This is only a preview of the March 2009 issue of Silicon Chip. You can view 32 of the 96 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "A GPS-Synchronised Clock":
Items relevant to "New, Improved Theremin Mk.2":
Items relevant to "Build A Digital Audio Millivoltmeter":
Purchase a printed copy of this issue for $10.00. |
Vintage Radio
By RODNEY CHAMPNESS, VK3UG
The deadly & the difficult – when
to say no to a restoration
The “Pilot Little Maestro” is a 5-valve medium-wave (MW) and long-wave
(LW) receiver made in 1939. It’s potentially deadly, as revealed in the text.
Deadly equipment and difficult faults can
present real challenges when restoring
vintage radio gear. Sometimes, you just
have to say “no” to a set that’s just too
dangerous to use unless it’s correctly
modified.
G
ENERALLY, we expect a vintage
radio to be intrinsically safe due
to its inherent design and as a result
of either careful restoration or proper
maintenance throughout its life.
Of course, if we are restoring an old
set to working order, then nothing can
be taken for granted. In fact, it is only
to be expected that some faults may
have developed in the set, particularly
82 Silicon Chip
if it has been stored for many years in
a garage or shed in less than perfect
conditions.
In particular, equipment that’s been
attacked by rodents and various creepy
crawlies will need careful attention to
ensure a successful (and safe) restoration. This same goes if the set has been
subjected to damp, dusty and hot/cold
conditions.
I have seen pictures of receivers
that have been restored from absolute
wrecks to pristine condition by dedicated enthusiasts. However, when the
ravages of time and storage inadequacies have taken their toll, it is time
to sit down and determine whether
restoration really is worth the effort.
Of course, an extremely rare piece
of equipment will be well worth it,
provided it isn’t like grandad’s axe.
You’ve probably heard the saying – it’s
had five new handles and three new
heads but it’s still grandad’s original
axe!
Well hardly and the same applies
to rare vintage radios if the parts
aren’t original or, at least, the correct
replacements.
If only a few parts of a set are recoverable, will it really be a genuine
restoration or just a replica that happens to use some original parts? Not
that there’s really anything wrong with
replicas. They are sometimes the only
way of showing us how technology
achieved things in times gone by.
Making the assessment
Assuming that the power cord itself
is OK, the first thing to do when assessing whether an AC-powered radio is
worth restoring is to check the power
transformer. There are a couple of
tests that will quickly reveal whether
a power transformer has withstood
the ravages of time. But first, if the set
has been stored in a damp location,
it’s worthwhile heating the chassis
with the transformer still attached in
a kitchen oven at around 50-60°C for
a few hours.
This will help to dry out any moisture in the transformer and make the
siliconchip.com.au
An above-chassis view of the Pilot Little Maestro. As originally designed, it was a AC/DC set with one side of the mains
connected to chassis but this particular unit has been modified to run from an external power supply for safe operation.
following tests more realistic.
The next step to test the insulation
between transformer’s primary winding and the frame and also between the
primary and secondary windings. The
secondary high-tension (HT) winding
also needs to be checked to ensure
that it has no shorts to frame. This
will usually involve lifting the centre
tap of this winding from the chassis
(or lifting the low-value resistor in
series with it).
I do this insulation test at 1000V
using a high-voltage, high-impedance
tester. Mine was obtained as a kit
from Altronics several years ago (Cat
K-2555). If this test shows the insulation resistance to be greater than
around 200MΩ between the various
sections, the next test can be carried
out.
This involves removing the rectifier
valve, then connecting power to the
transformer and running it for about 30
minutes. If, at the end of this period,
the transformer is only slightly warm,
it is fairly safe to say that it is in good
order, provided there is continuity in
each of the windings.
By the way, don’t leave the equipsiliconchip.com.au
ment unattended during this test. If
there is a fault in the transformer (eg,
shorted turns) it will quickly overheat and will need to be turned off
promptly.
From my experience, faulty power
transformers are quite rare even after
several decades in storage, often in less
than ideal conditions. I have, however,
come across a number of transformers
which have perished leads emerging
from the windings. Left as they are,
disaster is just around the corner in the
form of short circuits and a burnt-out
transformer.
Depending on how the transformer
is made, I have in some cases taken
off the cover plates and installed
new lead-out wires. Alternatively, if
that is not possible, I have carefully
removed the old insulation from the
various leads and replaced it with
fresh insulation.
Usually, the insulation has become
hard and it can be cracked off using
a pair of pliers. I then slip on new
spaghetti insulation over the bare
wires to make the transformer safe
to use. I also often tie some of the
leads together with plastic spaghetti
and also use some clear nail polish
to hold the sleeves in position where
they emerge from the windings on the
transformer.
If the power transformer is faulty, it
may mean that restoring the receiver is
not an economic proposition. Alternatively, it may be impossible to restore if
a suitable replacement isn’t available.
However, substituting a transformer
from another set is often practical,
providing it has a similar rating and
fits the available chassis space.
Dial scale
The value of a receiver drops dramatically if the dial scale is broken or
missing. To get around this problem,
some restorers have become quite
skilled at using computers to make
new dial scales.
In some cases, they lay the broken
pieces out in position and then scan
the dial into a computer. Then, using
a drawing program, they use this as a
template to make another scale which
can be printed out onto plastic film
and fitted to a glass backing.
Some restorers even provide a service to others by supplying dial glasses
March 2009 83
This below-chassis view shows the Pilot Little Maestro after it had been restored and modified. It’s basically close to
original except for the power supply wiring. Note the long metal control shafts. They protruded through the front of
the wooden cabinet and, with the original power supply arrangement, could deliver a potentially lethal shock to an
unsuspecting user if one of the push-fit knobs came off.
for a whole range of sets.
Most other items in a receiver can
either be repaired or replacement parts
salvaged from other wrecked sets. Of
course, the closer those parts are in
appearance to the originals, the better.
Cabinet restoration is often a big
problem for many people, myself
included (although I can do minor
repairs successfully). The fact is,
major cabinet restoration work is a
craftsman’s skill. It’s a skill that some
have though and I’ve seen some magnificently restored cabinets over the
years.
It’s important to consider all of the
above factors before taking on a major
restoration job. But that’s not all you
have to consider. You also have to
think about safety, especially when it
comes to AC/DC sets (ie, sets without
a mains transformer).
A deadly receiver
We come now to vintage radios
which, due to their design, are inherently dangerous or, in fact, even
deadly.
Not that long ago, I was looking at
a couple of sets that belonged to a fellow restorer. I was rather keen to write
them up for SILICON CHIP, as they both
84 Silicon Chip
looked quite interesting. One was a
Philco AC/DC mantel receiver and
this was featured in the January 2009
issue. The other was a “Pilot Little
Maestro”, a 5-valve medium-wave
(MW) and long-wave (LW) receiver
made in 1939. We only used LW for
a short time in Australia but LW had
been used in Europe for quite some
time before the war.
This set is a British 240V AC/DC
unit and was apparently adapted from
an American design that ran on 110V
AC/DC. As with nearly all AC/DC sets,
the valves heaters are all in series.
The valve line-up included a 6A8G,
a 6K7G, a 6Q7G, a 25AG5 and a 25Z6G.
These valves all used 0.3A heaters and
the total voltage drop across these heaters was around 69V. This meant that
a further 171V needed to be dropped
across a resistor in series with the
heaters, so that the latter would not
draw more than 0.3A from the mains.
By contrast, the high tension (HT)
current would have been around
50mA, so the total power drawn from
the mains would have been about
85W, of which 51W would have been
lost across the heater series resistor.
In practice, this resistor was actually
formed into the mains power cord
and care would have been necessary
to ensure it had adequate ventilation
and that it wasn’t placed near flammable material.
In this set, one side of the mains is
connected directly to the chassis via
the on/off switch. That’s par for the
course with AC/DC sets but in this
case, the controls protrude through
the front of the wooden cabinet. This
means that if one of the push-fit control
knobs were to come off, the exposed
shaft could well be sitting at 240V!
In addition, the cabinet back is held
in place using just four wood screws.
There are no warnings on the back of
the set about the possibility of electric
shock, if the back is removed.
In short, I consider it to be a very
dangerous set.
Modifications
A close inspection revealed that this
particular set had been considerably
modified by a previous owner. First,
the 25V valves had been removed.
A 6BW6 had then been substituted
for the 25A6G, while a silicon diode
replaced the 25Z6G rectifier. The filament supply was provided by a 6.3V
filament transformer.
The set still had the mains consiliconchip.com.au
Fig.1: the circuit for the Hotpoint
P65ME (or AWA 565MA) receiver.
It had to be modified to drastically
reduce the amount of IF signal
getting into the audio amplifier
stage – see text.
nected to the chassis and the HT was
produced by using the silicon diode
to rectify the incoming 240VAC. But
that wasn’t all – the standard of the
revised wiring was atrocious.
In view of this, I refused to work
on the set as I didn’t think I could
make it safe without spending a lot of
time on it. However, a fellow vintage
radio buff (Marcus) did have the time
to make the set both safe and usable.
He agreed that it was a death trap as
it was and so decided to convert the
set to AC operation only by using an
external power supply. That way, the
mains could be completely isolated
from the chassis.
To cut a long story short, after quite
a bit of effort restoring the set and
making up an external power supply,
the set is now working satisfactorily.
It might not be completely authentic
but the main part of the set is close to
original condition with only the power
supply wiring altered extensively.
This conversion is stage one, as Marcus has suggested to the set’s owner
that he obtain a power transformer
from a defunct set that will fit the chassis. That way, they can eventually do
away with the external supply.
The performance of the set as modified is good and it also works quite well
on long-wave. It’s just a pity that this
AC/DC set could well have caused a
fatality in its original condition.
is relatively straightforward for those
with reasonable woodworking skills.
Next, a careful examination of the
chassis will soon reveal any mechanical items that need attention, while
many component faults will also be
quite obvious. These faults include
capacitors with cracks, bulges or extruded melted wax and resistors that
show obvious signs of overheating.
A close inspection will also soon
reveal perished wiring, shorts, poor
soldered joints and any damage due
to rodents and insects.
Of course, capacitors and resistors
also need to be electrically checked
to make sure they are in good order.
Old paper capacitors, for example,
are usually leaky and many will need
replacement. Valves are more reliable
than many people believe and I find
that I only have to replace them occasionally.
Once all the faults have been fixed,
some sets will also require alignment
– especially where a previous owner
has had a bit of a fiddle.
However, while most restorations
are routine, occasionally a particularly
difficult fault will be encountered.
These can cause a range of symptoms
including instability, distortion, tuning and alignment difficulties and
other weird faults. Let’s look at a
couple of examples that I’ve had to
deal with.
Difficult faults
Hotpoint P65ME/AWA 565MA
Most vintage radio restorations follow a fairly routine path. The cabinet
is easily assessed and the work on that
A friend of mine (Richard) had been
having problems with a couple of his
sets. Both were unstable, with multiple
siliconchip.com.au
whistles across the band and generally
just sounded unpleasant. They were
also experiencing interference problems on the shortwave band.
The first set we tackled was a Hotpoint P65ME. This is quite a nice
looking set and a good job had been
done on its restoration.
If there is instability and it appears
to be due to feedback in the intermediate frequency (IF) amplifier, the
first step is to make sure that little or
no IF energy is getting into the audio
amplifier stage. Amplification of the IF
signal by the audio stages can easily
generate sufficient positive feedback
to make a set unstable.
Fortunately, I’d had previous experience in solving what is basically a
design shortcoming in this set. This
involves several simple circuit modifications. Fig.1 shows the relevant
circuit details of the set.
The first thing I did was to cut the
connection between the bottom of L8
and the top of resistor R7. A 47kΩ
resistor was then fitted between these
two points and a 100pF mica capacitor
added between the bottom of L8 and
the chassis to provide additional IF
filtering (see Fig.3).
In addition, a 47pF capacitor was
connected between the grid of the
6AQ5 and the chassis (Fig.4).
Together, these modifications drastically reduced the IF signal on the grid
of the 6AQ5. The set was now much
more stable but a whistle could still
be heard when tuning across the band.
Next, I removed the 6BA6 IF amplifier and checked the AGC (automatic
March 2009 85
Fig.2: this Philips 1252 circuit also required several modifications to make it stable. This involved decoupling the HT
line to the early stages and fitting a screen-stopper resistor to the EL3NG audio output valve – see Fig.4.
gain control) voltage. This measured
0V so I reinstalled the 6BA6 and
removed the 6BE6. The set now had
several volts of AGC bias when there
should have been none.
Simply touching the 6BA6 or placing a finger on its grid altered the
AGC voltage level, so the IF amplifier
was obviously going into oscillation.
This signal was being detected by the
diodes in the 6AV6 which in turn
provided the AGC voltage.
We substituted another 6BA6 and
that completely fixed the problem. So
a new 6BA6 and the added IF filtering
made the set better than ever.
I later suggested to Richard that he
try refitting the old 6BA6 in the set,
along with an earthed metal shield for
this valve. My reasoning here was that
the 6BA6 has an internal shield that
is wired to pin 2 of its base. If a weld
had broken in the set’s original valve,
the shield would not be functional and
so the valve would oscillate.
However, when Richard plugged
the old 6BA6 back into the set, the
instability was absent. There are two
possibilities here: (1) the valve has an
intermittent break in the shield line;
This external power
supply was built
specifically for the Pilot
Little Maestro. It delivers
180V DC (for the HT line)
and 6.3VAC (for the valve
heaters) and completely
isolates the set from the
mains, making it safe
to use until a suitable
transformer can be fitted
to the chassis.
86 Silicon Chip
or (2) the socket itself might have
had some contact resistance which
reduced the efficiency of the shielding.
Philips 1252
Richard’s Philips 1252 is a very attractive console set but its performance
was also woeful. In particular, the IF
coils could not be peaked without the
IF amplifier going into oscillation, the
tuning had many nasty unstable signals right across the shortwave band
(7-22MHz) and the audio quality had
a harsh edge to it.
Once again, Richard’s restoration
looked good. And as with his Hotpoint receiver, it appeared that the
instability in the IF stage was due to
excessive IF signal levels finding their
way into the audio amplifier stages. As
before, the cure was to add additional
IF filtering.
Fig.2 shows the circuit details. In
this case the line going downwards on
the circuit from the bottom of L18 was
broken and a 47kΩ resistor inserted
into the break. A 100pF mica capacitor was then wired from the bottom
of L18 to the chassis. Finally, a 47pF
mica capacitor was wired from the
siliconchip.com.au
SECOND
DETECTOR
AUDIO OUTPUT
LAST IFT
WINDING
AUDIO
INPUT
OUTPUT
TRANSF
4.7nF
47k
100
500 –2.5k
47pF
4.7–16 F
47k
100pF
VOLUME
HT TO ALL
EARLIER
STAGES
TO AUDIO
OUTPUT
BLUE CROSSES
INDICATE WHERE
CIRCUITS ARE CUT
100nF
(LOUDSPEAKER
FIELD COIL)
HT FROM
RECTIFIER
Fig.3: the circuit at left shows how to install additional IF filtering before the
audio stages while the circuit at right shows how to decouple the HT line plus
fit a screen-stopper resistor (100Ω in this case) to the audio output valve.
grid of the EL3 (EL3NG in this set) to
the chassis.
This simple modification drastically
curtailed the amount of IF signal being fed to the audio amplifier but that
didn’t cure all the set’s ills.
Many sets using field coils (as in
this set) have very little decoupling
of the HT line after the field coil. As a
result, any variation in current drain
by the output valve plate circuit will
slightly vary the HT in the early stages
of the receiver. And this in turn can
cause instability.
The answer here is to increase the
decoupling of the HT line. This involves decoupling the HT line to both
those early stages and to the screen of
the output valve using a 1kΩ series
resistor and an electrolytic capacitor
(typically 16μF) – see Fig.3.
I fully expected this modification to
finally cure all the set’s problems but
we got a rude shock. On the positive
side, the IF could now be tuned to a
peak without the set spilling over into
oscillation. However, the nasty “birdies” on the shortwave band were still
there and the audio was still harsh.
Acting on a hunch, I touched the
body of the resistor that’s used to
decouple the front-end and the audio
output stage and this caused the nasty
sounds to alter. This indicated that the
audio output stage was still bursting
into supersonic oscillation despite the
extra filtering that had been added.
This was confirmed when we found
that placing a hand near the EL3NG
output valve had a similar effect.
siliconchip.com.au
16 F
16 F
Photo Gallery: Healing Golden Voice Console Radio
T
HE CONSOLE radios of
the 1930s and 40s were
typically fine examples of the
furniture-maker’s art. A good
number of them shared exactly the same chassis as a
large mantel radio from the
same manufacturer or were
only slightly modified.
Consoles had a number of
advantages, including a large
cabinet which had plenty
of space to mount both the
chassis and a large speaker,
the latter delivering better
sound and volume than the
smaller unit found in its mantel
counterpart.
The valve lineup in this radio
is 6J8G, 6U7G, 6B6G, 6V6G,
5Y3G. Photo by Kevin Poulter
for The Historical Radio Society of Australia (HRSA).
Phone (03) 9539 1117. www.
hrsa.net.au
Richard had a spare EL3NG and
substituting this gave a noticeable improvement but the set was still a little
“edgy” in its audio quality. As a result,
I decided to try fitting a screen-stopper
resistor, as some valves will oscillate
at all sorts of supersonic frequencies
if a screen-stopper is absent.
In this case, a 100Ω screen-stopper
resistor was added directly between
the screen and the output of the HT
decoupling network that had been
fitted earlier. Fig.4 shows the details.
Once this had been done, the audio sounded clean with either valve
inserted into the audio output socket.
In addition, the “birdies” (whistles) on
shortwave also disappeared.
In short, the manufactuers didn’t
SC
always get it right.
March 2009 87
|