The
R1116 Receiver
The first picture shows the
receiver as it was delivered here. Compare
with end result... |
 |
 |
This interesting
old receiver belonged to the late George Williams, whose friend
Howard James kindly passed on to me George's collection of radio
bits and pieces together with the R1116. It has a 4-core cable
soldered to the pins of the power supply connector plus a flying
lead which I assume was for its loudspeaker.
This example is quite old, being
the version without the "A" suffix. There are a few
very indistinct markings on the metalwork, possibly ?.R.W. Could
this be Pye Radio Works? Condensers look earler than those used
in typical WW2 radios and the resistors also look earlier and
I spotted one dating from the early 1930s. |
|
This ex-RAF superhet receiver
is pretty rare nowadays and I knew very little about it when
I first saw it. Various sources said it was used in the Fairey
Swordfish and the Short Sunderland and from the valve line-up
I think it must have been designed around 1933/34 and was probably
in production up to WWII. There was a matching transmitter, the
T1115, and both were superseded by the R1155/T1154.
From the type number, and general
construction, it's the same vintage as the old RAF wavemeter,
the W1117
The valve line-up uses old 2-volt
British-based battery-types and these, as can be seen below,
are mostly contained in metal cans.
|
AM CODE |
CV, Post 1942 |
COMMERCIAL |
TYPE |
FUNCTION |
|
V1 |
VU33 |
CV1033 |
- |
Diode |
Transmit Side Tone |
|
V2 |
VR82 |
CV1082 |
220TH |
Triode heptode |
1st Mixer |
|
V3 |
VR83 |
CV1083 |
220VTP, VP21 |
Pentode |
1st IF Amplifier |
|
V4 |
VR82 |
CV1082 |
220TH |
Triode heptode |
2nd Mixer |
|
V5 |
VR83 |
CV1083 |
220VTP, VP21 |
Pentode |
2nd IF Amplifier |
|
V6 |
VR44 |
CV1044 |
210DDT, HD24 |
Double diode triode |
Detector/AVC/Audio amplifier |
|
V7 |
VR21 |
CV1021 |
210LF, L21 |
Triode |
Push Pull Audio Output |
|
Waveband range switching is slightly
odd. There are two tuners each provided with a waveband switch
and tuning dial so that either can be left set to a given transmission.
The user can listen to either of two transmissions by selecting
either LF or HF. The various controls are coloured yellow and
green for easy identification although some of the knobs in my
example are missing their coloured inserts..
The wavebands, calibrated in
Kc/s and Mc/s are as follows:- |
|
Range |
Lower |
Upper |
|
|
|
1 |
142Kc/s |
315Kc/s |
LF |
Yellow |
|
2 |
315Kc/s |
700Kc/s |
LF |
Yellow |
|
3 |
700Kc/s |
1600Kc/s |
LF |
Yellow |
|
4 |
2Mc/s |
4.4Mc/s |
HF |
Green |
|
5 |
4.4Mc/s |
7.3Mc/s |
HF |
Green |
|
6 |
7.3Mc/s |
12.0Mc/s |
HF |
Green |
|
7 |
12.0Mc/s |
20.0Mc/s |
HF |
Green |
|
The outer case is made from aluminium
because the receiver was designed primarily for airborne use.
Further clues to this are the provision of remote volume control
and a DF loop facility.Ranges 1, 2 and 3 are available for DF
working |
|
 |
|
Underneath, the receiver
chassis is made from steel and is nicely compartmentalised for
rigidity and performance.
Two separate aerial tuners are
provided and, together with the two sets of waveband switches,
my guess is that this was done to give rapid switching between
traffic reception and direction finding. Without this feature
the operator would have had some difficulty maintaining contact
with base. As can be seen, if one studies the front panel layout,
there are still a number of controls to set before the two features
can be selected.
You can see from the
simplified circuit diagram that the receiver is a double
superhet with the first IF 1.7Mc/s and the second 100Kc/s.
A quaint touch is the provision
of a grid bias battery near the rear of the chassis. The reason
for this feature is the use of directly heated 2-volt valves.
Circuits for this type of valve cannot readily cater for automatic
bias and consequently, to reduce HT consumption and provide peak
performance, provision has to be made to apply a negative voltage
to several of the stages.
The two groups of reception
bands use colour-coding, particularly reminiscent of the dials
used in the later T1154 transmitter. This was probably designed
to help the operator, who may have been in semi-darkness, although
the layout of the controls is pretty logical.
Scale illumination is provided
by small lamps fitted under the hoods over the tuning dials. |
 |
 |
|
Compare this picture with
that at the top of the page. I tried various cleaners but couldn't
get the front panel looking respectable until I rubbed a light
machine oil on it and it was transformed. |
|
|
 |
|
I puzzled over what was
connected to the two tuning dials until I peered inside. A separate
front-end is supplied for HF and LF reception so the operator
could leave these tuned and switch between the two, but frequency
selection is accomplished by permeability tuning which is relatively
unusual. Above, you can see some of the gear wheels. I couldn't
see what those semi-circular shiny metal pieces are for unless
a couple of screens were fitted to keep dust out of the gears.
Another possibility is some sort of dial lock could be fitted,
however, reading the AP all is almost revealed. A pair of 4-1
slow motion drives can be added to the receiver. The clue is
perhaps in the outer edges of the tuning disks which are machined
rather than smoothly finished. What these look like is anyone's
guess.
Originally the two duplicated
sets of receiver controls were coloured yellow and green, but
only a few remain preserved. Top right has a 25pF variable condenser
wired into the DF loop circuit under a metal shield, but other
pictures, and AP1186 show a second aerial socket (and there are
drillings for an aerial socket under the new control so either
this was an official modification or, much more likely, an amateur
modification perhaps published in a magazine), and lower down
a different knob has been fitted.
Below, the parts list stuck
inside the receiver outer case (not entirely helpful unless you
know AM codes off by heart). Does "G.P." stand for
General Purpose? Many components seem to be mounted in completely
inaccessible postions. |
 |
|
Everything appears to be in
place. Some of the valves are a bit wobbly and a couple have
loose top caps but on the assumption it used to work, and it
seems complete I decided to see if it actually works. The first
task is to replace the power lead which is perished rubber covered
cable. I have umpteen mains leads with 2-pin mains plugs (every
time I bought a computer monitor it came with two power leads
so I built up a collection of a few dozen). Cutting off
both ends revealed a standard brown/blue/green-yellow cable which
is a change as years ago it was a US style lead with confusing
colours.. eg black is live. As HT- and LT- share the chassis
as a common connection I chose brown as 120 volts, blue LT +
and green/yellow as HT- and LT-. I have a home-made mains power
supply and I used that. It gives me a range of HT, LT and grid-bias
voltages. I also removed the audio cable and cleaned it up before
refitting it across the small panel labels T+ and T-. Two other
connections are available but these are used for the T1115 transmitter
microphone because the designers cleverly? used the R1116 audio
valves as a microphone amplifier (involving yet more tricky circuitry).
Below are the resistor and condenser
values. |
 |
At first I tried a loudspeaker,
but reverted to headphones because nothing much could be heard.
It seems there are lots of problems (hopefully minor), coupled
with complete ignorance of its operation, but I've listed things
as I noticed them... with comments which were covered in the
restoration work.
|
The volume is working
at maximum in non-AVC mode (fixed) |
|
The volume control rotates
end to end but does precisely nothing non-AVC mode (fixed). |
|
The control marked AE
RES doesn't do anything (it works in D/F mode). |
|
All the rotary switches
are intermittent (switch cleaner worked OK). |
|
On the LF ranges the D/F
position gives better audio than Traffic (fixed) . |
|
The LF aerial tuner seems
to have a bent plate and shorts at two-thirds clockwise setting
(fixed). |
|
A loudish hum was present.
This improved a lot when I added extra smoothing (OK on good
PSU). |
|
In CW the result was a
hummy rumble superimposed on stations (fixed). |
|
In R/T-AVC stations were clear but
without AVC stations were uncontrollably overloading the receiver
(fixed). |
|
Other findings... |
|
The aerial tuning controls
are very critical and in some HF ranges can tune the image (design
failing). |
|
Apart from the uncontrollable
volume control everything appears to work.. after a fashion (fixed). |
|
All the valves are serviceable
with filaments at 2.0 volts (glass and tops super-glued). |
|
The added upper right
trimmer seems to be an antenna trimmer (later modification). |
|
Everything works at 90
volts. I tried 120 volts but the output was deafening (fixed). |
|
Dial readings seem quite
accurate eg Irish Radio One tuned at 250Kc/s and Radio 4, 200Kc/s. |
|
|
|
I read the description and checked
the circuit diagram. My guess is the volume is uncontrollable
because there's a faulty automatic switch for disconnecting the
internal volume control when the external volume control plug
is in place. That switch is almost certainly in poor condition
as are all the other switches in the receiver. Yaxley switches
and most other types are self-cleaning when they're used, but
of course without plugging in the external control the volume
cntrol switch has just gone high resistance. Fortunately, I bought
a new can of switch cleaner recently so that should sort out
most of the faults. Now that the switches are getting less unreliable
I can see some other problems more clearly. For example... if
I switch to R/T (AVC) the volume control works, albeit a trifle
crackly, but after a few score wiggles it should quieten down.
There's no chance of using switch cleaner everywhere because
the wavechange switch is in a sealed compartment from which the
lid is unremovable due to complete inaccessibility of an invisible
screw.
Switching to CW mode produces
a loud hum that reminds me of something shorting out and this
may be the case as when turning back to R/T it takes about 10
to 20 seconds for audio to recover. Previously turning to CW
behaved differently but since then I reduced the HT voltage from
120 down to 90 to stop overloading due to the inoperative volume
control. Looking at the circuit diagram one candidate is C47
which is connected across the BFO valve filament and if this
was killing the LT it may take a little time for that to recover?
V8 might also be duff? (Later I found the cause was that the
BFO was miles off frequency).
Another puzzle is that I get
better results from Radio 4 with the bandswitch on D/F rather
than traffic.
As I delved into the faults
I realised I was looking mainly at the effects of dirty switch
contacts. There are so many involved in each receiver function,
it needs only one high resistance contact set to prevent an important
feature from being activated. This became apparent when suddenly
the volume control started working on Range 3 CW Traffic. However
this was not the case if AVC was off..When the receiver is not
employing AVC it uses the -10.5 volt bias supply to provide a
cut-off source for the RF/IF valves (=a manual voltage level
for AVC) as shown below. Switches S8-A and lower S7 select the
wiper of R16 pot which is fed by the 10.5 volt (-ve BIAS) via
S8-C and upper S7. R16 provides the cut-off voltage with respect
to ground via R21 (R21 sets the minimum bias level or maximum
volume). In AVC mode the volume control R16 wiper is grounded
and the live end of R16 connected to the audio output (junction
of R22/R31). R22 appears to provide a rough and ready balance
of audio, by shunting a proportion of audio to ground, to minimise
the change in volume as S8 is moved from AVC to manual RF gain
control. Because R22 is 50Kohm it's likely to have drifted high
in value so switching to non-AVC mode will result in an increase
in volume, however this method of equalising audio is a bit hit
and miss anyway. R22 measured as around 70Kohm which doesn't
help in keeping the operators hearing in good shape. |
 |
|
During testing I was using grid
bias supplies from my home-brew power supply but I decided to
make a grid bias battery to avoid unnecessary hum. These
batteries were used in many early receivers with the express
purpose of significantly reducing HT battery drain at essentially
only the cost of a grid bias battery (around a shilling or 5p
in modern money). Current drain from these batteries was insignificant
and they usually lasted the life of a radio, or until the contents
just fell apart from corrosion. As I have dozens of used AA cells
with around 1.4 volts terminal voltage I used a set of seven
because the R1116 needs, a not very critical, minus 10.5 volts
for volume control, minus 6 volts for the audio amplifier and
minus 1.5 volts for the output stage. To accept the R1116 plugs
I drilled holes in the top of a set of M5 screws and used these
with solder tags for battery connections. The new battery fits
the holder which I measured as taking a battery 5.75" x
2.75" x 0.875" or if you prefer 146mm x 70mm x 22mm.
To complete this I need to print some paper labels. |
|
|
 |
Here's a similar 15 volt grid
bias battery and, from its general details, I made labels. The
length of the 15-volt battery is proportionally longer because
it uses 10 cells compared with 7 for the R1116 battery. The size
of the photo can be adjusted in printer settings to exactly fit
the new battery (shown below), allowing for overlapping of the
ends and the bottom edge. The extra brightness of the image is
necessary to get a decent print and I made two matching sides
with numbering reversed. The paper labels can be stuck to the
thin card used to cover the battery.
Click for the PDF versions. |
|
|
|
 |
Picture of battery with
labels to be added later.
|
|
|
One reason for making the ersatz
grid bias battery was to remove one of the possible causes for
lack of volume control (and absence of audio) in R/T and CW reception.
After switch cleaner had been squirted around I did get volume
control in R/T (AVC) and a little in CW mode but switching to
the centre, R/T position, resulted in no audio or a deafening
blast in the headphones when ether the grid bias + or -10.5V
plug was removed. Comparing information in the AP (Air Publication
AP1186) reveals several errors and, with component accessibility
generally next to impossible, this makes fault finding difficult.
Looking at the circuit kindly drawn by
Pete, G4GJL shows the 10.5 volt supply from the grid bias
battery is connected across the volume control, R16. Turning
this control anti-clockwise in non-AVC modes should place an
increasing cut-off voltage onto the control grids of V2, V3,
V4 and V5 in much the same manner as AVC is applied to these
valves via S8A Position 1. This latter wiring is in place and
works correctly. Oddly the third position of S8A also works but
S8A Position 2 fails to work. Looking at S8 and the adjacent
tagstrip it looks like some resoldering has been carried out
so the question must be whether at sometime in the history of
the receiver a wiring error was made, otherwise a faulty switch
might be to blame. In the worst case a new switch can be fitted,
but studying the circuit diagram reveals that S8D is only utilised
in Position 3, so swapping the wiring for this with any section
faulty of S8 is a simpler option. The logic behind the switch
wiring is simple but may be confusing so the next step is to
confirm that the wiring is correct. One simple test is to measure
grid bias current into the -10.5 volt terminal when the ON/OFF
switch is turned to ON. I measured 8uA in both R/T and CW positions
so at least the volume control is drawing current through R21.
Below a wiring diagram showing S8. |
 |
|
Now comes a perfect example
of Sod's Law. I'll detach the itinerant 4-pole 3-way switch so
I can either swap it for a good one, or get better access to
rewire the contact sets to avoid the bad contact. The knob has
been replaced in the past and by turning it to one specific point
I can almost insert a screwdiver to slacken it, but first I needed
to remove the AE RES knob which came off easily.. so far so good
except the screwdriver, now aligned properly didn't seem to bite
the grubscrew.. that's because it was broken.. no slot and the
knob was firmly in place. I can now either lever off the knob
(which I then cannot re-use) and detach the switch, or try and
rewire it in-situ, or even work out why A2 is open circuit...
I measured between the 10.5V
grid bias plug (unplugged from the GB battery) and gound and
saw 29Kohm in CW mode. This must mean that the total resistance
of R21 (10Kohm), if it's present, and the pot must be 29Kohm,
although I've already mentioned that components are inaccessible,
and there are differences between the drawings in the AP. One
such difference are connections to R21 so all is not exactly
clear. This resistor is used to limit the maximum gain by limiting
the proximity to ground of the RF and IF AVC line and is shown
as wired direct to R16 in the main circuit diagram and to S8
in both the simplified circuit and in the component wiring diagram.
But where is R21 as it does not appear to be in the location
shown in the wiring diagram ? Not only this, but the grounding
point on the back of the cover over S4 is missing. It was in
place at some time because a QA check mark is visible at the
screw hole, but was either removed when an official modification
was made or by a previous owner. As the duplicate aerial socket
is also missing, being replaced by a trimmer, this also is in
the "mod-by somebody" category. Could R21 have been
removed in order to increase overall gain? |
 |
 |
 |
 |
 |
The 50Kohm volume control
potentiometer
|
|
|
I did manage to fix the
uncontrollable volume in non-AVC modes, but the fault was very
strange although it did fit perfectly with the clues. The resistance
measurement of 29Kohm was correct but wrong. A 50K pot plus 10K
should measure 60K not 29K, but a 17K pot and a 12K resistor
does equal 29Kohm.
I decided to look at the 50K
volume control pot shown above. This is hidden below the AE RES
pot so I detached the two knobs, removed the top pot, then unsoldered
the near end of the 50K pot. It measured 17Kohm so I unsoldered
the wiper and noticed a resistor hidden inside a brown sleeve.
That's R21 and it measured 12Kohm instead of 10Kohm.. hence 17K
+ 12K = 29K. As luck would have it, the resistance of the volume
control is very critical. Because the bias feed is -10.5 volts,
the volume control potentiometer (acting as RF/IF gain) will
be feeding a bias of -4.4 to -10.5 volts to the RF and IF amplifiers.
Because of this range, almost certainly one or more valves will
be cut off hence virtually zero signal will get to the rectifier
hence no audio.
Once it was completely disconnected
and removed the pot still checked out at 17K, but I noticed the
metal cover, marked R16 in ink and 50,000 ohms on the side, must
have been swapped from the original and refitted to the wrong
value replacement... Maybe, I thought, the repairer hadn't realised
the importance of fitting the correct value pot as ordinarily
this doesn't matter too much? However, I detached the cover to
investigate further and found the pot had a solid carbon track
rather than being wirewound, but it was glazed over which somehow
didn't look right. The glaze was over about 80% of the area but
looked slightly patchy, so with nothing to lose, I attacked it
with switch cleaner and a cloth. After several tries the track
started to increase in resistance and, after several attempts,
the glaze had gone and a clean matt finish remained.. and....
it measured 49,700 ohms. I refitted the cover and put back the
two pots but not before changing the old 10K (=12K) to a modern
10K, fitting this inside the original brown sleeve. The 2K-less
ohms will allow more RF/IF gain at the maxim volume control setting.
Switching on the set proved
all was now well with all three modes working correctly. |
|
|
Now that the receiver
is functionally working correctly I moved it to the test bench
for alignment. I found the audio output fails to provide enough
power to drive a standard low impedance speaker but certainly
good enough to use headphones even with the volume control almost
fully anti-clockwise. The first step was to get the BFO working.
All that was necessary was to tweak the trimmer in the BFO can
located under the chassis. This was miles out but the BFO worked
perfectly once the trimmer had been adjusted.
I noticed that because of the
separate aerial tuning circuit, using a long wire resulted in
all sorts of spurious signals... for example amateur SSB was
clearly audible with the receiver dial set to around 350KHz.
At this point I decided to call it a day and move onto something
new, but first some pictures of the R1116 as it is today. |
 |
 |
 |
|
A few comments before
closing... the grid bias battery needs finishing off to make
it look more authentic and, when re-assembling the case, I discovered
there is no way to fix the top cover in place. I hadn't noticed
until now that the fixing points were either broken or missing.
It's possible that these were corroded and had sheared off at
some point in the history of the set? Also the under-chassis
shielding cover is missing (there are no less than nine screw
holes for it). You'll probably notice that I haven't tackled
the missing green and yellow coloured inserts on the various
knobs and two knobs appear to be the wrong type but, as the replacements
are no newer than the receiver and I have nothing better, these
will have to remain (and anyway I already mentioned that the
grub screw in the mode switch was sheared off).
At this point everything works
correctly and, much to my surprise, only one component needed
changing and that, the volume control limiting resistor R21 wasn't
that bad. Incidentally, those shiny semi-circular fittings under
the tuning dials were for additional slow-motion tuning mechanisms
but I have no idea what these looked like. Maybe someone will
solve this puzzle some day? |
