Marconi Model 730 TRF Receiver
|
 |
|
This TRF receiver,
designed in the early 1930s was used in ships until superseded
by more modern superhets.
Like a handful of manufacturers,
Marconi decided to use plug-in coils rather than a complex switching
arrangement. A few Eddystone and National HRO receivers had similar
arrangements.
It uses four old battery-powered
7-Pin British valves, type W21 pentodes popular during the mid
to late 30s before widespread use of octal types. One is missing
(my guess is that this is for monitoring the old 16KHz channel
from Rugby, now defunct). There's also a 6-pin socket that's
missing something. Looking at the circuit at the bottom of this
page, this was a coil for a preset circuit tuned to the old international
distress frequency of 500Kc/s or 600 metres and selected by the
mode switch on the front panel when in the "STAND BI"
position.
Arrayed along the top and under
the hinged top section is its set of plug-in coils. There are
10-wavebands, Range 8 coils are in place on the chassis, but
maybe the Range 4 set was surplus to requirements or just got
mislaid? I constructed the table below on the assumption that
Range 0 is the highest and Range 10 the lowest frequency range
and "M" is metres. Once I've tested the receiver this
will become clear. If my assumption is correct then Range 8 will
cover Radio 4 on 198KHz. I suppose I could weigh a Range 0 and
10 coil when the heavier will be for the longer wavelength? Indeed,
Range 0 was 90gm and Range 10 was 138gm. I used the wavelengths
given in a technical article which listed the 10 wavebands (below)
but inspection of the coil labels (0 to 10) surely shows eleven
wavebands. In that case could the missing Range 4 coil be the
one that plugs directly into the chassis covering a specific
wavelength in the 600 metre waveband? The number doesn't actually
make sense so some investigation is needed. I count ten pairs
of holes for the coils, so 10 wavebands seems right, however
I found on the Net, pictures of a couple of other 730 examples,
one of which had a pair of Range 4 coils, all the spares in place,
and a pair in use (making a total of 12 wavebands including the
fixed frequency coil. I'm sure that once the receiver is powered
up and the various coils tested, all will become clear... |
|
RANGE |
FROM |
TO |
RANGE |
FROM |
TO |
|
0 |
15M |
35M |
6 |
275M |
500M |
|
1 |
30M |
55M |
7 |
460M |
1100M |
|
2 |
50M |
100M |
8 |
1100M |
2300M |
|
3 |
85M |
180M |
9 |
2,300M |
8,000M |
|
5 |
170M |
330M |
10 |
8,000M |
20,000M |
|
|
|
 |
|
Here are the three valves
fitted in the receiver. The centre valve is an Austrian S217
(2v x 0.18A, Va 150v @ 3mA, Vg2 150v @ 1mA, gm 1.7) similar to
the variable mu VP2. Interestingly, the W21 and similar types
were introduced with a choice of B4 or B7 base which must have
resulted in more than a few puzzled customers. Both the VP2 and
S217 draw 180mA filament current compared with the W21 100mA. |
|
|
|
The set is built into
a wooden case fitted to rubber shock mounts. As can be seen below
the top section, holding the spare plug-in coils, hinges upwards
for access to extra coils and maintenance. |
 |
|
|
|
I'll take a better photo
soon. This shows an early crystal detector, above the Reaction
control because, in an emergency situation, the receiver could
be pressed into use as a crystal set. Note the rather stout cat's
whisker!
 |
|
Another interesting facility
is a 16KHz fixed frequency setting originally tuned to Rugby
Radio Station. This station was built in 1925 for the transmission
of timing signals for keeping clocks in synchronism. Alas the
station was finally dismantled in August 2007, over 80 years
after its first transmissions on 1st January 1926. Transmissions
which ceased on 1st April 2003 were replaced by signals from
Criggion, Powys which for many years had acted as standby station,
however Criggion, in turn, was dismantled very soon afterwards
and transmissions taken over by Anthorn in Cumbria.
These low frequency transmitting
stations were also used, on slightly different frequencies, to
communicate with submarines as the extremely low frequency (ELF)
signals could be received under water. If you look carefully,
elsewhere on this website I've mentioned this in connection with
my work with Plessey during the "cold war".
Below is the circuit diagram
from the manufacturer's operating manual.
Power is provided from external
batteries. 2 volts for valve filaments, -10 volts for grid bias
and 110 volts HT.
A puzzle must be the hole in
the top left of the front panel (seen
in the picture above). This might be explained by the neon
lamp shown in the circuit diagram. In keeping with other receivers
used by the Royal Navy, or in a marine environment ( for example
the DST100 and the R1116), some form of limiter was placed across
the RF amplifier. This was done to prevent too great a signal
from a local radio or radar transmitter from damaging the valve. |
 |
|
Click
to see an excerpt, describing the 730, from a technical instruction
handbook
Read
about the problem of interference on marine receivers |
|
The Marconi 730 was also
used for listening to enemy transmissions until being superseded
by the more advanced DST100 borrowed
from the Army. |
 |
|
Below, I've marked component
identifications to aid testing. |
 |
|
Resistor |
Ohms |
Resistor |
Ohms |
Condenser |
Value |
Condenser |
Value |
Component |
Description |
|
R1 |
500K |
R13 |
25K |
C1 |
|
C11 |
100pF |
CR1 |
Crystal rectifier |
|
R2 |
5K |
R14 |
750 |
C2 |
0.01uF |
C12 |
0.1uF |
J1 |
Heaphone socket |
|
R3 |
10K |
R15 |
2M |
C3 |
0.1uF |
C13 |
0.1uF |
TC1-TC4 |
600 Metre Rx trimmers |
|
R4 |
|
R16 |
1K |
C4 |
|
C14 |
500pF |
VC1 |
HF stage tuning |
|
R5 |
20K |
R17 |
- |
C5 |
|
C15 |
0.01uF |
VC2 |
Main tuner |
|
R6 |
20K |
R18 |
- |
C6 |
|
C16 |
200pF |
VC3 |
Reaction control |
|
R7 |
30K |
R19 |
- |
C7 |
|
C17 |
100pF |
VC4 |
16KHz Rx tuning |
|
R8 |
500K |
R20 |
- |
C8 |
50pF |
C18 |
|
S1 |
Mode switch |
|
R9 |
500K |
VR1 |
25K |
C9 |
0.1uF |
C19 |
|
S2 |
16KHz Rx on/off |
|
R10 |
10K |
VR2 |
500K |
C10 |
0.1uF |
C20 |
|
S3 |
On/off switch |
|
R11 |
50K |
|
|
C11 |
100pF |
C21 |
100pF |
T1 |
Interstage transformer 1:4 |
|
R12 |
2K |
|
|
C12 |
0.1uF |
C22 |
0.1uF |
T2 |
Output transformer 12:1 |
|
|
I worked out the power
supply wiring. There's a large connector block from which emerges
a set of wires. Green which is chassis, white, LT+, orange HT+
and black, HT-. Like many receivers from the 1930s and 1940s,
without a separate grid bias battery, the HT return is taken
to the chassis via resistors, the current through which automatically
determines a more or less fixed bias voltage. In the circuit
above VR1, the RF gain control, allows a varying grid bias to
the control grid of V1. A second connection from HT- to R14 and
R12 plus R13 sets the grid bias for V3, the audio amplifier.
I applied power from my home-brew
PSU and set the HT to around 120 volts. Using the coils marked
"8" and a short aerial revealed two stations, Irish
Radio 1 near the high end of the band and Radio 4 around the
middle. I added a long wire and continued testing. There was
some interaction between controls but this is not unexpected
as this is a TRF receiver. Everything appears to work as it should
except the audio output varied dramatically depending on the
control settings and it was easy to get deafened as tuning and
reaction were set to optimum. Many old receivers prevent this
by fitting a limiter in the headphone circuit.
Because it's been decades since
the set has been used I needed to use switch cleaner on the mode
switch but once cleaned the set was relatively stable. Overall
though I found it to be not too pleasant an experience especially
when compared with a suphet receiver employing AVC.
I decided to check the circuit,
in particular to see if any components needed changing. During
this exercise I found an anomaly. RF gain control, in common
with other battery receivers (such
as the R1116) is achieved by placing a suitable negative
grid bias on the control grid of the RF amplifier (note...in
a receiver using valves having heaters and separate cathodes
the much simpler technique of using a variable resistor in the
cathode(s) is used). Anyway, I found the wiper and live end of
VR1 were transposed. This has the effect of progressively shunting
HT- to chassis as the gain control setting is increased and simultaneously
reducing the bias to the audio amplifier. The overall result
spells trouble for the operator (bearing in mind the interaction
of the various controls which already makes it easy to get deafened
when attempting to tune in a station) by accentuating any interaction
of controls. Because the set uses separate variable condensers
for tuning and RF amplification, headphone output can vary by
a huge amount and for example, having tuned a weak signal in
conjuction with the reaction control, retuning to an adjacent
more powerful station results in excessive audio output. |
 |
If my memory serves me
correctly the wiper is connected to the centre pin. The right
hand pin is minimum resistance clockwise and left minimum resistance
anti-clockwise.
Top right is ground and that
means the 5Kohm resistor circuit is reversed with that on the
centre pin.
Below, a picture of the tagstrip
located centrally on the back of the chassis. Note a modern electrolytic
capacitor added by a previous owner. |
|
|
|
|
Below, this terminal board
carries the power supply leads. Above this a tagstrip carrying
R1 (500Kohm), R14 (750 ohm), C19 and C20. The blue component
appears to be a modern capacitor added by the previous owner. |
 |
|
Below, T2 (the output
transformer which looks brand new) and T1 (the interstage transformer).
The rear of the headphone jack socket is top left and R16 top
right.
The brown condensers are, left
C15 (0.01uF mica), lower right C17 (100pF), lower centre C14
(500pF). The tubular metal condenser is C13 (0.1uF) and all these
are Type 4703 and without a maker's name. |
 |
|
Still no firm dating evidence...
1936 to 1945 ?? |
|
pending... |
