85 p

HOW TO BUILD YOUR OWN

BY
f.G. BAYER, T.Eng.lCfll, ASSOC. IERE

R : complete electronic and

I details for simple
je Historie v/d Radiol tion
I
BABANI PRESS No.32
I f
<
i!:

Il

;•

(■

1
K
ri

:.
'I
)

'
 YU K
HOWTO in*
BUILD YOUR OWN
METAL AND
TREASURE
LOCATORS

i \
\
 • : ■

HOWTO
S.V.H-FL
BUILD YOUR OWN
METAL AND
TREASURE
LOCATORS

by

F.G. RAYER, T.Eng.(CEI), Assoc.IERE

BABANI PRESS
The Publishing Division of
Babani Trading and Finance Co. Ltd.
The Grampians
Shepherds Bush Road
London W6 7NF
England.
Although every care is taken with the preparation of this book,
the publishers or author will not be responsible in any way for
any errors that might occur.

©1976 BABANI PRESS

I.S.B.N. 0 85934 035 X

First Published — September 1976

Printed and Manufactured in Great Britain by

C. Nicholls & Co. Ltd.
!
i
CONTENTS
N.V.H.R,
Introduction 7

PART I PRINCIPLES AND CONSTRUCTION

Non-Intcrfcrcncc.............................. 8
Stability........................................... 9
Regulatory Department, Sensible Use 9
Production of Hetrodync................ 10
Hctrodync Locator........................... 12
Metal Finding................................... 13
Practical Circuits and Construction . . 15
Soldering........................................... 16
Transistor Leads................................ 17
Reference Oscillator........................ 17
Testing and Fault Tracing................ ■T2
Search Coil Oscillator........................ 27
Coil................................................... 30
Testing.............................................. 31
Handle.............................................. 32
Iletrodyne Detector........................ 33
Audio Amplifier....................................... 34
Completed Locator................................ 37
Pitch Control.......................................... 38
Locator Use...................................... 39

PART II ALTERNATIVE COILS, DETECTORS,

AMPLIFIERS
Frequency Stability................ 43
Broadcast and Wound Coils . . 45
Air-Cored Coils...................... 47
SEARCH COIL ASSEMBLIES
Shielded Coil........................ ' 53
Two-Terminal Coil................ 55
Square Coil........................... 57
Basket Coil.............................. 59
Underwater Coil...................... 60
ALTERNATIVE DETECTORS
Transistor Detector................ 62
Detector Faults...................... 65
AUDIO AMPLIFIERS
Headphone Amplifier.............. 66
Two-Stage Amplifier.............. 69
Three-Stage Amplifier........... 71
IC Amplifier........................... 73
PART III CONSTRUCTIONAL AND OTHER
POINTS
Search Coils and Handle . . . . 77
Case...................................... 78
Very High-C Loop................ 78
450-470 kHz Working........... 80
Oscillator Stabilisation 82
Kind of Metal........................ 83
3-TRANSISTOR LOCATOR
Search Coil.............................. 83
Coil L,, Circuit and Tag-board 86
Alternative Coil...................... 87
Case........................................ 88

1
INTRODUCTION

The use of electronic metal detectors or treasure locators has become

<|uite an extensive hobby. During out-door seasons of the year, searches
are made on the seashore, or over waste ground or the suspected sites of
ancient hoards. Electronic devices of this kind detect the presence of
buried metal objects, but do not of course respond to objects which arc
not metallic in character, such as pottery.

hinds made by metal locators over the past few years are so numerous
that it is impracticable to try to list or mention them. It is for this
reason that there is quite an extensive trade in ready-made locators.
These arc often expensive items, and it is often not easy to see why
they are so costly. In fact, the heterodyne locator is a comparatively
simple piece of equipment, and no one should experience any real
difficulty in following either its method of working, or in constructing
such a device. A side by side comparison between a ready-made locator
and a comparable home-built detector will normally show that they
give exactly the same results.

There are also of course more mundane uses for such instruments, such
as finding buried pipes or manhole covers. Methods of using such
locators are fully explained here, as well as principles of operation,
and actual construction.

7
PARTI
PRINCIPLES AND CONSTRUCTION

A heterodyne metal detector must produce an emission which is present

beyond the confines of the device. This may he likened to the trans­
mission of a low-powered radio signal, and it is clear that in certain
circumstances this signal could cause interference to the reception of
broadcast programmes. For this reason, certain requirements need to he
met, if the chances of causing interference arc to be kept to a minimum.

NON-INTERFERENCE

If high power oscillator or radiating circuits were used in the detector,

this would obviously increase the chances of causing interference.
Where the metal detector would be used in locations far removed from
houses, the chances of actually causing interference to radio reception
would be small. But where searches are made at no great distance from
dwellings, interference would be more probable. This would take the
form of whistles or audio tones, heard in conjunction with wanted radio
programmes, and spoiling their reception.

Fortunately there is no disadvantage in the use of low-power oscillators,

as the range of detection of an object does not depend on the power of
the oscillator. The regulations for such equipment specify that the
radiated field strength of any signal produced should not be stronger
than lOOuV/m (100 microvolts per meter) at a distance of 100 yards
from the loop. It is easy to fulfil this requirement by using equipment
which is quite unable to produce a field anything like this figure.

Even very low powered equipment can produce interference if used

very near to radio receivers, and when the actual working frequency, or
a harmonic of it, falls on or near a broadcast frequency. So there are
also limitations on the frequencies which may be used.

If a detector operated at the same frequency as a broadcasting station,

and its radiations were picked up at sufficient strength, this would cause
interference. Whistles would in fact be expected if the interfering signal
were within 10kHz or so of the transmission frequency of the broad­
casting station.

A similar effect may arise where harmonics of the metal detector

frequency fall near the frequency of a broadcasting station. The actual
working frequency of a detector oscillator is its fundamental frequency,
and harmonics arc multiples of this frequency. Harmonics arc thus
2x, 3x, 4x, 5x and higher multiples of the metal detector oscillator
frequency. For example, if the detector oscillator operated at 100kHz,
its harmonics would be 200kHz, 300kHz, 400kHz, 500kHz, and so on.

8
As long wave broadcasting takes place in this country on 200kHz,
second harmonic interference could be caused.

Fortunately, the harmonics grow progressively weaker, and arc thus less
likely to cause interference. But working frequencies should nevertheless
be selected so that trouble from this cause is not likely.

STABILITY

If the elements that determine the working frequency of an oscillator

vary in character with changes in temperature, vibration, alterations in
battery voltage, or from other causes, this may move the frequency
away from that required. Clearly this is undesirable from the point of
view of reliable working of the detector. The change in frequency
might also move the frequency to one causing interference. So a suf­
ficiently high degree of frequency stability is needed, and is obtained
by suitable design of the equipment.

It is also necessary to arrange the equipment so that random or careless

adjustment, which would place the detector on an unsuitable frequency,
is unlikely. It is in fact convenient to have a minimum of external
controls, as this also helps easy operation when actually using the equip­
ment for searching. Other adjustments, which may be necessary when
first setting up the metal detector for operation, can best be with
internal, pre-set controls, which will afterwards be left alone.
With a suitable circuit and components, random shifts in working
frequency are only likely to arise if poor constructional work is present.
Loose turns on reference or search coil oscillator windings, or careless,
insecure assembly generally, can quite easily be avoided.

REGULATORY DEPARTMENT

Heterodyne metal detectors fall within the scope of equipment the use
of which is controlled by the Radio Regulatory Department of the Home
Office. This also deals with amateur transmitting licences for home or
mobile use, model control, ship and aircraft licences, and those for other
classes of apparatus.
The Department address to which application should be made for a
licence is Waterloo Bridge House, Waterloo Road, London, SE1.

SENSIBLE USE

Though treasure and metal detectors are clearly most generally used for
finding buried objects, they arc sometimes pressed into useful service
to find buried water pipes, metal conduit concealed in plaster or walls,
and for similar purposes. This may bring a detector very near a neigh-

9
hour’s radio receiver, or aerial lead, and possibilities of radio interference
are then increased. If interference arises, it is only sensible to make such
tests for metal as brief as possible, or at times when no nuisance is caused.
It is, in fact, illegal to interfere with anyone’s normal enjoyment of
reception free of interference.

Similarly, the possession of a metal detector clearly does not give anyone
the right to trespass on other people’s property, or to make themselves a
nuisance on the seashore, or to conduct random excavation on sites
where genuine and carefully controlled searches for objects of historical
value may be taking place. In short, any usage should be of a reasonable
and sensible nature.

MINE DETECTORS

Present metal detectors are largely descended from the earlier military
mine detector. Clearly a device which could locate a concealed metal
object was of enormous importance.

An early mine or metal detector of this kind was a large and relatively
heavy object, employing valves, with their associated low tension and
high tension batteries. Some were supplied with an additional audio
amplifier, and even this could need half a dozen dry cells for the valve
filaments, and a 60 volt or 90 volt HT battery.

Fortunately the use of transistors has made these cumbersome devices

unnecessary. A modern detector can be light, compact, and run on a
small dry battery at virtually negligible cost. Compact metal detectors,
now popular, would have been completely impossible before semi­
conductor techniques replaced the old thermionic valve circuits.

PRODUCTION OF HETERODYNE

If two different frequencies are mixed together, one resultant frequency

obtainable after doing this is the difference between the two frequencies.
Thus, in a superhet or superheterodyne receiver, mixing an incoming
signal of 1500kHz with a local oscillator frequency of 1970kHz would
make available a 470kHz signal for the intermediate frequency amplifier.

A similar method can be used to produce a heterodyne which is of

audible frequency. In Fig. 1 assume that the first oscillator is operating
at 90kHz, and the second oscillator at either 91kHz or 89kHz. Both
signals pass to some type of detector or non-linear device, and the out­
put from this is the difference in frequency of the two oscillators. That
is, 91 - 90kHz, or 90 - 89kHz, which is 1kHz or 1,000 hertz. The original
frequencies, which are 90kHz and 91kHz (or 89kHz) are inaudible, but
the heterodyne of one thousand cycles per second is an audio tone, and
could be heard with headphones, or could be raised in level by means of
an audio amplifier, to operate a loudspeaker.

10
11
 This principle is the basis of the heterodyne metal detector. The second
oscillator may be higher or lower in frequency that the first oscillator,
as explained. But to simplify matters it may be assumed that the second
oscillator operates on the lower frequency, 89kHz, but is also variable.

With the first oscillator (1) permanently operating at 90kHz and the
second oscillator (2) operating at 89kHz, the difference is 1.000 hertz,
as already explained. Suppose, however, that the frequency of (2) changes
to 891kHz. Tlie difference between (1) and (2) is now 90 - 89-1 =
900 hertz. So the output frequency has changed from 1,000 to 900 hertz,
which would easily be apparent to the ear.

Should (1) be on 90kHz as before, and (2) be set to 90kHz also, there is
no difference, and no audio frequency is produced. But as soon as (2)
is altered, the difference becomes audible. The actual change in frequency
of (2) may be exceedingly small, in terms of a percentage of the frequency
of (2), and yet still be in the audible range. This is more easily seen if the
frequencies arc written in hertz, or cycles per second. Thus if (1) is
operating at 90,000 hertz, and (2) at 90,020 hertz, the difference or
audio output will be at 20 hertz.

When the heterodyne or audio output is ol high frequency, a small change

is not so easily heard. As example, a change from 5,000 hertz to 5.010
hertz is far less obvious than a change from 50 hertz to 60 hertz. The
extent to which the human ear can distinguish very small changes in
frequency varies with different persons, and also depends somewhat on
practice. Extremely low frequencies, of only a few cycles per second,
will be heard more as a plop-plop or ticking sound, than as a tone.
Somewhat higher frequencies are heard as a buzz or hum, such as “mains
hum” at 50 hertz. Higher frequencies move up the musical scale, and will
eventually become so high that they are no longer audible.

HETERODYNE LOCATOR

With the heterodyne treasure or metal locator, the first oscillator can be
the fixed frequency, or reference oscillator. The frequency of oscillation
- of the second oscillator is then controlled by the presence of metal
objects. The outputs of (1) and (2) are combined in the manner des­
cribed, and the resultant output provides an indication of the presence
or otherwise of metal.

The effect may be used in more than one way. As example, if (1) and
‘(2) are set to the same frequency, with no metal present, the heterodyne
will rise in frequency as the metal is approached. The closest position'
is thus shown by the highest pitch. This method is often used. Its main
disadvantage arises only with poorly-designed circuits in which one
oscillator "pulls” the other due to coupling in the detector ciruit. As a
result, small changes in frequency in (2) are followed by corresponding
small changes in frequency in (1) and no audio output, or indication, is
obtained. ,
12
An alternative method is to have (1) and (2) working on somewhat
different frequencies. The presence of metal will then move (2) to
either a higher or a lower frequency, causing a rise or fall in pitch in
the audio tone.

It is probably better to use an actual locator to observe these effects,

f ortunately adjustments, or the manner of use, only become critical
when the locator is operated at the extreme limit of its detection range.
At shorter ranges, strong, easily-apparent indications will be obtained,
even with some incorrect adjustment.

The matter of the range which is obtainable with a locator is very

important. It will be found that various high-priced locators, though
possibly of sounder construction, and perhaps with weatherproofing,
or other advantages, may not offer any improvement at all on a much
simpler, cheaper circuit. In fact, claims which are made about the
range of detection possible should always be investigated. This may
depend on all sorts of other factors. It is much easier to detect some­
thing of the dimensions of a dustbin lid, as example, at a given range,
than a small coin.

Other features may be more a matter of convenience or choice - such

as the use of a loudspeaker instead of headphones, or the presence of
an indicating meter.

MFTAL FINDING

The frequency of the second oscillator is controlled by a search coil,

or llat coil which can be moved along on or near the ground, or wherever
metal may lie. This coil is so designed that it has an extensive field
surrounding it, as in Fig. 2. When metal comes within the field of the
coil, the effective coil inductance is changed, and this alters the
oscillator frequency.

Tlie change in oscillator frequency depends on the size of the metal

object and its distance from the search coil, as well as on other factors.
l)y moving the coil about over the area concerned, it is possible to
decide whether the object is small and near, or larger but at a greater
distance. The position of the metal can also be decided fairly accurately.

The frequency of the search coil circuit, and that of the internal
reference oscillator in the locator, will be combined in the way explain­
ed, so that the heterodyne can be checked with headphones, or other
means. In this way the presence and position of the metal is found.

There is some advantage in having alternative search coils of different

type, if maximum possible utility is wanted. However, many treasure
locators have a single coil, suitable for general purpose work.

13
 The ability to detect a piece of metal depends also on some other factors,
over which there is generally little control. These include the presence of
other, perhaps larger, metal items, and also the existence of poor conduc­
tors such as saturated or wet sand or earth.

Some cheap and moderately-priced ready-made detectors have a coil

situated in an insulated casing, and changes in frequency can arise with
this, due to capacitance with the ground. This means that there are
shifts in the heterodyne tone with movements which bring the coil
nearer the ground, or move it away. An instrument of this kind is
. feasible for use, but this effect can be avoided by a form of screening
which prevents unwanted capacity effects to the coil turns, while not
interfering with the ability of the coil to detect metal objects.

It will be seen that although a heterodyne metal locator will always

' operate in a general manner to that explained, there are various improve­
ments or modifications which can be introduced.

14
PRACTICAL CIRCUITS

Practical locators can be extremely simple, or may employ quite a large

number of components in more complex circuits. The very simple
locators may, in favourable circumstances, give a performance almost
equal to the more complicated devices. But usually there will be some
loss of operating convenience, as with single transistor locators which
have to be used in conjunction with a portable radio receiver.

The circuits which are included here should not present any particular
difficulty, either in construction or adjustment. Details of “trouble
shooting” are also given where appropriate. Generally little more than
a multi-range test-meter will be wanted to find out which section of a
non-working locator is at fault. By dividing the circuit up into sections
- such as reference oscillator, coil oscillator, and detector - it is
possible to simplify the location of any fault, so that it can be easily
traced. This means that the constructor need never be faced with a
whole, non-working device, with no indication of where the fault lies.
Transistors and similar items are of easily-obtained types, but this does
not mean that alternative or similar items could not be used. Naturally
any substitutions of this kind ought to be made only when it is known
that satisfactory working is still possible.

PRACTICAL CONSTRUCTION

Probably the simplest way to assemble the circuits shown is to use

perforated board, placing resistors and other small components on one
side of the board, and having wiring on the reverse side. Plain perforated
board with a matrix of 0.1 in and 0.15 in between hole centres is readily
available, and is easily cut to the wanted size.

A saw with small teeth is recommended for cutting, and the board should
be supported on a fiat surface near to the cutting line. Larger holes, such
as may be needed to fix coils or other items, or which are necessary to
mount the board, should be drilled before mounting any components.

As it is not usually necessary to reduce dimensions to the absolute min­

imum, the 0.15 in matrix board is preferred, as allowing a little more
free space between leads and joints.

Resistors and tubular capacitors may be fitted flat on the board, or

vertically, as necessary, by shaping the wire ends to pass through the
correct holes. In many places the wire ends of such components will be
long enough to reach to other connecting points. Excess wire is snipped
off. Elsewhere, it may be necessary to use connecting wire. The gauge is
not important, but about 24 swg is convenient to handle.

15
 f

In many places leads can run from point to point under the board by
routes which will avoid any unwanted contact with other leads or joints.
Elsewhere, or where wires cross, insulating sleeving should be slipped >
over the bare wire as necessary. Small diameter sleeving, such as 1mm,
is most suitable. It can be very helpful to have two or three colours as
these will allow positive and negative circuits to be identified, as well as
transistor or other leads, when required.

Thin flex should be provided for battery connections. Use red for I
positive, and black for negative, and fit the proper type of battery clips.
A battery must never be connected in the wrong polarity.

Other.items required, such as nuts and bolts, tags, jack socket and plugs,
will be readily available from stockists of radio components. This also
applies to the larger items, such as headphones, loudspeakers, volume
controls, and other parts.

A case will be required for the circuit board, battery and associated items,
and it is not easy to improve on the ready-made plastic boxes which can
be purchased from various shops and stores. In general, these seem to be
available in two types. One is of a flexible material which is easily cut or
drilled with no chance of cracking. The other is a rigid type of box, quite
strong, but brittle. This type is preferred, but some care is necessary to
avoid cracking the material, when drilling it. The drill should be sharp,
and should be applied with only light pressure, the box being supported
on a piece of scrap wood. Small holes can be enlarged with a reamer, file
tang, or by filing, but again care is needed as too much pressure will snap
the material.

Gear plastic boxes can be painted, if wished. If so, the best finish is
obtained by painting the box inside, before any assembly work. This will
give a perfectly uniform and protected finish. Use one of the quick-drying
paints as available for painting models, and be quite sure it is hard before I
replacing the box lid.

SOLDERING
I '
The construction of devices of this kind is not really practicable without
soldering. Anyone except an absolute beginner will be able to solder.
The beginner will also find that he can do so, if he uses a radio or elec­
tronic wiring cored solder, an iron which has been allowed to reach its
proper temperature, and all joints arc clean and bright.

Place the solder on the joint, and the iron on the solder and joint, so i
that the flux core of the solder can take effect where it is needed. The
solder should melt in two or three seconds and flow over and around the
-joint. Remove the iron at once, especially when soldering transistors, as i
lengthy heating is unnecessary and may damage some components.

16
Do not try to solder to dirty or oxidized tags or leads - even if solder is
built up round the joint it will probably be defective electrically. New,
bright leads and tags will solder at once, but otherwise they need to be
scraped bright first.

TRANSISTOR LEADS

Difficulty sometimes arises in identifying the leads of transistors, when

they are mounted on a circuit board. They may be concealed by other
components, and often a top or side view will remain, whereas lead-outs
are coded when viewing the device from below.

Any difficulty from this can be avoided by placing short pieces of

coloured sleeving on the leads in advance. These need only be 14 in to
3/8 in or so long. With Nl’N transistors, it is convenient to use black for
emitters (which goes to negative), red for collectors (which go to positive
circuits) and some other colour for the base.

REFERENCE OSCILLATOR

The audio tone indicating the presence of metal is obtained by beating

together the outputs of two high frequency oscillators, in the way ex­
plained. The frequency of one oscillator is controlled by the metal­
sensing search coil. The second, or reference oscillator, has a fixed
frequency, or may have a small “trimming” adjustment so that it is
easily set to the most suitable frequency.

It is convenient to base the design on a centre frequency of 100kHz.

If adjustments allow trimming to about 10kHz each side this frequency,
the available coverage will be about 90-110kHz. In use, 100kHz will
not be employed, because of possible 2nd harmonic interference to
200kHz radio reception. With this coverage, the actual frequency can
be off-set from 100kHz, so that, as example, the oscillator is operating
on 95kHz, or 105kHz. It should be sufficiently stable to maintain its
frequency without any need for repeated adjustment.

Fig. 3 is the circuit of an oscillator which may be used cither at a pre­

set frequency, or with a trimming control. It uses a ready-made induc­
tor, though home-wound coils can be fitted as shown later.

R1 supplies base current for the transistor, and Cl provides feedback

from the oscillatory circuit to the base in the correct phase to maintain
oscillation.

LI is the primary of the coil, which is a Denco IFT13. This is a

465kHz type intermediate frequency transformer, and the additional
capacitor C2 across the winding shifts this to approximately 100kHz.

H
18
C3 is the capacitor fitted internally across LI. With an IF transformer
of this kind, the normal working frequency depends on both the value
of C3, and the inductance of LI. This means that transformers of other
manufacture might have a winding of different inductance, even though
intended for 465kHz. As a result, the value shown for C2 would not
necessarily be correct if a component other than the 1FT13 were fitted.
Should such a change be made, and it is found that no possible adjust­
ment of the 1 I T core gives the wanted frequency, then C2 would have
to be changed in value to suit.

The variable capacitor VC1 is optional. Without it, the circuit can be
set to frequency by adjusting the core of LI. When it is used, a small
band of frequencies can be covered, without needing to adjust LI core.
The purpose of this will become clear in iater circuits.

The secondary winding L2 allows coupling the oscillator to a later

circuit, where the heterodyne will be produced. A coupling winding is
not essential for this purpose, but it is already present on the 1 FT 13
and helps isolate the oscillators so that changes in frequency of one
will not pull the frequency of the other, thus causing the difficulty
described earlier.

With circuits of this and similar type, oscillation is not produced if the
transistor has insufficient gain for the feedback provided, or if feed­
back is too low with the correct transistor, or if feedback is of the
wrong phase. This means that primary connections and values should
be as shown.

Another set of unsatisfactory working conditions is encountered where

the level of feedback, transistor gain, or values allow squegging. This
arises from cutting off of the collector current in such a way that the
oscillator frequency spreads over a wide band, producing a hissing, or
audio whistles which are not the result of a heterodyne beat between
reference and search coil frequencies.

ASSEMBLY V

This circuit may be wired either as a separate unit; or on part of a board

which will carry the other stages required. A suitable layout for compon­
ents and wiring is shown in Figs. 4 and 5. The board is large enough to
accommodate the other stages also.
First drill the board to take the coil pins and two tags which are in
contact with the metal can. The finished board is most easily mounted
by two 6ba bolts. These can be about Zi in long, so that washers or
extra nuts will raise the board sufficiently for its underside wiring to
be clear of the case.

19
 < X
JO
o ° ° 2 -X o

Wiring is very straightforward. Remember not to “cook" the transistor

joints, when soldering them. There is little point in compressing or
crowding items to achieve the smallest possible piece of equipment,
and this is only likely to make assembly and wiring difficult.
Where external leads are required, wires or pins are provided. Flying
wire leads are best colour coded - as example, red for battery positive,
and black for negative, with some other colour for other wires which
may be necessary. If pins arc used instead, obtain those intended for
0.15 in matrix board. They are best put in with a proper pin-insertion
tool, while supporting the board adequately below as near to the hole
as possible. Leads of suitable length can then be soldered to the pins

20
later. Similar means of soldering on external leads can be devised by
leaving about Vi in of stout wire projecting above the board, or by
twisting a loop of wire through two adjacent holes.

A beginner will often find it helpful to mark each component outline

with coloured pencil or ink, as it is fitted, and to mark each connection
as it is put on and soldered. There is then no chance of omitting any­
thing. Also check to see that there arc no short-circuits from bare wires
or joints, and that all joints are properly soldered.
TESTING

If each part of a circuit is tested when finished, this avoids the situation
where a complete locator has been made, but does not operate, and its
builder has no idea of where the fault may lie. It is assumed that a multi­
range test meter is to hand, as this is a quite inexpensive item, and will be
of great service.

Oscillation can be checked by placing the meter in series with one battery
lead. A range is selected which will indicate about 2'/imA. This will prob­
ably be the 0-5mA or O-lOmA meter range. Take meter positive to
positive on the 9v battery. Oscillator negative goes to battery negative,
and oscillator positive to meter negative.

Current should be around 1mA or so, and should rise to about 2mA or
so if pins 1 and 3 of LI are temporarily shorted with a piece of wire or
metal tool. If there is no change in current in this way, the stage is not
oscillating, and it is necessary to look for omitted or defective connec­
tions.

It is then necessary to check the oscillator frequency. This is readily done

wjth any portable or other radio which has a long wave band, or which
has a switch position giving reception of the 200kHz programme. Tune
in this programme (it must be on long waves) and place the oscillator
near the receiver. Rotate the core of LI until an audible whistle begins,
accompanying the programme. A point will be found for the core where
this whistle ceases (or falls to a very low frequency). Turning the core a
little cither way from this point will cause a whistle which rises in pitch.

The oscillator is off-set from 100kHz in a low frequency direction by

screwing the core into LI, or by closing VC1, when present. It is off-set
in a high-frequency direction by moving the core a little out of LI, or
by opening VC1. When it is off-set to an amount which has caused the
tone to rise very high in pitch and cease to be heard, its harmonic no
longer interferes with 200kHz reception.

FAULT TRACING

Faults which cause failure are very unlikely, but a few points which may
be checked are worth noting, as they will apply also to later stages.
Checking R1 with the meter will make sure no error was made in reading
the colour code. Capacitors such as Cl and C2 can only be tested for
short circuit with the meter. So if there is any reason to suspect them,
the solution will be to replace them with new components. The winding
LI can be checked with the meter. This leaves only the transistor. If the
meter shows that 9v is present from emitter to collector, but no collector
current flows, then a new transistor is required, if R1 is connected.

22
In general, trouble from defective components is only likely when using
surplus items. It is very unusual indeed to encounter a defective item
among those purchased new from a reliable supplier.

An alternative transistor which may be used in the reference oscillator

stage is the 2N3704. Note that its leads are not in the same relative
positions as the HF194. No circuit changes are required. When testing
with a meter in the battery circuit, expect a current of around 2mA
with the 2N3704, rising to ; bout 7mA when the stage is not oscillating.

SEARCH COILS

The search coil is generally of quite large diameter, and it is this item
which is moved about near the ground, to detect metal objects. There
are several methods of constructing efficient coils of this type. That
adopted may depend to some extent on the ciioice of the constructor,
or on what materials are to hand, or whether a screened coil is required
or not

23
Mechanical or weather protection for the coil is not important if it is
used with care, and only in dry conditions. But otherwise protection is
necessary to keep rain or spray off the turns. This is generally easy to
arrange. Coils suitable for under-water searching can also be made.

The self-supporting winding shown in Fig. 6 is among the easier coils to

make. The turns are wound in a pile on an object of a suitable diameter,
and are then removed and bound with thread, to keep them together.
The object used as a temporary former must be slightly tapering (as
example, a plant pot), or a strip of card must be put round, and the
turns wound on this. Without these precautions it may be impossible
to get the coil off its former, when wound. It is afterwards cemented
to a disc of thin wood or other insulating material, of slightly larger
diameter. This supports and strengthens it, and affords protection when
moving the coil about at ground level.
Another method is shown in Fig. 7. This provides a square or rectangular
coil. The base is a piece of plywood or similar material. Four projections
near the corners take the coil turns. The projections may be pieces of
dowel or wood set in holes, or glued and screwed in place, or 6ba or 4ba
bolts can be used, with insulated sleeving to avoid contact with the turns.
The wire itself will of course be insulated, but enamel insulation in
particular is easily scratched away so that a short-circuit to unprotected
bolts is likely. An advantage of this type of winding is that the number
of turns can be readily changed.

/
24
Fig. 8 is another type of coil, of robust construction. The turns of wire
are wound in a shallow channel provided in the perimiter of the wooden
disc. Tliis former can be made from three discs of plywood, cemented
together. The middle piece is of smaller diameter, so that a channel is
available to accommodate the turns. The discs are marked out with a
compass, or pencil and string pivoted to a small nail, or by drawing round
a suitable object. The important dimensions will be the diameter and
thickness of the middle board, as this will determine the diameter of the
coil, and the extent to which turns pile up on each other. The outer discs
arc merely to keep the turns in place, and can be thin wood or paxolin.
(This type of former is rapidly and easily made from a single piece of
wood if a lathe is available.)

25
The coil shown in Fig. 9 requires only a single piece of thin insulating
material, such as l/16lh in thick paxolin. An odd number of slots is cut
in this, to an equal depth. As winding progresses, pass the wire down
through one slot, and up through the next. This forms a kind of basket
coil, with half the turns on top of the disc, and half below it.

It will be realised that with any kind of home-wound coil, the tension
of turns, or their exact diameter, may be expected to vary a little. To
compensate for changes in inductance which arise from this, trimming
to frequency can be provided in the locator. By this means any reason­
ably variation in the actual winding can be allowed for, even to the
extent of having a slightly wrong number of turns.

It may be noted that increasing the coil diameter, or having more turns,
or having turns in closer proximity to each other, will move the coil
frequency in a low-frequency direction, by increasing its inductance. A
similar effect is produced by increasing the capacitance across the wind­
ing. Opposite circumstances (fewer turns, smaller diameter, more space
between turns, or lower capacitance) cause the coil frequency to be
ngher than planned.

26
SEARCH COIL OSCILLATOR

The circuit in Fig. 10 is very similar to that used for the reference
oscillator. L3 is the search coil, and will often be at a little distance
from the other components. The tapping and component values are
chosen to give suitable feedback to maintain oscillation, and leads run
from the component board to C, Tap, and B, which arc collector,
tapping, and base circuit connections.

+ iv

> I
O'
o
u.
CL
CO
O CU Li-
CM
CL
H

u.
CL
CO O
CL .
f''

u. O
CL I'-' u.
o O <\ a.
u CO
CO 26
CL
<►
in
u o

X CL
.<
II
(J
CC _
ro <0 ,,-mDOMQOOQQQQ/-1CO
O)

27
C5 is the main capacitance across L3, but the trimmer T1 is provided,
so that the working frequency can be adjusted. T1 is a compression
type trimmer, and its minimum capacitance is about 150pF. Its correct
position should be about half screwed down, so there is latitude for
adjustment in cither direction.

The components (excluding the search coil) are assembled on the

perforated board in a similar manner to that already described. It is also
in order to use unperforated board, drilling small holes to pass the leads
from components.
 7r,nS for this part of the locator can be seen from Figs. 11 and
11. I he leads from C, Tap and B can be soldered on and cut to suitable
length later. The connection from C8 will run to the hctcrodvne
detector.

29
COIL

The search coil has fifty-four turns of 32 swg enamelled wire, IVi in in
diameter and / in winding width. To prepare the former, cut a disc of
Vi in plywood 7Zi in in diameter. This can be done with a keyhole or
pad saw. Smooth the sawn edge with glasspaper, if this appears necessary.
Cut two slightly larger discs - these may be 8 in in diameter. Apply
adhesive to the meeting surfaces, and fix these each side the IVi in disc
Figs. 8 and 13. Leave under pressure until the adhesive is hard.

30
As this locator is a straightforward design, which can give good results
but avoids needless complication, the search coil is not screened. Drill
three small holes 'A in from the edge of the top disc, and pass the end
of the 32 swg wire up through one hole. This is point C. Wind on thirty-
four turns, and take the wire up through the next hole, forming a loop
for the tapping. Continue to wind in the same direction for a further
twenty turns, and cut off the wire, taking the end up through the last
hole, for point 15.

Scrape insulation from the ends and tapping, for soldering. They may
be taken to a 3-way tagstrip, to which other leads can be soldered.
Alternatively, solder on wires for this purpose, and cover the joints
with sleeving or tape.

When the coil has been tested, wind a few turns of thin tape in the slot
to cover the wire, and varnish or paint the wood. The paint can be
taken over the edges of the discs and on to the tape, to help exclude
moisture, but the turns of the winding must not be saturated with
varnish or paint.

TESTING

If wished, this stage can be tested by itself, in a similar way to that used
with the reference oscillator. Temporarily connect the coil to its three
points on the circuit board. Note the current drawn by means of a meter
in series with the battery. This should be around 0.5mA with the stage
oscillating, rising to over 1mA if B and C on the coil are shorted with a
piece of wire. It is not necessary that currents be very close to these
figures, as they depend on the individual transistor and setting of trim-'
mer Tl. Any rise in current, with B and C shorted, proves that the
oscillator is working. Should there be no change in current, then it is
necessary to look for omitted connections, or other defects, as already
explained. These arc not very probable, but as only a few moments are
needed it is worth proving that the stage oscillates, by noting current
in this way.

The 2N3704 may be used in this stage, and the current will be approx­
imately 0.5mA, rising to 3ma or so when the stage is not oscillating.
No changes need be made to component values, but take care to use
the correct transistor connections, as these are not in the same relative
positions as those of the BE195.

Tuning to frequency can be carried out as already described. Alternatively,

wait until the other stages of the locator arc wired, and adjust Tl until a
heterodyne is produced.

The reference oscillator and search coil oscillator must be tuned to within
a few thousand cycles or kilohertz of each other, as already explained, so
that an audible heterodyne is produced. This is easier to accomplish if the

31
frequency of each has been checked. However, the locator will work what­
ever the frequencies actually obtained are, provided they are sufficiently
near each other. The most important reason for checking the frequencies,
and off-setting them similarly from 100kHz, is to avoid interference to
broadcast reception. That is, the choice of frequency is not bound up
with getting the locator to work, but is chosen so that its use can be able
to conform to the need to avoid interference to radio reception.

i
HANDLE

Nearly all locators have a handle allowing the search coil to be moved
over the ground while walking. A block of wood about 2 x 3 in in size
and I'h in thick has a hole to receive the handle, and is cemented to the
top disc. It is convenient to have the handle slope slightly. For this
locator, a piece of broomhandle about 2ft 6in long will do very well.

h- Ll_
03- CL
— cl a:
a O
3 3
< O o
O r*
o ^

CM
a Q

+ +
CM

mo.
co
o

o
u_

32
I r
HETERODYNE DETECTOR

Various circuits can be used for this part of a locator, which is a type of
mixer in which an audio tone is produced corresponding to the differ­
ence in frequency of the two radio-frequency inputs to the stage. The
circuit in Fig. 14 requires few components. Input from the reference
oscillator is by the secondary of the oscillator coil, L2. The search coil
oscillator input is from the small isolating capacitor C8. Output from
this stage is at audio frequency.

Any general purpose, AM detector, or similar diodes will be suitable for

this circuit. Examples are the OA81, OA91, and 1N67A.

Space is available near the screened coil for the few components, and
wiring for these is shown in Fig. IS. This places both oscillators and the
detector on the same board, forming a unit which gives an audio output
to take to an audio amplifier.

o o o o o o o o

33
This board may be checked by using medium or high impedance head­
phones, which are temporarily connected from the AF output point
VR1 to the negative line. The volume obtained will not be very great,
and is not sufficient for use out-of-doors. It will, however, allow the
functioning of the circuit to be checked. To do this, rotate cither the
search coil trimmer or reference oscillator core carefully, to produce
an audio tone in the phones. This is most easily heard with sensitive,
high-rcsistance phones, but as the purpose is merely to check working,
the audio level obtained is not important.

When the audio tone has been obtained, it should be found that thus
changes in pitch if a metal object such as a tobacco tin is moved near
the search coil. Final adjustments for best sensitivity are made later.

AUDIO AMPLIFIER

When a locator is in actual use out-of-doors, there will often be wind

and other external noises, and the audio tone has to be raised to a
sufficient level for easy listening in all conditions. For use with a head­
set, a simple one or two stage transistor amplifier will easily be adequate.
Tliis requires few components, and the total battery current allows long
periods of use, even with a small battery. Headphones are often used, for
this reason, and because the tone may be a nuisance to other persons, if
audible from a loudspeaker.

Despite this, there arc occasions when a loudspeaker is much more

suitable. Friends can then hear what is happening, while some people
find a headset tiresome, especially as it may hinder conversation. It is
thus quite a good idea to make provision for operating a small loud­
speaker, while also having a phone jack. A headset can then be plugged
in here, silencing the speaker. A gain or volume control is also of
advantage, though it is not essential. It will allow the volume level to be
adjusted to suit the user, or external noise conditions.

By having the audio section on a separate board, alternative amplifiers

are easily made, to suit the purpose in view, and without disturbing any
wiring on the oscillator board.

The gain of tire audio amplifier has no bearing on the maximum distance
at which objects can be detected, provided the audio tone is at a high
enough level to be heard easily. A relatively simple amplifier can thus be
used.

A 2-stage amplifier using the circuit in Fig. 16 will have enough gain and
power output for most purposes. VR1 is the volume control. It may be
omitted if a loudspeaker is permanently in use, but is necessary when
phones are also to be plugged in. Base bias for TR3 is by means of R4,
and R5 is the collector load for this stage. A suitable choice of transistors
and values allows direct-coupling to the output stage, which is an emitter
follower. Audio signals are developed across R6, and go to speaker or
phones via Cl 1.
34
This circuit needs very few components. It also has the advantage that
the speaker or phones will be coupled by Cl 1, so that the DC resistance
of the speaker unit or headset docs not upset working conditions. The
speaker can be of about 8 ohm to 75 ohm impedance, volume being
slightly increased with the higher impedance units (say 16 ohm and over).
Current drain is around 15mA.

35
BOARD

There is no reason why the whole locator should not be wired on a

single perforated board, as this need only be a little longer to leave
space for the audio section. Alternatively, if a separate board is used,
this allows a larger amplifier to be substituted later without disturbing
the oscillators and detector.

Fig. 17 shows the location of components, and Fig. 18 connections to

them. The BC108 transistors can be replaced by 2N3704 devices, with
no change in component values.

o o O T 0 o » o o

• o o R o o

-Loo 6 o o
C IO R Cll
4 o o o

o o o
VRI i o o TR4 o
4-
R5 O » O o
FIG, I ?

FIG.18.

- LINE SPEAKER

36
It will be convenient to leave a flying lead from CIO, to go to VR1. Also
a black lead for negative, a red lead for positive, and a wire from Cl 1
negative, to take to the speaker or jack socket. On-off switch SI is
integral with the volume control VR1.

A 3.5mm jack socket is suggested, but it may be any size to suit the
headphone plug. 'Hie positions of contacts may not be as shown. When
no plug is inserted, the circuit is from tag A to tag B, and thus to the
speaker, which is returned to negative. When a plug is inserted, A breaks
contact with B, silencing the speaker, and A then provides the circuit
to the plug tip, the plug sleeve being returned via C. Wire the socket so
that this is correct.

Other transistors than those mentioned can be satisfactory in this

circuit, but it might be necessary to modify some of the resistor values.
Because of the direct coupling, the DC operating conditions of the two
stages are linked.

Any fault or need for tests in this section is unlikely. It would only be
necessary to check connections etc. here if the heterodyne is heard with
the phones connected across VR1, but not when they are plugged into
the output jack socket.

COMPLETED LOCATOR

The operational circuit boards arc fitted in a small case which will also
carry the battery, a speaker unit, and the controls. These are placed as
in Fig. 19 so that the control knobs are on the right, when using the
locator. A box about 414 x 6Vt in in size and 114 in deep will readily
take the circuit boards, battery, and a small speaker unit. It should have
a properly fitting lid or top, and may be metal or plastic - the latter is
cheaper.

Drill holes for the volume and pitch controls, phone jack, and for leads
to pass out to the search coil. An opening for the speaker can be made
by drilling a ring of small holes, or using an adjustable tank cutter.
Alternatively, circles or rows of 14 in holes will form a grille. The speaker
can be bolted in place or fixed with adhesive, according to its type.

Two wood screws through the case hold it to the wooden handle. The
circuit boards arc held with screws, with extra nuts or other means of
spacing underneath, to clear wiring. If the box is metal, take care no
unwanted contact or short-circuit is made to it. With a plastic case, use
sharp tools and light pressure, to avoid breakage.

The speaker is reasonably protected by a piece of fabric over its

aperture. When the lid is fitted, the box should be weatherproof.

37
PITCH CONTROL

This is VCI, Figs. 3 and 19. Its purpose is to allow easy manual trim­
ming of the oscillator frequency. When first putting the equipment into
operation, set VCI half closed. Rotate the coil core so that this is the
central, or zero beat position. Tile control knob will then allow exact
re-setting to this, or off-setting slightly to produce a continuous tone.
/

38
HANDLE

The search coil and locator are positioned as in Fig. 19. A wooden
handle is most easily prepared, and it can have a top grip fashioned
from a piece of rubber tubing or hose, or a cycle handlebar replacement.
The main need is to have everything reasonably strong and rigid, with
an easily held handle, and to avoid unnecessary weight.

Connections are taped to the handle, where they run down to the
search coil. Parts should be painted, so that they do not absorb mois­
ture from wet grass, or become saturated.

LOCATOR USE

A few minutes spent witli the locator and metal objects which are to
hand will show the methods of operation. Begin with larger items,
which will produce a great change in pitch. From this, proceed to
smaller items, which have less influence on the tone.

With VC1 adjusted to zero beat with no metal near, a tone will com­
mence and rise in pitch as the search coil is brought near the metal
object.

If VC1 is slightly off-set from this position, so that a very low audio
tone is produced, this tone will cither rise in pitch as metal is approached,
or fall to zero, then rise. If the latter is the case, off-set VC1 the other
side the zero beat position.

Indications depend on the kind of metal, its area, shape, and its distance
from the coil. After getting accustomed to using the locator with test
objects which can be seen, it can be taken out to search for buried items.

COMPONENTS

R1 270k
R2 270k
R3 1.5k
R4 2.2 megohm
R5 10k
R6 330 ohm (All resistors 5% 'A watt)
VR1 5k log pot with switch SI.
J

Cl 330pF
C2 6800pF
C3 Present in IFT can
C4 O.OluF
C5 lOOOpF
C6 330pF
C7 O.luF

39
 =
C8 27pF
C9 4700pF
CIO O.luF
Cll 320uFlOv
VC1 300pF or 500pF small solid dielectric variable capacitor
T1 750pF compression trimmer
TR1 BF194 or 2N3704
TR2 BF195 or 2N3704
TR3 BC108 or 2N3704
TR4 BC108 or 2N3704
D1 OA91 or 1N67A
D2 OA91 or 1N67A
L1/L2 Denco (Gacton) Ltd. IFT13, or home-wound (see later)
L3 Made as directed.
Box, battery and clips, Vh in or smaller speaker, jack socket, knobs,
wood for handle etc., 0.15 in matrix perforated board.

40
: PART II
ALTERNATIVE COILS, DETECTORS, AMPLIFIERS

With the reference oscillator circuit shown earlier, a ready-manufactured

inductor was fitted. This is convenient to simplify construction, as it
avoids any need to wind a coil for this position. However, a home-wound
coil may be preferred in some cases; or it may be wished to press into
service coils of other inductance value. It is also possible to use coils
which do not need a tapping or feed-back winding, and to make other
modifications to the coil itself, or the circuit with which it is employed.

It is convenient to base the oscillatory circuit on a middle frequency of

lOOkllz, for the reasons explained earlier. Tills means that some adjust­
ment each side this frequency will be necessary, and a relatively large
change in frequency is obtainable by having a coil with an adjustable
core. The relative position of this core modifies the inductance of the
winding, which will be at a maximum when the core is fully in the wind­
ing. Inductance falls as the core is withdrawn from within the turns of
the coil, being at a minimum with the core entirely removed.

The inductance of the winding will depend on the number of turns,

their diameter, the type of core, and other factors.

The second factor which will modify the frequency of the circuit is the
parallel capacitance. With a given inductance, frequency falls as capac­
itance is increased. Alternatively, with a given parallel capacitance,
frequency falls as the inductance is increased.

The relationship between inductance and capacitance, for a frequency

of 100kHz, is shown in Fig. 20. From this, it will be seen that high
values of capacitance arc required, when the inductance is relatively
small. The chart could be continued for even higher values of capacitance
and lower values of inductance, but the capacitors rapidly become very
large. Should the inductance/capacitance ratio become too unfavourable,
the circuit will cease to oscillate. High gain circuits can operate with high
values of capacitance, but there is a limit in this direction in practical
terms.

At the other end of the graph, very high values of inductance become
necessary, as the capacitance is reduced. Again, the graph could be
continued in this direction (preferably on a different scale) but windings
with very many turns become necessary. The resonant frequency is also
considerably changed by very small alternations to capacitance, such as
may arise in circuit wiring, or in the transistors and other components
themselves.

41
It will thus often be better to adopt values which fall on the graph, and
other circuit considerations may further limit the choice of values. One
of these is the shift in frequency which can be obtained by means of a
trimmer or variable capacitor of particular value. This shift is very great
for high value inductors, but small for low value inductors with higher
fixed parallel capacitance. At worst, adjustment of the trimmer or
variable capacitor may be so critical that it is troublesome; or it may
give so small a change in frequency that its full rotation may not be
sufficient to compensate for small errors in setting the coil core.

These points need to be kept in view when pressing into service broad­
cast receiver or other ready-made coils, or when winding coils.

42
FREQUENCY STABILITY

Random, unwanted movement of the core, or turns, will change the

inductance of the winding, and thus the oscillator frequency. Where a
large number of turns will be used in a home-wound coil, it can be
helpful to place these between two insulated discs, previously cemented
on the former. This provides a small bobbin, to take the winding.
Discs for this purpose can be cut from 1/16th in thick paxolin or other
insulating material, the central holes being enlarged a little at a time
until they can fit tightly on the coil former. “ Bostik 1” and similar
adhesives are suitable for fixing. If the turns arc wound regularly
between these discs, with moderate tension, no other precautions will
be needed except a touch of adhesive at the wire ends.

Tlie regular winding of a coil in several layers by machine cannot be

duplicated by hand. If turns are taken off such a coil to bring down its
inductance, secure the outer turns with adhesive.

Windings having a fairly large number of turns can be wound in a com­

pact pile by hand. This is sometimes called "jumbled winding". When
the coil is finished, some adhesive will have to be applied to keep the
turns in place. Generally, such a winding should not be saturated with
an ordinary varnish or paint which penetrates between turns. Some
modification to the exact inductance values obtained can also be expec­
ted, and arises from the placement of turns, and width of the winding.

When the coil has an adjustable core, it is often possible to make some
modification to the gauge or type of wire, or coil diameter without
upsetting results, as the position of the core will be able to compensate.
A means of finding the oscillator frequency by listening for its harmonics
has been described. If the oscillator is working at lOOkllz, harmonics can
be detected at 200kHz, 300kHz, and so on. It is not always sufficient to
check at 200kHz only, if coil data are not closely followed, because if
tlie oscillator were working at around 67kHz, its third harmonic would
fall on 200kHz. This would be shown by other harmonics being found at
about 134kHz and 268kHz. If such a coil were fitted in a locator needing
an oscillator working at about 100kHz, no results at all would be obtain­
ed as neither the fundamental frequency of 67kHz nor the harmonic at
134kHz would be near enough to 100kHz. Yet an oscillator harmonic
could be detected at 200kHz, and if it were assumed this means the
oscillator must be working at 100kHz, a very great deal of time might
be spent in looking for faults which did not exist in the detector or
other circuits.

Another cause of drift is alteration in the values of capacitors with

changes in temperature. Unsuitable capacitors can cause quite large
changes in frequency, producing a steady rise or fall of the audio tone,
and making re-adjustment of the manual trimmer necessary. Silver
mica or other high stability capacitors should be fitted in the tuned
circuits.

43
 :

If necessary, capacitors may be connected in parallel to produce a

required value. As example, lOOOpF and 1200pF capacitors in parallel
would total 2200pF. The most suitable pre-sets arc approximately
50-450pF, 100-580pF, 150-750pF, 300-1 OOOpF, 400-1 250pF, and
500-2000pF. These offer considerable range for compensation or
adjustment of the total capacitance across a coil.

>
O
4-
o
u. O
Q. r>-
Z
o c\j
U_
o.
ro O
uO uO

<o
•-N

li. _ , vvV
1 * co cr
cq O
°6
HI—“
1L
_ CL
U CM f\J
O 03
—O O
cr oo
•A/WW* U.

44
BROADCAST COILS

An obvious choice when looking for a coil is to choose a long wave radio
tuning coil. This would be expected to cover about 150-400kHz or so,
with a 350pT or similar capacitor, so may have an inductance of around
2500uH. From Fig. 20, a parallel capacitor of about lOOOpF could be
expected to tune this to 100kHz. The exact value may depend on the
degree of adjustment provided by the coil core. A winding may be
present which can be used for feedback for the transistor oscillator. If
not, such a winding might be added, as it will need relatively few turns.

Fig. 21 is a circuit for the Denco (Clacton) Ltd., "Blue” valve type
Range 1 (long wave) coil. This is an aerial tuning coil. Here, the aerial
coupling winding, pins 8 and 9, is used for feedback purposes, and the
tuned winding, pins 1 and 6, has C3 in parallel.

With this particular coil, 960pF is suitable for C3, as well as lOOOpF. A
silver mica capacitor is preferred, though not essential here.

R1 provides base current, and RF feedback is by Cl. There is not a great

deal of latitude in the choice for Cl, as small values provide insufficient
feedback to maintain oscillation, while large values cause squegging to
develop, as described earlier. It is necessary that the winding 8-9 provide
feedback in the correct phase, so with other coils these connections
might have to be reversed.

Whether or not oscillation is present can be tested with a meter, if

wished. This is placed in scries with the battery, as explained earlier.
Current Ls expected to be around 0.25mA with the transistor oscillating,
and around 1mA when it is not doing so, due to wrong phase of feed­
back, or temporarily shorting pins 1 and 6. In any case, if shorting 1 and
6 results in no change of current, the stage is not oscillating and this
must be corrected before it is incorporated in a locator. In some circum­
stances it may be found that oscillation depends on the placement of
the core, as its position can influence the degree of coupling between
the two windings. If oscillation stops for some core positions, this will
be shown by the meter test described.

If a variable capacitor is required for trimming purposes, this can be

about 50pF. A larger value is not wanted here, for the reason explained
in reference to Fig. 20. The stage is readily assembled on an insulated
board, in the way shown previously.

WOUND COIL

Fig. 22 Is an oscillator using a home-wound coil. The former used for

the winding may be 'A in in diameter and 7/8 in long, or 7mm by 22mm.
A slight change in diameter can be compensated for by the core position,
while a longer former is of course suitable.

45
 >
■f O'
co Oit
.
UO
il
■ t a
IE—i"
oiOh.
Q. U cOq_ O 0.
< > IT)
I- rg
U
CD ^ooflfhaoo.

U_ vO
— CL _ o
Ut- a: r-
o
O'
h- z
CM
00

cr
♦

CD
■ '

UJ
CL ra
i-
<000*
•ti'if"; o
<J___
*
oj
1 r-'. i 04
i
V.
o
u CJ L
<LI •:
LL

i
46
As a large number of turns is used, they arc placed between two
insulated discs. These arc % in in diameter, and arc cemented near the
top of the former, as shown, with a space of 3/8 in between them.
Small holes are drilled for the wire ends to pass out of the bobbin.

The winding is of 34swg enamelled wire, and has 680 turns in all. Pass
end C through a hole at the former, wind on 600 turns, and make a loop
for the tapping. Wind on a further 80 turns, in tine same direction,
finishing at 11.

It is not particularly difficult to make such a winding by hand, but some

form of mechanical winder will greatly ease work. Tin is may be contrived
from a hand-operated geared drill. Clamp it in a vice so that it can be
rotated with one hand, while feeding the wire through the fingers of the
-other hand. Remove the coil former core, so that a screw or threaded
rod can pass through it, and be gripped by the drill chuck. Clamp the
coil former with a washer and nut. Count how many times the chuck
revolves, for one turn of the handle. Division then gives the number of
handle turns to count. As example, if the ratio is 5:1, turn the handle
120 times for 600 turns, and 16 times for 80 turns on the coil. If the
gear ratio is not exactly a whole number, work out the number of revol­
utions on paper, remembering that a few turns either way will not be
important. The spool of wire should be arranged to turn on a nail or
other pivot.

Another form of winder can be made by fitting the former to a horizon­

tal axle, which has a handle and runs in two bearings. This can be done
in a few minutes with constructional' toy parts, with a step-up gear drive
if wanted, anil if gears arc available.

The wire is fed on reasonably evenly, with light tension. A strip of

adhesive tape round the completed coil will keep the outer turns in
position.

Component values are shown in Fig. 22, and a check for oscillation and
frequency can be made after assembly, to prove that the stage is work­
ing correctly. The winding is near the top of the former, as shown, so
that it will be removed from metal which may be close by under the
insulated board; and also to allow maximum adjustment of inductance.
The core can initially be placed approximately as shown. Screwing it
upwards lowers frequency. Moving it down and out of the coil raises
frequency.

AIR-CORED COIL

If an air-cored coil is used, there is no means of adjusting its inductance,

except by adding or removing turns. So it is necessary to fit a parallel
trimmer for this purpose. The circuit in Fig. 23 uses such a coil, wound
with 32swg enamelled wire, on a tube or former 0.3 in in diameter and
1 in long.

47
Commence winding at 1 by securing the wire, either with a touch of
adhesive, or by passing the end through a small hole drilled in the tube.
Wind on 170 turns, keeping them reasonably even, and in a pile Zi in long.
Form a tapping loop 2, then continue winding in the same direction for a
further 50 turns, placing these turns on top of the earlier 170 turn section
and ending at 3. Secure the wire with a little adhesive. This coil can be
mounted by making the tube a push-fit in a hole in the insulated board
carrying other components, cementing it in position.

48
The lack of a core reduces inductance, and a relatively high parallel
capacitance is necessary. This is made up front two fixed capacitors, one
of 10,000pF and the other of 2200pF, and a 1250pF compression
trimmer. Since the placement of turns can influence the coil inductance,
and no core is present to correct this, it would be necessary to modify
the fixed capacitor values, if the circuit does not tunc to the wanted
frequency. This limitation is unavoidable with an air-cored, pile wound
coil of tills type, when made individually by hand.

Coil winding may be simplified by using a high capacitance, so that the

coil is of lower inductance and has few turns. Fig. 24 is such a circuit.
Emitter bias is also provided, and helps to stabilise operating conditions.
The base operating point is set by the relative values of R1 and R2,
which form a divider across the supply. The value of R3 is chosen so that
with normal operating current the voltage drop in this resistor results in

CD <
OJ

O u.
U-

Li.
uo
n—
0
CD h v CD M
O O
CD

Li.
itc\ u.
—o
v uo v °o
cr u) O
2S 6
-vWWV* i-WWA-------

PI
U

u
O

Ld

49
the base/cmitte: potential being correct. Should collector and emitter
current be low, due to falling battery voltage, the individual transistor,
or other reasons, the current through R3 falls. The voltage drop in R3 is
lower, and the basc/cniitter potential is modified, to help maintain
current. On the other hand, if emitter current is too high, the opposite
effect is obtained.

Fig. 24 shows a PN'P OC44 transistor, and A is positive, while B is the

negative line. This would be convenient for both oscillators if an audio
amplifier with a positive ground line were to hand, and is to be used

For the NPN transistor (2N706) A is the negative line, and U is positive.

The coil former has an adjustable core, is 7/16th in diameter, and allows
a winding length of 1 in. There are 130 turns of 32swg enamelled wire
in all. Commence winding at the top of the former, securing the wire
with cotton or adhesive. This is point C and a few inches of wire arc left
for connecting purposes. Wind on 80 turns, closely side by side, and
cover the turns with a layer of thin insulated tape. Scrape the wire here,
and form a loop for the tapping. Then continue winding, in the same
direction, placing fifty turns side by side on the tape, ending at point B.

Feedback through Cl to the base maintains oscillation. C3 is the parallel

tuning capacitance, and should be of 5% tolerance for preference. Two
10,000pF silver mica capacitors in parallel will be suitable here.

No particular difficulty should arise in assembling the components on an

insulated board, provided the coil is not immediately adjacent to a metal
box or panel.

Due to the large parallel capacitance already present, only a small shift
in frequency is obtained by VC1, and 500pl- is generally the largest
rcadily-availablc component of small.dimensions likely to be obtainable.
The oscillator is initially set on the wanted frequency by adjusting the
core, with VC1 half closed. The core should subsequently be scaled with
wax, or may be held by thin clastic run between core and former as the
core is screwed in. A precaution such as this is wise with any cored coil,
and is essential if the core is loose.

SEARCH COIL ASSEMBLIES

The size and type of search coil can be chosen to be most suitable for the
purpose in view. Coils of large diameter are better to locate objects at a
greater distance than are small coils, but there is clearly a practical limit,
if only from the need for reasonable portability. In addition, very large
coils arc not satisfactory for pinpointing the position of small objects.
For these reasons, a coil of about 5in to lOin diameter will generally be
chosen.

50
Many popular locators use a coil of about 6in to 8in or so in size. These
are reasonably portable, and the diameter is satisfactory for objects
with a range of average sizes.

The search coil is most often a fixed part of the equipment, and is
permanently connected. This is a satisfactory arrangement for all general
purposes. There are, however, some advantages for using coils which can
be rapidly changed. It is then feasible to employ a small, medium, or
large diameter coil, or one having an extension, and constructed so that
it can be immersed in water. For interchangeability, some form of plug
and socket connector can be used.
6in COIL

A locator with a 6in coil can be a compact and easily carried item, and
though its range of detection is a little less than when a larger coil is
used, it is quite practical for general purposes.

Fig. 25 is a circuit for a 6in coil, and it can be used witli cither a
BC108, or 2N3704, without changes to any component values.

The actual winding is 6in in diameter, and is placed in a channel l/8in

wide. Tliis need only be about l/8in deep, to accommodate the turns.
The former to take tire winding can thus be constructed from a disc of
l/8in thick plywood 6in in diameter, faced each side by discs about
6%in in diameter. One of the adhesives specially made for fitting wood­
work together will prove to be most suitable, and a sound, solid joint
should be made, so that turns of wire cannot slip down between one
disc and the next. The top disc can be V4in plywood, to give a little
extra strength.

If means to do the work arc available, the former can instead be a

6%in disc of wood about 3/8in to Vain thick, in which a channel l/8in
wide and l/8in deep is cut round the circumference. Whatever form
of construction is used, sandpaper if necessary to remove rough edges,
and varnish or paint the whole to keep out damp. Here, shellac varnish
or knotting will dry in a few minutes, though other varnishes can be
used.

Make three small holes, about '/jin apart, and sloping from the channel
up through the top board. The winding is 34swg enamelled wire, and
about 65ft will be adequate, so that 2oz (approx. 100yds.) would make
several coils. Thread the wire up through one small hole, leaving several
inches projecting, and mark this B. Wind on fifteen turns, and make a
loop, taking this up through the next hole. Continue winding in the
same direction for a further twenty five turns, cut off the wire and
take the end up through the last hole, marked C. Good protection for
the winding is by a length of outdoor insulating tape, lightly round the
perimiter of the disc.

TAG-STRIP

Fig. 25 shows wiring of this coil oscillator on a 7-way tag-strip. This may
be preferred to the use of an insulated board, though the latter is of
course satisfactory. No insulation is needed on the wires, except for
positive and negative connections. T1 can be located where convenient,
and leads run to it; or it can be soldered directly to the tags. It is
probably as well to leave the transistor until last, to avoid repeated
heating here.

52
Where the locator will only be used in dry weather, solder leads to a 3-
way tag-strip screwed to the search coil, from which stouter insulated
wires will extend to the locator oscillator. Otherwise it is necessary to
protect leads and joints from moisture. This can be done by forming
the junctions inside a small insulated box or block, which can be sealed
witli “Scelastik” or other means, such as binding with insulating tape.
Or solder on either thin flexible leads, or bare 24swg tinned copper
wires, of sufficient length to reach the tag-strip, and slip insulating
sleeving over the total length of the wires, including die joints. A little
sealing compound round the holes where the wires emerge will then
keep out moisture.

If rain or dew gets into die coil or connections, working is likely to be

upset. Salt water will cause corrosion. The oscillator is adjusted to its
working frequency by rotating Tl, as already described.

This coil, as with the others described, is intended for use at frequencies
near 100kHz. It may thus be employed in conjunction with any of the
reference oscillator circuits shown earlier. It will be remembered that
the detector can only produce an audio tone when the two oscillators
arc operating within a few kHz of each other.

SHIELDED COIL

A search coil equipped with a shield is a little more difficult to make

dian an unshielded coil. The shield is of a special type, intended to
reduce capacitative effects to near objects, and it operates in a similar
manner to a Faraday screen. It almost totally encloses the search coil,
except for a narrow gap which interrupts it electrically. If this gap were
omitted, it would act as a single short-circuited turn enclosing the search
coil, and the latter could not function.

Details of the coil can be seen from Fig. 26. As the turns are to be
enclosed by foil, they cannot be wound on a permanent former (as with
Fig. 25). Instead, it is necessary to wind the coil on a temporary former.
Eventual removal of the turns is greatly aided if this object tapers slightly,
as described earlier. The wire is 26swg enamelled. Temporarily fix the end
of the wire to the former with adhesive tape. Wind on twenty-five turns
in a compact pile. Form a loop for the tapping, and wind a further twenty-
five turns in the same direction. The diameter is approximately 6in.

The whole winding is moved a little towards the smaller end of the former,
and a few pieces of tape are put under its turns, and pressed together to
hold the turns together. The coil is then removed completely from the
former, taking care not to get it out of shape. It is then bound continuous­
ly with cotton or thread, the tape loops being removed as they are reached.
When this is complete the coil should be compact and fairly rigid.

53
l
Tlic screen is cut from kitchen aluminium foil. Mark and cut a disc 7in in
diameter, and cut away an unwanted centre piece Sin in diameter. Com­
pletely cut away a piece about l/8in wide so that the foil resembles a C.
Snip inwards, all round the foil edges, to a depth of about %in to 3/8in.

Put insulated sleeving on the tap and wire ends of the coil, and place the
coil centrally on the prepared foil. Fold the foil over the coil, working
systematically, so that the winding is completely enclosed, except for the
narrow gap where the leads emerge. Mere, the foil must nut touch. Leave
a small tag of foil projecting, so that the tapping can be connected to it
electrically. This can be done with a small bolt, with two washers and nut„
passing through a hole in the foil tag.

54
The foil can now be bound with thread, to keep all secure, and a little
cement can be spread on the insulated leads where they come out of the
foil screen, to prevent movement.

Cut a disc of plywood about 6'/iin in diameter, and varnish it. The screen­
ed winding is then cemented to this all round, by spreading adhesive
liberally on the foil and wood. When the coil is correctly positioned,
place a flat board on top of it, and hold this down with a moderately
heavy weight. Adhesive should not spread up on to this board, which is
removed when the whole is dry.

The circuit in Fig. 25 is used with this coil. Cl is changed to 33pF, and
C3 to 2200pF. A check of frequency should be made as explained.

It will be found that a coil of this type is likely to vary more in actual
inductance, than one wound as in Fig. 25. Tit is arises because of the
placement of turns, the extent to which turns arc compressed together
when binding with cotton, and the degree to which the foil closely
surrounds the turns, and thus adds extra capacitance to the circuit. All
these factors are likely to vary somewhat with a home-wound coil.

Normally, T1 should give enough adjustment to compensate for changes

in frequency caused by these effects. However, if it is found that the
frequency is still too low, with T1 fully unscrewed, the simplest solution
is to reduce the value of C3. Disconnect the 2200pF capacitor, and try
2000pF instead. Similarly, if it is found that the frequency is too high,
with T1 fully screwed down, then the value of C3 may be increased. It
may be changed to 2700pF, or a capacitor of about 500pF may be added
in parallel with the 2200pF component. In this way, the wanted
frequency can be made to fall within the range of adjustment available
by means of Tl.

The coil is used in the same way as an unscreened coil, but it will be
found that stray capacitance effects - such as those of wet soil - are
eliminated, and do not cause small shifts in the locator tone.

It is not feasible to change the number of turns on a winding of this kind,

once it has been constructed, and if the individual coil should prove to
be of slightly incorrect inductance. Nor is a check of frequency before
adding the shield of any aid, because this will differ considerably from
that of the shielded winding. For these reasons, constructional details
should follow those given as closely as possible.

TWO-TERMINAL COIL

With any form of oscillator, feedback from the output circuit of the
transistor to its input circuit has to be in the correct phase. In the
circuits shown, output has been from the transistor collector, and input
to its base. There is a 180 degree or half-cycle phase change in the

55
transistor used in this way. In order that each impulse shall arrive at the
base at the correct instant to maintain oscillation, a further 180 degree
change in phase is introduced by the feedback circuit.

With the search coils shown, this is achieved by having a tapped winding,
with collector input to one end, and base output derived from the other
end. The tapping is effectively earthed. That is, the required positive
supply is provided for it, and this is not earthed, but effective earthing
at radio frequency is obtained, either from the low impedance of the
battery circuit, or by using a by-pass capacitor down to the earth line.
The latter is actually the negative battery circuit, not an actual earth,
but it can be regarded as a virtual earth, or the ground or return point.

In these circuits, the emitter is also grounded. So the phase of oscillation

at base and collector, relative to the emitter, is correct to produce and
maintain oscillations. This means that a tapping has to be formed on the
coil, and a third lead run up to the locator circuits.

The need for a tapping can be avoided by using the circuit in Fig. 27.
The collector is grounded by means of C5. LI Is the search coil, with
parallel capacitor Cl, and trimmer Tl. The emitter is, however, connec­
ted to the junction of the capacitors C2 and C3, and an oscillatory
frequency voltage, in the correct phase, is developed across R2.

With C2 and C3 in scries in this way, the effective capacitance placed

across LI by these components is 0.01 uF. The exact values of these
capacitors thus help to set the working frequency. It is not essential to
use close-tolerance capacitors. Tl is expected to allow trimming the
circuit to the wanted frequency. But should this prove not to be the
case, frequency is easily raised by removing a turn or so, or by reducing
the value of Cl to 820pF. If frequency were too high, however, Cl can
be increased to 1200pF or 1500pF, or a turn or so may be added to LI •-
With this in mind, inexpensive 10 per cent tolerance capacitors can be
fitted for C2 and C3.

The search coil LI consists of twenty turns of 32swg enamelled wire,

placed in a slot 3/16th in wide, and has a diameter of 8 in. The construc­
tion of this type of winding has been described earlier. For this coil,
three discs of 3/16th in thick plywood may be used, the middle disc
being 8 in across, and the others a little larger.

Current figures, when making oscillation checks, have been given for
many circuits. These arc only approximate, as mentioned, as they
depend on individual transistors and other features. The most important
point is that a change in current should be found when the search coil
LI is temporarily short-circuited. This arises from the difference in
operating conditions, between the oscillating and non-oscillating state.
If there is no such change, the cause should be located and corrected
before proceeding any further, as the stage cannot operate.

56
 u_ i
V
ro cm
39O
u
I > —WWV-i

ec
O
E™ u. u.
OJ a.
-AAAAA/'—< i CM CM
Ll O O
3-
o o
f^-
o CM co
u O
CM U
o
u.
nCJ

U-
II o Ll
**

CL _ in q_
O
O li--
59
HI
mO0000/

SQUARE COIL

Circular coils made in the way shown for earlier circuits are robust, and
quite easy to make. However, a square coil with the turns supported on
short lengths of wooden dowel at the corners needs the minimum
amount of woodworking. Such a coil is shown in Fig. 28.

Tile base is a piece of wood about 7 x 7 in and preferably at least 3/8 in

thick. Mark the drilling points on the comers of a square 6 x 6 in. Drill
these holes V* in in diameter, to take pieces of V* in dowel, which project
about 'A in above the board. Fix with adhesive.
57
Tlie winding is of 32swg enamelled wire, and is kept in a compact pile
against the board. Use nineteen turns from collector to tapping, and a
further eleven turns to the base circuit end of the winding. Keep the
wire fairly taut, and add a little adhesive where the turns run round the
dowel pegs.
This coil is intended for use with the circuit in Fig. 25. A 2N706 is
employed. Cl is changed to 560pF, and C3 to 5200pF. Other values
are as shown.

58
A coil of this kind can be made very readily, but it is not protected
against rain or rough usage, so has to be treated with some care. There
is also likely to be more range in inductance, due to the extent to which
turns pile on each other, and how tight the wire is. This means that it is
possible that it may be necessary to modify the value of C3 slightly, or
to change the number of turns, if it is found the frequency is incorrect
and outside the adjustment provided by Tl. Fortunately it is easy to
cut a turn or two off; or to solder on the 32swg wire, cover the joint
with tape or sleeving, and wind a few more turns on. These can be at
the collector end of the winding.

BASKET COIL

A coil wound on cardboard will avoid the need for any woodworking
and is very easily made as also shown in Fig. 28. Stout card is required,
and will be found satisfactory provided it is kept dry. A stronger coil
could be made by using 1/16th in thick paxolin or other insulated
material, but this cannot be cut with scissors.

The card should be Hat and strong, and is 7 in in diameter. Also draw a
6 in diameter circle on it. Divide the perimiter into eleven approximately
equal sections. Cut from the circumference to the 3 in radius line, so as
to form eleven slots 1/8 in wide and Zi in deep.

Tlie winding is of 32swg enamelled wire. Pass the end through a small
hole, and bring the wire up in the nearest slot. Take the wire backwards
and forwards through the slots, as they are reached. Half the turns will
be seen one side of the card, and half the other side. Wind on fourteen
turns, and form a loop. Continue winding in the same direction for
twenty-three more turns, and cut off the wire and pass the end through
a hole.

The circuit in Fig. 25 is also used with this coiL Cl is changed to 560pF,
and C3 is 4700pF. The transistor is the 2N706.

In view of the liklihood of rubbing the unprotected turns along the

ground, some protection is required. This can be given by a further disc
of card or plywood, 7 in in diameter. Should this kind of winding be
used, and maximum protection be wanted for it, a shallow plant-pot
saucer may be placed over it. These are only 1 iri or so deep, and are of
plastic material, and may be cemented to the bottom, protective disc of
the coil, so that it is wholly enclosed.

COIL MOUNTINGS

A simple method is to use a block of wood, drilled for the handle

(Fig. 13). In some cases it is convenient to be able to adjust the angle of
the coil, as when walking up or down hill. This can be arranged by
pivoting the handle between two pieces of wood fixed to the search coil.

59
The pivot can be a 2ba bolt, with wing-nut. The coil can then be set at
once to any wanted angle, to suit the person using the locator.
With this type of fitting, and a tag-strip or terminal block, screws or
other small metal items will be permanently fixed in the proximity of
the coil. This makes no practical difference to working, but the coil
circuit has to be tuned after any such fitting is in place.

UNDERWATER COIL
It is not difficult to make a completely waterproof coil assembly, and
with this it is possible to search in shallow water. To case construction,
a plastic container with tightly fitting lid can be chosen, to take the coil.
The container should be reasonably strong, and of minimum depth. It
may be found most convenient to use it inverted.

The handle can be fitted to a disc of wood, which is well varnished and
fixed to the bottom of the inverted container with screws. Put an outdoor
type plastic sealer between wood and container, flexible leads of well
insulated wire arc taped to the handle, and pass through into the contain­
er. Plug and seal these holes.

The search coil is cemented to the inside of the container lid. A little
free lead is required, so that connections to the locator wires can be
soldered and taped with the lid off. Adhesive is then spread completely
round the rim of the lid and container, and the lid is put on. The result
should be completely moisture and water proof.

In other titan shallow water the top and bottom of the container may
be curved inwards by the water pressure. To minimise this it is necessary
to use a strong, non-flexible type of box. It will be found relatively easy
to make a coil assembly which is protected against spray and a few inches
of water, but more difficult to keep out water if the coil is much further
immersed. Such a coil is best made as an optional plug-in accessory, as it
can then have an extension handle, with a sleeve made from 1 in diameter
or similar rigid plastic pipe (as used to replace copper pipes in plumbing).
This can then be pushed on the locator handle at the bottom, and the
coil leads can be plugged into the locator.

It is necessary to re-trim the coil when immersed, unless it is screened.

The effects of water round the coil connections will be reduced if the
insulation is thick, as capacitance from conductor to water is thus
reduced. A screened lead can be used, but all conductors and electrical
items must of course be completely protected against water.

A moisture-proof coil can be constructed wholly of wood, using three

discs, or a disc with a slot turned in its rim, with the winding protected
by sealing compound and tape. It is not easy to keep water out of all
items, however. Again, trimming to frequency (by adjustment of
capacitor values in the locator) should be made after all constructional
work is finished.
60
 u. I
<
tr
>
WvW

©■
w J. \

u-
uO <n
Q Q
O
44 H 4i

IV

u
ru
u
1
<N
u U o
CO (/) Ol
o O
u.

ALTERNATIVE DETECTORS
Any metal locator using a reference oscillator anti search coil oscillator
assembled front the circuits and details given earlier will need a hetero­
dyne detector. This receives the two radio-frequency inputs, and provides
the audio-frequency output.

Two diodes, as in Fig. 14, arc suitable for this part of the circuit. How­
ever, with home-wound reference coils, there will be no coupling winding,
as shown, so the method of using the diodes will be slightly modified.

61
 Fig. 29 indicates how the signals from the two oscillators may be taken
to the diode stage. Cl and C2 are small coupling capacitors one being
used for each oscillator. Their value is not critical, but should be quite
small. Any capacitors of about lOpF to 27pF or so will normally be
satisfactory. If the values are very small, audio output is reduced.
Alternatively, if the capacitors are too large, there is a greater tendency
of one oscillator to “pull” the other in terms of frequency. With this in
mind, lOpF components can be fitted, and the need to use larger values
would only arise if the audio amplifier is a little inadequate in terms of
gain, so that volume is rather meagre. In fact, when plenty of amplifica­
tion is available. Cl and C2 may be wholly omitted. Instead, leads can
run from the diodes to the oscillatory circuits, and the insulated ends of
connecting wire can be given a turn or two round the base circuit
conductors, thus forming a small coupling capacitance.

Any detector diodes in working order are likely to prove suitable, and
they need not be the same. Inexpensive surplus point-contact diodes, as
used for detectors in radio receivers, are satisfactory.
The audio output from this stage is quite low - sufficient for a test of
the locator with a good pair of high resistance headphones, but insuffic­
ient for normal usage. So output will be taken to the audio amplifier.
Where an audio volume control VR1 is provided, this can be connected
directly across C3. With most circuits its value is not too important, and
a 10k potentiometer is suggested.

TRANSISTOR DETECTOR

Fig. 30 is a grounded-base emitter detector, and this will work satis­

factorily with the 2N3704 and many other transistors. Unless the
following amplifier is one giving very small gain and audio output, R3
and C5 will most probably need to be included, to avoid low-frequency
instability. This is a continuous whistle or howl, which usually ceases
when the audio gain control is turned back. It arises because of
unwanted feedback in the amplifier and detector, which causes
oscillation at an audible frequency.

The values of the emitter resistor R1 and collector resistor R2 are not
very critical, and this also applies to the capacitor values. C4 is an
isolating capacitor, added because of the positive voltage present at
the collector.
The main advantage of this circuit, compared with the diodes in Fig. 29,
is that it provides a fair amount of extra audio output. This can be use-
. ful when only a modest audio amplifier follows the detector, with the
result that volume tends to be inadequate. In such circumstances, where
a little free space is available, the diodes in Fig. 29 can be replaced by
the transistor in Fig. 30. There would of course be no point in using the
transistor in a circuit having diodes, and already giving adequate volume.

62
63
Circuits which are more complicated than this are seldom used, but
Fig. 31 requires only a few extra components, and affords good isolation
between inputs. These are from the two oscillators, via Cl and C2, as
before. This stage also gives some gain, and thus an improved audio out­
put over Fig. 29.

Component values are not critical, and others may be used or tried if
wished. Nor arc the transistors critical, and BC108s or 2N3704s may be
fitted, or similar types.

+ L.
<
Is-
7 U.
Ll O ** U-
a.
LO V ro 3 6 *o
(£ - uo
O

ro O
Tlh
ON
cc CD
<•—A/WW—

CN
cc

ON ON
cc O
r ’d- •
CC ON
1 ►—WW'A

*
O cc
— ON h- ro
cc <o O
—VsAAAA
Cl

ON

T 7
64
DETECTOR FAULTS
With a stage using a simple circuit such as that in Fig. 29, it is very
unlikely that any fault can exist. A fault might be suspected if phones
are used to check for audio output, and no tone is obtained.
If this should happen, first check wiring. If this appears to be correct,
tune the two oscillators in the way explained earlier, and check that
they can be placed on the same frequency. If both oscillators are work­
ing, and the detector stage is correct, but a tone cannot be heard, despite
this, when the oscillators are too removed from each other in frequency.

A check that the oscillators can operate on the same frequency becomes
more important with a circuit such as that in Fig. 30, or Fig. 31, if no
audio signal is obtained. In such circumstances, no actual wiring or
similar fault may be present, so repeated checking of components and
connections would be pointless.
Should the circuit in Fig. 29 operate, but Figs. 30 or 31 prove to give
no results in the same locator, then it may be assumed that some actual
fault is present in the new detector stage. This might be an omitted
connection, short-circuit between wiring or joints, or wrong use of
transistor leads. It is not very likely that resistors or capacitors will be
faulty, if new.

In general, the detector stage is likely to prove troublcfrec, because

there are no adjustments to be made to it. But it can only provide an
audio output when supplied with suitable inputs, which in turn depends
on the oscillators and their tuning.

AUDIO AMPLIFIERS
There is considerable choice in the type of audio amplifier which is
fitted after the detector. The use of headphones avoids irritation to
other persons nearby, and is quite popular. For this purpose, a very
simple single-stage amplifier can easily give all the amplification and
volume which will be necessary.

A comfortable headset, probably with padded ear-muffs, will be most

suitable, and will help to exclude external noise. For simple single-stage
and similar amplifiers, headphones of the kind intended for use with
crystal receivers and simple transistor receivers, and also for general
short wave listening, will be excellent. These have a direct current
resistance of about 500 to 2,000 ohms, and can be inexpensive. They
also provide a satisfactory collector load for a small transistor amplifier,
without any need for coupling or matching circuits. They may be
termed medium to high impedance headsets.

65
Low impedance headsets will generally have an impedance of about 4
to 16 ohms. These can be used with circuits whicli are designed to
operate with a low impedance load. Phones which may be to hand for
stereo reception can be used by operating them in parallel.

It is also possible to use single miniature earpieces, though these can be

less comfortable. They have the advantage of being very small to carry.
Medium or high impedance magnetic units of this kind will often be
about 300 to 1000 ohms, and can be employed with any amplifier
where a complete medium or high impedance headset would be used.
Low impedance earpieces, such as 12 ohm, require an appropriate
circuit, as for low impedance headsets.

Crystal earpieces, or crystal headsets, can only operate with a suitable

coupling circuit, or with an output which may take a crystal unit.
Magnetic earpieces have windings which provide a direct current circuit
for collector current, but crystal earpieces do not.

The choice of a headset may thus depend somewhat on the amplifier

to be used, but there is generally considerable latitude. It is often
practical to employ an earpiece or headset already to hand.

HEADPHONE AMPLIFIER

Where working will be exclusively with headphones, a single transistor

is adequate. Fig. 32 shows a circuit which can follow the diode or similar
detector stage. A BC108, or 2N3704, and similar transistors will work
well here.

Best results are obtained with medium to high impedance magnetic

phones, though these are not essential. Low impedance phones will
provide less volume.

Should this circuit be wanted because of its simplicity, and a pair of low
impedance phones be to hand, they can be pressed into service by adding
a matching transformer. This can be a small component as fitted in
transistor radio receivers. Its ratio will be about 8:1. Connect the trans­
former primary to collector and positive line, and feed the phones from
the transformer secondary. There is considerable latitude in both the
transformer and phones.

With a sensitive pair of medium or high impedance headphones connec­

ted as in Fig. 32, volume may be a little too great for some users. This
depends on the headset, person using the locator, and possible external
noise. If it is felt necessary, volume can be reduced to any wanted level
by fitting a potentiometer as in Fig. 33. It may have an integral on-off
switch, and need not be 10k, though about 10k to 25k is most suitable
here. No further components or changes elsewhere are required.

66
 :

,
M

;
i

.i

i
l!
1
5i
1

Fig. 33 also shows the addition of a resistor R2 as collector load, when a

crystal earpiece is to be used. Collector current Hows through R2. The
audio signal voltage developed across R2 reaches the output circuit via
capacitor C2. The value of R2 is not critical. C2 should not be much
smaller in value than 0.1 uF. With this circuit, direct current operating
conditions depend on R1 and R2, and are not influenced by the type of
earphones.

67
 3 I
+ Q.
> I-
Z>
o O
•>

C\J L-
u-r*
eg o O
CC IT)
iMVWWv
C
5 CL
— rg 1-
cc w
l-WAV^
Ll_
_ 3.

T u6
CO
CO
_ V :
o
LL LlI
Q
o
'f
Q

68 '
!
LARGER AMPLIFIERS
A 2-stage audio amplifier, such as that in Fig. 16, is of course adequate
for headphones, and will give moderate loudspeaker volume. Whether
or not volume is adequate with the loudspeaker at all times depends to
some extent on external conditions. If there is high noise or wind, and
phones are not to be plugged in when this is so, a somewhat more power­
ful amplifier is of advantage. The more powerful type of amplifier can
still be used with phones, at a lower volume control setting.
Larger audio amplifiers may be divided into three groups: (1) Two-stage
or 3-stage amplifiers of relatively simple type, and with a single transistor
in the output stage; (2) Amplifiers with two transistors in push-pull in the
output stage; (3) Those using integrated circuits.

Amplifiers of type (1) are often used, and can be perfectly satisfactory
for reasonable output levels, bearing in mind that high fidelity is unneces­
sary. Their main disadvantage lies in the fact that if they are to be capable
of a fairly powerful loudspeaker output they are not very economical in
terms of battery current. An amplifier with push-pull output (2) is
capable of any wanted output power. There can be some saving in battery
current, but this is not equal to the saving which would be obtained for
ordinary purposes (speech or music reproduction) because of the contin­
uous tone. An 1C amplifier (3) can have the advantage of enabling power­
ful results, with relatively few components.

It should be remembered that high volume loudspeaker operation is

likely to be a nuisance to other pcpple. No amplifier need be planned
with an output of more than about 200mW, except when quite unusual
levels of noise are possible.

TWO-STAGE AMPLIFIER

Fig. 34 shows a 2-stage amplifier which will give moderate loudspeaker

results. The loudspeaker should be of 75 to 100 ohm impedance, and an
efficient unit, fitted in a case, will provide appreciably more sound than
a very small diameter speaker. Units with a rather low impedance will
cause too much loss of volume.

The amplifier is also suitable for headphones, and these may be plugged
into a switched jack socket (sec Fig. 16), so that the internal speaker is
then silenced. There is enough output for low, medium or high imped­
ance headphones, while the capacitor coupling also allows a crystal
earpiece to be used instead, if preferred.

69
 + I
> u. H
O'
to
O •>
C
O
£ Z9%
*°If)
a S+T
° |
►—AAAA/V------ i-

aj
cr
H
ro O
CC !£)
I 'WvW 1 r

U.
CO 3
CM
cr ^ u6
O
>——AAA/VV'
U
CQ
C
cr
_ 2
cr —
AAAAAA

HuX. u.
ro

CM O o
U
<-> o
Ti
5 O
> u.

Tlie circuit requires few components, and draws about 15mA from a i

9v battery, so that reasonably economical running is obtained. Other

transistors than those indicated may be used. If so, it may be found that
volume is improved by modifying the values of R1 and R2, for TR1; or
R3 and R4, for TR2. These resistors can best be a compromise between
maximum volume'but higher battery drain, and reasonable volume with
some drop in the current required from the battery.

70
THREE-STAGE AMPLIFIER

A 3-transistor amplifier will provide substantially greater gain and output,

so that a loudspeaker can be operated at higher volume than can be
obtained from the 2-transistor circuit. This results in easier listening,
when using the speaker out of doors in windy or noisy conditions. In fact,
it is always worth while checking output under actual working conditions.
A locator which has only a small amplifier, with a very modest power out­
put, may seem satisfactory when first assembled and tested indoors, where
external noise is likely to be at a low level. Outside, wind and other sounds
may result in the same amplifier proving to be inadequate, so that it is
almost essential to plug in a headset.

Fig. 35 is a 3-stage amplifier which will give adequate volume for most
ordinary purposes, when used with an efficient speaker of about 15 to
100 ohm impedance. Though a relatively low impedance speaker may
be used if to hand, an impedance of 25 ohm or higher impedance is '
most suitable, and will give slightly more volume.

The circuit may be used with headphones. In this case, the impedance
is not very important. More output than required will be available, so
that any loss due to the use of an unsuitable impedance load is not
important.

The first amplifier TR1 has base current from Rl, and R2 is the
collector load. Audio input is from the slider of the volume control to
Cl. Should an alternative transistor be used here, it might in some
cases be worth trying alternative values for Rl and R2 also.

C3 provides the audio input for the base of the second amplifier TR2,
which has base current from R4, and R5 as collector load. Operating
conditions are arranged so that the collector voltage of TR2 is a suitable
base voltage for TR3, so that direct coupling is used. The audio output
of TR3 is developed across the emitter load resistor R6, and coupled to
the reproducer by C5. With this form of speaker (or headphone) coup­
ling direct current working conditions are not changed by the DC
resistance of the reproducer.

Many other small audio type transistors will prove to be satisfactory in

this circuit, but the working conditions of TR2 and TR3 arc set by the
values of R4, R5 and R6. This means that if transistors of rather
different characteristics than those shown arc fitted in the TR2 and
TR3 positions, changes might have to be made to these resistor values.

TR1 is supplied from a decoupled point, with decoupling capacitor

C2 and resistor R3. This Is to prevent audio-frequency voltages in the
positive circuit from reaching TR1, and causing instability, or continu­
ous audio oscillation. Such decoupling becomes necessary with an
amplifier having several stages.

71
Where oilier than a very small signal voltage may arise on the positive
line, it is also wise to supply the high-frequency oscillators from a
decoupled point, as in Tig. 35. A capacitor of lOOuF or larger value,
across the battery supply (C6) is also helpful in avoiding instability.

Low frequency instability is generally shown by a continuous whistle

or howling. When this is present for all settings of a volume control
such as VR1, it is arising in the audio amplifier itself. But when it only
begins with VR1 turned towards maximum volume, it is caused by Ill
(
feedback into earlier stages - oscillators or transistor detector, if the
latter is used. Such feedback is generally through the positive supply
circuit, and is avoided by C2 and R3.
i
Components may be assembled on a tag-strip or perforated board. The
layout should keep TR3 and output circuits reasonably clear of VR1,
TR1, and input circuits. v

With a 9v supply, battery drain is around 20mA. The circuit gives a

reasonable performance with a 6v supply. The output points may be
connected permanently to the speaker. Alternatively, if a phone socket
fi'
is required, wire this so that the circuit to the speaker is interrupted
when the phone plug is inserted. Headphones can be a very important
asset in avoiding irritation to other people.

IC AMPLIFIER

An integrated circuit consists of a compact assembly of semi­

conductors, and with an IC audio amplifier several stages of amplifica­
tion may be present. For a metal locator, the main advantage lies in
selecting an 1C which will need few external components (resistors and
capacitors). A large output power is not wanted. Nor is there any
advantage in using an IC which needs so many other components that
there is just as much wiring as if individual transistors were used.
Furthermore, the very good frequency response characteristics which
may so readily be obtained with an IC arc of no benefit in the present
circumstances.

For these reasons, the audio integrated circuits which actually offer
advantages in a metal locator arc somewhat limited in number. For an
1C to be worthwhile, it must give more'gain or volume than the simple
transistor amplifiers which may be used in locators, while not imposing
a heavy load on the battery, and not needing so many other compon­
ents that there is no eventual advantage in terms of simplicity of con­
struction and wiring.

Audio ICs designed for use in small radio receivers have complementary
push-pull output stages. As a result, the battery drain depends largely
on the volume required. This is an advantage, as current is quite low
when using headphones, yet more power is available (at the cost of
increased battery drain) when a loudspeaker is used.

73
 + I—
> + BB I 10
G*

to u. ■

U A
O
<N
<N
ro
Q. v:
O co O
v- cc
C\|

-J z°%
1C 1
u.
co a.
HP- i
i
* u“
0\J
O
O
cc _
I MVVWV'

c
2
u. — 01
cc OJ
A
O
OJ
u2
7*h
+1
T+ 3
co o
I-
o
u
a
u
Q
UJ
h- &
<
a:
O
UJ sA/V—
t-
z
VWV

ilri
74
The MFC4000B is a suitable audio IC for a locator. It needs very few
external components, and is economical in battery current. This is
likely to be around 5mA or so when operating headphones, rising to
12mA to 15mA or so when providing greater output for a loudspeaker.

In Fig. 36, the audio IC is used in conjunction with a single transistor

pre-amplifier. The 1C itself contains the equivalent of six transistors, in
a directly-coupled configuration.

Input is derived from the slider of the volume control. TR1 need not
be the 2N706, as a 130 08 and similar types will operate here instead,
if preferred.

The most suitable output load is 16 ohms, and a 3’/: in 16 ohm speaker
will give good results. Volume falls off if much higher impedance
speakers are used instead. With headphones the loss of output power
is not important, so that medium or high impedance phones can be
connected. Phones or speakers of under 16 ohms should not be employ­
ed. though S ohm phones can be used if a resistor of about 8 ohms is
connected in scries with them. (This is not recommended with a speaker,
in view of the unnecessary loss of audio output in the resistor.)

'Die pre-amplifier stage, with IC and its periphery components, can be

assembled on a perforated board, and the IC pins arc spaced to fit a
0.1 Sin matrix. No particular difficulty should arise in wiring, Fig. 36B.

75
 PART III
CONSTRUCTIONAL AND OTHER POINTS

Many details of the mechanical arrangement of items will have become

clear from earlier sections. But in the interests of convenient operating
and reliability, further notes on the mechanical construction of various
parts, and the assembly of a locator, are worth while.

Reliability depends not only on sound wiring and joints and good work­
manship when preparing the circuit boards, but also on the strength and
durability of such items as search coils, handle fixings, and cases.

It is as well to keep in mind at all times that unnecessary weight is a

nuisance, and will become increasingly so over a long period of using the
locator. The overall weight can be quite small. It is well worthwhile to
avoid an unnecessarily large and heavy battery or speaker, and to keep
down the weight of the search coil, handle, and case too.

SEARCH COILS

An 8 in disc of % in thick wood, with its winding and an unnecessarily

large mounting block for the handle, can approach 11b in weight. Lighter
forms of construction, using thinner wood, will halve this. Some kinds of
coil are inherently heavier than others, and a heavy coil, situated at the
end of a handle ns it is. always grows tiresome eventually. Material such
as paxolin sheet allows the weight to be reduced, because thin material
such as 1 /16th in can be sufficiently rigid.

For reliability, the windings have to lie in a slot, or be so fixed that outer
turns cannot move. It is also necessary that the whole coil is protected
against damp. Untreated wood will take up moisture and probably warp.
If spots of water reach the coil turns, and are absorbed between them,
this will upset tuning, and may also cause corrosion later. It should be
assumed that the coil assembly will almost certainly become wet, if only
from dew or damp grass, so it must be made to resist this. This means
at least two coatings of outdoor varnish or paint for wood, with protec­
tion for the winding itself by means of tape, or a closely-fitting cover
made from a plastic plant saucer, or similar methods.

HANDLE
i
A piece of broomstick handle is probably most easily obtained. It can be
of smaller diameter dowel, but will then need a handle or grip at the top
which is large enough to hold easily. This can be fashioned from wood,
a binding of tape, or similar means.

77
Suitable handles can also be made from the type of plastic pipe which is
rigid, but can be shaped when immersed in very hot water. It is available
from ironmongers and plumbers’ merchants, and can be chosen in a
suitable diameter to take a cycle handlebar grip at the top, which may be
cemented on. It may also be bent over here, for a better holding position.
A coil assembly can have a plug of wood which will pass into the bottom
of the pipe.

A plastic case for the circuit boards can be fixed with screws, or by a
strap of thin metal round the plastic handle. The material needs no paint­
ing to preserve it, as does wood.

CASE

One of the small plastic boxes available in various sizes will generally do
wclL Remember it must accommodate die speaker (if provided) and
battery, as well as the circuit boards.

Various other boxes and household containers can also be pressed into
good service here. For those who like woodwork, a box of any wanted
size can be constructed from 4mm or similar plywood, with joints
cemented, the whole being sanded and painted afterwards.
\
Aluminium boxes can also be obtained, either as complete units, or in
the form of flanged members which are bolted together. These may be
left bright, or painted.

Where a speaker is to be fitted, this is best placed on die box facing the
user, as it will slope downwards a little and be less likely to become wet
from rain, while being easily heard in this location. Controls can be in
any convenient position, according to the layout of items inside die case.
The case needs a closely-fitting lid. Should the lid be loose, one or two
screws should be provided for it. In general, it should be possible to
carry the whole device rather like a walking stick, without any disturbance
to components. The battery can be packed round with foam rubber, or
held by elastic or a shaped clip.

VERY HIGH-C LOOP

When the search coil Ls of very low inductance and impedance, it can give
some of the advantages of a shielded coil, but without the extra construc­
tional difficulty of the latter. A coil having very few turns needs to have
exceptionally high parallel capacitance, and as a result ground capacitance
effects become negligible.

The circuit in Fig. 37 will operate with a 3-turn coil 9in in diameter. Two
turns are required from collector to tapping, and one turn from tapping to
Cl. These turns are of 18swg insulated wire, side by side round a wooden
disc.
78
To obtain resonance at 100kHz, a total capacitance of 0.45uF, made
up from 0.25uF, 0.1 uF and 0.1 uF capacitors in parallel, was found
necessary at Cx. There is no easy means of trimming the coil circuit
to the wanted frequency, as even a 1250pF parallel pre-set capacitor
will have negligible effect on the frequency. It is thus necessary to
experiment with various capacitors for Cx, or to separate the turns
slightly, until a suitable frequency is reached. Because Cx is so large,
handling the coil turns will have virtually no effect on frequency,
while slight changes in inductance due to the proximity of turns to
each other will modify frequency considerably. If the winding is of
single core mains type cable, it will need no protection, but turns
should be secured with adhesive.
79
It is necessary to use a transistor offering sufficient gain, or oscillation
will not be obtained. Tlic BC109 or 2N3704 may be used.
The transistor and other few items of the oscillator stage should be
assembled on a tag-strip, which is situated on the search coil disc,
immediately adjacent to the ends of the coil. Some trial and error can
be expected with this oscillator, especially as it is impracticable to
provide a trimmer of large enough value to have any significant
influence on frequency, for adjustment purposes.

450470kHz WORKING

In tire British Isles, frequencies allocated for devices of this type lie
between 16kHz and 150kHz, so operation cannot be allowed on
frequencies around 450-4 70kHz, and a locator using such frequencies !
would not be licenccable here. However, this does not apply in some
other countries, so brief details may be given of circuits for 450-470kHz.

In these circumstances, this frequency band is often chosen because

most popular superhet receivers have an intermediate frequency within
these limits. This means that an inexpensive, ready-made portable
receiver can perform as detector and audio amplifier, with output to
the internal loudspeaker (or to earphones, where a socket is provided).

The simplest locator using such frequencies consists of a single transistor

as search coil oscillator. Fig. 38A is a suitable circuit for this purpose.
The search coil is constructed as described for Fig. 28, and has 23 plus
14 turns, wound in slots. Here, the parallel capacitance is much lower,
so that the frequency is raised.
Though the receiver may be carried separately, it is more convenient
to have this fixed with clastic bands or similar means to a handle, which
carries the search coil and oscillator. The necessary heterodyne is
obtained by tuning in a broadcast signal, then adjusting T1 until the
oscillator produces a beat note, in a similar manner to that previously
described. If the oscillator is tuned to near the receiver intermediate
frequency, this beat note will be heard on all transmissions. If it is
heard on one transmission only, but not accompanying other transmis­
sions of different frequency, then a harmonic of the oscillator is reaching
the signal-frequency circuits of the receiver. The receiver must be close
enough to the search coil, and so placed, that a strong enough oscillation
is picked up. Failing this, an insulated lead may be run from the
transistor collector to near the IF amplifier of the receiver. If tuning in
a strong, local station swamps the search coil oscillator input, select a
more distant, weaker signal.

The oscillator and receiver are used in the same way as the type of
locator already described.

80
Though this arrangement can give satisfactory results, it has the limita­
tion that the radio programme is present, and its volume will vary due
to directive effects of the receiver ferrite aerial. Fig. 38B shows a
reference oscillator adjustable around the range 450-470kHz. This uses
the coil shown in Fig. 24. Adjustment can be by the core alone, or by
means of a small trimmer (say 50pF) as shown earlier.

The two oscillators can be assembled and coupled to a detector, as

shown in Fig. 29, and followed by an audio amplifier.

As mentioned, equipment operating in this band of frequencies must

not be used in the British Isles.
81
OSCILLATOR STABILISATION

Long term stability of the frequencies of the oscillators will depend

on no changes arising in the inductance of tire search coil or reference
oscillator coil, or in the values of the capacitances. With reasonably
rigid construction, and no loose turns, the coils should prove to be
satisfactory.

With high stability silver mica or similar capacitors, no particular

trouble should arise here. The provision of a manual trimmer allows re­
adjustment over a narrow range.

Another cause of unwanted frequency changes may arise from the

transistors themselves. The effective capacitance of these devices varies
with current, and thus also with battery voltage. In the circuits shown,
there is a large parallel capacitance, and changes to oscillator frequency
due to a failing battery are generally negligible. Some oscillators, on the
other hand, will produce quite a significant change in frequency for a
small drop in battery voltage. This is not important where a steady
current is drawn from the battery, as re-adjustment of the manual
trimmer will allow compensation for it. With the amplifiers shown,
current changes very little so there is a steady load on the battery. But
if a more powerful amplifier is pressed into service it will almost
certainly have an output stage in which current varies considerably
between silent and tone conditions. If this is combined with an
oscillator whose frequency varies with voltage, and a failing battery,
the oscillator may be pulled by audio output conditions.
This difficulty can be avoided by running the oscillators from a
stabilised supply. Fig. 39 shows a 6.2v zencr diode added for this
purpose, with a series resistor Rl. A 400mW diode could allow a
current of 60mA, but the stabilisation current needs to be kept as
low as reasonably possible, to avoid an unnecessary load on the battery.
For this reason, Rl can be 270 ohm, or 330 ohm, the latter giving a
little more economy, with a stabilisation current of under 10mA. The
best value for Rl depends somewhat on the current drawn by the
oscillators, which also passes through Rl. Cl will normally be present
as part of the oscillator circuit.

vVWW
OSC. R! 9V +

4-
J.CI ZDI
T FIG .39.

82
KIND OF METAL

When a heterodyne metal locator is adjusted so that it is producing an

audio tone of a few hundred hertz, with no metal near the search coil,
it will have been noted that some metals produce a rise in pitch, while
others cause a fall in pitch.

Ferreous metals have an opposite effect to non-ferreous metals. The

ferreous metals include iron, steel, and numerous other iron-like mater­
ials, while the non-ferreous metals include silver, copper, aluminium,
and various alloys. Ferreous metals are attracted by a permanent magnet;
non-ferreous metals are not. The lists of both arc too long to include,
and the division is in any case an obvious one.

Whether or not the ferreous metal (or the non-ferreous) causes a drop in
frequency (or the reverse) depends on whether the search coil oscillator
is set above or below the reference oscillator frequency. A check with
known objects is the simplest means of noting which change is produced
by non-ferreous metals, as example.

It will be found that long, thin, or small objects produce much less change
in pitch than objects of larger area. For localising small objects, a very
large coil is generally less convenient and satisfactory. With a quite small
coil, small objects can easily be detected at very limited distance, and this
can be useful for finding conduit in plaster or for similar purposes.

3-TRANSISTOR LOCATOR
Fig. 40 is the complete circuit of an easily-made three-transistor locator,
designed to operate headphones. By arranging components on a tag-
board all wiring and parts are visible, and this may be found easier for a
beginner, than having leads under the board. In fact, very little construc­
tional work is necessary, and all is very straight forward.

SEARCH COIL

This is of basket type, as shown in Fig. 28. It can be wound on a stout

card disc (which can be cut with scissors) provided it is not allowed to
become wet in use. The card is 7 in across, and has eleven slots each
Vi in deep. The winding is 32swg enamelled wire. The beginning of the
winding is point 3. From here, wind fourteen turns, and form loop 2.
Continue winding for a further twenty-three turns, ending the coil at
point 1.

Fix a card or plywood disc about 7 in across under the coil, for protec­
tion and extra strength. A light handle is fixed to a piece of wood in the
way shown earlier. Scrape insulation from loop 2 and ends 1 and 3, and
solder on connecting wire leads about 18 in long. Tape the joints, and
bring the leads up the handle to where the locator will be fitted.

83
 >
V O
u
X
+ O I

l
m
m cc
a. H

O' co
u u

Q
n CM
Q

—HH-H*
(j

CM
CC
CM
cc
cm
’ ►vww-c1
_i o
u
^-\QOO^OOOOO m
UT “

*i •
•t
u
II
u
>

CC
H
CE

P eg

u
II cm P-H*
co
O
I •'4

O
<O
iu o li­
en

84
85
 C0ILL2

Details of this can be seen from Fig. 4!. It is wound with 34swg enamelled
wire, and the former is 7mm in diameter, with an adjustable core. Three
discs of insulating material about 1 in in diameter arc cut - these may be
1/16th paxolin or perspex, or made from strong cardboard. Each is a
tight fit on the coil former, and is secured with adhesive. Place them to
leave two equal winding spaces, each 5mm wide.

Bring the wire out through a small hole, for point 1, leaving enough for
connecting up later. Wind on three-hundred turns. This section ends at
point 2, and tape over the coil will prevent any turns coming loose. Cut
off the wire, and thread it through a small hole in the top disc to begin
winding the top section - this end is later soldered to point 2. Wind one
hundred and thirty-five turns in the top space, in the same direction as
the lower winding, and finish off at point 3. A narrow strip of insulating
tape will again hold the turns in position.

CIRCUIT

In Fig. 40, TR1 is the search coil oscillator, with feedback to the base by
C2. Cl and Tl, in conjunction with the winding, determine the working
frequency.

TR2 is the reference oscillator. As this has an adjustable core for initial
adjustments,only the fixed capacitor C5 is necessary, with the manual
control VC1 for fine adjustments.

C4 and C7 take the high frequency signals to the diodes D1 and D2, and
the audio output reaches TR3 via C9. The socket is for medium or high
impedance magnetic headphones.

TAG-BOARD

This has two rows of eight tags. Fig. 41. First drill a hole to mount Tl,
and for small bolts to secure L2. Connect together tags 2, 10, 14 and 6,
and run a black lead from tag 6 to go to the switch and battery negative.
Tags 1 and 9 arc also joined, with a red lead for battery positive, and
connection for the phone socket.

Components can then be fitted, cutting down the ends of capacitors and
resistors where necessary. Wires can generally be well spaced from each
other, so that no insulated sleeving is required with these items.

The transistors can be left until last, as this avoids repeated heating of
the leads while soldering other connections. Emitter, base and collector
wires can be identified from the diagram. The transistor types shown
arc readily obtainable, though other devices of similar type might also
be satisfactory.

86
The board requires a small plastic box, large enough to hold the battery
too. VC1 may be a solid-dielectric variable capacitor of ordinary type,
or can be a 250pF compression trimmer, fitted with the type of
threaded 14in shaft which allows a control knob to be used. Place VC1,
switch, and phone outlet socket conveniently on the box, which is then
screwed to the handle, bring the three leads from the search coil
through a hole, and solder them to T1 and tag 1. The leads should be
correctly identified.
1
T1 sets the search coil on frequency. Rotate the core of L2 until a
strong heterodyne is heard, with VC1 half open. Final small adjustments
can then be made by VC1, when using the locator.

Al.Tl RNATIVFCOIL

The cardboard search coil, though very easily made without wood­
working tools, is rather fragile, and susceptible to the effects of
moisture. The locator can be improved by making an alternative coil
on a wooden former.

The winding is GVi in in diameter, and 1/8 in wide. Assuming that the
former is to be built up from three discs of wood, in the way shown in
Fig. 8, the middle disc will thus be 6 Vi in across and 1/8 in thick. It can
be cut from hardboard, or thin plywood. If no other means are
available, it can be cut with a fretsaw, padsaw, or other narrow saw,
while gripped in a vice or clamp, or held flat on the table or bench.

Doth outer discs can be of any thickness, but there is little point in
using hardboard or plywood much over 1/8 in thick, as this will only
make the coil unnecessarily heavy. These two discs should be about
3/8 in to Vi in larger in diameter than the middle piece. Glasspaper the
edges of all three pieces, as rough projections may catch up with the
wire during winding. Glue the pieces securely together.

The coil can be 34swg or 32swg enamelled wire. Make a small hole and
pass the wire up through this, and wind on nine turns. Form a loop,
and bring this up through another hole, or a notch, taking care not to
damage the insulation. Continue winding for a further twenty-two
turns, in the same direction, and secure the wire with a notch or small
hole.

This coil is intended to operate with the same component values as the
basket type coil. For best weatherproofing, apply two coats of outdoor
varnish before winding. After testing the coil, place a turn or two of
insulating tape tightly round the circumference of the coil, to cover
the slot holding the winding.

87
WOODEN CASE

It may also be preferred to construct a wooden box for the locator.

When tliis lias been made on a tag-board approximately 2% x 3% in in
size, a box 4 in high, 3Vi in wide, and 1% in deep inside will easily
accommodate the board and a battery.

Dimensions may be taken from Fig. 42 using either 4mm or 6mm

plywood. The exact size is clearly unimportant, but remember to allow
for overlap when cutting the pieces. Glasspapcr the edges so that they
will meet correctly, and clear away dust. They can then be fitted
together with a woodworking adhesive. Wien this is hard, glasspapcr
tlie edges and corners, and varnish or paint the box.

Two screws will secure the box to the locator handle. The box back is
attached with small screws, so that it can be removed to replace the
battery. The same method of construction is suitable where a loud­
speaker will be provided, an aperture being cut of the same diameter as
the speaker cone. Gauze should be stretched over this opening, inside
the ease, before fitting the speaker.

88
VALUES FOR FIG. 40

R1 390k
R2 220k p

R3 2.2 megohm (Resistors 5% !4 watt)

Cl 4700pF
C2 560pF
C3 0.0 luF
C4 8.2pF
C5 1200pF
C6 lOOpF
C7 8.2pF
C8 2000pF
C9 O.luF
CIO 0.01 uF
T1 1 250pF or 1 500pF compression trimmer.
VC1 200pF or similar value variable.
LI, L2: see text.
Tag-board, coil former, jack socket, slide or toggle on-off switch, etc.
Dl, D2 OA81. 1N67A or similar diodes.
TR1 ISCI08 or 2N3707
TR2 BCI08 or 2N3707
TR3 11008 or 2N3707

89
SPECIAL NOTES FOR OVERSEAS READERS

1) If you should experience any difficulty in obtaining any

"DENCO" products locally, then these can always be obtained
direct from the manufacturers, whose address is shown
below: -
DENCO(CLACTON) LTD
355 - 359 OLD ROAD
CLACTON ON SEA
ESSEX C0153RH
ENGLAND
Tel: Clacton 22807

2) To help our overseas readers, the following list of

possible semiconductor equivalents is shown below:-

BC107 CV9780, AM251.BC147-167-207-317, MPS6566,

SK3020-3122, ZTX107, TT107, RS276-2003.
BC108 CV10541, AM252, BC148-168-208-318, MPS6520,
SK3020, ZTX108, TT108, RS276-2009.
BF194 BF115-121-1 63-173-184-189-197-224-234-238-247B-
594, G13694, MPS3694, SK3018, 40238, SE5056, 2SC460,
2N3694, RS276-2009.
BF195 BF11 5-125-1 63-185-196-173-225-235-237-273/D-595,
G13693, MPS3693, SK3018, SE5056, 2SC535, 2N3693 -
3932, RS276-2009.
MFC
4000B MCI 306-1454-1 524-1 554, MFC6000-8010, SN7 6010.
OA81 AA117-118-132-144, AAZ15, OA91 -95-1 61, SD38,
IN38-56-476-618,1S33, RS276-1136.
OA91 AA117-118-132-144.0A81-95-161, SD38, 1N38-55-63-
617, RS276-1136.
OC44 CV5710-7003, 0070-171 -410-613, 2N1303, BC126,
ASY55, SK3005, 2N36-1191/2-1352-1373, RS276-2003.
1N67A AA117-118-144, OA31 - 95,SK38,1N618.
2N706 BSX19, BSY62/A, MPS706, 2N706A-964. RS276-2009.
2N3704 BC140-337/16, MPS3704, 2N2222, SK3024, RS276-2009.
2N3707 BC149-149B-1 67B-169B-173-173B-184-209-209B-
237A-370, MPS3707-6520, 2SC458, SK3020, RS276-2009.

Although equivalent semiconductors may have similar electrical

properties, physical dimensions may be different and this must
be born in mind if space is tight and for mounting details.
Remember polarities if replacing PNP with NPN types and vice
versa. If in doubt always be advised by your local dealer for
suitable equivalent semiconductors.

\
i
 3) Wire Sizes - Nearest equivalent size is shown:
S.W.C. B & S METRIC
24 23 5. 5
26 25 5
28 27 4
30 28 3
32 29 2.8
34 31 2. 4

4) Remember 1" = 25. 4 mm

i
 S
1

ALSO BY THE SAME AUTHOR

TWO TRANSISTOR ELECTRONIC PROJECTS
BABANI PRESS NO: 30 PRICE : 85p
I.S. B. N. 0 85934 033 3
An ideal book for the beginner and more advanced
hobbyist alike. This book includes 28 Two Trans-
ister Projects designed around relatively inexpen­
sive and freely available components. Clear cir­
cuitry and full construction details, where neces­
sary are shown.
Projects included in this book are:-
Code Practice Set; Door Tone Unit; Door Siren;
Audio Trouble Shooter; Model Mini-Flasher; Double
Blinker; Polarity Protector; Bridge; Metronome,
Minano; Controller for Models; Variable Power Supply;
Priority Indicator; Lie Detector: Automatic Light;
Touchlight; Intercom; VHF Generator; 10.7 MHz Gen­
erator; Megahertz Marker; H. F. Bands Amplifier;
Superhet Tuner; TRF Tuner/Receiver; Push-Pull
Amplifier; Dual-Boost Mixer; Audio Pre-Amplifier,
Impedance Buffer; A. F. Amplifier.

I
•

i
 BERNARDS ft BABANI PRESS RADIO AND ELECTRONICS BOOKS
BPI First Book of Transistor Equivalents and Substitutes 40p
BP2 Handbook of Radio, TV and Industrial ft Transmitting Tube ft Valve
Equiv. 60p
BP3 Handbook of Tested Transistor Circuits 40p
BP4 World's Short, Med. ft Long Wave FM ft TV Broadcasting Stations
Listing (International Edition) 60p
BP5 Handbook of Simple Transistor Circuits 3Sp
BP6 Engineers and Machinists Reference Tables 30p
BP7 Radio and Electronic Colour Codes and Data Chart I5p
BPS Sound and Loudspeaker Manual 50p
BP9 - 38 Practical Tested Diode Circuits for the Home Constructor ??P
BPIO Modern Crystal and Transistor Set Circuits for Beginners 35p
BPI 1 Practical Transistor Novelty Circuits 40p
BPI2 Hi-Fi, PA, Guitar ft Discotheque Amplifier Handbook 75p
BPI3 Electronic Novelties for the Motorist 50p
BPI4 Second Book of Transistor Equivalents 95p
BPIS Constructors Manual of Electronic Circuits for the Home 50p
BPI6 Handbook of Electronic Circuits for the Amateur Photographer 60p
BPI7 Radio Receiver Construction Handbook using IC’s and Transistors 60 p
BPIS Boys and Beginners Book of Practical Radio and Electronics 60p
BP22 79 Electronic Novelty Circuits 75p
BP23 First Book of Practical Electronic Projects 75p
BP24 52 Projects using IC74I (or Equivalents) 7Sp
BP2S How to Build Your Own Electronic and Quartz Controlled Watches ft
Clocks 85p
BP26 Radio Antenna Handbook for Long Distance Reception ft Transmission 85p
BP27 Giant Chart of Radio Electronic Semiconductor ft Logic Symbols 60p
BP28 Resistor Selection Handbook (International Edition) 60p
BP29 Major Solid State Audio Hi-Fi Construction Projects 85p
BP30 Two Transistor Electronic Projects 85p
BP3I Practical Electrical
............................
Re-wiring ft Repairs 85p
BP32 How to Build 1 our Own Metal and Treasure Locators 85 p
BP33 Electronic Calculator Users Handbook 95p
100 A Comprehensive Radio Valve Guide - Book I 40p
121 A Comprehensive Radio Valve Guide - Book 2 40p
126 Boys Book of Crystal Sets 25p
129 Universal Gram-Motor Speed Indicator (Combined 50 ft 60 co model) I0p
138 How to Make Aerials for TV (Band 1-2-3) 25p
143 A Comprehensive Radio Valve Guide - Book 3 40p
150 Practical Radio Inside Out 40p
157 A Comprehensive Radio Valve Guide - Book 4 40p
160 Coil Design and Construction Manual SOp
161 Radio TV and Electronics Data Book 60p
170 Transistor Circuits for Radio Controlled Models 40p
177 Modern Transistor Circuits for Beginners 40p
■78 A Comprehensive Radio Valve Guide - Book 5 40p
183 How to Receive Foreign TV Programmes on your Set by Simple
Modifications 3Sp
196 AF-RF Reactance - Frequency Chart for Constructors I5p
200 Handbook of Practical Electronic Musical Novelties SOp
201 Practical Transistorised Novelties for Hi-Fi Enthusiasts ?5p
202 Handbook of Integrated Circuits (IC's) Equivalents and Substitutes 75p
203 IC’s and Transistor Gadgets Construction Handbook 60p
204 Second Book of Hi-Fi Loudspeaker Enclosures 60p
205 First Book of Hi-Fi Loudspeaker Enclosures
206 Practical Transistor Circuits for Modern Test Equlpmen
207 Practical Electronic Science Projects
•55 Pr»«i<al Stereo and Quadrophony Handbook
209 Modern Tape Recording Handbook
210 The Complete Car Radio Manual
211 First Book of Diode Characteristics Equivalents and Sub
213 Electronic Circuits for Model Railways
N
214 Audio Enthusiasts Handbook
215 Shortwave Circuits and Gear for Experimenters and Ra'
216 Electronic Gadgets and Gaines
217 Solid State Power Supply Handbook
E
218 Build Your Own Electronic Experimenters Laboratory
220 Bu/ld Your Owm SolldScate Hi-Fi and Audio Accessorial BibliOtheek Ned. V6f
RCC Resistor Colour Code Disc Calculator
The Grampians, Shepherds Bush Road, London W6
Telephone: 01-603 258117296