This is only a preview of the May 2021 issue of Practical Electronics. You can view 0 of the 72 pages in the full issue. Articles in this series:
|
Practically Speaking
Hands-on techniques for turning ideas into projects – by Jake Rothman
Restoring old equipment – Part 3
Fig.19. Complex wiring on the input
sockets. I bypassed the switches
because they couldn’t be replaced.
Fig.16. Easy servicing was an important criterion for old equipment. Full access
by undoing one thumb screw and sliding off the sleeve – unbelievably convenient
compared to modern designs.
L
ast time, we introduced the
famous Leak Stereo 30 amplifier,
an important milestone in British
Hi-Fi history. This month, we’ll move
on to switches, components and the first
application of power.
You only need to undo one thumbscrew
(shown in Fig.16) to get inside the Stereo
30, allowing the metal cover to be slid off.
Both sides of the PCB are then accessible.
This must be a world record in simplicity.
Some Japanese amplifiers require about
40 screws to be removed.
The front panel can be detatched by first
taking off the knobs by loosening their
grub screws and then undoing the pot
nuts with a 0.525-inch across-flats potspinner, as shown in Fig.17. I had to do
this to gain access to the front panel slider
switches and then bend back the corner
of the drop-damaged front panel (Fig.18).
Dodgy switches
A constant problem on these amplifiers
are the black Cinch slide switches which
use plain copper contacts. These fail
open-circuit as they oxidise, effectively
forming copper-oxide rectifiers. The rubber
behind the sliding contacts can sometimes
shrink and go hard, reducing the wiping
pressure. The signal goes through two of
them: gain switches on the back and the
tape monitor switch on the front.
Access on the Leak is not all perfect
however, there is some awful ‘spag’ (pointto-point) hard wiring around the rear input
sockets and switches (see Fig.19). They had
been wired after they had been screwed in;
Fig.17. The diecast chromed front panel from my eBay Stereo 30.
62
maybe the designers thought they would last
forever. As usual, the switches had failed. I
had to remove the big smoothing capacitors
to reach them. Even then, I still couldn’t
get them out, so I ended up bypassing them
out of circuit. The phono input has to be
set to high sensitivity anyway for (phono/
turntable) magnetic cartridges. Nobody
uses high-output ceramic or crystal ones
anymore – are they even made?
Getting to the front switches was also
a battle. I found after removing the front
panel bezel and knobs that they were
riveted in. I can’t stand rivets, they may
be good for aircraft wings, but they make
it difficult to replace components because
they have to be drilled out. This tedious
procedure is shown in Fig.20. I then
replaced the rivets with M2.5 countersunk
screws, countersinking the holes first.
The switches can be disassembled by
unbending their tabs with side cutters.
There is no need to disconnect them.
The dirty contacts can then be filed clean
with a fine file or sandpaper (Fig.21).
Fig.18. Bending the front panel – note tape to prevent scratches.
Practical Electronics | May | 2021
Figs.20. a) Rivets have to be drilled out to get to switches. The drill has to be a bit bigger than the rivet head; b) The rivets are replaced
with M2.5 countersink bolts; and c) Fitting the nuts is tricky; use a small rare earth magnet on a screwdriver to position them.
Seal them quickly with contact lubricant
such as Contralube (Fig.22). This trick is
often needed for wave-change switches
on old radios. If you have no luck with
such a repair, C&K do a suitable switch,
the S202031SS03Q from Mouser. Rotary
switches are generally very reliable,
but give them a good squirt with Super
Servisol 10 switch contact cleaning
lubricant while the front panel is off
and all the contacts can be accessed.
The rivet/bolt issue above neatly
demonstrates the restorer’s eternal
dilemma – do I ‘destructively improve’
or attempt to preserve an inferior
original aspect of a design? It applies
to the restoration (or preservation) of
everything – from medieval timber tithe
barns to early computers and vintage cars.
Only you can decide for yourself what
to leave as is, what to restore to original
spec and where to improve with modern
technology. There is no right answer, and
it very much depends on what you are
restoring, the availability of spare parts,
up-to-date safety thinking and why you
are restoring it in the first place.
Tape-monitor switch
Fig.21. a) (top) The troublesome Cinch
switches. Take the top metal part off by
bending up tabs on back; b) (bottom) The
blackened contacts of the switches are
gently filed to reveal clean metal.
Fig.22. Contralube is applied to the contact
to stop them oxidising again. This stuff
glows under UV light to aid even application.
Practical Electronics | May | 2021
There was no serial number on the back
of the unit. Looking at the eBay unit, it
had a proper serial number plate on the
back number – Z/18777 – and it was also
written in the same place with another
signature, which looked like ‘J Quinn’.
Corrosion
The phono connectors on old amps such
as this one are often badly corroded, as
shown in Fig.24. These can be replaced
with gold-plated ones, but it is a difficult
job, necessitating the machining of new
mounting plates. However, although they
often look bad, after a good scrubbing with
wire wool, they seem to work fine (Fig.25).
Sometimes though, radical surgical
excision is necessary! I had to do this with
It is possible to bypass the preamplifier
and feed a signal directly to the power
amplifier input by switching the tapemonitor switch on the front to ‘on’
(left-hand side). The signal is fed into
the tape replay input sockets, where it
then goes straight to the volume control.
Unfortunately, this is through another
Hi/Lo sensitivity switch which has to be
shorted out. Bypassing the preamplifier
noticeably improves the distortion and
high-frequency response. This is the best
way to use this amplifier with a digital
source, such as a CD player or computer.
Hidden history
I was pulling around the cable loom
dealing with the switches when I noticed
some pencil writing behind it on the
chassis. It said ‘158-4-65’ followed by
what looked like Harold Leak’s signature
(see Fig.23). I guess the ‘158’ was the
serial number, quite an early one then.
Fig.23. Signed by Harold Leak?
Fig.24. These input sockets are badly
corroded, and need to be replaced or
thoroughly cleaned.
Fig.25. Sometimes a good scrub with
a brass brush and wire wool can get
contacts usable again.
63
Fig.26. Radical excision of the input
assembly was deemed necessary on the
eBay amp.
Fig.27. Input selector switch looked so
corroded that I chopped it out. In fact,
I didn’t need to. It was sealed and the
contacts were fine when I opened it up.
Next time, I’ll test before condemning it
because it was a lot of work to replace it.
the input socket assembly in Fig.26 on
the eBay amp. The input selector switch
also looked beyond hope, as shown in
Fig.27. This is then a good excuse to fit
some decent sockets and rationalise the
whole input system for greater reliability.
I made a version with CD input and aux
inputs that went into the tone control
input, avoiding the excess gain and noise
of the original input stage, which was
then permanently wired for phono input.
still fully functional when tested on the
Peak Analyser. This is the sort of thing
to expect in old electronics and I always
take corroded components out of circuit
to test. Normally, it’s best just to replace
corroded components, because the rusting
may carry on. In this case they had to go
back in to maintain originality. In fact,
later during load testing the right hand
AC127Z blew, luckily it took nothing
with it.
Germanium transistor issues
Speaker connection
Most germanium transistors have thin
iron wires to stop soldering heat getting
into the device. These are especially
prone to corroding right through with
spilt acidic drinks, such as red wine.
The two GET538s germanium transistors
in Fig.28 have lost a couple of wires.
The AF118s and AC127Zs were badly
corroded as well (Fig.29) but they were
The Leak stereo 30 has unique speaker
plugs, shown in Fig.30. Fortunately,
Andy’s came with them. If missing, a
standard 4mm banana plug will fit in the
big hole and an M3.5 bolt can be screwed
into the smaller hole as shown in Fig.31.
Fig.29. The AF127Z (T7L/R) and AF118
(T5L/R) transistors are difficult to
substitute and replace. Note P7L/R which
are the quiescent current presets. Set
to minimum resistance (anti-clockwise)
before powering up.
64
Testing
There are a number of components that
can be checked with a multimeter before
applying power; for example, rectifier
diodes and resistors. Most of the latter are
carbon composition types which often go
high in value as the carbon grains shrink
with time. These resistors are hated by
most audio engineers – unless they’re
Fig.28. This GET538 driver (T6L/R)
transistor’s legs had corroded through.
This failure also killed its associated
output transistor.
making fuzz boxes for guitarists. I would
take all of them out in critical locations,
such as collector loads, and feedback
and bias networks to reduce noise and
distortion. Resistors can to some extent
be tested in-situ, changing any that read
higher than their marked value. Any
that read significantly lower must have
a resistance in parallel and will have to
be tested with one end lifted. I found
R53L/R and R54L/R had gone high in
the power amplifier. Later models had
the Erie ‘pluggable’ types illustrated in
Fig.32 which ease assembly, but age more
quickly than the traditional wire-ended
composition resistors. To aid desoldering,
simply crack open the body with side
cutters. The thermistors (Fig.33) also use
a granular material, and these go high as
well. I had one TH2 measure 764Ω when
it was meant to be 500Ω. However, in
some case readjusting presets should be
able to compensate for the drift.
Capacitors
The red Telegraph Condenser Company
(TCC) Elkomold electrolytics shown in
Fig.34 are often dried-up, going opencircuit or high ESR, and need to be
Fig.30. (left) The Stereo 30 had strange speaker plugs. Luckily, this unit had been
stored with its plugs; Fig.31. (right) On the eBay amp I had to improvise by screwing
an M3.5 bolt into the smaller hole.
Practical Electronics | May | 2021
TH2R
R2L
Fig.35. The Erie tubular ceramic
capacitors are reliable. They have unusual
colour coding. This one is the feedback
capacitor C36L and is 3000pF. The white
band represents ±10% tolerance.
T1L
Fig.32. Unfortunately, some old
equipment is a forest of Erie pluggable
carbon composition resistors. All had
increased in value by 30%. Even the 5%
grade ones on the RIAA network.
When I have to remove a lot of them, I
crack them in half and de-solder one leadout at a time. (See R2L and also note T1L,
the AC107 transistor in a black SO-2 glass
package). Note also the presets (P1L) for
adjusting the pre-amp DC point.
replaced. Given their plastic cases and
strange rivetted connections it’s amazing
any worked at all. All of these capacitors’
positions were marked on the trackside of
the PCB to enable testing with the Peak
ESR meter. Ten were found to be faulty. If
you don’t have an ESR meter, just take all
of them out. In general, I don’t normally
believe in ‘re-capping’ (ie, removing and
replacing all electrolytic capacitors), it’s
wasteful, but here it’s justified. If some
appear high in value, remember that the
initial tolerance was –20/+100%. The
non-polarised capacitors were mainly
SRC (Stability Radio Company) polyesterfoil capacitors. These are similar to the
Mullard ‘mustard cap’ C296 series and
are very reliable. The remainder of the
capacitors were Erie tubular ceramics,
such as the 3nF feedback capacitor, which
are also reliable. These have unusual
colour coding, as shown in Fig.35. If
Fig.33. (top) a) The part on the left is a
thermistor TH2R to stabilise the quiescent
current in the output stage. They go high,
but are usually reliable; (bottom) b) TH1L/R,
the DC point stabilising thermistors on the
power amplifier input stage.
you have old equipment with the blue
Mullard/Philips axial electrolytics, these
are usually fine, and Vishay and Kemet
still make equivalent long-life axial types.
Powering up
All Hi-Fi amplifiers need the same
standard sequence of checks, strictly in
the following order:
n Visual inspection, including earth
safety check
n Power supply isolation and voltage
check
n Power amplifier check
n Preamplifier check (there is a link
to isolate the pre-amp power supply
under the board if necessary).
AC check
First check the continuity from the earth
pin on the mains lead to the chassis.
This is to ensure the mains AC fuse
blows if there is a short from the live to
the metalwork. When first turning on
an old piece of electronics it is essential
to limit the input current. I did this by
removing the mains fuse and soldering a
2.2kΩ 5W wire-wound resistor across it,
as shown in Fig.36. The DC fuse (in the
DC power rail) was also removed. This
was to enable the power supply to be
checked and reform the main smoothing
capacitors slowly. Any problems with
the mains transformer or the metal ‘tophat’ Semics 20A5 rectifiers will also be
exposed at this stage and no damage
will occur, the resistor taking the strain.
If it starts smoking you know there is a
problem. Once all is well, the fuse can
be replaced and the resistor taken out.
I prefer to do this start-up test procedure
with resistors. The service books say use a
variac (a variable mains auto-transformer).
This works fine with valve amplifiers, but
transistor amplifiers can suddenly ‘snapon’ as the voltage is increased, taking a
massive current and thereby destroying
themselves. Also, a variac is quite an
unusual piece of equipment, which most
people don’t have.
DC check
We do the same trick again, but this time
putting a 100Ω 5W resistor across the DC
fuse, shown also in Fig.36. If it gets hot
turn it off and check for faults.
Next month
In the next installment we’ll complete
the restoration and look at a few optional
amplifier upgrades.
WARNING!
Fig.34.These old TCC Elkomould
electrolytic capacitors have all dried up.
They were the latest thing in PE, March
1965 p.359.
Practical Electronics | May | 2021
Fig.36. 2.2kΩ 7W vitreous wire-wound
resistor across the mains fuse holder to
limit current on first turn on. Take care,
exposed mains! Also note the siliconecoated 100Ω current-limiting resistor (‘RS’)
in place of the DC fuse for when power is
first applied to the power amplifier.
When restoring electronic equipment,
do ensure you work with:
• Eye protection
• Fume extraction
• Safe wiring/earthing
These are not nice-to-have optional
extras – you must follow all safety
guidelines to protect yourself and
those around you.
65
|