Auxiliary Unit radio

[Brian Drury]

Hello – this is my first post.

My area of interest is the British stay-behind force known as the Auxiliary Units during the 1940 – 1945 period. More specifically I am interested in a branch of the AU known as the Special Duties Section which maintained and operated a clandestine radio network during the same period. It is the radios, aerials, batteries and feeder cables that particularly interest me.

Recently I have been in involved in an archaeological excavation of one of the underground hideouts known as an Operational Base that was used as accommodation for one of the AU patrols in the South of England. During this excavation I dug up an HT battery consisting of 80 primary cells that I guess were wired in series to produce ~ 120V to power a radio of some sort. My guess is that this battery was probably used to power a broadcast radio rather than a military communications set.

Today’s project is to learn what I can about this battery in the hope of dating it. If any members can add any information regarding this battery I will be very grateful.

This image is of the top.
This is the side (could be upside down)
This is the base

Brian

[paulsherwin]

If the battery was found without radio equipment, it was probably a spare or a discarded used battery. 120V was a standard HT voltage for battery receivers in the 1930s but there's no knowing what it would have been used with without additional info.

[Cobaltblue]

From the little that's left looks as if it could have been a
Drydex battery similar to this one:
http://www.vintage-technology.info/p.../drydexbat.htm
Have fun Mike T

[Brian Drury]

Brilliant! that's the one, thanks a lot.

So far I have found two of these batteries and there is probably another 2m to go before we reach the floor. This hideout has totally collapsed but I think the batteries were on a shelf with a lot of other stuff which is why they were so high up. It would be nice to find the LT battery and the radio but they will have to wait as we probably have to shift another 30 - 40 Tons material (by hand) before we have much chance of seeing them (if they are there)

The radio transcievers that I mentioned before as used by the Special Duties Section of the AU were known as TRD sets. Does this name mean anything to anyone? The principle designer was Ronn Dabbs I believe.

Brian

[geeoboeh2s]

A few years ago I was a member of a team which set out to recreate a TRD transceiver. The work was done for the Museum of the British Resistance Organisation at Parham, Suffolk.
We succeeded in producing a space model which was delivered to the museum but determining the likely circuitry proved far more difficult. As I am sure you are aware all hardware and drawings were disposed of at the end of the war and all that remained was people's often conflicting recollections.
I would be surprised if any attempt was used to power the TRD from dry batteries as we do know that heavy duty mains valves were used e.g. 6V6, DET19/2C34 etc.

Chris

[Brian Drury]

Thanks Chris that's very interesting. I should make clear that the hideout we are excavating is consistent with the type used by the operations side of the Auxiliary Units and the patrols who manned them were not provided with two way communication but records at TNA show that they were provided with Murphy B81 and B93 receivers. The batteries provided to power these receivers were described as batteries dry HT 120V and batteries secy, port 2V LT (No grid bias battery mentioned). I believe that the batteries I found may have been associated with the Murphy receivers although the date of the batteries may have been in the 1950's.

The TRD transceiver is of course a different thing entirely and it’s extremely interesting to hear that you were involved in the VMARS project to reconstruct this set. I may be treading old ground here, Ken Gibson (president of the Mid Sussex Radio Amateur Society) and I are trying to learn what we can about these sets and the network in which they were used. You may already be in touch with Ken so forgive me if I am duplicating what may have been done already.

John Warwicker’s book provides a block diagram of the TRD – is this based on your groups findings? Some reports suggest that your group succeeded in producing a working TRD design – is this true? Finally is there any truth in the idea that the TRD was based on the single frequency duplex design published in the September 1936 issue of The Wireless Engineer?

Apologies for so many questions.

Brian

[geeoboeh2s]

It must be at least 3 years since we did any work on the TRD but people continue to contact us. I am not aware of Ken Gibson but he may have spoken to one of the others.
I haven't seen John Warwicker's latest book but it highly likely that he used information provided by us as he was the main point of contact.
However a word of caution. We were not able to come up with a definitive block diagram or circuits, only possible solutions, so he may have selected one of a number.
At the time we did the work there was only one known surviving team member, Ken Ward the team leader. His recollection was mainly on the external appearance of the equipment.

The composition of the team is interesting in that it was centred around amateur radio enthusiasts whose forte was construction rather than professional design. According to Ken there was no specific design as obtaining components was always a problem.
However there is no question that the TRD owed a lot to the September 1936Wireless Engineer paper "A Portable Duplex Radio Telephone" by Lewis and Milner. W B Lewis is probaly better remembered for his later work on radar at TRE.

We did manage to replicate the concept although not in the way the TRD would have operated. That would have required special permission from the authorities which was unlikely to have been given due to the very broad rf spectrum.

Personally I believe that should the equipment have ever been used in anger the operator, in fact the complete stay behind cell would soon have been eliminated.
The equipment was meant to provide secure communications but it certainly wasn't covert and it wouldn't have taken long for the enemy to locate the transmission.

I also have doubts about the message security. I built a solid state simulator for the transmitter and tested this with a potential TRD receiver design and a slightly more modern Army R216 VHF equipment.
At low signal strength speech was decipherable on the TRD but not the R216. However as the strength was increased some intelligiblity was recovered on the R216.

It could of course be that none of our potential solutions reproduced the original TRD but that is something we will never know.

Chris

[Peter.N.]

We were still selling these 120 volt batteries in the mid '50s along with a smaller 90v version and 9v grid bias types.

Peter

[Brian Drury]

Everything about the Special Duties Section of the Auxiliary Units has an element of mystery and intrigue to it. The TRD trasciever is of course the most intriguing thing of all.

I have tried to reverse engineer the network links. Starting with the network diagrams from TNA and then confirming the existence of Zero Stations and Outstations by inspecting the evidence on (or under) the ground. Then after analysing sections of the twin feeder used between the TRD and tree mounted aerials I have tried to estimate the losses with the hope of being able to make some attempt at judging how feasible these radio links would have been. There are of course several factors to consider:

1. Receiver sensitivity
2. Operating frequency
3. Transmitted output power
4. Method of modulation
5. Transmitted bandwidth
6. Aerial gain (If any)
7. Direction of radiation relative to receiver
8. Length of feeder between TRD and aerial
9. Elevation
10. Topography of the intervening terrain

Receiver Sensitivity
The receiver appears to be generally accepted as of the super-regenerative type. Ordinarily this would provide a good sensitivity relative to any other method at VHF using a single active component but in this instance because the transmitted signal is believed to have used a frequency modulated sub carrier at some tens of KHz the receiver sensitivity would presumably have been reduced as the bandwidth increased.

I am simply not knowledgeable enough to put a figure on the receiver sensitivity so if anyone would like to speculate please do.

Operating Frequency
The operating frequency is generally reported as being between 50 MHz & 65 MHz. There are two obvious advantages here. First VHF was not generally used for this sort of application – I believe that typically the usual range would be about 1 mile therefore there was a smaller chance of these frequencies being monitored. Secondly the tree mounted horizontal dipole antenna would be of sensible dimensions given the limitation imposed by the tree.

Transmitted Output Power
A figure of 1.5W is often quoted for the transmitter. This may be input power or output power but in any event the transmitter output will not all reach the aerial. The feeder cable has been inspected at several sites and all appears to be of the same unusual construction. See here:

https://www.vintage-radio.net/forum/...1&d=1303918126
This cable was buried in the ground between the underground hideout and the base of the tree in which the aerial was mounted. There is (was – may have gone now) a good deal of this feeder cable buried around the Zero Station at Shipley and I have liberated the very small piece shown above in an attempt to analyse it.

Shipley Cable
2 Core solid copper twisted cable
1 twist per inch
Conductor diameter 0.7mm
Inner insulation diameter 1.6m
Inner insulation colours Black/Grey. Material unknown
Middle insulation diameter 4.3mm colour white. Material possibly polyethylene. The layer encapsulating the insulated wires is surrounded by a further thin layer of what appears to be the same material
Outer insulation diameter 6.5mm colour black. Material probably rubber

Polyethylene would have been an excellent choice of dielectric material as described here:

Wiki

This section is interesting:
But because of World War Two commercial production was halted and secrecy imposed and the new ICI process was used to produce polyethylene for insulation of wires and cables in radar sets.

Assuming that this is indeed what the Shipley cable is made from I have calculated the resulting characteristic impedance using the following:

https://www.vintage-radio.net/forum/...1&d=1303918361

Where
D is the distance between the two conductors
d is the diameter of the conductors
Epsilon is the dielectric constant of the material between the two conductors
The dielectric constant for polyethylene was taken from here:
http://my.execpc.com/~endlr/dielectric_const_.html

Substituting:
D as 1.7mm (1.6mm measured but allowing 0.1mm gap between twisted pair)
d as 0.63mm (Imperial 25mil)

Zo = 75.6Ω

This would have made a good match for the roughly 80Ω impedance of the horizontal split dipole that was located in the aerial tree. It is perfectly possible that empirical measurement would have shown the cable to be nearer to 80Ω.

An obvious question now is what effect did burying this cable in wet soil and then inside the surface of a tree have on the cable losses? Also why does the cable have a twist?

If we consider a 1940’s 75Ω coaxial cable it might have a capacitance rating of perhaps 70pF to 120pF per metre. A quick calculation for our twin feeder cable is probably higher than this maybe 150pF/m.

The coax can be buried in wet soil with no impact on the electrical properties of the cable. Our twin feeder will be affected by the surrounding soil because each wire will see a distributed capacitance along the length of the cable. This capacitance is however is not shunting the RF to ground because our signal is balanced. Problems will arise if the cable becomes unbalanced which might happen if the material that the cable passes through is not homogeneous. Realistically most soils will be anything but homogeneous therefore I believe that the cable designers have put a slow twist in the cable to ensure that any stratification effects in the soil have an even distribution across the two wires of the cable and thereby minimise the loss of balance.

I believe therefore that this cable was very cleverly designed specifically for the purpose for which it was used but there will still be significant losses.

Modulation
Frequency modulation of an ultrasonic subcarrier seems to be the popular theory. For my part I simply don’t know.

Transmitted Bandwidth
This is related to the modulation technique and once again is something I am not sure about.

Aerial Gain
It seems feasible that aerial gain could have been achieved at some outstations although finding the right sort of tree where a reflecting element could be covertly erected would have been a challenge. At the In station or Zero station the geographical distribution of outstations meant that any narrowing of the beam would have been a disadvantage. There is some evidence that multiple aerials were used at some Zero station sites which must have provided even more challenges.

Direction of Beam
For a simple horizontal dipole this is not critical but some of the outstations were likely to have been well off centre and this would have been another factor to consider.

Length of Feeder
At the outstation dugout this could be kept reasonably short but if the Shipley Zero station is considered the feeder is very long maybe 60 ft in the ground and then maybe another 30 ft up the tree.

Elevation
Very important and yet at Shipley the ground elevation is only 30 m.

Topography
Again using Shipley as an example it is reported as being the receiving station for south coast outstations such as the one at Crossbush near Arundel. Crossbush elevation is again only about 30 m and there is about 600ft of chalk between the two. Generally speaking the ground wave operational range for VHF sets would presumably be limited to line-of-sight. If there is a very large hill in the way then communication will either be impossible or very rare.

The Shpiley to crossbush elevation profile:

https://www.vintage-radio.net/forum/...1&d=1303918361


So, the above shows that the network analysis is very complex. When considering the optimum location for a Zero station and the associated outstations the task is difficult enough when using modern mapping software – how it was done in 1940 with maps and rulers is even more of a mystery.

Brian

[neon indicator]

Only twin feeder in air can be linear. Any cable in ground or conduits is twisted to reduce unbalancing and minimise radiation/pickup. Hence Cat3 phone cable (3kHz Audio, 6MHz to 30MHz for DSL/ VDSL ) and Cat5 Network cable is twisted. The amount of outer insulation helps too. Might the very outer have been carbon loaded rubber for additional screening? I don't think there is a huge issue with burying that type of twisted pair with overall 2 layer sheath. I agree ordinary ribbon cable won't work buried.

The loss depends on the circumference AND area of copper as at 65Mhz there are skin effects, and on the dielectric properties of the insulation between the wires, not just the capacitance per meter.

RG58 is about 1.42mm between the inner's outer surface and outer's inner surface.

How it was done in 1940 was same as in 1960s and 1980s. Paper, Sums and Slide Rules. (well in 1980 we had calculators). What is 15mins today certainly was a lot more work then.

If the cable was even as low as 72 Ohm and aerial as high as 82 Ohm, the additional mis-match and loss isn't much. You can actually use 50 Ohm coax on 75 Ohm aerial and vice versa. Any aerial near a reflective object or wet may drop in impedance. It certainly varies. So your calculation may be perfectly accurate.

[trh01uk]

Folks,

just picked up this thread from Brian, with whom I have been in direct contact for some days following his post on here.

I was the leader of the TRD "re-creation" team that Chris Cooper mentions above. I also supplied most of the technical information to John Warwicker for his second book, and you will find my name mentioned a few times in there.

I am not going to attempt an answer to all the points above, however I will provide a few corrections. Firstly, the TRD frequency was given by Ken Ward, the leader of the original TRD design team to me as 65 - 76MHz. It started lower down (on the 48 - 65Mhz band) but following interference problems with the AA using WS17s in that band, the TRD team moved upwards somewhat.

There is a great deal of confusion generated by the later days of the Aux Units Signals. By 1944, their original purpose had virtually gone, i.e. the protection of the country during an invasion by the Germans (it simply wasn't going to happen, and everyone knew it). So what to do with them? One job they were given was to generate a lot of "spoof" traffic using about 200 WS17s, to try and convince the Germans that the D-Day landings were going to happen in the Calais area rather than Normandy.

They couldn't use the original TRDs for this purpose because the output of a TRD was impossible to demodulate using a standard AM receiver. So they had to have another set, and the WS17 was ideal and simple enough for civilians to use. Whether the Germans actually heard any of the traffic I don't know, but the propagation across the Channel should have been OK from coastal Aux Unit stations (and there were lots of them).

The 120V battery that Brian started this post with was probably the HT battery for a WS17.

I have already shown Brian that the Shipley to Crossbush elevation profile that he mentions as a difficult one radiowise, is not necessarily a show-stopper. I calculate a rough path loss of 125dB, and although that is not a very reliable figure, its a fairly modest one. Whether the TRD could cope with that is uncertain, because we don't yet have a working TRD receiver (to measure its sensitivity on the bench), nor do we have a mathematical analysis of how the receiver works. In fact we don't even have a circuit diagram as yet!

I am about to start an email list for those researchers interested in Aux Unit Signals (the technical aspects). If there is anyone on here who is a serious researcher in this area, you will be welcome to contact me direct for more details.


Richard

[CART researcher]

I have read the foregoing posts with great interest. I am a member of CART Coleshill Auxiliary Research Team and we are researching the history of Auxiliary Units - Churchill's Secret Resistance Organisation. My task is to study the GHQ which was at Coleshill House near Highworth. In the woods on the Coleshill Estate are a series of concrete bases for huts. At least three of these were used by Signals personnel. One base is larger than the rest and appears to have minor offices and stores and a large working areas all under one roof. This may the radios were developed and repaired.
Very little is known about the SDS, the Signals side of Aux Units and CART would like to have any information that forum members wish to share. To learn what we have found to date visit www.coleshillhouse.com
All help appreciated - Thank you

[Brian Drury]

Hello CART researcher

I looked at the CART website with interest and came across the following clip -
“In 1942, the Signals HQ was moved from Hundon to Coleshill where construction and improvements to the TRD sets continued including the TRM and TRF.”

This is very interesting and I wondered how you know that construction of the TRD took place at Coleshill? Have you found physical or documentary evidence to support this?

I would be very interested to know how this conclusion was reached.

Brian

[CART researcher]

Brian

'Auxiliary Units' had two branches -Auxiliary Units the Guerrilla resistance force and an even more secret set up, the SDS 'Special Duties Section' the communication/signals side of the organisation. Both groups were formed into small bands who would go to ground if the Germans had invaded. Neither knew of the other.
One of the tasks given to SDS was to develope their own radios. Now to answer your question we know that Captain Ward (Royal Signals) came to Coleshill from Hundon and lead a small team of about a dozen men who continued developing and maintaining the radios from a hut in the woods at Coleshill. We know a few names of men in his team and have some documentation but are looking for more.

[trh01uk]

Originally Posted by CART researcher

One of the tasks given to SDS was to develope their own radios. Now to answer your question we know that Captain Ward (Royal Signals) came to Coleshill from Hundon and lead a small team of about a dozen men who continued developing and maintaining the radios from a hut in the woods at Coleshill. We know a few names of men in his team and have some documentation but are looking for more.

A few years ago I met and interviewed Captain Ken Ward at some length. John Warwicker, who has written two books on Auxiliary Units, also met Ken Ward and got a great deal of information from him.

The trouble is that Ken never mentioned Coleshill to me. If he had gone to Coleshill for anything more than routine meetings, surely he would have told me about it? What he did say was that when the team finished at Hundon, production of the TRD was moved to Whaddon (where SOE were operating).

John Warwicker's books also fail to mention a role for Coleshill in the radio area. Again this is odd if there is real evidence of it.

I have no doubt that some SDS radio activity took place at Coleshill. It was a key location on the radio network for the flow of information collected by the network for one thing. So far hard evidence (eye witness reports, documents written at the time) of the nature of the "radio activity" at Coleshill have been lacking. You now say you have found some. Perhaps you would be kind enough to forward a copy of it to either Brian or myself by PM?

Richard

[Brian Drury]

Originally Posted by CART researcher

Now to answer your question we know that Captain Ward (Royal Signals) came to Coleshill from Hundon and lead a small team of about a dozen men who continued developing and maintaining the radios from a hut in the woods at Coleshill. We know a few names of men in his team and have some documentation but are looking for more.

Thank you for that but I have not yet worked out how you know the following statement on your website is true:

In 1942, the Signals HQ was moved from Hundon to Coleshill where construction and improvements to the TRD sets continued including the TRM and TRF.

The only reference that I know of to support the idea that Coleshill was used to manufacture the TRD appears in David Lampe's book The Last Ditch where he says:

Colonel Major was allocated £2,000 for parts, and production was begun. Partly to save time, partly for security reasons, the sets were built by Royal Corps of Signals men working on an assembly line at Coleshill.

The reason that I am keen to find evidence to support the idea that Coleshill was used as a manufacturing base for the TRD is that you can almost guarantee that some components of the sets will be left around the site and will make an archaeological investigation worth doing.

Presumably you have either found some of these components and or you have documentary evidence to support the idea of manufacture at Coleshill.

I will be very interested to know what evidence you have and if there is anything that I can help you with please let me know.

Brian

[malcolm1651]

Hi Brian,
I was interested in your mention of Murphy B81 and B93 receivers being issued to Aux Units as I haven't come across this before. I would be pleased to hear further details of the TNA reference.
Regards
Malcolm Atkin

[Brian Drury]

Hi Malcolm

The reference to B81 & B93 receivers was seen in the attached TNA document. The photo was taken on my first visit to TNA since when I have improved my technique considerably. The image came from one of the following documents:
WO260/9
WO166/3535
WO244/63

Unfortunately I had not perfected a decent method of recording source references on my first visit so I don’t know exactly which bundle.

The B81 & B93 reference relates to the batteries required to power them.

Brian

[Silicon]

On the third line of the first list (on the previous attached document) it mentions an R308 or R306 receiver.
The photo is a bit blurred so I cant read it clearly.

I would be interested to know which one it is. The R308 is a VHF intercept receiver working on mains or 12v batteries.

[trh01uk]

It's an R308 receiver. (There was no R306). As you say, the R308 was a VHF intercept receiver. Only two are recorded as being requested by the SDS. Presumably they were put on special listening watch for some reason.

Richard