Memoirs of a Boffin

The new site for our research establishment (TRE) was on St. Alban’s (otherwise known as St Aldhelm’s) Head, a couple of miles west of the town of Swanage in Dorset, on the south coast of England. Sir Frederick Treves (Highways and Byways of Dorset, Macmillan & Co, 1936) wrote:

“After leaving Swanage (to the west), the coast is drab and savage, the cliffs erect or scooped out in places by quarries. There is no beach. The rock rampart, cracked by sinister rents, rises from the sea above evil-looking ledges and hollow sea halls. At Dancing Ledge a sheltered bay has been fretted in the cliff. In the place of a beach a flat, acre-wide slab of rock slides into the sea, like the slope in harbours for the hauling up of boats........ Winspit, two miles westward is a lonely, tragic spot. It is here that the valley comes down to the Channel, to a cleft in the cliffs so narrow as to be like a Titanic hatchet cut. In this gap is a beach of boulders wedged in between high walls. On the strand are a boat and a wreck of another...... Inland from Winspit is the queer dun hamlet of Worth Matravers. built wholly of stone but boasting a few trees. The landscape around it is reduced to two elements only – bare grass and trees..... A little way beyond Winspit Gap is the promontory of St. Aldhelm’s Head. Standing boldly out of the sea, it faces the Channel with a bald cliff 350 feet in height. At its foot is a most inhospitable shore of rocks. Upon the summit of the promontory is hardy grass, so scoured by centuries of wind that it has become as smooth as a sand drift......”

It was on this “hardy grass” of St. Alban’s Head that our new facilities had been built. They consisted of a number of rectangular wooden huts, each about 50’ by 20’ in size, and a central building to house services such as the cafeteria.

The Dorset coast was quite a change from Dundee. The nearest village was Worth Matravers. The road from Worth went on down the hill through Langton Matravers until it joined the main road from Wareham, through Corfe Castle to Swanage. I got digs in Langton Matravers just above this junction, opposite Leeson House, which was a girls’ school. My hosts, Mr. and Mrs. Suttle, could not do too much for me. They gave me the front sitting room and the front bedroom for my exclusive use and provided breakfast and dinner seven days a week and lunch whenever I was there for it. All this was provided for the modest sum of one pound five shillings a week. And the food was excellent.

We equipped one of the huts for the IFF and radar beacon work. We had to start our experimental work before the workmen had put on the finishing touches. One of the features of the linoleum-covered work benches that lined the walls was a heavy copper earthing (ground) strip which was ostensibly connected to an earth rod stuck a few feet into the ground. One day I was soldering a circuit. I had a coil of resin-core solder on the bench, with the end reared up rather like a snake responding to the charmer. When I touched it with the tip of the soldering iron a huge arc, fed by the resin, formed between the iron and the earth strip on which the solder was standing. It was spectacular. We traced the strip and found that not only was there no ground rod, but someone had pierced the live side of the 230 volt AC line with a 3” nail, which had also gone through the “floating” earth strip. The grounded soldering iron had caused a major short-circuit.

It was here that F.C. Williams did his pioneer work on circuit development – inventions that permeated the whole of radar circuitry in Britain and, soon afterwards, on both sides of the Atlantic.

In a booklet published in his memory in 1985, I was quoted as writing:

“He was most prolific, enthusiastic and unselfish in his creativity. His sole concern was to see the desired electronic function performed elegantly, efficiently and reliably. His ideas were transmitted during informal, usually intense sessions. He was notorious for his tangled breadboard circuits, which often drooped over the edge of the bench towards the floor – a unique combination of conceptual elegance and material chaos It was thus that we saw the birth of a whole range of feedback timing circuits that brought precision into radar circuitry by the use of inherently linear, instead of exponential timing waveforms. These circuits carried typical Williams names, created on the spur of the moment in the laboratory. They included the Phantastron (fantastic!), the Sanatron (“sanitary” was his favourite description of a well-behaved circuit) and the hybrid Sanaphant. I remember improvements and extensions of these circuits, such as the cathode-coupled phantastron and the sanatron stepping divider. Most of these were made in the Pavilion at Malvern College, which, with F.C. assisted by N.F. Moody, S.W. Noble and F.J.U. Ritson was a Mecca for electronic types everywhere. Advice and help were given with F.C.’s unfailing, even unwitting generosity. Thus, in a very few years, his influence became apparent wherever circuits were designed. His contributions to the “Waveforms” volume of the (MIT) Radiation Laboratory “5-foot shelf” were as crucially important as his earlier contributions to the immediate circuit problems of IFF, ground and airborne radar, G, H, and many other radar and navigation systems.”

The early development of some of the circuits mentioned in this passage actually took place in the hut on St. Alban’s Head, before we moved to Malvern. However the circuit group as such was first given a life of its own, separate from any particular radar projects, in the Pavilion at Malvern College, a couple of years later.

We had a special Chain Home (CH) radar at Worth Matravers, for experimental work and operator training, with a 240’ wooden mast and a 350’ steel mast for the radar antennas. (The operational CH radar stations had four of each, in line). The cathode-ray tube displays and direction-finding goniometers were “manned” by WAAFs. Notable among them was ‘Smoky’ Wills, of the family of the W.D. & H.O. Wills tobacco company, who, like Nellie Jefferson, another pioneer RDF operator, married a TRE scientist. These and other WAAFs on the coastal radar stations did a brilliant job. These skilled operators could recognize an approaching raid, estimate its height and its numbers from the pulsating blips when the rest of us just saw a wavy line on the tube.

It was also during this period that the first experiments were made on a separate band IFF system and on the first radar homing beacons. Mark II IFF, which was now in service, responded directly to the radar signal and sent back an amplified, coded version of it on the same frequency – the radar frequency itself. When many of our own aircraft were doing this it tended to obscure or clutter up the echoes from hostile aircraft. Indeed we adopted the word “clutter” for the mutual interference between a large number of transponders. (I remember developing a theory of these interactions and writing several reports on “clutter”). Moreover, to extend the old Mark II system would have required a separate receiver-transmitter for each type of radar; and as the types proliferated, it would have become impractical for aircraft to carry IFF sets to respond to all of them. Thus the idea of a separate frequency band for IFF was conceived and pursued by F.C. Williams from 1940 onwards. This was to become my major preoccupation for the rest of the war, first under F.C. Williams, then under B.V. Bowden and finally heading the IFF Group myself. Both F.C. Williams and B.V. Bowden had a great and friendly influence on me and their names will appear frequently in this narrative. For now it will suffice to say that F.C. was a recognized genius at original electronic circuit design while Bowden was a brilliant, erudite physicist with a unique personality.

In the early days at Worth Matravers, we did our flight tests from a field in Christchurch which was the first airfield devoted exclusively to TRE work. The Commanding Officer (CO) of RAF Christchurch was Wing Commander ‘Jackie’ Horner. The Commander of our ‘Special Duty Flight’, as it was called then, was Flight Lieutenant Douglas Rayment. Frank (“Griff”) Griffiths, who was a leading member of the Flight writes in his book (“Angel Visits - From Biplane to Jet”, Thomas Harmsworth Publishing, 1986):

“....RAF Christchurch, on arrival, looked more like a country cricket pitch than an aerodrome. The ‘pavilion’ (Duty Pilots’ Hut) was a wooden hut with a verandah and hitching rail rather like a two-gun saloon in the Middle West. Parked round the tiny grass airfield, under the trees (Yes, trees, and aeroplanes didn’t like trees!) were aircraft of various types, Battle, Anson, Hurricane, Blenheim, Fox Moth, Whitley, Spitfire and a lone United States aircraft, a Boston (DB7). The majority had large toasting forks protruding from their wings and noses.”

I did a lot of flying out of Christchurch. I don’t know whether Griff would confirm it, but I swear that he once flew me out of that tiny field in two takeoffs. As he mentions, the field was small and very tight for the type of aircraft they had to handle, especially when the wind was unfavourable. There was no runway. I remember the Blenheim IV thundering down the field. The hedge was coming up at an alarming rate. Griff left it to the last moment and heaved the aircraft off the ground. It jumped the hedge but sank back to the ground for a brief touchdown in the next field before finally taking off. Cross my heart that is the way I remember it; I hope it isn’t all imagination.

During this period, I flew a good deal with the Flight Commander, Flight Lieutenant Rayment. In particular he insisted on piloting the low-flying trials himself. Their purpose was to test the coverage of the IFF system at very low altitudes. When I asked him to fly out 50 miles over the English Channel at a height of 50 feet he remarked that this was pretty dangerous sport. Nevertheless, he was clearly delighted to have an excuse to do it. We made several trips. There were only the two of us in the Blenheim, which was a twin-engined fighter-bomber, with a gun turret sticking up amidships. I occupied the gun turret, which could be moved up and down several feet by hydraulic power. Just forward of the turret, in the floor of the aircraft, was a 3-foot diameter camera hatch, made of heavy glass. For the low-flying trials, Rayment insisted that the hydraulic power to the turret be turned off. He didn’t trust a boffin not to fiddle with the hydraulic turret controls during flight, thus upsetting the balance of the aircraft which, at 50 feet, could have fatal results. There isn’t much time for corrective action at a height of 50 feet.

We were returning home from one flight when the camera hatch blew in with a crash, letting loose a tremendous gale in the fuselage, which entered the perspex turret and shot it upwards at a frightening speed – and me with it. I remember looking wistfully at my parachute pack as I was on the way up. I had placed it correctly in the rack provided for it, but now I thought how bloody silly it was to attach the rack to the main fuselage so that one left the parachute behind on the way out. However the turret stopped with a clang at its full extension, which was very high. It tilted backwards in the slipstream and jammed: and that is how it remained until we landed. Once we slowed down on the runway, it coggled back into an upright position and slid thankfully down to its lowest position. I would never go in a Blenheim turret again without hydraulic power.

It was on this same aircraft, during the same series of flights that Flight Lieutenant Rayment started to nag me over the intercom about moving around and upsetting the balance of the aircraft. I swore to him repeatedly that I was fixed in the turret and had not moved a muscle. He never really believed me, so I suggested he clear up the matter once for all by taking the Blenheim up alone to see if he still had the problem. He did! Only then did he institute a careful examination of the airframe. It didn’t take long. There were so many rivets out of the tail assembly that you could move it by hand a few millimetres or so back and forth. That is the aircraft we had been flying in for days at 50’ above the sea.

The pilots of the Special Duty Flight were remarkable fliers. They flew anything and everything without instruction on the specific type. I remember climbing into a prototype Mosquito aircraft which had been borrowed for a test flight on IFF before it went for operational tests. In order to make polite conversation, I said to the pilot “I’m flying in a Mosquito for the first time.” It was disconcerting, to say the least when the pilot answered “So am I”!  In his book, Griff Griffiths writes (loc cit):

“......we had to fly so many types without dual. A study of the pilot’s notes, a chat with the pilot who flew the aircraft in, if you were lucky, and off you went on your own.”

While Griff survived, accidents took the lives of several pilots of the Test Flight Unit. A Blenheim IV had been lost from Leuchars in 1940 with three or four of the pilots aboard. The aircraft had just been returned to the Station from overhaul. One pilot was to take it up for test but, such was their enthusiasm that two or three others went along for the ride against regulations. Apparently some of the reconditioning had been badly done and the craft went into an uncontrollable spin and crashed. But the loss that hit us hardest was that of Flight Commander Douglas Rayment, who was not only Griff’s Commander, but a close friend. Griff writes (loc cit):

“In July 1941, he was doing calibrations over the Channel........ He saw an aircraft on the sea and said he was going down to investigate. Then he reported that it was a German flying boat marked with red crosses. And that was the last that was ever heard of Flight Lieutenant Douglas Rayment.”

In the meantime IFF Mark III was under way. I had the job of developing the airborne transponder in close consultation with F.C. Williams. We were always working against the clock. Equipment was no use if it was too late. Research, development, production, testing and installation had to be compressed into the least possible time. That is why, as soon as the outline of the idea for IFF Mark III was there, I went up to the Ferranti Ltd research laboratories at their plant in Moston, near Manchester and acted as project officer, design authority, researcher, engineer and everything else concerned with the IFF Mark III transponder – a regular scientific Pooh Bah. We had chosen a frequency band about 30 Mhz wide centred at about 200 Mhz. Every radar, whether on the ground, in the air or at sea had to be equipped with what we called an ‘interrogator’ (transmitter) and a ‘responsor’ (receiver) within this band. The response was then blended on the display tube, as desired, with the direct radar response from the aircraft. But now only the one kind of transponder was needed in the aircraft.

To make the response intermittent, for easy recognition, and to avoid any problems of frequency stability, the airborne transponder was tuned repeatedly through the whole band by means of a cam on the generator shaft. The coding switches were also driven from the generator through a Geneva mechanism. The generator was there to provide the d.c. power for the electronics – all tubes, of course in those days. Preset coding caused the set to miss sweeps in a variety of patterns according to the “code of the day” which could be selected at the control panel in the cockpit. A further, protected switch fired a cylindrical detonator which was mounted in the heart of the IFF transponder. This was for the crew to activate in the event of a crash landing in enemy territory. It effectively destroyed the coding circuits. Occasionally it was activated by mistake, causing all kinds of recriminations.

The importance of the reliability of IFF is illustrated by a story related by A.P. Rowe. He writes:

Each of two pilots had felt uncertainty regarding the identity of the other and were discussing the incident. One said “You were lucky, I was about to open fire.” The other said “You were luckier; I did.”

One Story of Radar, by A.P. Rowe; Cambridge University Press, 1947.)

The IFF Mark III transponder used a superregenerative receiver which, unlike its predecessors in Marks I and II, was automatically stabilized for sensitivity to the incoming signal, thus making the equipment extremely reliable. The use of the superregenerative receiver in this way was breaking new ground. No-one really understood it fully up to that time and even we had no time to develop theories beyond the point of immediate practicality. However we filed a large number of patents on it during this period, which was perhaps as well because the Americans tried to obtain similar patents after we had handed over the designs to them for US production.

After the war, a friend and colleague at TRE, George Macfarlane and I published a theory of the superregenerative receiver in the linear mode [1] [2]. Soon afterwards I wrote the first (and still, I believe, the only) book devoted entirely to the subject [3].

My industrial counterpart during the IFF Mark III design period at Ferranti Ltd was Hubert Wood, the Chief Engineer of the Company. We became great friends. Like my mother, he was from Yorkshire and he had the same qualities of integrity and courage often known as “Yorkshire pluck”. We made a good team and saw the Mark III transponder into mass production (I believe about 85,000 a year in Ferrantis and, of course, many more in the United States, once the design had been handed over) as one of the most successful and reliable of radar ancillaries. It was installed, in one form or another, into all Allied ships and aircraft.

My stint at Ferranti Ltd. spanned the major air raids on Manchester. I was in Blackley, staying with the Davenports (a friend of undergraduate days) during the two intensive attacks in which about 40,000 people were rendered homeless. The incendiaries and the machine-gun attacks on civil defence workers and fire-fighters horrified us. A local cinema was hit, but, from our position under the table or beneath the stairs, we never quite knew at the time what was happening. It was not until we saw the huge casualty lists that were posted in the streets that we realized the extent of the damage.

As a physicist I was interested to discover that Sir James Chadwick’s original home was next door to the Davenports’ in the row of mill houses in Blackley. A great physicist, Chadwick was of course the discoverer of the neutron, for which he had already received the Nobel Prize in 1934. I was to get to know him pretty well in the late ’40s when I was a Member of Gonville and Caius College, Cambridge, during his Mastership. However, at the time of my visits to Ferranti Ltd., early in the war I only knew of him from physics books. Imagine my feelings when I heard his proud Mother gossiping over the back fence with Mrs. Davenport and saying, in a broad Lancashire accent “Aar Jimmie’s dun very well!” And so he had. The Nobel prize is no small achievement – even for a Lancastrian!

Just as the IFF Mark III production was under way at Ferranti Ltd, there arose an urgent need for a navigational aid to help the pilots of fighter and coastal command to navigate directly back to their bases. F.C. Williams had the idea of using a transponder beacon which the AI (Airborne Interception) and ASV (Air to Service Vessel) submarine-hunter radars could identify and use to find their way back to base. I was asked to design and produce about 125 units for this purpose with the utmost despatch. I decided the only way it could be done quickly was to pirate the components from the IFF production line at Ferranti Ltd. I had been told there was an empty factory in Manchester, a ‘shadow’ factory belonging to another electronics company, A.C. Cossor Ltd, which I could use for the job. I went to Manchester to investigate and found the empty building with a Manager and, of all things an AID (Aircraft Inspection Directorate) inspector, though what he was there to inspect, we never knew. His name was Arthur Levin.

I drove over to Ferranti’s and scrounged the parts I thought I would need to build a prototype beacon. The design had formed in my mind on the journey up to Manchester. Coding, to identify each airfield uniquely, was very important, so I first sketched out a coding wheel, much more elaborate than that in Mark III, but still driven by the generator gearbox. While Art Levin went off to get one machined, I set out to design and build the first beacon. We worked for 48 hours continuously, with six meal breaks a day. By the end of that time we had a fully tested production prototype. We made a components list and telexed London for authority to steal the parts from the Ferranti production line. The Department of Defence Production, DDP, was less than cooperative at first and our Headquarters, the Department of Communications Development (DCD) was also involved, in the person of a man called Ball. This enabled me to compose a Telex to DDP which began “Reference mine to you on the 15th, Ball’s to you on the 16th......” It ended the deadlock. I took the parts across to the Cossor factory in a station wagon.

In the meantime Art Levin had almost literally gone out on the streets to put together a team of young women for the assembly line. In spite of my persuasion, Ferranti had only been able to supply the gearbox parts, not the assembled unit, so here we had workers who had never seen a gearbox or an electronic circuit being taught not only to solder but to assemble complex mechanical parts. Nevertheless, in 30 days after my arrival, we had the first 125 radar homing transponder beacons ever made. Again we worked for 48 hours continuously to test and adjust every one of them against specification. I drove them down to Swanage, where two of my excellent 19-year-old technicians, Eric Kirk and J.B. Wray had been occupied building 125 antennas for them in the meantime. Wray and Kirk were sent off with a station wagon each with instructions to install them at Coastal and Fighter stations in the UK. They did it in an incredibly short time. These beacons successfully brought home our aircraft for 2½ critical years of the war, before they were replaced.

The laboratory site on St. Alban’s Head, near Swanage was very attractive and we enjoyed our two years there. Swanage is a seaside resort and, in spite of the military precautions against invasion landings, it was a remarkably pleasant place to live in.

We could walk along the cliffs or down to the “Square and Compass” – the local pub – for a ploughman’s lunch of bread, cheese and pickled onions and a mild and bitter.

As the establishment grew, it took over a private girls’ school, Leeson House, which was conveniently located directly opposite my digs below Langton Matravers, near the main Corfe-Swanage road. The IFF Group, now headed by Dr. B.V. Bowden (and I as his deputy) moved there and occupied the lodge at the main gates, from which there was a long uphill driveway to Leeson House itself. The girls’ dormitories were converted hastily into laboratories, sometimes with surprising results. One scientist found a hand-printed notice on the laboratory wall beside the light switch, which read “If you need a mistress during the night please ring the bell.” Even though TRE was a remarkable establishment in many ways, pressing that button produced no results.

But we were not destined to stay in the Swanage area, for a number of very good reasons. There was reluctance to expand the establishment on St. Alban’s Head because of its increasingly perceived vulnerability to air attack. During the Battle of Britain we would lie on the grass at lunchtime and watch dogfights overhead. It was inevitable that the Germans would eventually mount direct attacks. It was surprising they did not come sooner or more intensively. First there were sporadic machine-gun attacks by aircraft, usually in the cafeteria area at lunchtime. I can’t remember any casualties on the site but I recall a scientist who was eating a sandwich lunch in his laboratory being frightened practically to death when a bullet entered a metal stationery cabinet and spent the longest time rattling around in it before the racket subsided. Then there came machine-gun strafing of the main street of Swanage, usually early on Monday mornings. One of our scientists and his wife who lived above a store on the main street were killed while still in bed by bullets through the roof or window.

The timing was not accidental. These attacks were timed to catch Cabinet Ministers, Sir Henry Tizard, Commanders in Chief of the Armed Forces, Generals, Admirals and Air Marshals as they went to catch the train from Swanage to London after a weekend in Swanage. The attraction of Swanage was twofold. It was a resort in its own right, but there were many other resorts nearer to London where they might have gone. However, the real reason lay in the “Sunday Soviets”, which were held on Sunday mornings in the office of A.P. Rowe, the Chief Superintendent of TRE. The Sunday Soviets were a remarkable phenomenon, brilliantly conceived by Rowe, which probably had more influence on the outcome of the war than any other single military or civilian activity.

Rowe had realized at the outset that in order to give the military the tools they needed to win the war, the scientists of TRE and the strategic and tactical decision-makers of the three armed forces would have to act cooperatively with great speed and flexibility. It was obvious that this could not be achieved through bureaucratic channels. The intervention of bureaucrats as intermediaries between those who had the problem and those who had the power to solve it would always ensure that the solution came much too late. (It still does!) So the Sunday Soviets brought together top military and political figures – the decision-makers – with TRE scientists who could propose technical means of meeting their operational needs in a timely and effective way. There was little formality in these meetings. The Chiefs of Staff spoke openly of their plans and welcomed suggestions from youngsters like us on how to fight the radar war. If a suggestion appealed to the top brass, immediate authority was given, on the spot, to proceed with the work.

By virtue of the Sunday Soviets, we always knew the implications of our work for the conduct of the war, which led directly to the remarkable speed and enthusiasm with which TRE pursued its task during the entire war. These meetings were, of course, graded Top Secret and involved absolute trust on both sides. There is no record that trust was ever betrayed. The operation Red Queen to defeat the hit-and run bombing attacks later in the war, which is described in another chapter, is one that was conceived, conducted and successfully concluded within the context of the Sunday Soviets. Not a single piece of paper, including the formal authorization of the operation, flowed until weeks after it was all over. Lessons on the harnessing of science to national objectives could still be learned today from that experience, but it would need the mutual suspicion that now seems to exist between scientists and operational service personnel to turn once more into mutual trust. In my experience this can only be achieved with a minimum of bureaucracy.

While I was in Manchester completing the development of the Mark III and Mark IIIG IFF transponders, in May, 1942, it was decided to move TRE to Malvern in Worcestershire. I was still deputy head of the IFF group under Vivian Bowden. Concerning the choice of site, A.P. Rowe wrote:

“We made little progress towards the selection of a site until B.V. Bowden, one of the stalwarts of TRE, said he had heard that Malvern College was empty. .....There was no doubt.... that Malvern College was a far better site for TRE than was likely to be found elsewhere. What chiefly put it in a class by itself was its position on the side of the Malvern Hills, with a view over the Vale of Evesham and the Severn valley. There was room enough for laboratories and it was clear that some of the features of the college, such as the assembly hall and the gymnasium, could be put to good, if unusual uses.”

I dashed down from Manchester to Swanage in some consternation when I heard the news. Soon we were packing up equipment at all the sites near Swanage. Sidney Jefferson, who was one of the first scientists I met at Bawdsey, was managing the loading of a constant stream of vans which climbed the hill to Leeson House. He had asked a cockney technician of mine, one Ernie Prior, to stand at the main gate and direct the truck drivers up the hill to ask for Mr. Jefferson at the main building. I don’t know what Jeff had done to Ernie, but Ernie certainly bore a grudge. To every truck driver he said in his disarming cockney accent: “Drive up the hill and ask for Mr. Jefferson. When you see him speak up because he’s very, very deaf.” The result was hilarious; except for poor old Jeff, who was, of course puzzled, then furious when he realized why everyone was shouting at him.

With the packing up of the research establishment for the third time, so ended our stay in Swanage – a place I shall always remember fondly.

The spirit of Swanage continued, in Malvern, until the end of the war, but it had settled down to a steadier pace by then and most of the work was the development of ideas conceived in Swanage, not so much the evolution of new ones. I am surprised and disappointed that the sporadic attempts at the history of radar that have appeared since the books by Rowe and Watson-Watt have glorified the real achievements at Orfordness in the ’30s, but have mostly neglected the immense advances made during the Swanage years. In particular, the IEE conference on the subject in 1986 and subsequent contributions to the IEE Journal have shown remarkably little appreciation of the crucial importance of wartime TRE to the development and application of radar, or, at least, have reflected very little of the spirit of it.