March, 1962 PRACTICAL WIRELESS 1037

SERVICING
TAPE
RECORDERS
FAULTS, SYMPTOMS AND THEIR are able to work a loudspeaker, as if the original
REMEDIES FOR DOMESTIC microphone were connected to the input of the
EQUIPMENT amplifier, instead of the replay head.
By T. S. Smith Gain and Equalisation
There are two important points here. One is
N this new series for the experimenter the that the electrical impulses frorp the replay head
overall construction of tape recorders will be are extremely weak, so great amplification is
dealt with ; how a recording is made and repro- required -more, for instance, than is required for
duced; how the various circuits work. Various an ordinary medium-quality gramophone pick-up.
fault symptoms and conditions will also be means, then, that it is not usually possible
described. to connect a tape replay head across the pick -up
terminals of a radio or radiogram and expect to
Basic Knowledge obtain tape reproduction. A head amplifier and
some form of equalisation (see below) would, at
This approach is essential, since it is virtually least, be required.
impossible to repair a tape recorder successfully Secondly, the signal output from a replay head
without a basic knowledge of how the equipment is not constant over the whole of the audio-
functions. Indeed, the amateur recordist stands frequency spectrum. The output peaks ,towards
a far better chance of securing better recordings the centre of the spectrum and diminishes fairly
if he knows how the various items operate. quickly (depending upon the tape speed) towards
To play a tape record (sometimes called a pre- the high -frequency end. There is also a drop at
recorded tape) a "replay head ", an amplifier and the low -frequency end. In order to correct this
apparent shortcoming, the amplitude of the signal
has to be corrected against frequency, and this
is accomplished by an equalisation network which
gives, in effect, high -frequency lift, and a certain
degree of bass boost. Such a network has to be
included either before, or after, the head amplifier,
and, as this introduces an overall "insertion loss ",
an even greater gain from the replay amplifier
is required.

Tape
magnetism

Transfer
characteristic
Gap-) Magnetising
field
To output of record amplifier
Fig. 1 -How the magnetic field between the gap of
the record head produces small magnets on the coated
side of the tape.

loudspeaker and some mechanism capable of

causing the coated side of the tape to pass at a
constant speed past the replay head, are required.
When all these things are set up, the magnetic
," sound pattern " on the tape is converted back Distorted
across the replay head to the original electrical Input output
impulses, as were present at the microphone signal
during the recording.
The impulses are amplified first in terms of -A distorted tape signal can be produced by the
Fig. 2
voltage and then in terms of power so that they "transfer characteristic".
 1038 PRACTICAL WIRELESS March, 1962

the tape passing the pole pieces is coated with

Tape Transfer a substance that is influenced by magnetism, small
magnetism characteristic magnets are, in fact, formed on the tape. The
length of the small magnets is governed by the
frequency of the sound, while the strength is
Magnetising
governed by the loudness of the sound. The
field. general idea is illustrated in Fig. 1.
Here, it will be seen that the pole pieces are
in two sections with a gap at the top and bottom.
Non-magnetic shims are used to fill the gaps and
the whole assembly is clamped mechanically. The
small magnets can be seen on the tape, and the
wavelength of the recorded sound is related to
the length of the magnets. The lower the wave-
length, the higher the frequency, so for high audio
frequencies the magnets are very small indeed.
Distortion tree The replay head is of very similar construction,
output and on almost all domestic machines the same
head is used for both record and replay. As would
be expected, the top gap dimensions have quite
a bearing on record and replay (especially on
Bias superimposed replay), and to a certain limit, the smaller the
on input signal gap the better the high -frequency reproduction.
More will be said about that later.
Fig. 3 (above)-A supersonic bias superimposed on The Need for a Record Bias
the recording signal at the record head eliminates the Because the magnetism imparted on to the tap,
transfer distortion. by reason of the magnetic field set up between
Fig. 4 (below)-The basic record /playback switching the pole pieces of the recording head is not linearly
of a domestic recorder. related to the magnetic field, severe distortion
would result on replay from a recording pro-
Microphone Note: All switches shown in duced simply as described above. This is called
'playback' position
playback
" transfer distortion," and results from the resi-
dual magnetism retained by the tape during the
S3 recording cycle. The " kink " which produces the
distortion is shown on the transfer characteristic
in Fig. 2.
In order to counteract the effect of the
Take-up " transfer kink " a " supersonic " bias is superim-
spool posed on to the tape along with the record signal.
The bias, being slightly above audio-frequency,
cannot be heard on the recording. Fig. 3 shows
how this supersonic bias eliminates the distortion.

Recording
Tape - At this stage it should be understood that the
amplitude of the bias has quite an influence, not
only on the quality of the recording, but also
on its ' signal -to- noise " ratio. It is also very
important that the bias signal be as pure as pos-
sible. Excessive harmonic content makes it vir-
So much for replay. Now to deal with the tually impossible for the bias to rid the transfer
recording side. To make a tape recording, a characteristic of its kink completely, and another
recording head, an amplifier, an oscillator and, kind of distortion may also occur. It is for this
again, some mechanism to drive the tape at a reason that high -quality recorders use push -pull
constant speed past the recording head are needed. bias oscillators. Some machines have a control
Some programme material to record is also required. for adjusting the bias amplitude, but before hap-
If the programme is from a microphone, then hazard adjustment is made to this control it is
this would be connected to the input of the as well to refer to the instruction manual, as the
amplifier, while the output would be connected recording level and type of tape are related to the
across the recording head. The microphone converts bias amplitude. We shall have more to say about
the sound waves to electrical impulses, which are that later.
considerably magnified by the amplifier. The out-
put of the amplifier is designed in such a way
that quite large current changes occur in the
" electromagnet " of the recording head. These
Erasure
The bias oscillator also serves another purpose
it energises the erase head so that prior to mak-
-
current changes, of course, occur in direct sym-
pathy with the electrical impulses from the ing a recording the originally recorded material is
microphone caused by the sound waves. wiped off the tape. The erase head is rather like
Thus, across the pole pieces of the recording the record and replay head, but does not require
head occur variations in magnetic field, of polarity to be so exacting in its construction.
and strength determined by the original sound. As (Continued on page 1057)
 March, l96 PRACTICAL WIRELESS 1057

Letters to the Editor

The Editor does not necessarily agree with the opinions expressed by his correspondents.

Whilst we are always pleased to assist readers with their crystal detector. I would like to add that the
technical difficulties, we regret that we are unable to supply results are superior to those formerly obtained with
diagrams or provide instructions for modifying commerical
or surplus equipment. We cannot supply alternative details the ratio detector.
for receivers described in these pages. WE CANNOT Another effect which I have noticed, and which
r UNDERTAKE TO ANSWER QUERIES OVER THE TELE-
PHONE. If a postal reply is required a stamped and may be connected with this aspect of the question,
addressed envelope must be enclosed with the coupon from is that the VHF frequency modulated transmissions
-age iii of cover. (from Wenvoe at least) can often be heard weak
METAL CABINETS and distorted at points on the dial other than on
the normal readings. This effect has been noticed
SIR,-Your correspondent T. H. Hughes, who
writes of his difficulty in obtaining metal cases
on both the VHF /F.M. and on the TV trans-
missions, although it is more pronounced on the
to house his test gear (January issue), may be former. -T. G. DAVIES (Llantrisant, Glamorgan -
interested to learn that sheet steel boxes in a shire).
variety of sizes, from 3in. x 3in. x lain. to 12in.
x 12in. x 4in. and possibly larger, are readily VINTAGE MODELS
available as standard electrical hardware. They are
usually referred to, in the electrical contracting SIR, -I fully agree with all that Mr. J. Mansell
trade, as " pressed steel adaptable boxes ". These
I have used with various small items of test gear
etc., but one should however, be wary of using
a
says in the December issues
truly magnificent receiver-an H.M.V. 442
which was built in 1934. I purchased this set in
of P.W. I have
-
steel cases with instruments incorporating sensitive 1937 for £3 in working order and the only repairs
moving coil meters. I have had to pay for were a part rewind of the
No doubt the firm from which Mr. Hughes mains transformer and a MHD4 valve in 1940;
purchases his components could obtain such boxes and a PX4 valve in 1958.
for him through the usual trade channels,-D. J. The volume and bass response is still as good as
Motuus (Birmingham). ever and intend to maintain this set in working
1

order for as long as it is possible to do so. -P. J.

CORRESPONDENTS WANTED PLATER (Wellington, Surrey).

SIR, -I am interested in radio technology and

have a radio service shop of my own. I would
like to correspond with any radio serviceman from
anywhere. All letters received will be answered
Servicing Tape Recorders
promptly. -C. S. YEAR (61 Jin. Market, Bedong, (Continued frorn page 1038)
Kedah, Malaya). The alternating magnetic field created across
the gap of the erase head serves to demagnetise
F.M. QUALITY the tape. An ordinary permanent magnet wiped
SIR, -Your correspondent Mr. Van der Syde's along the tape would give more or less the same
letter in the January issue regarding F.M. result, and some old machines employ permanent
quality and his remarks regarding ratio detectors. magnet erasure, but the " noise " background re-
prompts me to add that I have noticed several maining on the tape is far less by the use of
unusual effects with these transmissions, including an alternating field than by a direct field. Again,
that of the attenuation of the lower audio fre- since the "erasure" signal is above audible limits
quencies. Even allowing for the pre- emphasis of a tape so erased is completely cleared of signal.
the system I feel that there are effects that are not The Basic Domestic Recorder
easily explained.
I have enjoyed and constructed amateur radio The record and replay functions are integrated
apparatus for over 30 years and I will be the first in most domestic machines. Already it has been
to admit that there are difficulties in making and mentioned that the record head works also as
aligning I.F. transformers and ratio detectors of the replay head and that the bias oscillator also
sufficient accuracy to make worthwhile F.M. serves as erase oscillator. In addition, sections of
receivers at home; the great point being to get the the replay amplifier are used for record.
spot on zero at the centre of the I.F. passband. The This means that there is considerable switch-
effects which I have noticed are, however, ing on changing from record to replay and vice-
apparently outside any explanation of this sort. versa. Moreover, the motor itself is invariably
For instance, I have noticed that an A.M. coupled in some way to the "record /replay"
detector will give a very good reproduction of switch, as also is the tape drive capstan. In Fig. 4
VHF/F.M. on either side of the centre frequency. is shoJvn a simplified version of the record /replay
As a matter of fact I have discarded the ratio change -over system.
detector and am now using a normal single diode (To be continued)
 1122
PRACTICAL WIRELESS
April, 1962

SERVI CING-
TAPE
FAULTS, SYMPTOMS AND THEIR
RECO D,DEB
REMEDIES FOR DOMESTIC (Continued from page 1057 of the
EQUIPMENT March issue)
By T. S. Smith
section-between
iN Fig. 4 (page 1038 last month)
a simplified
recording
l head. A he
components and wires feeding k hould becmadea on the
version of the record/playback change the bias signal to
was given. -over system the recording head. Switches are
and are often involved here.
In the "record " position, the sources of trouble.
the erase head, which is energised tape first passes Similar trouble would occur due to over-
oscillator via Sl. After the tape from the bias recording. This often happens when
is "cleaned " of gramme material is being taken the pro-
old recordings, it passes the recording receiver or record player, especiallyfrom a radio
is receiving an amplified signal head, which if the coup-
phone via switches S2 and S3. Thefrom the micro- ling circuit (between the radio or record
player
also receives a bias signal in recording head and the input to the tape recorder) is frequency
recording signal, via switch addition to the sensitive.
S4. Some recording -level
In the `replay " position,
amplifier is connected to the the input of the fairly that they show the level indicators
of the input
are designed so
switches S2 and S4. The microphone replay head via narrow band of frequencies, signal over a
so if one is
nected, as also is the erase is discon- over -recording at, say, a low- frequency,
head, the latter by may not be indicated on the recording -level the effect
SI. The output of the amplifier or meter. Nevertheless, over -recording "eye"
the recording head to the loudspeaker is switched from
If this trouble is suspected, a second would result.
With the above information by S3. recording
to diagnose a number of faults alone, it is possible should be made at a much lower
cording level. If the re-
a simple machine. For example,that may occur on should is "ow of a reasonable quality, attention
would now probably lead to serious distortion be given to the input signal coupling
the bias oscillator and associated an investigation of arrangements.
arrangements. switching No Replay
Record Distortion and No Erase If, after making recording,
Let us suppose that a " clean "
i normal recording, say from
the
tape is used for or in the replay he
could either in
section
tre replay,
ording section
(or both). The best thing
revere distortion is present a microphone. If to do is to try replaying a tape recording
on
should be done. Firstly, a tape replay, two things viously recorded tape. If that replays or pre-
recording known to then, of course, the trouble lies in normally,
)e free of distortion, or the recording
ape should be run througha the previously recorded section of the equipment.
eproduction is now completely machine. If the If there is still no output from
a previously
his is sufficient proof that clear of distortion recorded tape, the tape which
was used for
he amplifier system are freethe replay sections of original recording should
from faults. friend's machine. If that replays, n a
ut trouble lies in the recording then the
ackc on the machine
lade to erase it by switching antaá topotash should pbe machine. The trouble would section of the
be common to both
unning the tape back through to " record " and recording and replay, of course,
if the tape was
le gain controls turned the machine with proved to have no recording
;cording is still present on right down. If the Som. of these techniques areon obvious, it.
oubt that both troubles arereplay, there is little they are adopted they can most but if
caused by a fault a lot of time in searching definitely save
i the bias oscillator. The most usual cause of sections of the system. aimlessly in fault -free
to trouble is valve failure.

ase Normal Typical Record System

If it is discovered that the erase We must now graduate to a real
function is we can obtain very circuit before
orkmg correctly, the recording much more servicing data.
ill be caused by lack of distortion could Such a circuit is given in Fig. 5, and this applies
bias,
e trouble would exist somewherebut in this case essence to in
after the bias Amplifier. the Truvox Type "K " Tape Recorder
 1123
PRACTICAL WIRELESS
April, 1962
oscillator coil, where the
to the a different tap on the through C28 and switch
As drawn, the circuit is switched first voltage signal is slightly stronger,
" record " position. V1 (EF86) is thethat the micro-
amplifier, and it is to that valveJack J'The power supply system y is reasonably straight-
1. This is the mains is pro-
phone signal is coupled, via through an forward. Full isolation from
coupled to the first triode, V2A,to the second by the double -wound mainsadopted. transformer
vided H.T.
equalising arrangement. Coupling the volume or and full-wave H.T. rectification is C27
triode, V2B, is effected through smoothing is provided by the electrolytics
" recording gain " control, R18. stages by R38.
and C30 and H.T. for the earlier motor, and the
Signal to Record Head
Mains feeds are provided for the
solenoid is
It will be seen that switch via"D" couples the brake R39 and the appropriate contacts
on the tape
anode of V2B, C17, to switch
A.F. from the comprising C18, deck.
"A" through an R/C network R27. Switch "A" then Replay
C19, C21, R24, R26 and
the record /i
On " replay", the output fromof VI through
e de.
accontantcurr nta cord-
cRo/uC ens
e replay head is coupled to the input
ing signal to the record head 3104
C E 290V
A B
66Ök
All sections make Ri1?MKt
All switched before break ,Ok MKII
shown except E' R2S 7
More posRbn wX0 4.7M
V

40ÁF

32

R31 4.7k i Brake

0,OD1yF
30 2k s,,,
tRecordlge whack are common to both
outputaand bias oscilla orVonb playback and record
playback
operates asp
V4recorder
playback and record,awhi e stage respectively.
a voltage amplifier
are used for switch "A ". VI now acts asThe equalisation net-
Thus, only the first three stagesthe recording -level for the replay head signal.
the recording proper. V3 is variety, is altered slightly
and works work, coupling V1 to V2A, over and
indicator, of the magic eye of the A.F. by switch " C ". Switch " H" changes
in the following way. A portion R23 to the removes
recording signal is coupled through C 6 from hV2Bt and switcheshes incapacitors, e control,
A D.C. voltage is thus developed
rectifier WX6. the control R19 and C14.
Switches D" and "I" operate feed circuits,
across its load, R25, and appliedof tothe signal so to disconnect
grid. With increase in strength grid and a suit- the A.F. from the recording head anc
a greater bias is applied to the" eye ", this being
the remove the oscillator feedback componentsof V4, whick
able deflection occurs in couple the signal to the control grid
the
proportional to recording -level. C20 is simply is applied tc
now acts as the output valve. H.T.by switch "E,'
rectifier reservoir capacitor. the top of the output transformer conditions fo
and the normal screen operating " B", which als(
Bias Oscillator V4 are established by switch indicator.
V4 (EL84) acts as disconnects the recording -level
In the " record " position,The oscillator coil is
the bias oscillator valve. circuit, while coupling Fault Finding
stages, inpu
connected in the anode a feedback path, As there are several interconnected fault findin
to the control grid, to produceand switch "I ". The and monitor points on most recorders,
is accomplished by C23, R32 is considerably facilitated if a
logical approac
oscillator signal is fed to the head recording head (Continued on page 1145
erase is fed from is adopted.
through C24. Signal for the
 April, 1962 PRACTICAL WIRELESS 1145

(Continued from page 1142) the screws really screwed home tight, the legs are
certainly well able to bear the weight of the cabinet.
Fig. 3 shows the dimensions of the legs and Fig. 4
Diagonal the fixing positions on the cabinet base.
centre lines
Finishing
When the cabinet has been completed it is first
rubbed down with fine sandpaper and given a coat
of linseed oil. This applies to the motor board and
panel as well. When the linseed oil has dried the
whole is given a coat of " Lactoloid " which is
rather like the clear dope used for model aero-
3+ planes. It is recommended that this is carried out
in a warm atmosphere and applied quickly and
Fig. 4 -The positions of the leg fixing blocks on the
evenly with a fairly large soft brush. The cabinet
underside of the base. is once again rubbed down with very fine sand-
paper and given a final thin coat of linseed oil
units the width of the cabinet may have to be and allowed to dry.
increased to 18in. A three -way connecting block is screwed to the
When the motor board has been cut to the bottom of the cabinet near the rear panel and a
manufacturer's template, and fitted, the lid of the hole drilled alongside to take the three core mains
cabinet can now be fixed by means of a piano hinge lead. From this block, current is taken to the
about 35in. in length. The hinge is held by small amplifier, tuner, and motor. A coaxial socket is
¡in. brass countersunk screws. If the lid is fitted fitted in a convenient place at the rear of the
flush at the rear of the cabinet the front when cabinet to connect the F.M. aerial to the tuner.
closed will be found to overlap the front by
approximately }in. This overlap does away with Transfers
the need for a lifting handle. A lid stay is fitted to When the amplifier and tuner have been fitted
ensure that no damage is caused by accidentally and the knobs replaced, the functions of the con-
allowing the lid to drop, particularly when a record trols can be marked with transfers. A small pair
is being played. of tweezers is useful here and the transfers should
be applied after the panel has been varnished.
LIST OF MATERIALS In conclusion, the writer would like to
emphasise that the arrangement is very flexible.
Two faced -chipboard panels 36in. x ¡Sin. x ein. Some constructors might like to put a tape deck
Two faced-chipboard panels 36in. x gin. x áin. in place of the amplifier panel and make suitable
Two plain chipboard panels ¡Sin. x gin. x cut -outs in the front of the cabinet for various
One plain chipboard panel 13 +in. x bin. controls. This arrangement would be quite satis-
IOft softwood, lin. x Iin. (nominal) factory although care would have to be taken in
One piece kin. mahogany -faced plywood (to suit the positioning of the power supply unit in order
panel) to avoid hum. However, whatever arrangement
One piece ;in. mahogany -faced plywood (to suit individual constructors care to adopt, if care is
motor- board) taken with the construction and finishing, the
Four legs builder will have a cabinet of which he will justly
Four fixing blocks for legs feel proud.
Thirty -six l áin. x 6 countersunk brass screws
Eight I ;in. x 6 countersunk steel screws
One piano hinge, 34in. long and screws SERVICING TAPE RECORDERS
One lidstay
Fabric for end and centre panels (Continued from page 1123)
Model aeroplane transfers for control marking For example, if it were discovered that a
One 3 -way terminal block recorder replayed a tape recording, but would not
One coaxial socket record from a microphone, and yet the micro-
Expanded metal for ventilation cut -out phone were in order, a very speedy check would
be to apply a signal to the " radio input " or other
higher voltage input socket, such as Jack 2 in Fig.
Legs 5. If a recording were then possible, the voltage
It is the writer's opinion that the ready made legs amplifier (VI) would be an immediate suspect.
sold for home constructors do not do justice to The " monitor " socket is also useful in this
this cabinet. This applies either to the plain respect, as the signal applied to any of the input
wooden or sbonised legs. It is not difficult or sockets should be heard in a pair of headphones
tedious to cut four well -proportioned legs which connected to the monitor point, such as Jack 3
will look pleasing when fitted. No doubt there will in Fig. 5.
be those who favour ready made legs and the
choice is left to them. Level Indicator
The legs are cut to the dimensions shown. It Another good check is the recording -level
is preferable to make a cardboard template and indicator For instance, if a recording cannot be
mark the four legs with the aid of this. The four made and yet the recording -level indicator operates
fixing blocks are also cut and the legs glued and normally when a signal is applied, the trouble
screwed to these with 1$in. countersunk wood - would most likely lie in the recording head feed
screws. If care is taken with the mating faces and and switching circuits. (To be continued)
 52 PRACTICAL WIRELESS May, 1962

SERVICING
TAPE
EQUALISATION AND
RECORDERS (Continued from page 1145 of the April issue)
RECORDING SPEEDS
Now, if a recording is made at various fre-
By T. S. Smith quencies over the audio spectrum from, say, about
40c /s to 10,000c /s, and the recording is played
S mentioned in the previous article in this back again through a replay amplifier with a per-
series, the recording -level indicator gives a useful fectly flat frequency response with some means of
check on the recording circuits -if a recording monitoring the output in terms of level, a rather
unusual effect will be observed.
cannot be made and the indicator works nor-
mally, the fault probably lies in the recording
head feed and switching circuits such as the Top and Bass Cut
equalisation network and switches "D" and "A" From below 2,000c /s or thereabouts the out-
in the circuit (page 1123, last month). If the put will fall at the rate of about 6dB per octave.
machine also fails to replay, the record /replay Maximum output will occur at about 2,000c /s
head could be open -circuited. (depending on the tape speed), and the output
Poor quality recordings may originate from a will again fall at higher frequencies. The set -up
defect in the "record /replay" equalisation switch- for observing this effect is shown in Fig. 7 (a)
ing, such as switches " C " and " H" in the circuit. and the resulting recording /replay response is
Poor reproduction of a tape recording may be shown in Fig. 7 (b).
caused by a similar fault. The bass cut occurs because the rate of change
In the type of circuit shown, a definite defect of magnetic flux at the replay head decreases with
in V4 stage would cause failure of both the bias decrease of frequency. It should be remembered,
oscillator and replay output stage. The symptoms of course, that the replay head is rather like a
would be no replay and severe recording distortion. dynamo, the output of which increases with in-
accompanied by erase failure. The recording dis- crease in speed of the rotor. The dynamo gives
tortion, of course, would only be revealed by play - an output which is approximately proportional to
ing the tape on another recorder, in which the the rate of change of magnetic flux, which is
replay sections were working correctly. caused by the rotor revolving.

Frequency Response
Supply Take -up
The overall frequency response of a tape record spool spool
can never be perfectly "flat". Let us take the case
of a perfectly flat recording amplifier through Erase Recording
head head
which a tape record is made. The recording amp-
lifier is coupled to the recording head in such a
way that the recording current is constant at all
frequencies. This is called " constant -current "
recording. and is adopted in almost all recorders.
Let us now suppose that the tape is passing
~ a
=3
Constant
current

Frequency --
\
,

the recording head at a constant speed and that

an audio signal is applied to the input of the
recording amplifier, and also that the signal can o, Flat response\
be varied in frequency from the very low end
of the spectrum into the very high end at a con- 1
o
stant level. Frequency-i-
Characteristic
This means that, irrespective of frequency, the d
OT Constant
signal current in the recording head will always Audio output
be the same. The general set -up with frequency generator 7
0> 11 t11.11
characteristics is given in Fig. 6. Here the signal Frequency -0-
source is produced by an audio generator, and
if this test is ever performed it is most important Fig.6-The arrangement of tape recorder, amplifier
to ensure that the signal level from the generator and audio signal generator for producing a constant-
remains the same as the frequency is altered, current recording over the audio -frequency spectrum.

www.americanradiohistory.com
 May, 1962 PRACTICAL WIRELESS 53

The rate of change of magnetic flux at the and 30in. /sec while the home recordist usually has
replay head is, of course, a function of the fre- to make do with a top speed of 7 }in. /sec. This
quency of the recorded sound. The higher the should not be taken to mean that high -quality
frequency, the greater the rate of change of flux recording cannot be made at the lower speeds.
and the greater the output voltage from the replay Indeed, with recent improvements in tape, gap
head winding. Thus, the output goes on increas- dimensions and equalisation, recordings containing
ing at the consistent rate of 6dB per octave of frequencies up to about 10,000c /s are possible
frequency increase until it reaches a peak. at 7 1in. /sec. The very low speeds, however, are
essentially unsuitable for recording high-quality
Gap Width music and are used mainly for speech and dicta-
The output then starts falling very rapidly for tion purposes.
an entirely different reason. Actually, there is more Constant -Current Recording
than one reason for this. One is inherent to the
recording process in that the very short " mag- In most domestic -type recorders the replay out
nets " which are produced on the tape at high put valve is often also employed as the re-
frequency tend to demagnetise themselves, which cording amplifier valve, and a circuit after the
happens normally with very short magnets. style of that shown in Fig. 8 is often evolved.
Another reason has something to do with the Here Vl is the replay output valve with T1 as
gap size of the replay head-the smaller the gap, the output transformer. On "replay ", switch Si
the better the high-frequency response. This is connects the loudspeaker across the secondary of
because a very small gap is shorter than the the transformer in the ordinary way. A further
very small magnets impressed upon the tape at switch usually at point "X" on the diagram, dis-
connects the record /replay head from the output
valve and connects it in place across the input of
Supply Takeup
spool spool 4-20

Erase Playback F10

head head
O

Tape recorded 10
as In Fig. 6
Output .d+ 20
meter
u 30
>
4
Flat response

-
50
60
10 100 1000
Frequency Frequency in e/ls

Fig. 7(a) Fig. 7(b)

If a tape, recorded as shown in Fig. 6, were replayed as shown at 7(a), and the output monitored on an
output meter, a graph of relative output against frequency would appear as shown at 7(b).

high frequencies, whereas a gap of larger dimen- the replay amplifier. The output stage then works
sions may approach the size of the recorded in the normal manner, its signal being received at
magnets and thus give very little or zero output the control grid.
as the frequency is raised. On " record ", however, S1 connects the load
Other reasons are related to the "coercivity resistor R2 across the secondary of the output
of the tape and the speed of the tape past the transformer and the recording head is connected
recording and replay heads. Tape coercivity is a to the network comprising Cl and Rl. The loud-
measure of the ability of the tape to resist speaker, of course, is also disconnected by Si.
demagnetisation, so a tape of relatively high Bias is applied to the recording head through R3
coercivity is likely to result in an improvement and C2. Now, when the recording signal is applied
in high -frequency response, owing to the reduced to the control grid of V1, a fairly high peak
risk of demagnetisation of the higher frequency A.F. voltage occurs across the primary of TI.
magnets. Most modern tapes have a fairly high The output stage is loaded by R2 so there is
coercivity value. no possibility of damage to the output valve or
the output transformer.
Recording Speed
Varying Impedance
Speed is most important so far as the high-
frequency performance is concerned. Far better The recording head contains mostly inductance
recordings can be made at a tape speed of due tb its winding, plus a little resistance. It
15in. /sec than at, for instance, 3 #in. /sec or lower. will thus have an impedance value which increases
Professional recordists use tape speeds of 15in. /sec with increase of frequency of the signal applied

www.americanradiohistory.com
 PRACTICAL WIRELESS May, 1962
54

from the output stage. A typical high impedance unduly troublesome, for, in effect, this is the same
recording head may have an inductance in the as giving a little bass boost to the recording, and
region of 0.8H. Excluding winding resistance, this some circuits may require this.
would give an impedance of 50051 at 100c /s and On the other hand, it may be necessary to
50,00051 at 10,000c /s. provide a little more recording current at the
higher frequencies to compensate for some of the
high -frequency losses occurring in the overall
Signal to H..T+ system. This is sometimes accomplished by con-
recording level TI playback necting a capacitor of suitable value across the
indicator
St recording head feed resistor, as shown by the
Cc broken line in Fig. 8.
r--i---
LS
MERCURY SIX
(Continued from page 51)
L.W. Band
Put the switch in the L.W. position (towards the
right in Fig. 3) and tune in the Light Programme
Recording signal on 500m. Adjust TC2 until this station falls at
1

input roughly the centre of the band. Adjust TC3 for

R3 best volume.
10k It should then be possible to receive some over-
seas long wave transmissions. Choose one near
C2

T300PF
Chassis
TTo bias oscillator
the low wavelength end of the band (tuning con-
denser open) and re- adjust TC2 and TC3 as neces-
sary, for best volume.
Then, locate a station towards the high wave-
length end of the band, and move the long wave
section of the aerial along the rod, for best volume.
Fig. 8 -A circuit diagram giving the essential These stages can be repeated as necessary. to
elements for constant -current recording. bring about any final improvement.
However, since this is fed over such a range Important Notes
of frequencies via a high value resistor R1 The alignment procedure may seem very difficult,
(470k), the varying impedance is almost totally but in practice it is not generally so. Remember
swamped by the resistor, and so far as the signal that an insulated tool, as described, must be used.
source of the anode is concerned, there is a If all wiring is correct, local stations should usually
total impedance variation of only 470,50051 to be heard at reasonable volume almost at once.
520,00051, representing the very small ratio of This gives a signal for initial tests. Deal with the
about 1.2:1. Without R1 there would be a varia- circuits in the order given -I.F. stages, medium
tion of 50051 to 50,00051, which represents a ratio waves, then long waves.
of 100:1. With a meter connected as described. the " no
Clearly, then, the source provides an almost signal " current will probably be around 7mA or
constant current into the recording head network. 8mA, but may be a little outside these limits. If
What really happens, of course, is that the voltage the current is much under 6mA, and the speech
across the recording head rises with frequency and music sound distorted, R18 is probably too
(as the impedance rises), but because the impe- low in value for the transistors used. This would
dance rises at the same time, the current remains be possible if a different pair from those men-
constant. tioned were fitted.
The extent to which the meter reading rises will
Series Resistor Value depend on the volume. and will be around IOmA
The value of R1 is selected to suit the impedance to 15mA for average purpose, but may be up to
of the recording head, and when it becomes neces- 25mA or more for maximum volume. If R18 is too
sary to replace this component, one of equal high in value, current peaks will be much heavier
resistance must be used. If the resistor increases than necessary.
in value, the constant current function will not
be destroyed, but the current in the head will Cabinet Fitting
be decreased and under- recording will result. The receiver is inserted from behind, with the
A decrease in value will upset the constant control wheels projecting through the slots. Adjust
current characteristics and will probably cause the position of the receiver until the wheels can
over -recording at the lower frequencies. The bias turn correctly, then screw the receiver in place.
may also be disturbed, since some of the bias This is done by passing four countersunk screws
signal will be shunted by the relatively low through the holes in the cabinet front, and driving
impedance of the source (i.e., valve anode circuit). them into the hardboard.
Capacitor Cl in the network is essentially to To hold the battery in position, thread a piece of
block D.C. However, its impedance increases with fairly stout elastic through two holes in the
increase in frequency and thus tends to reduce panel. This will take either the square type of
the recording level a little towards the high - 9V transistor battery, or the slightly smaller round
frequency end of the audio spectrum. This is not 9V type of battery. u

www.americanradiohistory.com
 122 PRACTICAL WIRELESS June 1962

SERVICING
TAPE
THE MAGNETIC
RECORDERS (Continued from page 54 of the May issue)
RECORDING TAPE
Tape Tracks
By T. S. Smith There are three chief recording systems. These
are sometimes called " whole track ", "half track "
and "quarter track " (or " single track ", " two -
track " and "four-track " respectively) and are
illustrated in Fig. 9.
GENERAL picture of the recording Single -track recording (Fig. 9a) is used essen-
process and the various circuit sections involved tially by professional recordists for extremely high
has been given in previous articles in this series; quality studio work at a tape speed of 30ín. /sec.
this article will be concerned with a more detailed Two -track recording (Fig. 9b) is practised by the
discussion. Let us begin with the tape itself. It majority of amateur operators, and most domestic
is not intended to delve into the construction of recorders in present use employ two -track working.
magnetic recording tape in any detail, but one or This type of recording is also used professionally.
two points will be of interest. Magnetic recording Four -track recording (Fig. 9c) is gradually
tape was first made with paper as the base, but it
was not long afterwards that cellulose acetate took
its place. Then development was towards a Direction of tape travel
polyvinyl chloride (PVC) base. The overall
development in this direction is just about being
completed by the use of a polyester film base, and
this is now replacing PVC tape -in America it is
used almost exclusively.
Magnetic recording tape requires about six
major features. These are [i] freedom from
"pinholes ", thereby reducing drop -out effects
("drop-outs " refer to lack of recording due to
some imperfection in the tape); [ii] good flexibility,
which permits the tape to remain in intimate
contact with the heads (it also facilitates winding);
[iii] maximum strength to avoid frequent
breakages; [iv] high resistance to temperature (b)
changes and storage defects; [v] immunity to
curling and good dimensional stability over a wide
range of conditions of temperature and humidity;
and [vi] the best possible finish to avoid undue
wear of the heads and to enhance recording and
replay quality.
All these features are given by the modern tapes (e)
now available and, as already intimated, improve-
ments are always being made towards better and Fig. 9-Recording tracks and directions: (a) single
better quality. The tape base is, of course, coated track recording, (b) two -track recording and (c) four -
with an iron oxide of characteristics to provide track recording. (The arrows on the tracks indicate
low noise and good high -frequency sensitivity the top of each track during the recording process.)
(i.e., high coercivity -see page 53, May issue).
Where the base is of PVC, a homogeneous becoming more popular, but there are several
mixture of the plastic and the magnetic material schools of thought concerning the desirability of
is arranged in the form of a finish for the working dividing a }in. tape into four tracks.
side of the tape. Even so, some of the oxide There is no doubt that it is possible to secure
ultimately leaves the tape and often finds its way better results in a less exacting manner by using
on to the heads and associated mechanism, as we as much tape width as possible. With full -track
shall see later. The recording pattern, in terms of recording, the replay gap length covers about
magnetism, is thus imparted to the oxide side of 0.215in. of the oxide face. With half-track
the tape, and in that way the sound signals are operation this is reduced to about 0lin., with
stored. 0.03in. between the two tracks, and with four-track

www.americanradiohistory.com
 PRACTICAL WIRELESS 123
June, 1962

working the width of each track is about 0.043in., Four -track Compatibility
with 0.025in. spacing between the tracks. The two gaps of a four -track recorder are
However, at the time of writing it would seem arranged so that they will scan a part of the track
that four -track parameters have not been finalised, of a two -track mono or stereo tape record. Fig. 11
since some machines are arranged for 0.039in. and shows how this is accomplished. When replaying
(at least one) 0.031in. track widths. The track a two -track mono recording, for example, gap A
spacing also tends to differ somewhat, as also do works in conjunction with track 1. For the other
the equalisation recommendations. track gap A is still used and the tape is inverted
as for normal two -track working.
General Parameters
Tape direction Recording gaps
The generally recommended direction of tape
motion (for all systems) is from left to right with
the coated side of the tape away from the
observer and the recording being made on the
upper track. It is thus usually necessary to
remove the spools and invert the tape (by trans-
posing the two spools) for recording and replay on
the other track unless the machine features
switching to enable recordings to be made on the
lower track with the tape movement reversed.
Four -track operation is undoubtedly useful for (a)
stereo work as it gives the same playing time as
mono on a two-track system (assuming the same
speed, of course). A four -track record /replay
head has two gaps and two windings. Thus, in
Tape direction '-s
Head
Recording gaps

one direction of the spools two tracks are catered

for and by changing over the spools (as for two- ,`.$Track
track working) the other two tracks are produced.
This arrangement gives compatibility for playing gË513é,iï?ii-
back two -track tapes -as will be seen shortly.
The idea can be better understood by reference
to Fig. 10. In Fig. l0a it will be seen that with
the tape placed on the left -hand spool and travel-
,::Track 2
d;:^;;ï1
ling in the direction shown, tracks 1 and 3 can be
recorded or replayed. If we assume two -channel I (b)
stereo, then track 1 will accommodate the signal
of one channel (A), and track 3 the signal of the Fig. 10 -These diagrams show how the four tracks
other channel (B). In order to record on the other are put into order by the use of two gaps and tape
two available tracks the tape is simply inverted as reversal.
shown by the positions of tracks 1 and 3 in
Fig. 10b. Tracks 4 and 2 may then be recorded For stereo, gaps A and B are used simul-
or replayed. taneously, as is already the case with two-track
On four -track mono it is usual first to record on stereo. Although the gaps embrace only a part
track with gap A, reverse the tape physically by
1
of the recorded track, the signal is nevertheless
transposing the spools to record on track 2 with complete, but the output across the replay winding
gap B. reverse the tape again to record on track 3 will be of a smaller level as compared with that
with gap B and finally reverse the tape to record obtained from a head with a gap embracing the
on track 4 with gap A. Replay, of course, follows whole of the track width.
exactly the same pattern. Of course, one is sometimes tempted to arrange
There is, at least, one tape deck, shortly to a machine in a way that the second track is
become generally available, which will replay recorded by switching gaps, thereby avoiding
mono on four tracks, automatically switching removing the spools and inverting the tape. There
from the finished track to the next and so on is nothing against this technically provided the
through the whole tape. When the four tracks " recording sense " of the tape is clearly labelled
have been played in that manner the whole for future reference. (Some two -track and four -
process is automatically repeated if required. The track machines have switching and multiple heads
track configuration is somewhat different in this to enable recordings to be made on all tracks
case, since the tape direction changes on track without the need to transpose the spools, at the
change. same time keeping to the correct recording senses.)
It should be noted that the signal across the In future articles in this series we shall be
replay head winding of a four -track system is investigating the detailed mechanism of specific
appreciably below that on a single or two -track tape decks, but there are several items which are
system and, for that reason, greater amplification common to all decks and these will be considered
is essential fully to drive the power amplifier. first.
Moreover, for equipment of a given specification,
the signal -to -noise ratio on four -track working is Heads and Their Alignment
poorer than that on single or two -track systems.
The equalisation required for optimum replay To achieve optimum high-frequency perform-
quality also differs as the track width is made ance from any tape recorder the gap of the replay
smaller. head must be exactly at right-angles to the edges

www.americanradiohistory.com
 124 PRACTICAL WIRELESS June, 1962
of the tape except where a machine is used solely Face angle adjustment Azimuth
to replay its own recordings. In this case any adjustment
effects occur equally on record and replay and the
results of a small misalignment are counteracted.
However, where the same machine is used to Gap
replay either tape records or recordings produced
on another machine, misalignment will produce Head

Tape direction >

undue background noise and loss of treble.

Gap A
base plate

Azimuth
adjustment

Coil spring
Fig. 12-One arrangement used for the face -angle
adjustment and the azimuth adjustment of the
Gap B record/replay head.
Fig. 1I -How a four -track, two -gap head can be used cannot be obtained with the replay volume control
to replay two -track stereo or mono recordings. For set to about three -quarters, a more sensitive output
mono, gap A only is used and track 2 is replayed by meter or A.C. voltmeter is required.
inverting the tape, while for stereo both gaps are On two- and four -track machines the height of
used simultaneously. the head is of considerable importance to ensure
correct alignment with all tracks. It may be
Most machines provide adjustments for setting necessary, therefore, first to set the height for
both the angle of the face of the head to the maximum output on one track of a multi -track
vertical and the angle of the gap itself to the test tape (frequency is not so important here).
vertical (known as azimuth adjustment). These Then the face -angle adjustment must be set and,
adjustments are provided in a diversity of ways, finally, the azimuth adjustment for maximum
depending upon the design and type of machine, high- frequency response. All the other tracks
but a popular arrangement incorporates three should then be in correct alignment.
adjusting screws arranged appropriately around On some machines the height of the head may
the baseplate of the head as shown in Fig. 12. not be adjustable, as the azimuth adjustment may
Coiled springs or rubber washers are sometimes pivot on a small dimple on the head baseplate, in
used between the head baseplate and the panel of which case there may be only two adjusting
the deck at each screw to maintain a constant screws. On the other hand, some machines have
tension and to hold the head firm at all adjustment four adjusting screws, which are best dealt with in
settings. pairs. Normally it is not desirable to interfere
The idea is first to set the face angle adjustment with the overall height of the head, the main
so that when the tape is running past the head adjustment being that of the azimuth, which can
with the pressure pad released and with the tape usually be set without disturbing any of the other
guides in correct alignment the tape remains in adjustments.
contact with the centre of the gap area over its
whole width without curling, twisting or jumping. Magnetised Tools
Under this condition the face of the head will be
parallel with the back of the tape. Residual magnetism is a very real danger so far
as tape recorders are concerned, and it is most
Azimuth Angle Adjustment important that all adjustments in proximity to the
There is only one method of setting the azimuth heads and associated mechanism are carried out
with non -magnetic tools. Almost all steel screw-
angle for optimum results and that is by playing drivers. for example, used for radio work are
back a test tape carrying a constant level, high endowed with slight magnetism, picked up from
frequency recording. Test tapes of this nature are loudspeaker fields and so on.
available commercially. The azimuth screws Non -magnetic screwdrivers are available and a
should then be very carefully adjusted to give set is well worth acquiring if much work is done
maximum output of the test tone. on tape recorders. A bulk eraser can be used for
As small chances in level at high frequencies clearing magnetism from small tools, and there
are extremely difficult to detect by ear an A.C. are other methods of dealing with the problem, as
voltmeter or output meter. which is capable of will be given later. It must be remembered,
responding to the test tone, should either be though, that residual magnetism on a tape guide,
connected across the loudspeaker or across a mu -metal screen or head, considerably impairs the
suitable load resistor used in place of the loud- noise performance and can introduce hiss on to a
speaker. The load resistor should be of the wire - tape record which was previously noise free.
wound variety and its resistance should closely
approximate to the impedance of the loudspeaker. (To be continued)
The replay volume control should be adjusted REMEMBER TO BUY
to give a little under half -scale deflection on the
meter, but care should be taken to avoid overload- OUR COMPANION JOURNAL
ing the replay amplifier in order to achieve PRACTICAL TELEVISION
sufficient deflection. If the desired deflection 119 Every Month

www.americanradiohistory.com
 July, 1962 PRACTICAL WIRELESS 20Z

SERVICING
TAPE
RECORDERS (Continued from page 124 of the June issue)
THE RECORDING,
PLAYBACK AND ERASE HEADS.
By T. S. Smith fairly close spacing is nevertheless required as a
means of keeping the reluctance of the magnetic
circuit within the pole pieces as small as possible.
01-1AST month we investigated the basic The pole pieces themselves are composed of a
requirements of magnetic recording tape and the material possessing a high permeability and
need for correct head alignment; we will start Mu -metal is often used. The sensitivity of the
this month's article by looking in greater detail heads is also related to the magnetic permeability
at the heads themselves and the circuits that feed of the pole pieces as well as to the number of
them. turns on the winding and the current in the
We have already discovered that
three bead functions are required:
one for recording-converting the
signal voltages to magnetic flux
changes and transferring these
permanently on to the magnetic
I I 1

speed: 751n/sec. -
tape -the second for replay-con- o I

0 000lln.
verting the magnetic programme . gap space-
pattern on the tape back into 5
signal voltages -and the third for
erasing programme material that
is no longer required so as to
0.00025 n.
gap spacer -F `

leave the tape "clean " for the :

next recording.
The three heads have much in ;
common and, as has been told in
past articles, the recording and ,
playback functions are nearly -2
always carried out by the same 3 4 5 6 78 10 20 __ ,.
2
head on most domestic machines.
The average tape recorder thus Frequency (kc/s)
has two heads, one for recording
and playback and the other for Fig. 13 -How the high -frequency response of a replay head is improved
erase. The basic construction of a by reducing the width of the working gap.
recording playback head was
shown in Fig. 1 of the first article also governs the
of this series (page 1037, March, 1962), and this winding. The number of turns -a impedance head
illustration also reveals how the changes in impedance of the headturns high than a low impedance
magnetic flux across the working gap transfer has considerably more
magnetic signals on to the tape. head but, as the latter is fed to or from a trans-
former, the correct signal voltage step -up is
Gap Spacing achieved. In other words, the voltage across a
only one low impedance head is smaller than that across
In some cases, however, there may be a high impedance head, but in the former case the
winding (as opposed to the two windings shown on signal current is greater.
the diagram mentioned) and there may not be a
rear gap, depending upon the precise design of the
Frequency Response
head. The most important feature is the front
gap. This must be highly engineered so that it The high- frequency response at a given tape
is straight and uniform, and to facilitate these speed is very closely related to the width of the
requirements a non -magnetic " spacer " is inserted working gap in the playback head in particular.
between the two halves of the pole pieces. The dimension of the gap of the recording head is
Although the spacing of the rear gap is also important, of course, but not so much as that
nowhere near as critical as that of the front gap, of the playback head. However, since composite

www.americanradiohistory.com
 208 PRACTICAL WIRELESS July, 1962

recording /playback heads are invariably used, the

recording gap width requirement is adequately
catered for within the gap requirements of the
playback function.
From the high- frequency aspect, therefore, we Short for
can consider primarily the playback action. Past oscillator
articles have explained that the signal EMF Signal test
from
generated across the winding is proportional to the Record
rare of change of magnetic flux. Thus, with a amplifier
tape of constant recording over the audio
spectrum, the signal EMF across the playback
winding will rise with frequency at the rate of
6dB per octave up to a certain frequency which is
governed by the gap width and other factors, as
will be discussed.
6dB Per Octave
An increase of one octave is a two-to -one rise
in frequency which, since the output is propor-
tional to the rate of change of flux, gives a two-
to -one rise in signal EMF (voltage) across the
replay winding and, because a two -to -one voltage
ratio is exactly the same thing as a rise of 6dB, it
becomes perfectly clear why the replay signal
voltage rises at the rate of 6dB per octave. This Fig. 14-The basic circuit of a Hartley oscillator.
happens at all speeds, but the advantage of higher
speeds is that the point at which the 6dB per
octave rise ceases extends further into the higher claimed to have a top response approaching 15kc /s
audio-frequency spectrum. This is because the at 3Iin. /sec are now readily available, and quite
peak occurs when the wavelength of the tape good quality sound can be produced at a tape
signal is about twice the gap width. Thus, if the speed as low as 1 -in. /sec-using transistors.
gap has an effective width of, say, 0.0005in., the
wavelength would be 0.001in., which at a tape TV Recorders
speed of 7 1in. /sec would represent a frequency of Many of these improvements have resulted
7.5kc /s. At 3 }in. /sec the peak would occur at from developments in the field of video tape
approximately 3.75kc/s and at approximately recorders and, while we are still on the subject
15kc /s at 15in. /sec. of frequency response, a quick look at such
recording systems would not be amiss. The
Other Factors Ampex machine uses a replay head gap somewhat
Unfortunately, the effective gap width is some- less than one -quarter thousandth of an inch, and
what greater than the thickness of the spacer, and at a head -to -tape velocity of 1,500in. /sec a
because of this and other things like imperfect response up to 4Mc/s is maintained. The tape
contact between the head and the tape and losses itself is moving longitudinally at a speed of
in the head due to eddy currents and hysteresis, l5in. /sec, but the effective high tape speed is
the peak output usually occurs at a lower achieved by the head rotating and scanning the
frequency than that calculated in relation to the tape over its width.
gap width. The British system, on the other hand, operates
This is illustrated by the curves in Fig. 13 which in the more conventional manner and at a tape
show the relative replay responses for a head of speed of 200in./sec gives a response which at
0.0001in. gap and for a head of 0.00025in. gap 2.5Mc /s is only 3dB down. As would be
under equal recording conditions. It will be seen expected, considerable wear occurs on the head
that the peak occurs in both cases at about 5kc /s, at such high tape speeds, and on both machines
but that the curve of the head with the 0.0001 in. the heads need replacing after about 100 hours
spacer does not fall off so rapidly as the other of use!
and extends further into the high- frequency The gap width of the British system is
spectrum. These curves are taken from the 0.00002ín. and at such a small dimension the
Gresham range of heads which employs Mu-metal ordinary type of spacer is rarely used. Instead
pole pieces. quartz or some other non -magnetic material is
The main reason why the effective gap width is ` sputtered " or vacuum deposited between the
not equal to the thickness of the gap spacer is interfaces of the pole pieces. This technique is
because of imperfect contact between the spacer now being examined for use on sound replay
and the faces of the pole pieces. However, new heads and it should not be very long now before
manufacturing techniques now make it possible further improvements are found in the domestic
to achieve an effective gap width which is con- and professional recorder, leading to enhanced
siderably closer to the spacer thickness than was high- frequency response at low speeds.
possible on early heads. These have also made
possible the recent "low speed " recorder, the Poor Top Response
production of which several years ago would have The tape is held in closecontact with the head
been considered almost impossible. Machines either by pressure pads or by the tape being run

www.americanradiohistory.com
 July, 1962 PRACTICAL WIRELESS 209

through special guides either side of the head, so should not normally be visible to the naked eye,
that the tape is under slight pressure against the and if it is-except for the erase head gap -the
pole pieces. The former arrangement is the most head may need replacement.
popular, but the latter arrangement has much to Excessive head wear coupled with impaired
commend it and will possibly be found more in H.F. response may also be caused by incorrect
machines of the future. adjustment of the pressure pads or tape guides.
Apart from incorrect azimuth adjustment, as With the pinch roller disengaged from the
dealt with last month, another common cause of capstan, the tape should pass the centre of the head
poor top response is inadequate contact between (with the pressure pads released) without kink or
the tape and the head. The working surface of curl. If this does not happen then the guides
the head sometimes picks u? oxide deposits from should be adjusted in height to give the desired
the tape after considerable use which hold the effect. The guides often wear badly themselves,
tape away from the gap. As the deposits are but this can usually be overcome without cost by
likely to become very hard with time a magnify- rotating them so that a fresh surface is presented
ing glass is often necessary to detect them, and to the tape.
extreme caution should be exercised when
removing them. Optimum Pressure
Too much tension on the pressure pad will not
R2
-).-
HI+
hold the tape any straighter nor will it enhance
the high- frequency response; it will simply wear
Cl out the heads long before the end of their life is
really due. If excessive pressure appears to give
a better top response, then the trouble is caused
either by a worn head or oxide deposits adjacent
to the gap (or incorrect azimuth adjustment). On
the other hand, too low a pressure will impair the
H.F. response, but the pressure should be as low
as possible consistent with optimum H.F. response.
It is surprising just how the H.F. response falls
C3 when contact of the tape with the head is reduced.
The loss in dB can be computed approximately by
multiplying the gap distance between the tape and
the head by 55 and then dividing the answer by
the wavelength of the signal being reproduced.
Record;;OErase:'
Thus, with a one tenth of one thousandth of an
head he,l inch loss of contact on a tape running at 7-in. /sec
recording a signal at 7.5kc/s, the cut would be
5.5dB. Varying pressure of the tape against the
head is also a cause of noise modulation, and this
could result from a twisting tape or wear of the
Signal from Record amplifier head.
Erasing
Fig. IS-The basic circuit of a Meissner oscillator.
The erase head, although of similar design to
the recording /playback head, has a considerably
Removal of Oxide Deposits wider gap of some five thousandths of an inch or
so. The pole pieces may be either of laminated
On no account should a pointed instrument be Mu -metal or ferrite and the supersonic erase
employed to pick away the deposits from the pole signal is applied at fairly high power to a com-
laminations or gap. The best thing is to endeavour paratively low impedance winding. To supply
to dissolve the deposits with a good quality sufficient erase power the oscillator often incor-
lighter fuel or carbon tetrachloride. A piece of porates a pentode valve (or power tetrode) to feed
lint -free cloth should be soaked in the solvent and between 200 and 500mA of erase current into the
applied with reasonable pressure to the working winding of the erase head, and as we have
area of the head. Care should be taken to avoid already seen (Fig. 5, page 1123, April, 1962, issue),
either solvent coming into contact with the tape, the output valve in the playback amplifier may
and lighter fuel (petrol) should be kept well clear take over as bias and erase signal oscillator on
of rubber drive wheels. Carbon tetrachloride does record.
no harm to rubber, however, and is useful for In Fig. 14 is given the circuit of a Hartley
roughening and cleaning such drives. oscillator, which is very popular in tape recorders.
During the course of servicing the head in the The frequency is governed by the tuned circuit
foregoing manner it may be discovered that the comprised of LI and Cl, and a good test for
pole surface is badly stepped, indicating wear. oscillation is to insert a milliameter in series with
While consistent wear over the area in contact the H.T. feed to the anode circuit, as shown. The
with the tape is of little consequence (provided it reading should be noted and, if there is an
is not too severe, of course, and has not widened appreciable rise in anode current when Cl is
the gap), unbalanced wear, resulting in a sloped short -circuited, this is proof enough that the
face, nearly always calls for head replacement. circuit is oscillating with the short removed.
Again, a magnifying glass or small miscroscope is
useful for investigating head wear. The gap (Continued on page 227)

www.americanradiohistory.com
 July, 1962 PRACTICAL WIRELESS 227,

Short -wave Listeners' Log

ANY S.W. receivers, including those of Receiver Matching
'simple type, can perform well with inefficient Best results are obtained when the aerial feeder
aerials. For this reason poor aerials are often impedance matches the receiver input impedance.
used and probably give reception of stations over With a Zepp or doublet feeder an aerial tuner
a distance of thousands of miles. Despite this, will tune the feeder and allow matching to the
when really good S.W. results are wanted, an receiver. Such a tuner is any air -spaced variable
efficient type of aerial is worth while. For real capacitor and parallel coil, tunable to the
Dx working, such as reception of Australian and operating frequency. Each feeder has a clip and
New Zealand stations, such an aerial can make is tapped on to the coil equal distances from the
all the difference. Remote Dx may be inaudible centre tap, which is earthed. The receiver aerial
with a poor aerial but come in well with a good lead also has a clip which can be taken to any
aerial. turn on the coil. Aerial tuning and clips are
The requirements of a good aerial can be put adjusted for best volume.
under a number of headings. Very often all these
points cannot be met, but adhering to even only Directivity
one or two will greatly improve results if the
present aerial is poor. Aerials of the kind mentioned are not very
directive, so there is no need to orient the wire
Signal Pick -Up any particular way and good reception can be
This should naturally be as large as possible. expected from all directions. There is also
Height above ground and the distance from usually no great loss of results if the aerial is
earthed objects is important. Other things being sloping.
equal, a doubling of effective height will increase Materials
signal strength approximately four times. For
short aerials pick -up is roughly proportional to Stranded wire (about 7/26) or solid wire (about
length, so time spent in getting a reasonable 14s.w.g.) will do well. The aerial should be one
length of wire as high as possible will be more uncut length or any joints must be soldered.
than justified. A good ribbed glass or similar insulator should
be fitted at each suspension point. If the down -
Lack of Noise lead is not screened it should be well clear of
Local noise may blanket out weak signals and walls, etc.
so the aerial should be remote from mains wiring,
etc. and the downlead should also be clear of Earth
such wiring or be of the anti -noise type. The Finally a reasonably stout, short lead to a good
simplest anti -noise down lead is 750 coaxial cable, earth will always help. A copper or plated earth
taken to the centre of a dipole cut for a chosen spike or other non -corrosive metal object actually
band. An open wire transposed feeder is also buried in the ground will do well for this purpose.
helpful and allows the aerial to be used on several
bands. An aerial, a } -wave long, may be connected
to a 750 coaxial downlead, the outer braiding of
which is earthed.
Multi -band use
Servicing Tape Recorders
One aerial for all bands is often the aim of (Continued from page 209)
S.W.L.'s and an end -connected wire, taken to the The smaller current required for bias is fed to
receiver, will work on all bands. An excellent the recording head through C2 and R1, while the
all -band aerial is the tuned doublet, which is an larger erase current is fed through C3 direct to
aerial, cut in the centre, with two leads, held the erase head.
roughly 4in. apart with insulated spreaders,
descending from this point. Prevention of Surges
The Zepp feeder will also work well on all
bands and is fundamentally the same as the Resistor R2 and capacitor C4 give a fairly long
doublet feeder, but one feeder wire goes to the time-constant which ensures that the bias and
end of the horizontal aerial, the other ending at erase signal amplitude rises slowly on switching
an insulator. on and falls slowly on switching off, and in this
way transient surges of signals are avoided.
Single Band Use To conclude this article, Fig. 15 shows the
Listening on one particular band allows the Meissner oscillator which is also frequently
aerial to be chosen' to suit and dipoles, with a employed in domestic machines. This uses two
coaxial feeder as described, are largely used. The coils-one for feedback- instead of the tapped
lengths for the popular Dx bands are 22ft. for coil of Fig. 14. Otherwise the operation is
21Mc/s and 33ft. for 14Mc /s (the feeder length is similar to that already described.
unimportant). (To be continued)

www.americanradiohistory.com
 PRACTICAL WIRELESS 313
August, 1962

SEI[VICIN /.
TAPE
RECORDING AND
RECORDERS
PLAYBACK EQUALISATION
(Continued from page 227 of the July issue)
By T. S. Smith

shows what the replay response would be like by

OL ET us recapitulate a little and have another using treble boost on record and zero bass lift on
look at the diagrams on pages 52 and 53 of the playback. At around 3,000c /s. the bass falls as
May, 1962, issue. Fig. 6 showed how a constant before (Fig. 16) at the rate of 6dB /octave, but
current recording is made so that the magnetic
there is a definite indication of treble lift and the
induction on the tape is equal over the audio response is approximately flat from about 3.000c /s
spectrum. Fig. 7(b) showed how the output on to 10kc /s. It is still far from perfect, of course,
replay would rise with frequency to a peak due to
and would sound thin and lacking in lower
the tape signal passing through a replay amplifier
frequencies.
with a flat frequency response, as in Fig. 7(a).
The article last month made clear why the out-
put rises at a constant rate of 6dB /octave to a
maximum and why the output falls as the fre-
quency is further increased. The whole point in
question, therefore, is how to achieve an output
which is flat over the greater part of the A.F.
spectrum. A typical frequency response character-
istic is shown again in Fig. 16.
From a study of this, it becomes obvious that
two things must happen. Firstly, the lower fre-
quencies up to the peak (often known as the o
" turnover point ") must be boosted, and the higher
frequencies after the turnover point must also be
boosted. These things are carried out partly during 10
100 1000 10000
the recording process and partly during the replay Frequency (c/s
operations. For example, on ' record '. a con-
siderable treble boost or lift is applied to the Fig. 17 (above) -How the record amplifier response
amplifier at a frequency and magnitude depending is given a boost at the high -frequency end to avoid
on the tape speed and the exact characteristics of the treble falling too rapidly after the turnover point
the heads. (see Fig. 16).
The curve in Fig. 17 shows such treble lift on a
recording amplifier, while the curve in Fig. 18
20
40 m0
m
a
10
5
30 á
0 0
w
>
° 20
ct 10
m
10

-20 100 1000 10000

0 1000 10000 Frequency (c/s)
100
Frequency (c, s) Fig. 18 -The replay characteristics on a flat amplifier
Fig. 16 -A typical frequency response characteristic from a tape which has been recorded with treble
with the bass falling at the rate of 6dB /octave. boost, as at Fig. 17.

www.americanradiohistory.com
 314 PRACTICAL WIRELESS August, 1962

signal, but when the signal is

applied to the " radio " jack,
30 attenuation occurs due to the
potential- divider network R2
25 and R3. This constitutes a
form of " level " equalisa-
tion, as distinct from fre-
quency equalisation, which is
the prime subject of this
article.
The amplified A.F. is
176 In/sec, developed across the anode
load resistor R4, and from here
31h/sec. it is fed through C2, R5 and
R6 to the resonant circuit
LI /C1. Now, at frequencies
010 away from resonance, the signal
30 100 300 1000 3000 10000 level applied to the voltage
Frequency (c/s) amplifier is dependent on the
ratio of R5 to R6. The net-
Fig. 19 -Three degrees of bass boost corresponding to different tape work Cl /L1 as far as the signal
speeds. Such boost is usually applied to the replay amplifier. is concerned is low impedance,
meaning that the bottom end
of R6 can be considered almost
as connected to chassis. Let
In Fig. 19 are shown three degrees of bass lift us suppose that R5 and R6 are of equal value and
on playback corresponding to the three indicated that C2 is sufficiently high to avoid low frequency
tape speeds. In a similar pattern, three degrees of attenuation. Then, off resonance, half the signal at
treble lift when recording are given by the curves the anode of VI will be fed to the voltage amplifier.
in Fig. 20, while the curves in Fig. 21 reveal the What happens at resonance? Since the circuit is
overall frequency response characteristics at the parallel- tuned, the impedance across its terminals
three tape speeds of l , 31 and 7 +in. /sec. These, (A -B) rises sharply at resonance to a value
of course, are created by making the recordings governed by the " goodness value " (i.e., " Q ") of
according to the patterns in Fig. 20 and replaying the tuned elements. Let us suppose that Ll /C1 is
them with the responses of Fig. 19. tuned to 8kc /s and that at that frequency the im-
This, then, is equalisation, and the curves in pedance across A -B in series with R6 causes the
Figs. 19, 20 and 21 relate to a tape amplifier circuit ratio R5:R6 to rise from half to three -quarters.
designed by Mullard Limited and described in In this event, then, the voltage applied to the
their leaflet TP421 under the title
"3W Tape Amplifier Circuit ". 25
334
Treble Boost Circuit in./seç
20
The treble is usually boosted m 7%2
in the coupling following the in /sec, in,/seç
15
microphone amplifier or pre-
amplifier of the recording
Channel. Quite a reasonable ó 10
amount of boost is needed as >
may be seen from Fig. 20, and m 5
this is invariably accomplished cr
by a resonant circuit consisting
of an inductor and parallel- o
connected capacitor (sometimes
the capacitance is contributed 510
by that of the winding of the 30 100 300 1000 3000 10000
inductance). Frequency (cis)
In Fig. 22 is given the circuit
of a pre -amplifier which is Fig. 20 -Three degrees of treble boost, showing how the boost frequency
often the first stage in the is related to tape speed.
recording channel (this could
also be the first stage in the playback channel amplifier would also rise from half to three -
which, on " playback ", would have the head quarters -but only at resonance. In this way,
switched into circuit instead of a microphone therefore, the boost is applied in the recording
or radio). The pre -amp stage is usually response at the required frequency.
designed to accept either a microphone
signal or a signal from the detector circuits of a Correction for Different Speeds
radio receiver. When the signal is coupled to the
"mie" jack, it is applied direct to the As shown by the curves at Fig. 20, top boost
control grid of VI, across R I. This is because has to come in at frequencies to suit the tape speed
full gain is necessary on the low -level microphone (and also the characteristics of the head), and

www.americanradiohistory.com
 PRACTICAL WIRELESS 315
August, 1962

therefore, on multi -speed recorders, some means of Fig. 23-How the capaci-
tive element of the resonant
switching the recording equalisation is required. < R5

,
circuit is switch to give Positions for St
This is simply accomplished on the resonant type treble boost at the fre- K. t % in/sec
of circuit by switching in the appropriate parallel quency to suit the tape (2).33/yin/sec,
capacitor, as shown in Fig. 23. The values given speed. in/sec.
are taken from the Mullard amplifier referred toa
(3) .7x2
amWif
in the foregoing text, and Ll is wound on
Mullard Ferroxcube pot core, Type WF816. This
gives high efficiency (Q) and excellent equalisation
at the lower tape speeds. The frequency of
resonance of any tuned circuit is equal to
1

2H[106.L.C]
where C is in microfarads and L in Henrys.

10 Bass Boost for Playback

m 71/2i .
Now is the time to look at the
v 5 bass boost circuitry used in the
7y2 in,/sec, playback channel. Such a circuit
is given in Fig. 24 and, as with
0 the recording channel, the bass
ó
5 5
á 1%8, 33iá in./sec 1 g in/sec,
-
equalisation follows the pre-
amplifier in this case, the
replay head amplifier. Instead
of an inductor, a capacitor is
to used in the coupling network,
33/4 In/sec.
which is Cl in the circuit. What
,c
too 300 1000 ---- ----- happens is that the capacitor
to 30 has a low impedance at high
Frequency (c/s) frequencies and an increasing
of Fig. 19 is applied to the replay amplifier, impedance as the frequency is
Fig. 21 -When the bass boost
the decreased. Thus, a potential
boost of Fig. 20 is applied to the record amplifier,
and the treble divider is formed by R1 in one
resultant overall response is os shown here. rR` arm and by R2 and Xc in series
in the other arm, and the
In certain cases, L is made adjustable with a
output voltage (that applied to the voltage ampli-
be fier) is that which occurs across R2 and Xc in serves.
dust -iron core, and misadjustment here should Now, at the high- frequency end of the A,F.
suspected if there occurs a distinct loss of treble spectrum, Xc contributes but little to the bottom
which is definitely not caused by head misalign-
defects. In a case arm of the potential divider, and the output voltage
ment or other more obvious is almost proportional to the ratio of RI to R2.
such as this, however, it would be as well to check
tape or tape record However, at decreasing frequencies Xc adds to R2
the replay channel on a test progressively, thereby producing an alteration in
and if " top " were then present, then the trouble
lie in the record channel. the overall ratio. (To be continued)
would almost certainly
HT+ HT+
Average
slope=
C2 6dB/
Pre -amply tier S EF86
Freq. -+ Head
amplifier
octave

Microphone Frequency -

Jack
To
voltage
amplifier
R6 To
voltage
ampi if
R2

C1 (xc)

Fig. 22 -The pre-amplifier (or microphone amplifier) reply

Fig. 24-How the bass boost is applied to the
stage of a record amplifier with treble boost introduced
pre -amplifier coupling network.
in the coupling circuit by the tuned circuit LI and Cl.

www.americanradiohistory.com
 September, 1962 PRACTICAL WIRELESS 403

SERVICING
TAPE
RECORDERS
PROGRAMME SOURCES
(Continued from page 315 of the August issue)
AND SIGNALS
By T. S. Smith microphone to vibrate in sympathy, and then to
use the signal produced by the microphone for
recording (see Fig. 25). By far the best idea is to
utilise the A.F. produced by the set as the record-
CXAST month we dealt with record and replay ing signal (Fig. 26).
equalisation and discovered how a reasonably flat This, then, presents several problems; the best
overall frequency response can be obtained by way of extracting the A.F., etc. The obvious
applying controlled treble lift on record and' bass arrangement is to disconnect the loudspeaker and
boost on replay. This month we shall be consider- apply the two loudspeaker wires to the input of
ing the various programme sources and signals the recorder. This, although obvious, is a bad
used by the home recordist. thing to do for several reasons. One is that as
soon as the loudspeaker is disconnected from the
Recording from the Radio set, the output stage is operating without a load.
The radio is undoubtedly the most used source This immediately introduces quite a lot of extra
of programme material exploited by the amateur distortion, but not only that, it also causes the A.F.
but. unfortunately, full advantage is not always voltage to rise to dangerous peaks at the anode of
taken of this medium and consequently the quality the output valve and across the primary of the
of reproduction may be sadly lacking. output transformer. Before very long either the
The easiest way of getting a radio recording is valve would flash over or the primary of the trans-
to site the microphone in front of the loudspeaker former would short -circuit or break down.
and record in the usual manner. Although the
results so obtained are acceptable to some, this is
a very poor method of recording technically and Aerial Microphone
far better results are possible by the use of other Receiver
methods. With the loudspeaker/microphone Programme
sound
method, three basic distortions are introduced to
the recorder, and these are: (i) the inherent dis-
tortion in the output stage of the receiver (even
the best sets produce some distortion in the output
stage); (ii) the distortion and coloration produced
by the loudspeaker itself and by its baffle or
enclosure; and (iii) the distortion, reverberation
and coloration reflection reflected from the room. Signal
The first kind of distortion is present always and to tape
recorder
cannot be reduced or eliminated; (ii) is, again,
always present and depends much on the loud- Fig. 25-The simplest way of recording from the
speaker and enclosure employed by the receiver; radio is to site the microphone in front of the loud-
(iii) however, can be varied to some extent by play- speaker. This is technically poor, however, since
ing with the spacing between the loudspeaker and there are various distortions introduced to the
microphone and by cutting out excessive reverbera- programme signal as this diagram shows.
tion by covering the channel between the
loudspeaker and microphone with a heavy cloth. If this method of feeding is to be used, then it
The loudspeaker /microphone channel is virtually is 'essential for either the loudspeaker to remain
an electro- acoustic transducer, in which there are connected or a resistive load to be used instead.
two distortions to contend with: the " electro " The resistive load should have a value equal to
distortion and the acoustical distortion, as already the impedance of the loudspeaker, which, in most
described. ordinary valve -type receivers, is of the order of
3f2. With transistor receivers, the impedance may
Eliminating the Acoustics be entirely different and in some cases the loud-
It is rather pointless to use the A.F. signal speaker speech coil may be centre -tapped, in which
produced by the set to operate a loudspeaker, and case special precautions will have to be taken to
for this loudspeaker to cause the diaphragm of a prevent damage to the output transistors.

www.americanradiohistory.com
 404 PRACTICAL WIRELESS September, 1962
Where possible, it is best to operate without the example, say the radio is turned up to nearly full
loudspeaker for, apart from the disturbances output, using a resistive load instead of a loud-
created by a watt or so of audio during a recording speaker, so that 3V peak are applied to the
exercise, the varying impedance with frequency of recorder. This will mean that the record level con-
the loudspeaker can detract from the quality of trol will have to be turned well back to avoid
the recording and, in certain cases, upset the overmodulation as indicated on the modulation
equalisation. When a resistor is used instead, it depth indicator. Indeed, under such conditions
should be able to handle the full output power of the record level control will only be a fraction on.
the set or amplifier hook -up. As 3i2 resistors are Now, although the modulation or record level
rather difficult to come by, a length of resistance indicator will be showing that overloading is not
' wire (available almost anywhere) wound round the
apparently taking place, the recording will almost
body of a ceramic high -value resistor, and certainly be very poor indeed. The main reason
terminated at the lead -out wires, is adequately for this is not so much a question of matching, but
suitable for this application. one of overloading in the first stage. The recording
level control is usually connected after the first
Voltage and Matching valve, so the first valve is receiving a full 3V of
Next things to consider are the signal voltage A.F.! It cannot handle that, of course, and dis-
required at the input of the recorder to give full tortion will occur, the distortion being controlled,
modulation without overload, and the impedance in terms of recording level, by the level control
matching between the output of the set and the after the amplifier.
input of the recorder. The Best Settings
The best way of setting up such a combination
Loudspeaker is, first to turn on the recording level control almost
as monitor to maximum, then turn up the radio volume con-
if desired
trol for maximum record level as indicated. In

HTt

Programme
signal

To Radio
input on
-If the
Fig. 26 A.F. ,signal is taken from the loud-
recorder
speaker circuit, the loudspeaker must be left
connected or a load resistor equal to the impedance
Output transformer
of the speaker, must be connected instead.
-u- Output valve
Voltage Considerations
Tape recorders usually have two (or more) Fig. 27 -This circuit shows o "loudspeaker on /off"
inputs, one for microphone and the other for radio. switch, a dummy load R3 and an attenuator RI and
The microphone input is usually more sensitive R2. The attenuator ensures that the first stage
than the radio input. This is because the voltage of the recorder is not inadvertently overloaded and
at the output of a microphone is much below that also gives a theoretically better match between
available from a radio set. Typical levels for full the set and the recorder.
modulation are: microphone lmV and radio
100/150mV. Sometimes there is a second radio that way the first stage of the recorder will always
input suitable for low -level signals direct from the work well below the distortion level, but difficulty
diode detector, but this will be considered later. may be had in obtaining a sufficiently low setting
The question, then, is what sort of voltage is
present at the loudspeaker leads of an average
of the receiver's volume control -a slight touch
being sufficient to push the recording level
radio set? This depends very much on the im- indicator well over the limit. Also, if the receiver's
pedance of the loudspeaker circuit and the output loudspeaker is to be used as programme monitor,
of the set. The voltage is easily found, however, the low volume control setting will not give suf-
for any set, by using the simple expression ficient output, bearing in mind that only 0.001V
E= k/(W x R), where E is the r.m.s. output voltage or so is required to drive fully the recorder and
W the watts output and R the impedance of the that this corresponds to very little audio power
loudspeaker or load resistor in ohms. across 312.
A typical set, for example, might be giving, say
3W across 312. Three times three is nine, and the Attenuation
square -root of nine is three. Thus, it follows that,
at full output, 3V r.m.s. exist across the loud- Thus, an attenuator is a good thing to have
speaker load. This is well above the input signal between the set output and the tape recorder input.
required on most recorders, so the signal is either This will enable the set to be operated at near
turned down at the set (set's volume control) or at normal volume without overloading the first stage
the recorder -but this can cause trouble. For (Continued on page 418)

www.americanradiohistory.com
 41 PRACTICAL WIRELESS September, 1962

Short-wave Listeners' Log

5 OR the best possible listening results, a
general idea of the way in which short -wave signals
are propagated will prove to be very helpful. Two
Unless conditions are very bad, listening on
these bands will almost certainly furnish some Dx
(long distance) stations, and 15m and 20m are
often termed the " Dx bands ". When conditions
are good, signal strength from remote countries
may be good enough for satisfactory reception
types of signals are of interest-those furnished by with simple 2 -valve and similar receivers.
the ground wave, and those from the sky wave.
On the S.W. bands, the ground wave is soon
absorbed and lost, so long distance reception is
not possible with its aid. Frequencies in the 20 SERVICING TAPE RECORDERS
and 15m bands, or around 14 and 21 Mc /s, are (Continued from page 404)
most used for long distance reception, and ground
waves from stations in these bands may already of the recorder, whilst leaving a reasonable margin
have become too weak at 20 or 30 miles range. on the recording level control for low level signals.
For this reason, local stations may not be heard, Fig. 27 shows a very convenient arrangement,
even when remote overseas countries are coming where the attenuator comprises R1 and R2. The
in well. single -pole, two -position switch allows the set's
The sky wave travels upwards at an angle, :.nd loudspeaker to be used as programme monitor
reaches the ionised layers which surround the when required, while in the " loudspeaker 'off "
earth. When conditions are favourable the wave position, the output stage is automatically loaded
is bent or reflected by these layers, and returns to by R3. This resistor, of course, should' have a
earth, often at a considerable distance. The value equal to the impedance of the loudspeaker.
distance from the transmitter to the point where Values for R1 a..J R2 are well worth consider-
the wave again reaches the earth is the " skip ing. From the impedance aspect, there is not too
distance ". There may be short or long skips, much to worry about by connecting 3û across a
according to time of day, frequency, and other megohm or so of the radio input socket. On the
factors. The wave may be reflected from the face of it. there would appear to be a bad case of
earth, and again be deflected downwards by the mismatching, but in practice this has remarkably
ionised layers, reaching earth again at even greater little adverse effect. Overloading is the chief trouble.
distances from the transmitter. If conditions or However. when an attenuator is employed, the
frequencies are unsuitable, the wave may pass impedance can be stepped up to the tape recorder
through the ionised layers, and be lost in space. input so that the match is more theoretically exact.
RI and R2 simply form a potential-divider.
Long Distance Reception That is, all the signal is applied across the two
Long distance reception is by waves that have resistors in series and only a fraction of it is tapped
been reflected from the ionised layers, possibly off from'across R2. Ignoring the loading effect of
travelling a thousand miles or more at each hop, the recorder input circuits, then in proportion the
when skip is long. So best long distance reception voltage across R2 is equal to R2 divided by the
will be around those times and frequencies giving sum of R1 and R2. For example, take R1 and
long skip conditions. 100k and R2 as 50k. then R2 would be 50 divided
The 20m or 14Mc /s band is extremely popular by 150, which is one -third, meaning that one third
for long distance reception, and can give world of the full voltage from the set's. output stage
`wide coverage. The ionised layers are not stable, would be fed to the recorder.
but are influenced by sunspots, magnetic storms, In practice, the applied signal would be a little
and solar radiation, so results change hour by below one -third, since R2 would be shunted by the
' hpur. Usually, most remote stations (Australia, input impedance of the recorder which, in effect,
- New Zealand) will be audible early in the morn- would reduce the value of R2 in relation to R1,
ing. Somewhat later in the morning, there may be but this should not make a lot of difference in the
a* period when only relatively near stations majority of cases. Thus, the attenuator can be
(European) can be heard. Range may increase made any required value simply by working out
after noon, with Near East, Far East, South suitable combinations of resistors, as explained
African, and other distant stations coming in above. The value of 100k and 50k may well
during early afternoon. These are likely to fade be used in practice, and this would mean that the
out later in the afternoon, when Far East stations input of the recorder would see a resistance
-

are liable to disappear, and many American almost equal to 50k and 100k in parallel, which is
stations will begin to come up in strength. With approximately 33k.
evening, very considerable- distances can still be The signal should be conveyed_ ,via screened
covered. cable, the R2 side of which should bé connected to
",

A somewhat similar pattern is repeated daily, the braid and earthed. But beware 'At this point,
except that conditions vary day by day, and month since the set may have a " live " c1Sassis: that is,
-

by month, due to seasonal and other changes. With connected to one side of the mains supply to
the 15m or 21 Mc /s band, rather similar results follow the now popular A.C. /D.C. technique. This
can be expected. but with even more irregular aspect of recording. however, will be discussed in
fluctuation. So this band may be excellent one next month's article. - -

day, and almost dead the next. (To be continued)

www.americanradiohistory.com
 486 PRACTICAL WIRELESS October, 1962

SE i

RECORDING FROM A.M. /F.M. (Continued from page 418 of the September issue)
AND A.C. /D.C. RECEIVERS It is not proposed to delve into the detailed
operation of these stages as such information is
By T. S. Smith available in past issues of this magazine. However,
the A.M. detector load resistor is R15 (330k) in
LTHOUGH the method of obtaining a radio series with the "earthy" side of the secondary
signal for recording purposes from across the (L19) of the A.M. I.F. transformer. The A.F.
secondary of the receiver's output transformer signal from across this load is fed to the
probably introduces a smaller distortion 'content A.M. /F.M. change -over switch Sla.
than by placing the microphone in front of the Now, in the "F.M." position, C44 (680pF) acts
loudspeaker, the overall distortion is still, never- as the load and this, as will be seen, is connected
theless, higher than it need be. This is because the to the other side of the A.M. /F.M. change -over
receiver's output stage is designed to supply power switch, via C45. Thus, on the slider contact of the
at a low impedance, and it is far more difficult to switch (point 5) we have an audio signal which is
secure a distortion -free signal that way, than a either A.M. -or F.M.- derived, depending upon the
low -level signal at high impedance, which is far setting of the receiver. These factors are common
more suitable for the input of a tape recorder, as to all A.M. /F.M. receivers and, as with the circuit
was revealed last month. under discussion, the A.F. is always fed to the
top -end of the volume control (R29), in this case,
Detector Signal via C51.
All receivers, valve and transistor types, produce For normal operation of the set, the required
a source of audio -frequency signal which is as free level of A.F. is taken from the slider of the
as it can possibly be of distortion. This signal is volume control and fed to the grid of the A.F.
developed across the load resistor of the detector triode amplifier. However, for normal driving of
stage (A.M. or F.M.). This signal is of a relatively the average tape recorder, the A.F. signal across
low level (in terms of millivolts), and in the the volume control is of adequate level, and there
receiver itself is fed first to an A.F. amplifier and is absolutely no need to amplify it further in the
then to the output stage. It is obvious, of course, receiver. Thus, by extracting the signal from
that the A.F. and output stages will introduce some across the volume control, the inherent distortion
distortion to the signal, even if only small, so in the receiver's audio stages is eliminated.
connecting to a tape recorder before those stages Furthermore, this component is common to both
goes a long way to enhance the recorded quality. the A.M. and F.M. audio, so extra switching is
A receiver typical of the type used by home not required.
recordists is given in Fig 28. This shows the final
I.F. amplifier, the A.M. and F.M. detector circuits, Avoid Shorting the Negative Feedback
the magic-eye tuning indicator, the first A.F. stage, In some receivers the bottom end of the volume
the output stage and the power supply circuits. One control is returned direct to chassis, as shown in
rather important point to notice is that the mains Fig. 29, but in other sets, as in Fig. 28, the bottom -
transformer has a totally isolated mains primary end of the control is loaded with a fixed resistor
winding. This means that the chassis of the set may to facilitate the application of negative feedback.
be connected direct to an efficient earth point, and In the former case, there is no difficulty what-
that a tape recorder may be connected direct to the ever in obtaining the .A.F., for it simply resolves
receiver without any danger of electric shock or fear to the connection of a screened cable, with the
of damaging the equipment. inner conductor to the top of the control and the
V3 is the final I.F. amplifier valve, while V5 braiding to chassis, as revealed in Fig. 29. In the
serves as both A.M. and F.M. detector and also as latter case, however, similar connection would
the A.F. amplifier. One diode in the valve (the one almost certainly effect the negative feedback
connected to pin 6) is used as A.M. detector, and system and result in somewhat confusing
the other two diodes are used in a ratio detector symptoms.
circuit for F.h_. The triode section of the multiple Where possible it is always best to retain a
valve is the common A.F. amplifier. The amplified direct connection between the braid of the tape
A.F. signals from the anode of the triode feed the recorder signal output cable and the chassis of the
output valve V6, and the bridge rectifier WI receiver. But if that is done in Fig. 28, the input
supplies H.T. voltage for the whole of the set. V4, resistance of the recorder would appear across
of course, is the magic -eye tuning indicator. R29 and R30 in series. In certain cases this may

www.americanradiohistory.com
 October, 1962 PRACTICAL WIRELESS 487,

not matter much, but in other cases instability and whole or part of the detector load. This means
hum may result, depending upon the exact nature that in addition to the A.F. across the resistor,
of the receiver's negative feedback loop and the there occurs a direct -voltage which is the rectified
input impedance of the recorder. To be on the I.F. carrier. This voltage is negative with respect
safe side, therefore, the braid should be connected to chassis (positive in transistor sets) and increases
to chassis, but the inner conductor of the feed -out with increase in signal amplitude. It is thus
cable should be connected to the top of the employed as an AGC bias.
volume control via an isolating capacitor of 0.1µF Care must, therefore, be taken to avoid shorting
(see Fig. 30). the voltage out or applying it to the input of the
The A.F. signal across the volume control is tape recorder. This would not harm anything
upwards to a maximum of about 1V peak, and is particularly, but it might either introduce distor-
adequate for almost all tape recorders. The tion due to bias disturbance in the first stage of
frequency response at this point is also very good, the tape recorder or, if the radio signal is passed
being limited essentially by the passband of the through a volume control in the recorder, noise
I.F. circuits and the design of the detector. With as the volume control is adjusted.
A.M. /F.M. receivers, de-emphasis usually takes In all cases where the detector diode is in direct
place to some degree between the ratio detector connection with the volume control, as in the
or phase discriminator and the volume control, so transistor circuit in Fig. 31, an isolating capacitor
that the A.F. across the volume control should be should be connected between the top -end of the
in reasonable frequency balance. However, when control and the inner conductor signal lead -out
a coupling capacitor is used, as in Fig. 30, the (Fig. 30).
value of that can be reduced if necessary to Even in a transistor set, it is desirable to main-
provide a certain amount of bass cut (i.e., top lift) tain connection between the braid and the chassis
to suit the recording characteristics of the tape or common battery circuit. This will prevent
recorder. random earth effects which could produce hum or
Isolate D.C. instability. In transistor sets, the volume control
In some A.M. -only receivers, including is of a lower value than that in valve receivers,
transistor sets, the volume control acts as the (e.g., 5k in Fig. 31 as compared with 500k in Fig.

o -'
Fig. 28-A typical circuit of an A.M./F.M. receiver from which a "recording" signal can easi y be obtained. The
audio from both A.M. and F.M. transmissions is developed across the volume control R29, and from here it may
be directed to the "radio input" of a tape recorder, as shown in Fig. 30. This set also features a fully isolated maint
transformer, which is a requirement for recording purposes.

www.americanradiohistory.com
 488 PRACTICAL WIRELESS October, 1962

From From At this stage, it should also be remembered

detector' detector that large transients are likely to destroy
circuits circuits
transistors, and that transients of destroying
r=- magnitude can occur due to the connection of
mains equipment to a transistor circuit. This can
mainly be avoided, however, by ensuring that all
Volume the connections are highly efficient and that they
control To Al are made before switching on the tape recorder
amplifier and transistor set. A poor connection causes
To AF in
amplifier receiver crackles in the loudspeaker or on the tape, and
Volume receiver these crackles are really transient surges of
control current. Further, on no account should a transistor
Negative set be coupled direct to A.C. /D.C. equipment,
feedback but this does not often apply, since almost all
tape recorders are for A.C. mains only and feature
a fully isolated mains transformer.

Mains Isolation
rve
Ea Tape recordists often require to record from an
To tape tape A.C. /D.C. type of receiver. These sets have no
er recp erreco

mains transformer, and due to that there must exist

input Input
Fig. 29 (left) -When one side of the volume control is a connection direct from the mains to the H.T.
connected direct to chassis, and provided there is no negative line or chassis. This presents no trouble
direct- voltage due to carrier rectification, across the under not mal operating conditions, but if internal
control, then the audio may be fed direct to the tape connections are made to this kind of receiver, one
recorder as shown. side of the mains supply is, in effect, brought out
to the connecting leads and applied direct to the
Fig. 30 (right) -If there is a direct -voltage across
the volume control or if a fixed resistor is connected
tape recorder.
Since, for the sake of hum and stability, it is
at the "earthy" end to facilitate negative feedback, necessary to connect the braid to the chassis of
then the tape recorder signal must be directed the receiver (Figs. 29 and 30) and to connect the
through an isolating capacitor C. braid at the other end to the "earth" side of the
radio input on the tape recorder, the mains supply
28). This does not matter unduly provided the is connected straight to all the metal work of the
coupling capacitor C (Fig. 30) is no less than recorder. if that happens to be connected to the
0.44F. It may even pay from the bass response
point of view to increase the value to 0.25µF. (Cont(nued on page 525)

-9 volts

Ti

Cr

0
8 e C24
OF4
Y
J

AGC line C20 R15

8YF 10k
1R1.3

RI2 R17 C23 R20 R21

5k
Volume
C22 1k loon 10C1
50
control pF
TOOOpF

Fig. 31-Although a transistor circuit uses a lower value volume control (Rl2) than valve circuits, the recorder
signal can still be picked up from that point, bût a higher value capacitor may be required for isolation. The
capacitor is necessary to isolate the D.0 from the recorder input circuits (see Fig. 30).

www.americanradiohistory.com
 PRACTICAL WIRELESS 525
October, 1962

expressed by his correspondents

The Editor does not necessarily agree with the opinions

Whilst we are always pleased to assistare readers with their wave and particle!
technical difficulties, we regret that we unable to supply
commercial Also, if a photon were, as Mr. Craske put it.
diagrams or provide instructions for modifying to "explode ", its energy would be dissipated
or surplus equipment.
UNDERTAKE TO
We
ANSWER
cannot supply
for receivers described in these pages. OVER
QUERIES
PHONE. If a postal reply is required the
alternative
WE CANNOT
THE
details
TELE-
a stamped and
among several secondary photons and the fre-
quency would be lowered in a random manner!
K. ROBBINS (Coventry).
-
addressed envelope must be enclosed with coupon from
page iii of the cover.
CORRESPONDENT WANTED
STRIPED SOUND
SIR, -I am years of age and I have been
15
interested in S.W. radio for some time. would
-I haveforbeen
SIR.magazine
an avid reader of your
the last 15 months and. although
1

like to correspond with amateurs of my own age

the
many of the articles are still above my head,have or anyone interested in 38 and 18 sets. -S. Antos
home construction bug has bitten hard. I 53 Meldon Terrace, Newbiggin -by-Sea, Northum-
imported a number of the kits advertised in -a berland.
Practica! Wireless and my most recent attempt a
stereo amplifier -is now working well, albeit SERVICING
little distortion in the left-hand channel. work
I now wish to attempt some experimental
with 8mm. striped sound. using the above amplifier
TAPE RECORDERS
and the necessary pre- amplification. (Continued from page 488)
Therefore. I am appealing for information from "live" pole of the mains, the whole equipment
any readers who may already have attempted this becomes highly dangerous. There is not only the
sort of thing. I wish to keep the record /playback possibility of fatal electric shock, but a random
unit as compact as possible, using the existing earth connection made on the recorder could
film drive sprockets of my projector. If any
cause severe damage to the equipment.
readers could let me have any such information toI Some operators feel that adequate isolation is
would be very grateful and would endeavour possible by interposing two capacitors, one in each
reply to all correspondence. conductor, at the set -end of the signal lead -out
I am currently busy constructing
a stereo pre-
cable. Such isolation is pretty futile for several
amplifier and am hoping that this will also be of reasons. One is that an isolating capacitor must
use with these experiments. -A. P. RAATS (6 not exceed 0.00511F, since a higher value would
Elizabeth Road, Bedfordview, Tyl., South Africa). pass too much alternating current due to its lower
COPIES OF P.W. WANTED reactance at 50c /s and would thus be almost as
unsafe as a direct connection. Another is that
-I require
SIR.Practical
certain copies of
urgently
from 1948 onwards.
Wireless dating
such a relatively low value capacitor would .

attenuate the lower- frequency audio signals too

I will pay the full price for copies in good much for good quality recording. A third reason
condition. -B. E. WASHER (15 Bower Road, is that the impedance across the isolating capaci-
Harrogate, Yorkshire). tors gives rise to a 50c /s voltage, which is
ELECTROMAGNETIC RADIATION extremely difficult to remove from the low -level
letter in signal circuits.
-With reference to Mr. Craske's
SIR,the There is only one real answer to this problem
August issue, I fear that he has miscon- and that is to connect a 1 -to -1 ratio isolating
strued the Quantilm theory of electromagnetic transformer between the mains supply and the
radiation (EMR) entirely. A.C. /D.C. set. In that way the chassis is at zero
Mr. Craske proposes to emit EMR as a particle, potential and can be retained that way by connect-
then convert it into a wave at a later stage. treats ing it to a good earth point. The external tape
While it is true that the Quantum theory recorder circuit may then be connected as shown
EMR as particles, reference to the theory will the in Figs. 29 and 30 without damage. On no
show equations involving the frequency of fact, account, however, must an earth be connected to
EMR and only waves have frequency. In only an A.C. /D.C. receiver without an isolating trans-
the " wave " and " particle " conceptions are former. It should also be noted that some sets,
analogies used to explain certain observable although for A.C. mains only, may have a mains
phenomena of EMR. Actually EMR is neither transformer to supply only the valve heaters, with
wave nor particle and should not be considered the H.T. mains circuit still in connectionin with
as such. chassis. Such models should be handled the
It follows, therefore, that a photon cannot be
versa. same way as A.C. /D.C. receivers.
emitted and converted into a wave or vice- of both (To be continued)
It simultaneously exhibits the properties

www.americanradiohistory.com
 633
PRACTICAL WIRELESS
November, 1962

SERVICING
TAPE
TYPICAL FAULT
SYMPTOMS AND
THEIR CURES ECORDERS
By T. S. Smith (Continued from page .525 of the October issue)

N past articles of this series, we have even with the volume control turned right down,
investigated the principles of tape recording, the trouble must then lie after VI stage, since VI
is always operating at full gain and the volume is
examined circuit sections in some detail and controlled prior to V2A.
discovered the best ways of recording a radio VI is biased by the voltage dropped across the
programme. In this final article of the present cathode resistor R4 -this making the cathode
series we will examine a typical recorder circuit,
find out exactly what the components do and more positive than the control grid, which is the
finally discuss the various fault symptoms which same as the control grid being more negative than
are characteristic of tape recorders. the cathode. Heavy cathode decoupling is pro-
The circuit to be discussed is given in Fig. 32. vided by C5 to avoid degenerative feedback.
Here VI is a high -gain low -noise pentode which Should C5 go open-circuit, then the gain on both
serves as a preamplifier for both the microphone record and replay would be well below standard,
on " record " and the tape head on "replay". and there would be a definite lack of bass
Failure of this valve or associated component response. A short-circuit would cause bad dis-
would thus result in lack of both record and tortion and excessive microphony.
replay. This is typical of most recorders, and well The stage would fail completely, of course, in
worth bearing in mind. the event of either the screen feed resistor or the
With switch SIA in the " record " position, the anode load (Rl and R3) going open -circuit. The
microphone signal is coupled direct to the control same would happen should the screen decoupler
grid of VI, across the grid leak R2 and the equalis- C2 or the main decoupler CI short-circuit, but in
ing capacitor C3. the latter case R7 would overheat.
The amplified signal at the anode of VI is
High Gain Amplifier coupled to the volume control VR1 through the
coupling capacitor C4. The required level of
V1 is arranged as an ordinary A.F. amplifier to signal is taken from the volume control and fed
provide a high gain, this being necessary to boost direct to the control grid of V2A. This valve is
the very weak signals from the microphone and also arranged as a voltage amplifier and, in con-
replay head. Since it is running at high gain, the junction with the following stage V2B, gives
valve may be prone to microphony (e.g., it may suitable equalisation via the negative feedback
produce a ringing noise in the loudspeaker when loop coupled through C7.
mechanically disturbed), and for that reason its It will be seen that the feedback signal is in-
holder is often mounted on rubber buffers on the jected at the cathode of V2A, and to facilitate that,
chassis. the cathode is not decoupled. R6 is the anode
After some time in service, the valve may load, and from here the signal is fed to V2B grid
become "noisy "; that is, frying and crackles may through the coupler C6. Because the overall gain
be superimposed on replay or on the recording, provided by V2B need not be very great, degenera-
but this can usually be checked by switching to tive feedback (e.g., current negative feedback) is
replay, turning the volume control to maximum introduced by leaving the cathode resistor R10 un-
and gently tapping the valve with the handle of a by- passed.
screwdriver. If this action results in loud noises The H.T. supply for both VI and V2A is
from the loudspeaker, then the valve should be decoupled by R7 and Cl, these components also
checked by substitution. Note, however, that even giving extra H.T. smoothing, which is essential in
a ' new valve will give some noise when high -gain stages to eliminate all traces of mains
mechanically disturbed owing to the very high hum.
circuit gain.
If the valve is reasonable, yet a disconcerting Radio and Pick -up Input
background noise remains, but only with the
volume control turned up, the cause must lie in On the majority of domestic recorders, there is
VI stage. Typical troubles are bad connections provision for the injection of a radio or pick -up
between the pins of the valve and the sockets on signal for recording purposes (see Parts 7 and 8).
the holder, a "noisy " grid leak (R2), a "noisy " This does not require to be a low-level input since
anode load resistor (R3) and poor soldered con- the signal from a radio or pick -up is usually far
nections on the valveholder. 1f the trouble persists stronger than that from a microphone or replay

www.americanradiohistory.com
 634 PRACTICAL WIRELESS November, 1962
head, so the signal is fed in after the preamplifier as a bias oscillator. Feedback takes place between
VI. the screen grid and control grid circuits via C15.
In the circuit of Fig. 32, it is applied across the Erase on this model is by permanent magnet but
volume control. This, then, gives another way of on models where an erase head is used instead the
checking stage VI, for if the microphone input is oscillator also feeds that head in the "record"
dead and yet normal operation is possible by position.
injecting the signal at the radio and pick -up input
socket, we can be absolutely sure that the trouble
is in stage VI. The signal at this socket can also Record Level
be monitored, if required, on a pair of high V3 is a "magic eye " valve which is arranged as
resistance headphones. a record level indicator. This comes into opera-
The amplified signal at the anode of V2B is fed tion only on record, the anode being switched to
to the negative feedback loop, via R11, R12 and the H.T. line by SIC.
R13, with frequency correction being given by C9. The grid of the valve is fed by a D.C. voltage
The signal is also fed to the wiper of switch SIB the level of which is proportional to the peaks of
through R14. Moreover, the signal is also directed the signal at the anode of V2B. The signal is
to the "extension amplifier" socket. The signal at rectified by M3 and smoothed by C12, which also
that point is at a fairly high level, so it can be endows the circuit with a time- constant. RI5 and
used to drive a separate hi -fi amplifier, for C11 form an A.F. filter circuit and contribute
example. towards giving a clear -cut shadow on the eye.

Fig. 32 -The circuit diagram of a typical domestic tape recorder.

With switch SIB in the "record" position the The recorder is powered from the mains supply
signal is taken back to the head and a "constant via a fully -isolated mains transformer T1. The
current " record characteristic is secured by the two primary windings give 200 -250V operation
signal being passed through the high resistance of when points B and C are connected and 110 -115V
' R14. With the switch in the "replay" position,
however, the signal is directed to the control grid operation when points B and C are disconnected,
of V4 through R20, with R19 being the grid -leak. A connected to C and B connected to D.
The H.T. primary winding feeds a full -wave
On "replay" V4 is the normal output valve, it rectifier V5, while a separate 6.3V winding energises
being loaded in its anode by the speaker and all the parallel- connected valves. The reservoir
speaker transformer T2. Under that condition capacitor is C18, with C17 as the main smoother.
SID disconnects the bottom of the cathode resistor The tapped primary of the speaker transformer
R18 from chassis, the return circuit then being gives hum cancellation, since the hum voltage in
provided through the secondary of the speaker the top section is opposite in phase to that in the
transformer. This is done to facilitate switching bottom section, and lower current smoothing is
the speaker out of circuit on " record ". provided by R21.
Also the screen grid of V4 is connected direct to
the H.T. positive line, since SIC shorts out the Replay Normal -No Record
oscillator coil L1. Thus V4 works as a conven-
tional output stage. On the type of instrument detailed in the fore-
On "record" several things happen. going this symptom would almost certainly mean
SID that a faulty connection exists between the switch
changes over and puts the bottom of R18 to
chassis, thereby shorting out the speaker. SIC section SIB and the record /replay head. The best
changes over and puts LI into the screen grid thing to do would be to switch to "record", inject
circuit of V4, in that way causing the valve to act
(Continued on page 637)

www.americanradiohistory.com
 PRACTICAL WIRELESS 637
November, 1962

(Continued from page 634) any cathode by-pass or decoupling electrolytics,

a signal and endeavour to monitor it either on a such as CS in Fig. 32, by shunting each one in
high resistance A.C. voltmeter or pair of head- turn with a test capacitor known to be in good
phones first at SIB and then at the head. order (remember that electrolytics are polarised
Sipce "replay" is normal R14 must be working and must, therefore, be connected the right way
correctly and there would be a good connection at round).
the wiper of the switch, since the signal is con- If the symptom is accompanied by distortion
veyed correctly to the control grid circuit of V4. and high background noise, check the heads and
Replay Normal -Distorted Recording gaps for tape deposit. Any deposit left on the
This symptom is typical of bias oscillator failure. record /replay head will greatly impair the repro-
One check for oscillation is by metering the screen
duction, while an unclean erase head will upset a
current at the H.T. tap of LI. The tap should be subsequent recording made on an already recorded
disconnected from the H.T. line and a milliammeter tape. The heads can be cleaned with a little
(25mA full scale) connected between the tap and switch cleaner or carbon tetrachloride applied on a
H.T. positive (positive of meter to H.T. positive).
matchstick or a small piece of cloth.
Also make sure that there is reasonable pressure
If the stage is oscillating a distinct change in screen
current will occur when LI is short -circuited.
If the stage is not oscillating C15 would pro-
applied to the heads via the pressure pads and
check to see that the tape is tracking correctly
e.g., that it follows dead in line from the record/
-
ably be open -circuit or there may be shorting turns replay head to the erase head. Otherwise a small
in LI. If the stage is proved to be oscillating, lack part of the recording will be left on the tape where
of bias at the record head could be caused by C13 the erase head does not cover.
being open -circuit. Note that if the heads are checked by making a
Record Normal -No Replay resistance measurement of the windings a small
Again, as all the valves themselves must be residual magnetism may occur. This can cause
working reasonably well to give normal record noise on replay, the solution being to demagnetise
facilities the trouble would most likely be in the with a defluxer. For that reason, therefore, head
head switching or in the speaker circuit. A pair continuity readings should not be taken until the
of headphones could be used to trace a tape signal last resort.
through the preamplifier and triode voltage ampli-
fiers right up to the control grid circuit of the Hum Troubles
output valve. If hum is present with the volume control turned
Indeed, a pair of headphones can be extremely right down it is usually caused by low value or
useful to trace the signal through any audio open -circuit of one of the electrolytic smoothing
section whether it be on record or replay. The capacitors.
bias signal, of course, will not be heard since this If hum occurs only when the volume control is
is at a frequency just outside the audible range. advanced it is probably getting in at the preampli-
Low Record /Replay Gain fier. Check the valve by substitution and ensure
If this is accompanied by the record level that the screening (if fitted) on the head is properly
indicator not closing fully on signal peaks, even bonded to chassis and also that the braid of the
with the record level control (gain) turned to wire connected to the head winding is earthed to
maximum, attention should first be directed to the chassis. Check the preamplifier decoupling
preamplifier, and voltage amplifier valves them- capacitor and screen by -pass capacitor (such as Cl
selves. Low emission of one or more of the valves and C2 in Fig. 32).
is a frequent cause of the trouble. Displaced heater leads to the preamplifier valve
If the valves are in order it would pay to check is another cause of hum injection.

work, and terminated at the mains switch S3.

Mayfair pre -amplifier Because of the close proximity of PLI to other
(Continued from page 612) components and wiring, the protruding pins must
be covered with insulated sleeving.
now be wired in position and the remainder of the With the mains wiring in position the earth bus-
wiring to S1 completed. bar can be completed. This involves running a
Screened Leads length of 18s.w.g. tinned copper wire between the
It now remains to make the connections between two jack sockets JK1 and JK2 with further con-
top and underside of the chassis. Lightweight nections to TB3, TB4 and TBS. As this wire runs
plastic covered screened cable should be used for very close to the mains switch and the lamp holder,
this purpose. Each screened lead is clearly shown suitable lengths of insulated sleeving should be
in the wiring diagrams Figs. 2 and 5, and is marked threaded on to-the straight portion of busbar so
with a code letter to distinguish it from the other that, except for the points where a branch con-
leads that pass through the same opening in the nection is made, the whole of the wire running
chassis, thus simplifying cross -references from one from JK1 to JK2 is covered.
diagram to the other. Connect wire "D" to TB3 and feed this wire
Final Wiring through to the top of the chassis and secure it by
The mains input circuit should be wired next. means of a solder tag to a fixing screw of one of
This requires four conductors -two lengths of the input sockets.
twin -core mains lead will be satisfactovy -and This completes all the internal wiring. The four -
these should be routed from the input plug PLI core power cable and the coaxial output cable
across the end of the chassis then along the front should be brought in through grommet holes and
edge of the chassis close in the angle of the metal- connected up according to Figs. 1 and 5. Ill

www.americanradiohistory.com