to 
impose. Phrases such as “deep cut” and “high 

rate of feed” are exceedingly unhelpful in pro¬ 
viding guidance, and it becomes a question of 
how deep is “deep”?. 

In the descriptions of the basic turning proc¬ 
esses presented in Figures 15.17 and 15.19, a 
cut of .OlOin. or .015in. is suggested for turn¬ 
ing bright mild steel. Provided that the tool is 
sharp, a cut of .OlOin. should be supportable, 
even on a small lathe, and this can be consid¬ 
ered as a suitable value from which to start. On 
a metric machine, this means taking a cut of 
0.25 mm. 

Once the depth of cut has been set, the feed 
into the cut is applied using the other feed¬ 
screw. If this is turned by hand, the operator 
obtains ‘feedback’ to judge how the cut is going 
- the ears can detect a change of speed of the 
mandrel very easily, which is often the indica¬ 
tion that the cut is too heavy. 

The most important requirements are a firm 
feed and a steady rate, since it is only by main¬ 
taining the deflections of tool and work at the 
same values that an even cut and smooth finish 
can be obtained. The feed is always stated as the 
distance the tool travels into the cut during one 
revolution of the work i.e. the feed per rev. If 
.OlOin. is suitable as the depth of cut, it is also 
suitable as the feed per rev. Taking the example 
of the pivot bolt illustrated in the sequence of 
Figure 15.17, the bar is turning at about 300 
rpm and .OlOin. per rev means 3 inches 
(75 mm) in each minute, or 20 seconds to cut 
along 1 inch (25mm) of the work. 

It is not necessary to use a stopwatch, but 
this figure does give a guide to the rate of feed 
which can be used on a bar turning at 300 rpm. 
If it is turning faster, the rate of traverse along 
the work can be faster to achieve the same rate 
of feed per rev. 

The figure of .OlOin. per rev (0.25mm per 
rev) for the rate of feed can be considered as 
one which may be used for finishing the work 
(a finishing cut). If the object is simply to 
rough-out the shape of the work (a roughing 



cut) then the rate of feed may be greater, up to 
the maximum which the power available will 
allow. The depth of cut may also be greater 
than .OlOin. (0.25mm), and .020in. or 0.5mm 
can readily be used when the need is to remove 
metal more effectively. At the other extreme, 
cuts less than .005in. (0.125mm) should not 
generally be used since small cuts can give rise 
to chatter (see below). A large lathe can natu¬ 
rally support a deeper cut and 0.1 in. or 
0.25mm might well be used during a roughing- 
out operation. 

Backlash and endfloat 

When considering the feedscrews of the cross¬ 
slide and topslide, which are used to ‘put on’ 
the cut, it must be appreciated that there is 
inevitably some free play in the arrangement. 
This means that the cross-slide, for example, 
has a small amount of endfloat which may be 
large enough to be detected by pulling it back¬ 
wards and forwards. 

If the gib strips and screws are correctly 
adjusted, the endfloat should not be discernible 
simply by attempting to move the cross-slide 
directly, but if a to-and-fro motion is given to 
the feedscrew handle under these circum¬ 
stances, a dead zone will be detected in which 
handle rotation does not result in motion of the 

This lost motion, or backlash, means that if 
the feedscrew is fitted with an engraved dial to 
indicate its setting, the dial reading for a given 
position of the tool relative to the work, will 
not be the same when moving the cross-slide in 
as when moving it out. When setting the cut or 
tool position using a feed handle and relying on 
the dial to indicate position, the feed must 
always be in the same direction. It is natural to 
take out the backlash before commencing the 
cut and it is usual to move the tool into the 
work to take the first cut, initially just advanc¬ 

ing the tool to touch the work and then rotating 
the feedscrew handle further in the same direc¬ 
tion each time a cut is taken, not forgetting the 
need to retract the tool before returning it to 
the start of the cut. 


Consideration of the deflections of tool and 
work, and the backlash which is always present 
in the feedscrews, naturally brings us to the 
turner’s number one enemy - chatter. This is 
characterised by a singing noise from the tool 
when turning, and a ridged surface finish to the 
work on completion. Once the surface has 
become ridged, it is frequently very difficult to 
render it smooth once more, and if you are 
nearing its final size, the work may conse¬ 
quently be spoiled. 

There are several root causes for chatter, but 
the ridged appearance of the work is caused by 
springiness in both the tool and the work. If the 
tool and the work are both large and substan¬ 
tial, the deflections which occur when the tool 
is cutting are small, and consequently cause no 
problem. However, any deflections do tend to 
push the work and the tool apart, which 
reduces the depth of cut and thus also reduces 
the force being applied to both items. 

The scenario can go something like this: the 
cut starts and the tool and the work both bend 
under the load. As the two move apart, the cut 
is reduced slightly, reducing the load on work 
and tool, and allowing them to come closer 
together. This naturally increases the cut, 
which... and so on. 

Since the cut is increasing and decreasing, 
the work naturally has a ridged surface. This 
produces a varying depth of cut on the next 
pass along the work, which varies the load on 
tool and work etc. The chatter thus tends to be 
self-perpetuating, since the conditions which 
first caused it are still likely to be present. 



Figure 15.5 The situation to be avoided. Both tool and work 
are projecting too far from their holding devices, and large 
deflections will result if an attempt is made to take a cut on the 

The general cure for chatter - prevention 
might suggest the true requirement - is to 
reduce the deflections as far as possible. This 
means supporting the work, supporting the 
tool, and only using a correctly ground and 
sharp tool, properly aligned with the work. 

Figure 15.5 illustrates a highly undesirable 
situation. There is a very long extension of the 
work outside the chuck and a long overhang of 
the tool outside the toolpost. Under conditions 
of large overhangs, the deflections caused are 

Figure 15.6 The ideal situation. The projections of tool and 
work are the smallest possible. 

significantly greater than for short overhangs, 
so the first requirement is to reduce any long 
extensions, or provide support when this is not 
possible. Figure 15.6 shows a vastly improved 
situation, which should be the one aimed for. 

Extra support for the tool is not very con¬ 
veniently arranged, so stiffness here must be by 
using a more robust tool. Support for the work 
can readily be provided either by using a centre 
in the tailstock, or by using a fixed or travelling 

In spite of achieving sensible support for the 
work, and using a stiff and short tool, chatter 
can still occur, which means that the forces act¬ 
ing on the tool and the work are still too high. 
This arises because there is too large a contact 
between the work and the tool, accounting for 
the chatter which is frequently produced when 
using a parting-off tool. This naturally has a 
large contact with the work since a groove of 
material is being removed. 

An ordinary tool can, however, easily 
present a large contact with the work, princi¬ 
pally if it is incorrectly ground, is blunt, or is 
incorrectly set in respect to the work. These 
conditions are illustrated in Figure 15.7. At A, 
the tool has been ground with insufficient end 
and side relief and as a consequence both front 
and side lie at too shallow an angle in respect to 
the work. 

At B, the tool, although correctly ground, 
has become worn, producing a flat on the front. 
At C, although the tool is correctly ground, it is 
not set with proper end relief of between 5 and 
10 degrees. All of these conditions increase the 
contact between the work and the tool, conse¬ 
quently increasing the load and the deflection 
and encouraging chatter to start. In these cases, 
the remedy is obvious; grind the tool properly 
and set it correctly in the toolpost. 

If care has been taken in setting up the work 
and the tool to minimise overhangs and ensure 
correct clearance and relief angles for the cut¬ 
ting point, chatter arises most commonly from 



small changes in the overhangs of the tool and/ 
or the work, assuming these are possible. 

The singing noise (chatter) which accompa¬ 
nies the production of the ridged surface is 
produced by vibration (oscillation) of the work 
and the tool. The frequency of the vibration is 
set by the length of the tool overhanging the 
support provided by the toolpost (or the similar 
overhang of the work outside the chuck) 




Figure 15.7 Faults in the 
grinding and setting of the tool 
which may lead to chatter. 

Figure 15.8 This knife tool is blunt and is producing a 
significant burr while facing a gunmetal casting. 

blunt tools. Bluntness can often be felt if the cut 
is being applied by hand, but it comes on gradu¬ 
ally and the increased force which needs to be 
applied may not be noticed. In some situations 
the effect is noticed visually, rather than by feel, 
as shown in Figure 15.8. A knife tool is in use 
for facing a rectangular block, but its traverse 
towards the centre is producing a significant 
burr. That this is due to a blunt tool is without 
doubt, for the result of the same operation after 
the tool has been sharpened is shown in Figure 

If all of the above points (support, tool 
grinding and setting) appear to be in order, but 
chatter still occurs, it can often be cured by 

Figure 15.9 Grinding the tool produces a lean cut, without a 
burr, on the same casting. 



and this frequency determines the distance 
between successive peaks of the chatter marks 
on the work. The speed of the work is also a fac¬ 
tor here, since, as the tool vibrates, it cuts less 
and more on the work and the pitch of the peaks 
and hollows is determined by how far the work 
rotates past the tool in each cycle of vibration. 

So, if chatter occurs, it might be reduced on 
the next cut by changing the overhang of the 
tool and/or the work, or by changing the speed, 
preferably a reduction for both factors. 

One other reason for chatter to develop is 
that you may be ‘tickling’ the work rather than 
taking a proper cut. The tool must cut at all 
times. If it is allowed to rub on the work it 
quickly becomes blunt, creating the conditions 
to allow tool vibration. If a larger cut is taken, 
and the tool pushed firmly into the cut, both 
tool and work are maintained in the deflected 
condition throughout the entire cut, thereby 
preventing the conditions for producing chat¬ 
ter developing in the first place. 

So, try not to be forced into the position 
where you have to take a light cut when 
approaching the final size. Stop the lathe well 
before the final size is reached and measure the 
work so that the remaining cuts to the final size 
can be planned in advance, aiming to take a cut 
of at least .005in. (.125mm), i.e. a diameter 
reduction of .OlOin., as the finishing cut. If you 
are doubtful about this, take a practice finish¬ 
ing cut when still well above the final size to 
check that this will be satisfactory. 

If the contact area between the tool and the 
work is very large, as for instance when using a 
form tool, the possibilities for producing chat¬ 
ter are much increased and the tool must be 
made as strong as possible. An example of this 
is shown in Figure 15.10. A tool with a large 
radius has been used for turning the tread of a 
cast iron wheel for a model locomotive. While 
turning the parallel portion of the tread, the 
radiused tip of the tool is in contact with the 
work, but as the flange is approached, the tool 

Figure 15.10 Chatter can be produced If a large part of the 
tool is used for cutting, as here, where a round-nosed tool has 
been used to form the side of the flange and the root radius on a 
wheel casting. 

cuts on a large part of the left-hand side, 
increasing the load on the tool dramatically and 
creating ideal conditions for the production of 
chatter, with the inevitable results shown in 
Figure 15.10. There is little which can be done 
in these circumstances, beyond employing a 
very large and stiff tool and reducing the speed 
as much as possible, even resorting to pulling 
the mandrel round by hand in extreme circum¬ 

Using the tailstock 


The tailstock is most commonly used for 
centring and drilling, but tap and die holders 
are also used on the tailstock, and so are turrets 
which provide a mounting for several tools. 
The purpose of the tailstock (as a holder) is to 
bring tools up to the work in alignment with 
the lathe axis. The tailstock is also used to 
support the end of the work, particularly work 
which is mounted between centres. The princi¬ 
ple of its use in this role is described in Chapter 
13, and practical aspects of using the tailstock 
for support are considered below. 



Since the sleeve of the tailstock may be 
advanced towards the headstock by the feed 
handwheel, a drill mounted in a tailstock chuck 
can be advanced for drilling an axial hole. To 
do this, the tailstock is brought towards the 
headstock and locked to the bed and the 
feedscrew handwheel used to advance the drill. 

A turret holding several tools may also be 
advanced in the same way as for drilling. A tur¬ 
ret may hold six tools in all and can be fitted 
with a set of items which is needed for a par¬ 
ticular job. It might hold a centre drill, a pilot 
drill, a second drill, a reamer or D-bit, and so 
on, which are used in sequence, the turret being 
rotated and locked to bring the tools succes¬ 
sively in line with the lathe axis. 

If the turret is in use for a production run of 
similar items there is much slow-speed screw¬ 
ing of the sleeve in and out of the tailstock 
barrel, and for some lathes an adaptor can be 
fitted so that the sleeve may be advanced and 
retracted more rapidly by use of a lever. 

Threading may be performed manually but 
tailstock-mounted tap and die holders greatly 
assist proper alignment of the cut threads. They 
are used with the tailstock locked to the lathe 
bed, but in these instances the tool holder is 
usually free to slide along a sleeve-mounted 
arbor and is prevented manually from rotating, 
or turned with respect to the stationary work. 

The operation of putting on screw threads 
from the tailstock is so convenient that the use 
of a tailstock die holder is described later in 
order to establish the general principles of 
tailstock-mounted tool adaptors. 

Centring and drilling from the tailstock 

The tailstock sleeve is normally bored to a 
standard, shallow-angle taper so that accesso¬ 
ries mounted on suitable tapered arbors may be 
mounted directly in it. The most common of 
these accessories is the drill chuck, and by far 

the most common operations performed from 
the tailstock are centring and drilling. 

Since the tailstock sleeve can accommodate 
a chuck, a workpiece revolving with the man¬ 
drel can be drilled. Just as with conventional 
drilling, a long and relatively flexible drill 
cannot be expected to produce a hole along the 
axis of rotation of the workpiece, unless it is 
provided with a centre into which it can com¬ 
mence drilling. For creating such centres, a 
short, stiff centre drill is used. This has a small 
diameter pilot, ground with straight flutes 
which extend away from the point into the 
body of the drill, the end of which is ground at 
60 degrees. The drill therefore produces a 60- 
degree cone-shaped hole in the end of the work 
and the pilot provides clearance at the point of 
the cone. Centre drills are available in a range 
of sizes to suit large or small work and Figure 
15.11 shows a set classified by the references 
BS1 to BS4, which range from '/sin. to 3 /sin. 
(3mm to 10mm) in diameter. 

Centre drills are sufficiently stiff to start a 
hole without initial guidance, but to assist a 
true centre to be formed, the end of the work 
must be faced off, if this is possible, to present a 
smooth, true surface. 

Figure 15.11 A set of centre drills, BSI to BS4, /am. to s /i6in. 
(3mm to 8mm) in diameter. 



Figure 15.12 Putting a centre drill into a large steel disc using 
the tailstock chuck. 

Figure 15.13 Following up the centre drill with a '/»in. (6mm) 

Figure 15.12 shows a s /i«in. (8mm) centre 
drill being used for putting in a centre, an 
operation usually simply called centring. It is 
not necessary to make the centre very deep, it 
simply needs a wide enough ‘mouth’ to accept 
the tip of the drill which is to be used, or to fit 
the centre, if the work is to be supported in this 
way. If a centre is being created in a part for a 
model, it may need to be made the correct scale 

Once a centre has been made, a drill can 
readily be started in the work, and the hole 
opened out as required. Figures 15.13 and 
15.14 show the initial drilling of a !Ain. (6mm) 
hole and its subsequent opening out to Viin. 
(12.5mm). With this larger drill, notice how it 
is possible to judge whether both lips of the 
drill are cutting equally. 

When drilling from the tailstock it should be 
remembered that, on many lathes, the spindle 
speeds do not approach those available on the 
drilling machine, and much ordinary drilling 
can be carried out at the highest, or second- 
highest speed, particularly for drills below !4in. 
(6mm) in diameter. It is usual to open out large 
holes progressively as for normal drilling, since 
this reduces the amount of metal to be removed 
with each cut. 

One point must be made in relation to drill¬ 
ing - the drill does not necessarily remain 
exactly on centre and can be expected to wan¬ 
der from the truly axial path. The wander usu¬ 
ally increases as the drilled hole becomes deeper 
and this should be borne in mind when deciding 
on the best method by which a hole should be 
produced. As a consequence, if a true, axial hole 
is required, it must be produced by drilling 
slightly smaller than required, followed by bor¬ 
ing to the required size. Naturally, a sized finish 
can also be produced by boring to suit a reamer 
and then reaming from the tailstock. In both 
cases, the boring operation is vital since it is this 
which creates the truly axial hole. 

Centre drills are naturally used to drill the 



cone-shaped housings for the centres in the 
ends of work which is to be mounted between 
centres. The centre drill’s pilot provides clear¬ 
ance for the point of the centre and since the 
body of the drill is ground to the same angle as 
the centre (60 degrees included angle) accurate 
location of the workpiece on the centre is 

Thread cutting from the tailstock 

Turned work frequently requires internal or 
external threads to be cut and these are usually 
best done while the workpiece is still held in the 
chuck, or whatever, attached to the mandrel. If 
the thread is required to be truly concentric 
with other turned diameters, it must be cut 
using the screwcutting technique described in 
Chapter 16. 

If the threads are required simply as fixings 
or fastenings, then the use of a tap or die is 
appropriate. Since a small amount of misalign¬ 
ment can usually be tolerated under these con¬ 
ditions, the tap or die can be applied by hand, 
but starting a die squarely is relatively awkward 
and some assistance in maintaining alignment 
is helpful. The time-honoured way to assist the 
alignment of a die when forming a thread in the 
lathe is to lock the tailstock to the bed and jam 
the die holder between the tailstock sleeve and 
the job. Then, with one hand pulling the chuck 
round and pressure maintained on the tailstock 
feed handle with the other, the die may be per¬ 
suaded to start, after which it is just a question 
of whether you wind the mandrel round or turn 
the die holder, or some of each. 

It has to be said that it does work, but it isn’t 
really ‘engineering’ and a much better 
approach is to use a tailstock die holder. This 
consists of an arbor, tapered at one end to fit 
the tailstock sleeve and having a parallel por¬ 
tion on which a cylindrical die holder can slide, 
as shown in Figure 15.15. The die holder is 

about 3in. (75mm) long and knurled over its 
central portion to allow a good grip. This 
particular holder accepts ,3 /i«in. diameter dies 
at one end and lin. diameter at the other so it is 
suitable for the BA sizes and other small- 
diameter threads. 

The holder is drilled to accept a tommy bar 
but this is only needed when cutting larger and 
coarser threads. Adequate torque can be 
applied manually to the body to cut threads up 
to VUin. (5mm) or so in diameter, and the usual 
method of use is to cut the thread while the 
work is revolving, even at quite high speeds. All 
that is required is firm pressure into the job and 

Figure i 5.15 The components of a tailstock die holder and 
the sliding body fitted to its arbor which is mounted in the 
tailstock sleeve. 



the die leads itself along the workpiece quite 
nicely, especially if opened up slightly. You 
should not be foolhardy in relation to speed 
since you need to let go of the knurled body 
soon enough or a nasty friction burn results as 
the die reaches the shoulder on the job (or the 
chuck jaws) and is forced to rotate. However, 
for small threads in brass, or such things as 
boiler stays in phosphor bronze, the one-pass,