This is only a preview of the April 2022 issue of Silicon Chip. You can view 47 of the 120 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. Items relevant to "500W Power Amplifier, Part 1":
Articles in this series:
Items relevant to "Railway Semaphore Signal":
Items relevant to "Update: SMD Test Tweezers":
Items relevant to "Capacitor Discharge Welder, Pt2":
Purchase a printed copy of this issue for $11.50. |
Vintage Radio
Monopole D225 tombstone radio
from 1934
By Assoc. Prof. Graham Parslow
Made in France in 1934, this ‘French Cathedral’ style radio was also sold
separately as the model D25, which consisted solely of the chassis. The
model D225 is a superhet design featuring five valves, with a total weight
of nearly 16kg. Its original price was 1850 French francs.
G. Bouveau et Cie Constructeurs
started business in 1925 in Paris. Its
name was changed in 1928 to Societé
des Établissements Monopole and in
1934, it moved to the Montreuil-sousBois area of Paris. It manufactured
a range of radios through the 1930s,
ceasing after the German occupation
in 1940.
The radio featured here was one
of their prestige models and nicely
brings together form and function in
the prevailing tombstone style. Radios
of this era typically came with internal
speakers, rather than requiring separate speakers as in the previous ‘coffin box’ era.
This radio came to me for electrical restoration via Darren McBride,
a French polishing professional with
110
Silicon Chip
Hecdar Heritage in Melbourne. He
restored the case magnificently but the
electrical, fabric and mechanical restoration were my challenges.
I have restored many timber cabinets using polyurethane, but French
polishing is far more labour intensive
and is justified by the unique quality
of the outcome. As Darren McBride
relates on his website, this traditional
technique results in a high-gloss surface with deep colour and a striking
three-dimensional vibrancy.
The process of French polishing
consists of applying many thin coats
of shellac dissolved in alcohol, using
a rubbing pad lubricated with oil (or
even a microfibre cloth without oil).
It is a lengthy and repetitive process,
requiring a specific combination of
Australia's electronics magazine
various rubbing motions, waiting, and
repeating, building up layers of polish.
One advertisment poster (shown at
the end of the article) states “Vague
de Puissance et Harmonie etc” which
translates roughly as “A wave of
Strength and Harmony [will be delivered by your Monopole Radio], made
by hand in France with the latest technology.”
Monopole claims to have used the
latest technology for 1934, which is a
reasonable claim. The D225 is a superhet with fundamentals that would
continue to the end of the valve radio
era in the 1960s. It was not so much
the basic superhet design that would
evolve after this, but the efficiency
and performance of the valves would
increase.
siliconchip.com.au
The valve sockets for the D225 were
designed to match the number of pins
needed by a valve, rather than using
a standard socket. The octal base was
only released onto the market in April
1935 and did not immediately gain
traction.
The photo of the top of the chassis
shows four different valve bases of
4 to 8 pins. It also shows the mains
transformer without the top cover and
reveals the unusual pattern of winding
and lamination.
Circuit details
The model designation D225
describes the cabinet; the chassis
mated to it is the model D25. The
Monopole circuit diagram is among
the clearest to be found from the early
1930s, with only one significant use
of French notation – “H.P.” for haute
parleur (high speaker), the primary of
the output transformer.
All of the valves have indirect heaters driven by 4V except the mains rectifier, which has a directly heated 4V
cathode. In the radio featured here,
the superhet mixer valve was an AK2
(equivalent to AK1 indicated on the
circuit, but with an alternative base).
The IF amplifier was an AF2, while
the detector-audio preamplifier was
a type TE44 (equivalent to the E444).
The output valve was missing.
The full-wave rectifier was type
AZ1. The high tension filter choke
is the field coil of an electrodynamic
speaker.
siliconchip.com.au
The restored tombstone radio, along
with some shots of the chassis during
restoration. Note that the photo below has
the dial already repaired.
Australia's electronics magazine
April 2022 111
The RF section has a tuned aerial
coil primary as well as the usual
tuned secondary. The radio has good
selectivity for tuning, helped by the
double-tuned aerial coil. More significantly, the double tuning (preselection) improves image rejection generated by the intermediate frequency
(IF) of 120kHz.
Images (a second tuning spot) would
exist at 120kHz × 2 = 240kHz above
the transmission frequency. I did not
find images generated by this radio.
The third gang of the tuning capacitor is linked to the local oscillator,
configured as an Armstrong tuned
grid oscillator. They possibly chose
an intermediate frequency of 120kHz
because valves of the time were more
effective amplifiers at lower frequencies.
The normal MW range is calibrated
on the dial as 200-550 metres. Shortwave is tuned by shorting sections of
the coils used for MW.
The radio is not highly sensitive,
and strong local stations are noticeably
louder than medium-strength stations.
This is despite automatic gain control
(AGC) mediated by the 1MW resistor
feeding back from the audio detector
to the AF2 valve. That 1MW resistor
also feeds back to the mixer valve via
a 250W resistor and the secondary of
the aerial coil.
The circuit diagram shows that the
voltage to be expected at the first filter
electrolytic is 328V DC, with 248V DC
after the choke. These values proved
useful in restoring the radio.
Electrical restoration
The top of the chassis was grubby
and stained with a resinous film that
is thought to be from material in the
mains transformer that sublimates
(turns from a solid to a gas) to cover
surrounding components. Otherwise,
its condition was fair.
The temptation to immediately
clean the components is one I try
to resist because the radio probably
worked in this state before. Sometimes
cleaning introduces new problems, so
I leave it for later unless the presentation is severe.
The bad news at the top of the chassis was that the E463 audio output
pentode was missing. The Historical
Radio Society of Australia (HRSA)
valve bank listed the E463 valve, but
had no stock.
This is not surprising for an old
European type. I later came up with a
work-around for this problem.
The elegant AZ1 double diode
mains rectifier was wrapped with bandage material at the base, but this no
longer held the glass envelope to the
base, and any knock may have separated them. I ran a line of thin-CA
glue around the base. This glue sets
relatively quickly and strongly (CA is
cyanoacrylate, a form of superglue).
Unfortunately, some leaked into the
base where it continued to leak down
the pins and glued the valve to the
socket. The recovery from this mistake
was tedious, but eventually successful. A lucky circumstance is that these
veteran valve sockets have elongated
claws rather than sockets, and I could
prise the claws apart.
Below the chassis was a mostly
pleasant surprise. Someone had
replaced all capacitors and many resistors with 1960s-vintage components.
The circuit diagram for the D225 Monopole radio.
112
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
The electrodynamic speaker was
a replacement Australian 8-inch
(200mm) Rola of 1930s vintage, with
the cone in perfect condition.
The niggling thought when encountering such a comprehensive component replacement is to resolve whether
it was motivated by the need to fix a
difficult fault, and if so, whether it
was successful.
The HT filter electrolytics were contemporary black sheathed types (other
replacement capacitors were from the
1960s). It appears that more than one
person had worked on restoring this
radio. For a radio of this vintage (with
uncertain integrity) I decided to ramp
up the AC input using a variac to avoid
self-destruction.
I monitored the HT voltage at the
first filter electrolytic, knowing that the
circuit diagram indicated 328V. There
was no surge of power consumption,
but the HT reached 350V at 205V AC
from the variac. On reflection, this
made sense because the E463 output valve was missing, and this load
would generally reduce the HT due
to an increased drop from the internal
resistance of the AZ1 diodes.
Was the radio working? Linking a
signal tracer to the volume control
input gave instant gratification that
audio signal was coming out of the
RF section.
This was a qualified joy because
the signal crackled and intermittently
dropped out. Tapping anywhere on the
chassis upset the signal.
After many hours, I located one dry
joint and another joint that was merely
a wire resting on a solder lug. Fixing
these improved matters considerably,
although, even at the end, there was
still an intermittent crackle and sensitivity to tapping the radio. I concluded
that one or all of these old valves was
susceptible to microphonic instability. Unfortunately, no swap-in valves
were available.
To allow safe operation without the
variac, I inserted a 400W 20W resistor
in series with the primary of the mains
transformer. The result was a drop of
40V across the resistor and an HT voltage after warm-up of 310V DC.
At one point, the radio stopped
working, and the HT rose to 395V. I
traced this to a 5kW 2W resistor loading the screen grids of the AK2 mixer
and AF2 IF amplifier. It had gone
open-circuit, so I replaced it without
difficulty.
siliconchip.com.au
As the power supply was weak, and the output valve missing, I replaced that
valve with an LA4160-based amplifier module at the top of this photo (taken
from an old Sanyo cassette radio).
Unexpectedly, the HT fell to 110V
DC after this replacement, but this
could be reversed by removing the
AF2 valve. The screen voltages of the
AK2 and AF2 valves were originally
derived from an 8.5kW series resistor
from the second filter electrolytic. A
previous restorer had increased this
series resistor to 15kW, a change that
lowered the current drawn by the AK2
and AF2 valves.
Even with this limited screen current, the power supply was not coping.
Australia's electronics magazine
Increasing this limiting resistor to
27kW brought the HT back, and the
radio worked well.
The severe limitation of the old AZ1
diodes to supply current was reinforced when I rigged up a 6V6 output
tetrode to replace the missing E463
valve. I provided the heater current for
the 6V6 from an external 6.3V source
because the radio didn’t have any suitable windings.
Initially, this replacement produced
absolutely nothing, leading to the
April 2022 113
discovery that the output transformer
(labelled HP on the circuit diagram)
had an open-circuit primary. This
could explain why the E463 valve
was missing. When the anode has no
HT because the output transformer
is open-circuit, a high current flows
through the screen grid, which can
destroy the valve.
With a good output transformer, the
6V6 again produced nothing, this time
because the HT had fallen to 88V. It
was evident that the AZ1 dual diode
valve had such low emission that it
was not up to providing more than a
few milliamps. Any output valve was
going to over-tax the power supply.
Looking through my collection of
1980s cassette radios (that my wife
wonders why I keep), I selected a
Sanyo model M2553F to sacrifice for
the greater good. This model has a
discrete amplifier section, separate
from the radio module, that runs at
7.5V DC rectified from a small mains
transformer.
The Sanyo LA4160 amplifier IC on
the module is good for 1.2W audio output. This is comparable to the 1.5W
output from an E463 valve. Bench testing proved that this was a workable
solution and that the old electrodynamic Rola speaker was in good condition. I mounted both the small power
supply transformer and the amplifier
module under the chassis so that the
radio continued to look original.
The other option would have been
to replace the AZ1 with a solid-state
rectifier, but the resulting inrush current can cause problems, so it isn’t as
simple an option as it first appears.
Restoring the dial
A French advertising poster showcasing the D225 radio. The advert is stated to
be from “Damour-Editions”, and measures 120cm high and 80cm wide.
The earlier chassis
photo had the
transformer cover
removed so that
its windings and
laminations could be
seen. This is what it
looks like with the
ventilated cover in
place.
114
Silicon Chip
Australia's electronics magazine
The celluloid dial was discoloured
and cracked. It was so brittle that an
attempt to glue the pieces together
fragmented it even further.
The chemistry of this is interesting.
Celluloid dials are fabricated from
nitrocellulose with an added plasticiser like camphor that makes the
product supple. With age, the camphor evaporates, leaving a brittle sheet.
UV light also catalyses denitrification
of the cellulose with the release of
nitrogen oxides that give the celluloid a brown colour (note how the area
exposed to light is darker).
The only solution was to completely
redraw the dial at a larger scale and
reduce it to size. I used PowerPoint to
create the text and lines on a yellow
siliconchip.com.au
background, then printed it on 60gsm
paper with adequate transparency to
allow the dial light to shine through. I
mated this to a rectangle of clear polycarbonate for support.
The original dial lamp was open-
circuit, so I replaced it with a 3.5V
torch globe.
Finishing touches
The original speaker grille cloth was
in tatters. Fortunately, I had material in
my fabric drawer that closely matched
the original, with a brown and gold
pattern (unlike the plain fabric shown
in the Monopole poster).
I carefully installed the chassis into
the cabinet to avoid damaging the
French-polished finish. I then added
a ventilated rear panel along with a
warning of the high voltage hazard
inside – the top caps on two of the
valves are the anodes, not low-voltage
control grids. I also included a note
telling the user that an aerial must be
installed (which you can see in the
adjacent photo).
I needed to give the radio a final
check before all 15.7kg of this hefty
unit could be returned to Darren.
After warming it up, I was listening
to only crackle, dreading the need to
start again. My first check was to see
that the wave change switch was still
set to MW. It was not, so a click later,
happiness prevailed.
The radio produced excellent
sound, with the speaker now baffled
SC
in its resplendent cabinet.
This shows the rear panel and label I fabricated. Also present are (from left-toright) the power cord, mains voltage selection switch, speaker socket (hidden),
a ‘pickup’ audio input socket, Earth connection and antenna connection.
This speaker was not original, it was
instead an Australian-made 8-inch
Rola speaker of the same vintage.
The celluloid dial originally came cracked and yellowed, due to the way
celluloid degrades as it ages. It was much easier to make a new dial than fix it.
siliconchip.com.au
Australia's electronics magazine
April 2022 115
|