Two R1155s

 

R1155F S/N 80966 (steel)

R1155L S/N 81961 (aluminium)

 

As a kid, I was interested in radio and electronics and in the early 60s, someone gave me copy of 'The Boys Book of Wireless' from the 50s. There were pictures of a microwave oven which occupied an entire 19" rack (the caption was, "A microwave oven cooking a laboratory steak"), a quartz controlled clock (yes, in an entire 19" rack) and modern communications equipment in an aircraft. The comms equipment was an R1155 and T1154 transmitter. I liked the looks of the R1155 and wanted one. The clock and the oven, well...

As it happens, I got a microwave oven and a quartz clock long before an R1155.

In recent years, I've collected two modified examples, which I've worked on, and a third which is presently a wreck.. I've put this web page together for anyone interested in working on an R1155 who wants to get it to a working state. It's my hope that it will be of some use, if only in avoiding the mistakes I made.

 

 

The R1155 - overview

The history of the set has been detailed in various articles and on websites. In summary, the R1155 was developed in the very late 30s for the RAF. It was a vast improvement over the existing equipment and entered service in1940. It was retrofitted to aircraft with previous equipment. It was not developed as a straight comms receiver but was part of an aircraft system intended for both communications and direction finding, and therefore had circuitry and controls for D/F (Direction Finding). Only headphone output was provided, as loudspeakers were not practical in an aircraft with four roaring Merlin engines. The set was powered externally from the aircraft electrical systems. The complete system for heavy bombers included two R1155s (one for the radio operator and one for the navigator) an 1154 transmitter, an intercom system for the crew and three aerial systems

The aluminium R1155 was used in RAF bombers, the Mosquito fighter/bomber, and Coastal Command aircraft. The steel versions were used in Air-Sea rescue vessels, vehicles and ground schools. It continued essentially unmodified through its production life, apart from the addition of various traps to counteract interference from RADAR and MW stations. There were problems with the original tuning knob and Ekco developed an improved epicyclic slow motion drive, which was fitted as a replacement to many existing sets. The BFO valve tended to overheat and the problem was solved by making a hole in the BFO compartment cover to let the heat out. Orders were issued to make these holes in the BFO cover in existing sets in the field.

Approximately 80,000 R1155s were made.

After W.W.II the R1155 was sold off as surplus and found its way into hands of radio amateurs. Mostly, the D/F circuitry was stripped and the sets were heavily modified by the addition of an audio power output - often with an internal speaker, an S-meter, and sometimes an internal mains transformer and rectifier valve. The Radio Constructor mod involved fitting a new front panel. Other mods included newer, lower noise RF valve types, a product detector, noise limiter, bandspread tuning etc.

It is a single conversion superhet comms receiver with:

The KTW61, 62 and 63 are of course, not strictly speaking, pentodes; they are small-signal beam tetrodes.

There is also a D/F circuit which used:

The magic-eye was shared between the comms and D/F circuitry.

AP2548 suggests that the VR99A valves were selected VR99s which allowed a balance to be made with the D/F controls.

The sets were wired with rubber covered wiring. The insulation was of very variable quality from the point of view of tolerating aging. Some sets still have serviceable and safe wiring, but many have wiring with insulation which has perished and falls away at the slightest touch. No doubt, as well as the fact that the quality of the original wiring was variable, the conditions in which a particular set has been stored have a bearing on the state of the insulation. The set also used paper caps. They were of good quality, but as a breed, paper caps do not age well and tend to become leaky. They can fail short circuit and damage other components. Carbon composition resistors tend to increase in value or go open circuit with age. The set is compact and the wiring is cramped and rather untidy, which makes it hard to work on. It gives the impression that in-depth servicing was was not a consideration

The R1155 has a peculiar power requirement in that the 220V HT is floating. The set acts like a potential divider with the valves and screen supplies acting as one part, and a couple of big carbon resistors acting as the other part. The middle of the potential divider is earthed to the chassis, as is one side of the heater supply. The -ve side is used to for the valve grid bias supplies. With a supply voltage of 220V, the +ve HT line is at 184V to190V and the -ve HT line at -36V to -28V. The actual voltages depend on the setting of the RF gain and the particular values of the resistors. If the RF gain is high, the -ve bias on the valve grids is low and the valves conduct, lowering the voltage drop across them.

The R1155 puts the dilemma faced by the restorer into sharp focus. Do you try and return it to original condition, or do you just get it working with whatever parts are easily obtainable? Even if the set has been kept in storage since W.W.II in its transit case and never powered on after it left the factory, it's my bet that the insulation and some of the components could have deteriorated seriously. Restoring a heavily modified set to original condition would tax Barny from 'Mission Impossible'.

The First R1155

 

 

This was given to me, complete with power supply in 1998. It's an R1155F (steel) and was heavily modified in the 60s. It came with AP2548A vol1, documentation on the mods made and several articles on modifying the set from the 50s/60s.

I fixed the worst bodges and replaced components which were clearly defective. The wiring was mainly good. I aligned it, and it's a nice set to use. I never got round to reconnecting the meter which had been fitted as an S-meter. I didn't refurbish the paper caps in aluminium enclosures. The reason was that I was forced to do very little rewiring. If you are going to refurbish the caps in aluminium tubes, then in many cases, you have to do a great deal of dismantling e.g. the caps in the coil box. I didn't want to disturb the coil box if I didn't have to.

The set was modified by the previous owner to use the Radio Constructor audio stage based on a 6G6. Details are given on Maurice Woodhead's site.

The power supply was 250V - a little too high. I built another PSU to give the 220V which the set was intended to operate on. The PSU is solid state and delivers 220V constant under load with no hum; definite overkill.

The Second R1155 - First impressions

 

I bought the set complete with an external PSU in summer 2001. It is an R1155L with the 3.0-1.5MHz band, rather than the less useful 200KHz-75KHz band, and was intended for Coastal Command aircraft; Sunderlands etc. It is aluminium. It was claimed that it worked, and was in use as a workshop radio until sold.

The Second R1155 - Second impressions & second thoughtsons and second thoughts

I realised it would involve a lot of work when I bought it. When I looked at it at home;

No great surprises, but I certainly wasn't going to power it up. Had it been a wise purchase? Almost certainly not, but there we go.

What was I going to do with it? It didn't seem sensible to do the least work to get it working. It wouldn't have been safe. Restoring it to a state fit to go into a Sunderland would have meant tracking down the missing D/F components and finding or making the external D/F equipment. Searching out the missing parts seemed like a huge undertaking. Even the Jones plugs aren't easy to come by. If I'd got all the missing parts, I'd do it.

So, was I going to make it into a useable set, or leave it in boxes with the worthy intention of returning it to as-new condition, but with no realistic prospect of doing so any time soon?

I decided that the way to go was to make the least invasive modification into a comms receiver that was reasonably possible, with no drilling of fresh holes, no making mods that were impossible to reverse - and definitely no S-meter. It was clear that the set needed complete rewiring. The wiring in the coil box was perished and dismantling the coil box had to be done. As this involved a fair amount of work, it made sense to go the extra mile and replace the paper caps which could only be accessed by removing the coil box. The few remaining D/F components were retained in case a proper restoration became a serious prospect in the future, or another set turned up which they could be used to restore fully.

Phase 1 - Dismantling and preparing

The set was photographed and dismantled, apart from the IF cans and the aerial tuning cans for the three lower ranges;

The front panel was very dirty. It was carefully cleaned with a damp cloth; most of the silk screened printing was cleaned away at the same time. Filthy with some lettering, or clean with less lettering was the choice. Some sets have the control legends on metal plates which were bolted to the front panel.

The tuning scale is a common problem with old sets. The ink perishes and becomes no more permanent than the dirt covering it. I didn't risk any sort of washing, but just brushed it over with a soft paint brush which got rid of the surface dust. The epicyclic drive was taken apart, lubricated and reassembled. A screw had worked loose and was causing the problems. It's suffering from slight play, but it's OK. The perspex panel was cleaned with a sponge and came up very nicely. I was lucky because in some sets, the perspex seems to have become opaque over the years.

Some people say that the perspex was originally flat and bulged with heat from the valves, others say it was vacuum-formed with the bulge. My theory is that it was vacuum formed with the bulge and that the bulge became slightly irregular with heat and age. There are definitely signs of shrinking and pulling through the top mounting bolts, but the bulge looks far too consistent for heat alone to have caused it. Maybe some sets were made with a flat perspex panel (some later sets reportedly had a black & white tuning scale). However, this is very much a side alley of R1155 lore.

I got as far as rewiring the BFO unit and first IF trap then became distracted by other matters. It did hit me that a huge amount of work was involved, far more than in dealing with a domestic set, and more than you normally have with a comms receiver. I carefully collected all the parts, put them into boxes and stored them in the airing cupboard. The set stayed there until late last year when Maurice Woodhead mentioned that he had just acquired an R1155 and intended to get it working. I felt a bit guilty about the set in the airing cupboard which I couldn't bring myself to get rid of, and never had the motivation to set right. I determined to do something with it.

The logical way to approach the rebuild was to work from the back of the set to the front. Power supply, audio stage, IF amplifier, BFO, coil box. In my view, this represents an increasing level of difficulty, and the reconstructed sections can be tested and used to test the next section. If the audio section doesn't work properly, what's the point of worrying about the IF amp? When the audio section works, a modulated signal can be fed into the primary of the last IF, and the detector can be checked easily, then you can work forwards.

Phase 2 - Fitting a power connector

I already had an external PSU for an R1155. I used a six-pin Jones plug for the PSU connection, as this was compatible with the other R1155 I have, and fitted it in the the position occupied by the original Jones plugs.

Phase 3 - Deciding on the audio output stage and getting it working

I decided to use the DH63/6V6GT mod published in Short Wave Magazine in early 1994. There were no serious difficulties. The transformer fitted neatly under the tuning cap at the front of the set, and there were already screw holes which happened to fit. The extra components were fitted using a tag board and solder tags mounted on a standoff using holes already drilled in the chassis in manufacture. The filter board and the tagboard on the front panel behind the wavechange knob were removed, the components tested, then the board was rewired and replaced. Definitely, the tagboard is a pain to deal with. The valves in this section use automatic biasing, and the full 220V. I tested it with an audio oscillator and speaker and it worked. The filter worked and the voltages looked right.

With hindsight, a better approach may have been to build an amplifier using an audio power op amp as a headphone substitute and use the redundant transformer from the first set to produce headphone output. Using headphones directly with a 60 year-old piece of mains powered equipment does not strike me as a very good idea.

Blue wire was used for the HT+ line, yellow for the HT- line, thick blue for the heaters, and black for earth. Other colours were red, green, white, grey etc. There was no particular scheme for the second set of colours; they were used as seemed easiest for identification. I used a mixture of solid core and stranded modern PVC wiring as seemed appropriate. Most of the original heater wiring was perfectly good and was left. Modern PVC covered wire doesn't quite look right, because the thickness of the insulation is much less than the original rubber covering. Afterwards, I saw that there was a colour scheme for the wiring in AP2548A. In the sets I've seen, the wiring has been so faded that it's hard to tell what the original colour was.

In these photographs, the audio stages were nearing completion. The O/P transformer was not yet fitted, but the 3rd IF can had been rewired. The chassis-mounting caps have all been refurbished and replaced.

 

Phase 4 - The IF stages

The third IF transformer differs from the others in that it has no top connection and has stranded core wires coming out through holes in the base which is covered with a shield. The wires were perished, and this is an area where bumping the set could easily cause an HT short. There's no option but to remove the transformer, open the cover, and replace the wires. If you look at the circuit diagram, it's fairly easy to decide which is which if you get lost. C19 is inside the can in the sets I've seen, although the documentation suggests it's mounted in the shield.

One of the mounting studs broke off. Originally, they were tightened then sealed with varnish. Sometimes the varnish just breaks, and sometimes it binds. A possibility is to try and dissolve it with a solvent or paintstripper. Another possibility is heating it with a soldering iron before trying it. Anyway, I broke one off. I heated the stump in a gas flame to burn out the varnish and removed the nut, then butt-soldered the stump back with a semi-hard (96% tin,4% silver) solder.

New wires were soldered in place, (all different colours) and it was noted which colour was connected where in the can. There is a D/F connection which is useless unless the D/F circuitry is to be restored.

The cores were originally set in a gooey white putty which has hardened over 60 years. They may be easily moved or they may be stuck fast. Mine were stuck. A couple of drops of 3-in-1 oil was put into each and the set was turned so that the oil would seep around the cores by gravity. They were left to soak overnight. Carefully turning the cores back and forth with a screwdriver freed them. In one case, a section of the core housing broke away and was glued back with superglue. A metal screwdriver which FITS the slot exactly, is fine for loosening the cores. As an aside, it's generally a mistake to use a screwdriver which is not an exact fit in the slot with any screw. A plastic potentiometer shaft was filed into a tool and a small knob was put on the end. When the cores had been extracted, the putty and oil was removed by turning a plug of tissue through the core housing a few times. You don't want oil getting onto the coils. They were very aware of the differing dielectric properties of various waxes in the 30s, and Lord knows where 3-in-1 stands as far as losses are concerned. The oil is also likely to degrade the wax.

The coils were checked for continuity and it was assumed that the mica caps were good. Generally, this is a safe assumption, but they can catch you out. The transformer was roughly aligned using a sig-gen and scope at 560KHz, then refitted. The set was powered up; not a peep from the speaker. A wiring error was found and corrected and the AF tone was produced.

The IF cans are not transformers in this set; they are resonant LC circuits which are capacitively linked. The coils are mounted at right angles to avoid magnetic coupling. The IF frequency is 560KHz as opposed to the more usual 450 to 490KHz.

The decoupling caps are mounted in aluminium cans on the topside of the chassis. Usually, they were wired with PVC wiring (if only all the set had been wired that way), but the paper caps are at best a dubious bet. Because the cans are a part of the character of the set, and are a neat mounting solution, I decided to open them and fit the new MKT1813 400V caps inside, replacing the red and black wiring. Waxed paper caps, as a breed, are prone to deteriorate and leak with age. They can fail short circuit and damage other components. Whether you replace them or not, is a matter for judgement. In some positions, e.g. DC blocking caps to the grids of output valves, any leakage can do enormous damage. In other positions, it doesn't matter that if they have a DC resistance of a MegOhm.

The rest of the IF section was fairly straightforward rewiring. Usually carbon composition resistors go high. Sometimes they can show strange behaviour, in that they have their right value when cold, but can drop enormously or go open circuit when they warm up. The rule I follow is to check the resistors and leave them if they are within 25% of their original value, unless of course, their value appears critical or they would be hard to replace at a later date. Mostly, the resistors in this section were pretty well spot-on, and they are easy to reach, so I left them. Only one or two needed replacing. It pays to monitor the behaviour of old resistors carefully when they are first powered up after a long period of not being used.

The cores in the other two IF cans were freed as before.

One of the resistors (R1) governing HT-ve had been disconnected, so it was assumed that as the RF and mixer valves were not in place, the HT- ve would be about right. It was checked it and was roughly correct.

A separate RF gain control was fitted and its operation was roughly checked. The Function switch (figure-of-eight, AVC etc) was used to switch between manual and automatic gain control.

Manual RF gain was selected and turned to maximum. A 560KHz (30% modulated at 400Hz) signal was fed to the primary of the first IF can. The cores were adjusted to maximise the tone from the speaker.

Phase 5 - The BFO Compartment

The wiring had deteriorated and was replaced. There were no real problems. There was a capacitor in an aluminium housing which was opened and the contents replaced with a modern cap. The resistors were good and in any case, the BFO compartment is reasonably easy to remove and service.

The compartment was refitted. The AGC seemed to work as far as it could be checked at this stage. The AGC diodes are part of the DL63 valve in the BFO compartment.

The BFO frequency is 280KHz. The second harmonic of the fundamental heterodynes with the IF. This was to avoid the oscillator and IF stages 'pulling'. Once the BFO compartment is reinstalled, it is easy to confirm the operation of the BFO. An unmodulated 560KHz signal is applied to the IF amp and adjusting the variable cap so the vanes are about a third interleaved. The core is adjusted until the heterodyne note from the speaker drops in pitch to a slow warble or disappears. In use, adjustments can be made using the variable cap with a screwdriver.

The BFO valve is hard to remove. Unless the compartment is out of the set you are forced to grab the glass and pull and wiggle, with a fair chance of damaging the valve. Once the valve is out, future removal is made easier by placing a loop of thin nylon string so that it passes between the spigot and three or four of the pins. When you have to remove the valve again, you can apply all the force you need by putting a finger through the loop and pulling gently. In cases where octal valves are more accessible, they can be removed by pushing the spigot with a pencil from below, or grasping the base firmly if there's more room.

Phase 6 - The Magic Eye

The magic eye is operated from the AGC line and gives an indication of the strength of the IF signal. It thereby operates as a tuning guide. Magic eye tubes generally fade quickly because of the target being poisoned by small amounts of material emitted from the cathode. In this case, the magic eye was serviceable, if not in its first flush of youth.

The wiring was redone and was tested. The magic eye seemed to work as varying the strength of an IF signal from the sig gen caused the shadow to open and close.

The phrases 'seemed to work' and 'roughly checked' come up a lot. They mean that the sub-circuit was tested and showed some sort of functionality rather than not working at all, however, whether it was capable of working correctly with the rest of the set, was not yet clear.

The disconnected HT-ve to earth resistor (R1 in the Air Ministry circuit diagram) was 1.3K rather than 2K, so it was replaced with a wire-wound 2K which looked 50s vintage. The resistor removed was a thick carbon stick. It hadn't drifted; it was manufactured as a 1.3K and appeared to be itself a replacement.

Phase 7 - The Tuning Heart (Coil Box)

 

This was the hard bit. All of the wiring was crumbling apart from the heater supply.

I wondered whether it was better to try and do a wire-for-wire replacement, or dismantle the coil box completely and rewire it from scratch. I decided that while a wire-for-wire replacement seemed attractive from the point of view of lessening mistakes, there were difficulties in desoldering and resoldering joints in very cramped conditions, which made it harder than complete disassembly. Even with the wafers removed, wire-for-wire rewiring of the coils still wasn't practical. It was better to draw a diagram of the coils and the wiring between them, then remove the existing wiring and work to the diagram, reconciling it with the circuit diagram and component layout diagram. There was also the other - slightly different set - to check against.

The wafer switch was dismantled and the sections were removed. The wiring between the coils was sketched and the cores were freed. In a couple of cases the slots were chewed up and the heads of M3 nylon screws were superglued to the tops. If you try this, glue the head to the chewed up slot end of the core so that the glue has a 'key' and give the glue ample time to harden. The solder connections to the wafer switches were cleared of solder. The coils were checked for continuity with a meter and rewired, checking against the sketches and the circuit diagram from AP2345A. Coloured wiring was used liberally as it made tracing and checking easier. There is no HT-ve in the coil box, so yellow wiring was used for coil connections. I found the diagrams I made more useful than the photographs.

There are two patterns of capacitors in aluminium cans in the set; single 0.1uF caps with a solid screw base and a solder tag set on a pitch-covered disk; single 0.5uF or triple 0.1uF caps in a larger can with a substantial aluminium base. The last type has a hollow threaded shaft on the base through which the wires to the caps pass. The earth connection to the caps is brought out by a black wire and the hot end of the caps is brought by red wires.

The first type of cap:

The wax, foil and paper contents are topped with a plastic disc which rests on a pressed groove. The lips of the container were rolled over the disc and pitch was set over the top. There were two easy ways of approaching opening the can; just to pry open the seal, which is like a Smartie tube bottom closure; go round the edge with a needle so as to cut through the disc. The second option takes some work. It might be possible to put the cap on a lathe and cut a hole in the bottom; I didn't try that option. The capacitor's earth contact looked like nothing more than touch contact. The earth contact for the replacement was wrapped in a ball of kitchen foil and forced into the bottom of the can. Checking with a multimeter confirmed that it was a good contact. The cap was packed in with cling-film and the wire to the solder tag was insulated with sleeving. The disk was replaced and pitch was melted over the top with a gas blow-lamp. This is an operation that has to be carefully done. The cap was checked with the capacitance range on a multimeter and found to be right. I threw the caps on the carpet and rechecked them after. This was to make sure that the contacts could not easily be broken by shock. I didn't really trust this method of making the earth contact, but at least it seemed no worse than the original method, and has caused no problems so far.

The second type of cap:

The topside tubular cans have a threaded aluminium base which was crimped into the can. You can carefully peel back the seal with a thin, strong blade, like a large jeweller's screwdriver, then refitting the nut, pull off the base with pliers. There's a pitch plug then the paper, foil and wax. Pull out the wires with pliers, leaving a hole, then screwing a corkscrew into the hole left by the wires, gain enough purchase on the innards to be able to extract them. It may take a few tries. Wire up the replacement caps with suitable insulation and tie the wires into a loose knot to relieve strain, then reassemble and turn the closure down carefully, using pliers and a screwdriver. When the can is refitted, the pressure of the nut finishes off the closure and if you've been careful, you have to look very hard to tell that the can has been touched.

The most inaccessible resistors were replaced with modern metal-film 2 watt components. In most cases, replacing a resistor in the coil box at a later date would be a nightmare.

The wafer switch sections were checked for bent contacts then rewired and connected to the coils. Usually, it makes sense to have the coil end of the wire in place and left long for later connection to the switch. This is an area where prior thought can lessen the number of connections which have to be soldered by parting and poking through renewed wiring with significant risk of damaging work already done. The rods and spacers supporting the wafers were refitted as the wafers were replaced. This was another occasion when having another set for reference was very useful.

The coil box was refitted. There were three or four fairly easily accessible joints to make to components mounted on the chassis and external leads in the frequency changer and RF valve section. There were more in the RF grid tuning circuit compartment, but there was far more room to work. A bit of bent cardboard is useful for keeping the heat of the iron away from items you don't want to damage.

The tuning capacitor was refitted and reconnected, the set was checked for shorts with a multimeter and powered on.

The set worked a little on ranges 1, 2 and 4, but was dead on the middle ranges, 2A and 3. Checking with a scope showed that the local oscillator did not work on these ranges. Oh dear! Out came the tuning heart and the wiring was examined. Sure enough, there were a couple of wiring errors. They were corrected (the coloured wires were extremely useful in tracking down the wiring faults), and back went the coil box. The set was tried again. The set now worked, after a fashion, on all ranges. Performance on Range 1 was very poor.

The cores for the coils in the coil box are colour coded blue and yellow. The blue ones are for high frequency and they had been replaced in the wrong coils. That corrected, things were better. The IF stage was more carefully aligned and then the local oscillator and RF stages were aligned. No matter how the core of L18 was adjusted, the local oscillator amplitude on Range 1 was low, as checked with a scope, and dropped off to zero towards the low frequency end of the range. It seemed better with the core removed completely. I read up on local oscillator design in The Radio Designer's Handbook, and came to the conclusion that although it was generally well understood, there remained something of a black art about it. The decline in local oscillator amplitude towards the bottom end of the highest frequency range is a common problem. An article on Maurice Woodhead's website stated that L18 was intended to make the local oscillator output level more even across ranges 1 & 2 and that it shouldn't be touched.

Eventually, the poor HF response was traced to the first IF trap being rewired incorrectly. That fixed, things improved considerably. The grid capacitor to the local oscillator was very sensitive to any contact and was replaced. Experiments were made with a few values, but nothing showed a significant improvement over the 200pF value originally used. The new component was more stable. The core in L18 was tried again, but seemed better left out.

The set was now working to the specified sensitivity according to my fairly crude tests, but drifted noticeably when receiving SSB. The local oscillator anode voltage and mixer screen voltage varied depending on the level of RF gain. I decided to use a stabilised supply for these. I used a VR105 stabiliser tube to provide constant voltages, fitted where one of the D/F valves had been. R42 was disconnected from HT+ve by cutting the wire connecting R35 and R42 and bending the ends up. R36 and R37 were desoldered from the valve holder, separated, and the bare leads insulated with sleeving. The stabilised supply was taken to the wafer switch (connection zr12) , and thence to the screen of the frequency changer.

Drift and flutter were far better, but the stabilsed supply from the VR105 was 110V, which is beyond the recommended maximum for the X65 triode section , and it was replaced with a 90V zener in an octal base from a dead valve, (I'll try a VR90 or VR75 when I come across one).

The alignment was adjusted again. SSB reception was notably more stable, although with no proper BFO frequency control, tuning was a bit like safe cracking. When warmed up, the set can keep tuned to the Shannon Volmet (5.505MHz, USB) for hours. It might be worth providing a stabilised supply for the BFO too.

The set still drifts slightly on SSB until it has warmed up properly. I think it's largely down to the BFO drifting until the thermal gradients in the set have settled. The components in the BFO are close to the valve and are warmed by it. In some R1155s the wax on the caps has softened and shifted. In other designs, these components were mounted under the chassis and away from the heat of the valve. The BFO frequency changes as the RF gain control is altered and HT+ve changes. This isn't a big problem. Eventually, I found a shaft coupler and 3mm shaft which fits over the end of the pre-set cap screw fitting and goes through the Het Adj hole without its needing modification; this is a far better way of managing the little tweaks than simply using the tuning knob. It also allows swapping between USB and LSB without using a screwdriver.

The R1155 just wasn't designed for SSB reception and wasn't a no-expense-spared design which happened to come into its own when SSB became popular. While it's nice to make small mods to enhance its SSB capabilities, there's a definite limit as to how far you can go.

Various valves were tried - KTW61, KTW62, KTW63 and 6K7G. These are all roughly equivalent and all worked fairly well. The KTW63s seemed to be the best. It can't be claimed that the valves were rigorously cross-compared. Be careful if you use KTW61Ms - the ones with the conducting paint as a screen. The paint is connected to pin1 which may have been used as a connection point. Metal octal valves have the same problem. Using these types carries the risk of a shock or a short circuit.

The four IF traps were adjusted to exclude 560KHz interference. A 500uV 560KHZ modulated signal was applied to the aerial and the traps were adjusted in turn until the audio output was minimised. It became almost undetectable when the traps were properly set.

Phase 8 - Tidying up

The aerial tuning mod, from an article on Maurice's site, was tried. It wasn't an improvement and was removed.

Bearings and moving parts were lubricated.

Dial lamps were fitted. The dial lamps are really convenient.

A paxolin panel for the aerial and loudspeaker sockets was fitted and loose wiring tidied.

The coil box cover was replaced and the radio was left running for a couple of days before the final careful adjustment to the alignment. I've found that old sets tend to settle down with use, and a period of burning in adds to one's confidence.

The trimming caps were locked with a touch of nail varnish, and the cores with a small piece of candle wax melted with a cool soldering iron. Locking the presets is an interesting question. You want them fixed so that they don't change with vibration, but you don't want them glued in place for ever. I'm not sure about the candle wax, but I haven't found a better way. I've used bits of dental floss and strands of rubber in the past. Both the nail-varnish and the wax are easy to get rid of later if required.

Just sweeping across the Medium Wave (1.5MHz to 600KHz) range, the set fluttered at 1.120MHz - twice the IF frequency. I tuned the R1155 to a station and tuned my domestic radio around 560KHz. I heard the station the R1155 was tuned to clearly. As I swept the dial on the R1155, the output from the speaker was repeated on the domestic set. Wow, I was now operating a pirate Medium Wave station. No surprise that the R1155 was picking up harmonics of its own IF.

Touching the R1155's speaker lead made the sound on the domestic set louder, so this rubbish was coming out through the speaker lead, which was floating. One side of the speaker lead was earthed and the IF radiation stopped, furthermore, the instability at 1120KHz was reduced enormously. I'm still not sure about this. The fluctuations are nowhere near as severe and seem to occur across a wider range, so this may be simply a matter of MW fading being particularly bad around 1100 KHz.

A 0.01uF tone correction cap was soldered across the primary of the output transformer. This cut a lot of the hiss and noise from the set. It also makes life easier for the output valve if the speaker becomes disconnected.

Conclusion

 

     
     

I've worked on several ex-services receivers and commercial comms receivers. Usually, there are things to do like make sure the set is safe, sort out the PSU, replace any bad components and maybe, strip out codges done by previous owners or redo them properly. Possibly, there will be some mechanical work to do to the tuning mechanism - often restringing the cord - and then aligning the set properly according to the manual. I've also worked on scopes and other test equipment. Without doubt, completely rewiring and overhauling this R1155 has been the biggest job. The very poor condition of the wiring caused most of the work. The paper caps all appeared good, but I prefer not to trust them. I've heard that the front panel was welded on after the set was wired. The compactness and untidy wiring makes the R1155 a difficult set to work on.

At it stands, the set uses 85mA at 220V and approximately 3 amps of heater current with the lamps on. The lamps use 0.6 amps.

I thought the 6V6GT was a doubtful option in terms of heat generated, but it seems fine. Maybe the holes left by the missing D/F controls give sufficient ventilation. The set doesn't get overly warm even after few hours use.

The tuning scale isn't perfectly accurate across the range, but isn't bad either. It's within the spec. I'm wondering whether I should have checked the padding caps with a bridge while I had the chance.

I might have tried harder to find wire that was closer to the original in diameter. The problem is that modern insulation is much thinner than the original rubber was, and to get something of the same overall diameter as the rubber you'd have to cope with a very stiff and awkward conductor; at least if you wanted solid core wire. Originally, the sets were wired with a mixture of stranded and solid core.

It would have saved time to reinstall the coil box when the local oscillator was completed, and when that was workiing, turned to the aerial tuining compartment. As it is, when I removed the coil box to fix the local oscillator problem, I also had to desolder the aerial tuning section.

It 'birdies' slightly around 6MHz. The first set also does this. Generally, this is caused by inter-stage feedback caused by inadequate screening or careless wire routing. The seemingly paranoid screening and earth arrangements in these sets weren't done to waste wire and aluminium in a war economy, they were there for a reason. However, as far as I can see I've put all the odd earth wires and screens back. Moving wires and trying extra earthing with croc clips makes no difference. Possibly, it was a defect of the set which was a minor nuisance on a tuning that wouldn't normally be used, and there were other problems to solve.

You have to remember that the R1155 was designed as an airborne set, so there were restraints of size and weight which probably lead to other design restraints.

The American equivalent (the BC348) was better IMHO and certainly, the R variant looks like a class act: 2 RF stages, 3 IF stages, cast aluminum chassis, stabilised supply for the local oscillator, separate local oscillator valve in a ceramic holder, temperature compensated caps in the local oscillator circuit, antenna tuning, variable BFO and crystal gate all built in. It was wired with cotton covered wire which is still fine. On the other hand, the USAAF had a different viewpoint (and budget) and the BC348 didn't have a load of D/F equipment in the same enclosure. The R1155 doesn't seem to have been disliked by its users and although there was a postwar replacement, the R1155 continued in service up to the early 60s. According to documentation I've seen, the BC348 was bought by companies after the war, modified by adding a mains PSU and wiring the heaters in parallel, and was sold as a maritime radio for merchant ships. I haven't heard that there were similar companies reprocessing R1155s. In part, this must have been because the R1155 was very much a dedicated airborne receiver, whereas the BC348 required comparatively little work to adapt it.

Was it worth rebuilding the R1155? It was certainly better entertainment than watching TV.

This wouldn't have been a good first project.

When a suitable candidate turns up, I'd like to do a proper restoration to W.W.II condition.

I'd like to know something about the equivalent German, Japanese, Russian and Italian airborne sets of W.W.II The story I've come across is that the Luftwaffe radios were designed by Zepler, who fled Germany in the late 30s and worked for Marconi on the R1155, so there may be a certain amount of family resemblance there.

 

You may agree that the original wiring was terrible..... And you you may also agree that the new wiring is much better..... But, you could say that people who live in glass houses shouldn't throw stones.

Joking apart, comparing the R1155 with similar equipment from the period with nicely bundled and tied cable runs, it's cramped and messy.

The third R1155. S/N 94467

Aluminium, variant unknown, caps dated 1944.

The third R1155, was bought on impulse while the second was still in bits. Its wiring is surprisingly good, but it's missing the tuning panel and knob and the case. It has an internal mains transformer added. Definitely a challenging restoration.

References

Websites:

http://www.duxfordradiosociety.org/equiphist/r1155/r1155-hist.html

http://www.shlrc.mq.edu.au/~robinson/museum/R1155.html

http://www.vintageradio.me.uk/military/military.htm

http://www.zepler.net/about

These were the most informative and useful websites I found. Try 'R1155' in Google.

 

Hardcopy:

Air Publication 2548A Vol1.  
  Comprehensive R1155 and T1154 cct diagrams, theory of operation, component layout, parts list, installation, operation and field servicing. No factory setup details.
Short Wave Magazine:  
January & February 1994.
'Restoring an 1155'
Parts 2 & 3 of a restoration article. I used the audio circuit. Interesting practical comments. The remarks on the AF gain control are worth noting.
  Part2 page1 Part2 page2 Part3
This three part series of articles 'Restoring An R1155' was originally published by Short Wave Magazine and remains copyright ©PW Publishing Ltd. The material is featured here by kind permission of Short Wave Magazine.

August 1994.
'Wireless Telegraphy Set Type T1154/R1155'

Description of the T1154/R1155 system. Operational history and methods of use. List of the variants .
Practical Wireless:  
September/October 1959.
'An S-meter for the 1155'
Well, if you have a set that has been modified by the addition of a meter, it's easier to get it working than to fill in the hole. Otherwise, don't do it.
Wireless World:  
July 1946.
'Ex-RAF Communications Receiver : Type R1155 Modified for Civilian Use'.
Describes an external PSU and audio output unit. Few modifications to the original set recommended.
The Radio Designer's Handbook:  
4th Edition. No specific 1155 references, but a definitive reference for all aspects of valve technology. Very helpful in understanding why certain circuits were designed the way they were.

Practical Wireless, Radio Constructor and Wireless World published numerous articles on modifications to the R1155 from the late 40s to the 60s. The R1155 was cheap in comparison to commercial comms receivers or most other ex-military sets but some mods were necessary to make it useful as an amateur receiver. See http://www.vintageradio.me.uk/military/military.htm for an interesting selection.

©Peter Tranter 2004