This is only a preview of the November 2012 issue of Silicon Chip. You can view 22 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. Items relevant to "High-Power Class-D Audio Amplifier, Pt.1":
Items relevant to "High-Energy Ignition System for Cars, Pt.1":
Items relevant to "LED Musicolour: Light Up Your Music, Pt.2":
Items relevant to "Hacking A Mini Wireless Webserver, Pt.1":
Purchase a printed copy of this issue for $10.00. |
Vintage Radio
By Rodney Champness, VK3UG
The HMV A13B 4-Valve
Twin-Chassis Mantel
Produced by HMV during the early 1950s,
the A13B 4-valve mantel receiver was
small in size but big in performance. It’s
also an easy set to work on and a simple
modification to the oscillator circuit makes
it work even better.
dial scale which looks quite attractive,
although it is relatively small. The
loudspeaker is located to the left and
is partly behind the dial scale.
The cabinet has four large holes
towards the top of the rear panel,
designed to accept four fingers so that
the set can be easily lifted and moved
from one location to another (after
first disconnecting the antenna and
unplugging the set from the mains).
The antenna supplied with the set was
around 6-7m long and this was typically draped along a picture rail or run
along the skirting board in the room.
As an aside, radio receivers of this
era were often supplied with a shortwire antenna. This could be used in
good signal areas instead of the set
being connected to a large, outside
antenna. Of course, that was before
ferrite-rod antennas came into common use.
In practice, most people soon abandoned the idea of shifting such sets
from room to room, since relocating
the antenna each time was a nuisance.
The advent of the ferrite-rod antenna
made shifting valve sets easier but it
wasn’t until transistor receivers arrived that sets became truly portable.
The advent of transistor receivers
also eventually made it possible for
households to afford multiple sets. By
contrast, at the time the HMV A13B
was produced, receivers were expensive and the average household could
only afford one receiver.
Circuit details
I
FIRST SAW one of these receivers
at my grandparents’ home in the
early 1950s. I’m not sure why I was
so intrigued with the set; maybe it was
because it was so small compared to
other radios I was familiar with at the
time (mainly large vibrator-powered
receivers that ran off batteries). Or
perhaps I was impressed by the performance delivered by such a small set.
Of course, by modern standards, it
isn’t all that small. However, at the
84 Silicon Chip
time, it was the smallest I had seen and
my grandparents’ set was even housed
in a brown Bakelite cabinet, just like
the A13B receiver featured here.
A little history
HMV has always built interesting
receivers, both from a technical viewpoint and in terms of appearance. The
cabinet of this receiver is much smaller
than other 4-valve superhet receivers.
It features rounded edges and a central
Fig.1 shows the circuit details of
the HMV A13B. It’s a 4-valve superhet
design (broadcast band only) but its
performance matches that of many
5-valve receivers due to the fact that its
IF (intermediate frequency) amplifier
valve also acts as the first audio stage.
This particular circuit technique is
called “reflexing” and was common
in Australia from the 1930s to the mid
1950s. However, it was not used as
much in other countries.
The antenna input circuit (top, left)
siliconchip.com.au
Fig.1: the circuit is a 4-valve superhet
design with valve V1 functioning as a
converter and V2 (EBF35) acting as both
an IF amplifier and an audio amplifier (a
technique known as “reflexing”). V2 also
includes the detector and AGC diodes. V3
(6V6GT) is the audio output stage while
V4 (6X5GT) is the rectifier.
was designed to extract the maximum
possible signal from a relatively short
antenna. As shown, the antenna input coil L1 is tuned using parallel
capacitor C1, the resonant frequency
of this combination being just below
the bottom end of the broadcast band.
By doing this, the maximum possible
signal is extracted from the antenna
at the low-frequency end of the band
and this signal is inductively coupled
to coil L2.
At the high-frequency end of the
band, capacitor LC1 couples the antenna signal direct to L2. Either way,
the signal is fed into a secondary tuned
circuit comprising L2, TC1, VC1 & C2.
C2 is the AGC bypass capacitor and
although it’s included in the tuned
circuit, it has little effect on its tuning.
The signal from the antenna tuned
circuit is fed to the signal grid of V1,
a 6A8G converter valve. The local
oscillator is a little different from
normal in that it uses “padder feedback”, achieved by connecting bypass
capacitor C4 to the tuned oscillator
winding instead of to chassis (earth).
siliconchip.com.au
This view inside the receiver shows the unusual “twin-chassis” arrangement,
with the parts mounted mainly on the two horizontal sections.
This ensures that the oscillator operates reliably at the low-frequency end
of the tuning range.
However, I don’t particularly like
the design of this circuit. The circuit
used in the A13B was slavishly followed by many man
ufacturers but
other manufacturers used the circuit
shown in Fig.2. This is a more dependable circuit that will oscillate
November 2012 85
The top section of the chassis supports valves V2 & V3 (EBF35 & 6V6-GT) plus
the two IF transformers. Note the shield over the IF/first-audio amplifier valve,
which minimises IF signal radiation and hum pick-up.
This view shows the parts layout under the bottom section of the chassis. This
section mainly supports valves V2 & V4 (ie, the 6A8G converter & the 6X5-GT
rectifier) and the mains transformer.
reliably across the entire tuning range
without resorting to tricks like padder
feedback. It maintains a more constant
degree of feedback between the priV1
50pF
50k
C3
425pF
VC2
TC2
L3
L4
R3
+HT
C4
10nF
This slightly revised oscillator circuit
uses the same parts but provides better
performance than the original.
86 Silicon Chip
mary and the secondary windings than
the circuit used by HMV in the A13B.
To prove the effectiveness of this
slight circuit modification, try modifying a set using the HMV-style circuit
to that in Fig.2. Receivers with 6A7
converters appear to benefit a great
deal from this modification and the
sensitivity of the set often improves
noticeably.
The output from the converter (at
the plate of the 6A8G) is fed to the first
IF transformer (IFT1) and then to the
grid of IF amplifier stage V2, an EBF35.
From there, the IF signal is fed to a
second IF transformer (IFT2) which
then feeds the detector diode in V2.
As an aside, note that C6 and C10 in
the first IF transformer have different
values, ie, 100pF and 50pF respectively. This means that, unlike IFT2
where the values are equal (100pF),
the inductance of IFT1’s secondary
is double the value of its primary. As
a result, it’s not a good idea to swap
these two IF transformers (or use an
incorrect substitute for IFT1), as the
IF gain and hence the sensitivity of
the set would be degraded.
The detected audio signal from V2
is fed to the top of volume control
VR1. It then passes via C14, R10 and
the secondary of IFT1 to the grid of
V2, where it is amplified (along with
the IF signal). The resulting amplified
audio signal is then fed through the
primary winding of IFT2 to resistor
R16 (20kΩ) and from there to the grid
of V3, a 6V6-GT audio output stage,
via C18 and R18.
V3 in turn drives the loudspeaker
via a speaker transformer.
Note that because V2 acts as both as
an IF amplifier and audio amplifier,
some compromises have been made
in regards to some of the component
values around this stage. This means
that it may not provide the maximum
gain that would otherwise be possible,
either as an IF amplifier or as an audio
amplifier. The usual compromise is to
restrict the audio gain to around 15,
whereas if the valve had been used
purely as an audio amplifier, its gain
could be well over 100.
Getting back to the output stage, the
6V6-GT’s grid has -10V bias applied to
it from the power supply’s back-bias
network. In addition, negative feedback is applied from the secondary of
the audio output transformer to the
bottom end of volume control VR1.
Tone control
The tone control is extremely simple and consists of switch S1 which
switches capacitor C20 in or out of
circuit. In addition, resistor R19 and
capacitor C21 between the 6V6GT’s
plate and chassis form an elementary
fixed tone control. The capacitor has
a reactance of about 5kΩ at 3.5kHz,
giving a combined impedance for the
series resistor-capacitor combination
of 10kΩ at this frequency. This impedance drops to just 7.5kΩ at 7kHz.
Power supply
The power supply is quite conventional and is based on a power
transformer and a 6X5-GT full-wave
rectifier (V4).
As shown in Fig.1, the power transformer primary is tapped for 200-225V
mains supplies and for 226-250V
supplies (40-50Hz). There are two
secondary windings: a heater winding
siliconchip.com.au
of 6.3V and a 520V centre-tapped HT
(high-tension) winding (ie, 260V either
side of the centre tap). Note that the
centre tap is connected to chassis via
two series resistors (R14 and R7) and
these are used to generate the back-bias
for the various valves.
The 6X5-GT rectifier produces nearly
270V DC at its cathode and this is
applied to the plate circuit of the
6V6-GT via the speaker transformer’s
primary. By contrast, the HT voltages
for the plate circuits of V1 & V2 and
the 6V6-GT’s screen are obtained via
parallel resistors R12 and R13, which
limit it to around 185V. The screen circuits for V1 & V2 also have additional
filter components connected to their
supply lines.
A back-bias voltage of about -2.2V
is applied to V1 and V2 and this is
obtained across resistor R7 (40Ω).
This back-bias is applied via R9 and
R8 to the AGC diode in V2, so this
receiver has delayed AGC (automatic
gain control). The back-bias voltage
is also applied via R2 to V1, which
receives the full AGC bias developed
at the AGC diode.
V2 also has -2.2V of back-bias applied to its grid. However, it only receives around 9% of the AGC control
voltage compared to V1.
The reason that only a small percentage of AGC voltage is applied to
V2 (the IF amplifier-cum-audio amplifier) is simple. Its operating conditions are a compromise and any major
variation in these conditions could
result in distortion and overload. In
addition, because AGC reduces the
gain of the IF amplifier stage, it’s obvious that it also reduces the gain at
audio frequencies as well. So if too
much AGC voltage is applied, the
audio output could become quite
weak in the presence of strong station
signals.
However, with careful circuit design, it’s possible to come up with
a good compromise to maintain a
constant audio level regardless of the
incoming signal strength.
Servicing access
Access to the chassis is gained by
removing two screws from the rear
section of the cabinet and then slipping the back off. Once this is done,
the unusual layout of the receiver
is immediately obvious. It has a Cshaped “twin-chassis” arrangement,
with the parts mounted mainly on two
siliconchip.com.au
The loudspeaker is attached to the front vertical section of the chassis and
sits partly behind the dark backing material for the dial scale.
A label inside the cabinet indicates the alignment points and the valve types
(and their locations). It also shows the dial-cord arrangement.
horizontal sections, one at the top and
the other at the bottom. The vertical
section carries the loudspeaker and
the dial-drive components.
To remove the chassis, it is first
necessary to remove both knobs and
centre the tone control switch between
the two rotary controls. It’s then just
a matter of removing four mounting
screws, after which the chassis and
cabinet front can be separated.
Once the chassis has been removed,
it’s easy to access all the parts, including the dial-drive mechanism. Most
of the larger parts, including the coils
and transformers, would rarely (if
ever) require replacement. Only the
occasional valve replacement would
be necessary.
Two large holes in the bottom section of the chassis provide access
to the bottom tuning slugs of the IF
transformers. Note that the circuitry
around the IF/first audio valve is
shielded to minimise IF signal radiation. This shielding also helps prevent
the audio stage from picking up mains
radiation, which would cause audible
hum in the output.
The chassis layout and wiring of
this set are quite logical and access for
normal service is a dream compared to
many other sets. A label on a curved
section inside the cabinet shows the
valve types, the dial-drive layout and
the locations of the alignment adjustments. It also shows the mains winding taps for the transformer primary.
November 2012 87
had also gone low in value and so all
the electrolytics were replaced as a
matter of course. A number of out-oftolerance resistors were also replaced
but all the valves checked out OK.
This work solved an annoying intermittent crackling in the audio that had
previously been evident. In fact, the
set then performed so well that the IF
stage alignment was left alone. It may
have been possible to wring just a little
more performance out of the receiver
if an alignment had been done but it
was thought that this was already was
close to optimum.
Keep the leads clean
The rear section of the cabinet has four large “finger holes” so that the set can be
easily picked up and carried. Undoing the two screws allows the rear section to
be removed and provides good access to most parts with the chassis in-situ.
So even without a circuit diagram,
it’s not particularly difficult to find
your way around this chassis. However, because it is a reflex set, a circuit diagram is handy when working
around that IF/audio amplifier stage
(V2), as this stage is more complex
than in many other receivers.
Restoration
This old HMV A13B was overhauled
and restored to full working order by
its owner (Mark) and one of his friends.
First, the figure-8 power lead was replaced with a 3-core cable (securely
anchored using a cable clamp) so that
the chassis could be safely earthed.
That done, the capacitors were all
checked and quite a few were found to
be electrically leaky, with resistances
of just a few megohms when they
should have been greater than 200MΩ
(and preferably in excess of 1000MΩ).
Several of the electrolytic capacitors
As an aside, my own supply of resistors and capacitors is quite extensive.
Most of these parts are new-old-stock
(NOS) and often 20-30 years old,
which means that their leads have
tarnished in many cases.
As a result, when using these parts,
I have to carefully clean the tarnish off
using sandpaper and sometimes even
a scraper to get down to bright metal
which can be soldered. Neglecting
to do this would result in bad solder
joints and could easily introduce numerous new faults into equipment that
was being serviced.
Mark didn’t have this problem
because his replacement parts were
purchased new. However, it’s still
something to keep in mind if using
parts that you’ve had stashed away
for some time.
Summary
Another view inside the old HMV A13B with the rear section of the cabinet
removed. Two holes in the bottom section of the chassis provide easy access
to the adjustment screws of the IF transformers.
88 Silicon Chip
HMV has always produced welldesigned receivers and this set is no
exception. However, as stated earlier,
the oscillator circuit has some minor
shortcomings and I much prefer the
circuit shown in Fig.2. The components used in this revised circuit are
the same as those used in the original
but from my experience, it offers better
performance.
The oscillator circuit used by HMV
in the A13B wasn’t unusual though.
The same configuration was used by
other manufacturers, including AWA,
and “Radio & Hobbies” magazine also
used it in many of their AM receiver
designs.
Apart from my beef about the oscillator circuit, the rest of the circuit is to
HMV’s customary high standard. The
performance of the set is also very good
and I would be happy to have one in
SC
my collection.
siliconchip.com.au
|