A Touch of Magic

Alan Dower Blumlein...

As a young development Engineer working in the Laboratories of a large industrial research and development unit in the early seventies, I listened avidly to the old hands talking about their early experiences in both peace and war. One had the good fortune to have been at Bletchley Park, during WW2, and would modestly mention names such as Enigma, Turing and the Colossus digital computer project.

Another was at Ferranti in the fifties and had worked on Pegasus, one of Britain's first commercial main frame computers. Others had stories about early television manufacturing with Ekco and some talked at length on the development of radar.

What was interesting, however, was that they all knew a name I had never heard of, a man called Blumlein — Alan Dower Blumlein — and they hardly ever failed to mention this man other than in terms of admiration, deference or endearment. They said he was a genius and added a touch of inventive magic to anything he set his mind to. Even now, they said, people in EMI speak of him in hushed tones!

Being young and with no background information I half suspected they were making more of matters than was justified. After all, if he had been that smart I would have heard of him wouldn't I?

It was to be a good few years before I found out why I hadn't heard of him and that the regard my colleagues showed for Blumlein was wholly justified. For all my skepticism, they had actually understated his brilliance.

Related further reading...

Further Reading:

Barrie Blake-Coleman

A name known to few, but as significant to the progress of technology as any of the "greats"

BLUMLEIN GENIUS

Few will ever match A. D. Blumlein in his insight, breadth of knowledge or ability to solve problems in electronics and electrical engineering. If one defines genius as "original thought" then Blumlein had it in abundance. In a life that was cut short during the Second World War, he created more in a career lasting only 17 years than the vast majority do collectively in a lifetime — one eminent author has said that he was more versatile, prolific and inventive than even Faraday.

Blumlein either originated or made profound contributions to Telephony, Monaural and Stereo Recording, Television, Radar, Amplifier technology, and Signal Processing circuits. In all Blumlein held 128 patents — on average "one for every six weeks of his working life" (Burns 1992). Of this it has been said (Lynch 1993) that "the sheer number of Blumlein's (successes and) patents should have made him famous — but it didn't!"

That he was hardly known by his contemporaries outside his small group, or by those he left behind after his untimely death, is more to do with historical circumstances and bad timing than any judgment of his incredible ability. Add to this a culture which is slow to acknowledge engineering celebrities (and tends to assume that modesty equates to lack of ability) and his lack of recognition is understandable.

Blumlein's unsung standing had nothing to do with the true priorities in history. If Blumlein had lived longer he would undoubtedly have become famous for his achievements — but after he was killed wartime restrictions meant no immediate public recognition. Though the few academic papers he did publish were outstanding, they were sporadic (restricted by his employers) and he had insufficient academic credentials and background to be feted by academic peers.

Similarly, the esoteric nature of his work meant that few understood, let alone admired, his abilities — and at his death there was no one with the immediate opportunity or mandate to collate or champion his successes and accomplishments. As often happens, it took decades before his reputation emerged from anonymity.

BEGINNINGS

Born in London on 29 July 1903 at Netherhall Gardens, Hampstead, London, Alan Dower Blumlein was the son of a naturalised British father, Semmy Joseph Blumlein who had left his home in Alsace (then Germany) and, after training as a mining engineer, had lived in Manchester and South Africa. In South Africa he had met and married Jessie E. Dower, the daughter of the Reverend William Dower, a Scottish missionary.

Semmy had travelled extensively to survey appropriate mineral sites and to find the finance for their mining and exploitation. Semmy and Jessie Blumlein had settled in London before Alan was born.

Alan had the good fortune to find the right schools. The first, Gothic House, was a preparatory school at which he was very happy, and in later years he would re-visit the school. On one such occasion he was to meet his future wife who was teaching there. After Gothic House he attended London House, described as a "progressive" — meaning that it allowed its pupils to do what they found stimulating.

This suited Alan's intellectual development and he progressed well enough to be sent to Ovingdean, a "crammer" near Brighton. He was later offered scholarships at both Aldenham and Highgate and decided on Highgate. Highgate was one of the great London boarding Schools and under the leadership and direction of the Head, Dr. J. A. H. Johnston, it developed an outstanding reputation in science.

One contemporary of Blumlein spoke of him as needing little in the way of instruction — whilst others built model boats, Blumlein built a breathtakingly professional balance cabinet. Whilst others laboured with prep and experimental procedures, Alan was often seen relaxing having completed his work seemingly without effort.

However, though Alan was not particularly happy at Highgate he matriculated with ease and entered the science sixth form without difficulty. He left Highgate in July of 1921 and with a Governors' scholarship entered the electrical engineering course at City and Guilds Engineering College (C&GEC), one of the then constituent colleges of Imperial College.

Blumlein, aged 4, taken at home in Hampstead, 1907

Alan Blumlein as a boy

Highgate School

INTELLECT

Alan's intellect was immediately apparent, a friend (I.L. Turnball) noted that for the first four months he teamed up with Alan to share the engineering exercises but had to drop out of the partnership — he found it simply impossible to keep pace. As at Highgate, Alan hardly needed to examine experimental directions — he tended to finish as others in the class were still reading the notes.

Alan was not, however, the kind to create jealousy, he was "quite popular — mixed pretty well — but did not suffer fools gladly". He swam well and obtained Imperial College colours for swimming.

Alan Dower Blumlein obtained his third year passes and was admitted as ACGI (Associate of the City and Guilds College) before graduating in July 1923 with a first class honours degree (external) in electrical engineering. Immediately following graduation, Blumlein, along with A.R. van Warrington another graduate student, was offered and accepted, a post at C&GEC as assistant demonstrator.

He was insightful enough to choose to work under Professor E. Mallet on telephony and wireless telegraphy, even though his degree was in heavy engineering. Two research students working under Mallett were A.H. Reeves and G.F. Dutton. Reeves eventually joined Western Electric and went on to invent Pulse Code Modulation. Dutton was investigating the electrical reproduction of sound. Both men were involved in research which influenced Blumlein profoundly in later years.

That Blumlein was broader and more comfortable with "light engineering" and electronics had not been lost on his College. With Mallett he devised a novel high frequency resistance measurement and read a paper covering its development at the Institute of Electrical Engineers in January 1925. The paper was awarded a premium. However, Blumlein had reservations about an academic career and laying aside his demonstrator's mantle he subsequently joined International Western Electric, which was to become International Standard Electric Corporation (ISEC) and then (late in 1924) Standard Telephones and Cables (STC).

TRANSMISSION LINES

Long distance telephony was becoming pan-European but it had its problems. Blumlein travelled to various locations to learn at first hand what problems existed in attempting to eliminate the "out of balance problems" in long transmission lines.

Blumlein's vital work on the reduction of cross-talk and interference in telephony and his investigation of unbalanced capacity in transmission lines led to the development of the transformer ratio arm bridge — a device which was to become a crucial piece of test apparatus in many electrical laboratories and field testing in telephony.

In Switzerland he was highly regarded and thought of as quite brilliant. In England the value of his work was immediately recognised and resulted in his employers awarding him a year's salary (£250) for his efforts. This in any era would be valuable, but with the onset of the economic depression of the twenties the award was manna from heaven.

In 1929, after five years with STC and a number of important investigations and patents in telephony, Blumlein found himself invited to talk to Isaac (later Sir Isaac) Schoenberg, then technical director of the British Columbia Graphophone Company.

Columbia, like everyone else at the time involved in sound recording, were subject to licensing agreements with Bell Laboratories. In 1924 Bell Labs had introduced the first all-electric recording system for the gramophone industry and had created a clean sweep of the whole technology.

The Bell technology (the "Western Electric") monophonic recording system was composed of a capacitive (condenser) microphone, linear high gain thermionic amplifiers and an exceptional moving-iron wax (disc) cutting unit with "an excellent (mechanical) damping system" for the cutter. Each licensee paid Bell Labs approximately one old penny (actually between 0-875d and l-25d -approximately 0-36p to 0-52p in new pence) per pressing.

Columbia, resented the royalty imposition and were not prepared to go on paying the fee - it was agreed that they should develop a royalty-free disc cutting and record pressing system of their own. In attempting to extricate his company from the royalty trap, Schoenberg initially invited E.K. Sandemann from ISEC to join Columbia and tackle the development of a novel recording system to bypass the Western Electric patents.

Sandemann declined, but spoke to Blumlein about it. Remembering Dutton's work at Imperial, Blumlein wrote to Schoenberg.

E. A. Nind, then at Columbia, recalls that two applications were actually received but, recollecting Blumlein from his college days, gave him preference and recommended him.

At the interview, Schoenberg was immediately impressed with Blumlein and expressed his satisfaction with the young engineer's potential to resolve the sound recording problem. It is reported (Lodge 1993) that Blumlein, on hearing Schoenberg's comments, retorted that Schoenberg might revise his opinion about his suitability when he heard what salary Blumlein expected — Schoenberg's riposte was to offer substantially more than Blumlein was asking and Blumlein left Schoenberg's office a more than willing recruit.

A H Reeves

Standard Telephones and Cables (STC) - Cable Manufacture

Sir Isaac Schoenberg

Columbia Recording Equipment

SOUND REPRODUCTION

Working "at phenomenal speed" with a small team which included an inspirational mechanical engineer called H.E. Holman, Blumlein set about circumventing the Western Electric Patents. The whole Western Electric system was subject to one patent specification or another, and so they had to address the whole system from scratch — including the wax master disc cutting apparatus and the microphones.

The Western Electric system was reliant on mechanical damping and Blumlein found no way around this. Instead, he decided on electromagnetic damping and began to develop a full moving coil drive which had a very light cutter made from a single turn of aluminium. This in turn lay in the field of a wound coil which was able to deliver enough power without any serious upper frequency losses. The method, in effect, was the same as a moving coil meter with the needle indicator being, in this instance, the cutter head. Initially, the axis of the field coil was horizontal but a vertical axis was eventually demonstrated as more effective. Spurious resonances were carefully designed out — a small shaft resonance at 3kHz being eliminated by replacing steel nuts with ones made from platinum, thus increasing the mass/spring ratio above audible resonance. Blumlein also chose the moving coil principle for his microphone.

In April of 1930 he began work on designing a microphone with a stiff ultra-light diaphragm and incorporated a thin sheet of balsa wood between two very thin aluminium sheets, much like a sandwich. The restoring force of the diaphragm was simply produced by allowing the edge of the aluminium sandwich to overlap and then terminating at the edges by clamping the overlaying aluminium sheets in a two part frame.

The voice coil mass and winding was also carefully managed — in all Blumlein tried over 20 different surrounds and cavities before he was satisfied. Initially, the magnetic field was provided by a mains energised coil but by 1937 the later models of Blumlein's design incorporated recently improved high field strength permanent magnets.

The main resonance of the microphone was at 5.0kHz and emanated from the diaphragm surround and the coil mass. This was set as the standard microphone resonance and could be adjusted by a screw on the diaphragm surround.

The Western Electric Patents prevented the use of an air space between the diaphragm and magnet for damping, so the box dividing the microphone from its amplifier was used. All microphones were adjustable in terms of their base resonant frequency and resonant peak value, so interchangeability was easy.

Subsequently, Blumlein designed all the intermediate audio amplifiers and gain controls, providing not only a highly innovative system but a complete operational manual for end users.

In August of 1930, 22 master discs were cut using both the standard Western Electric system and the new moving coil system. Subjective assessments resulted in the moving coil system being preferred in just over half of the recordings. In December of 1930, tests were continued and a marked improvement was noted in March 1931 with redesigned microphones. The new system, designated HB1B (H for Holman) was just being proved when Columbia merged with His Masters Voice. (HMV).

EMI FORMATION

Columbia merged with His Masters Voice to form Electrical and Musical Industries (EMI) in April 1931. Schoenberg became director of research for EMI and Alan Blumlein was formally transferred to EMI from Columbia on the 1st of November 1931. Just prior to this (September 1931) Blumlein was awarded a premium of £200 in recognition of his outstanding work on the recording system.

HMV also had a formidable R&D team whose main interest (though sometimes unofficial) had been television. To the HMV team of C. O. Browne, J. Hardwicke and W. E. Tedham was added the brilliant Columbia group consisting of Blumlein, P. W. Willans, E. C. Cork, H. E. Holman and others.

Still concerned with his recording research Blumlein went for another redesign of his cutter in early 1932. This unit, denoted as MC4, embodied all that had been learnt to date and overcame the problem of location recording, made difficult by the bulk and power requirements of the earlier marques.

New record and replay amplifiers were designed, so too were high compliance moving magnet pickups, and Blumlein was also able to use the new PX25 audio power triodes just becoming available. He used negative feedback extensively (his undoubted forte) to match and load the new cutter assembly.

PATENTING STEREO

Schoenberg, now acting as Director of Advanced Development (and head of the EMI Patent department) was well able to take Blumlein's developments to their natural conclusion. In December of 1931, after a series of six initial patents on the wax cutter, pickup and microphone, EMI filed Blumlein's "monumental" system specification (No: 394325) on binaural (stereo) recording and playback. The specification, covering almost every detail of two channel stereo recording and reproduction, was intended for the cinema but was equally applicable to home sound reproduction. Blumlein's master patent contained some 70 claims.

He had either experimental evidence, or had pre-calculated, the effect of off-centre sound sources on microphones, phase ambiguities, and integration and difference effects. He catered too for the consequences of employing either velocity or pressure microphones and the justification for using velocity microphones with their axes at right angles.

In January 1932, Blumlein started stereo recording experiments and in March 1932 the first record/replay experiments were carried out in stereo. Further trials continued until August 1933 when a binaural wax cutter was incorporated and eventually produced (December 1933) six 10-inch stereo wax masters of EMI staff walking and talking. Shellac pressings of these masters still survive.

Later in December there was a demonstration given to Schoenberg and other senior staff. A decision to take the apparatus to Abbey Road for studio testing was followed by tests on binaural film (some of this too survives). Then, in June of 1934, the binaural amplifiers were taken to be tested in conjunction with early television systems. Likewise, some effort was directed at designing the sound system for the Alexandra Palace television studio.

This venture into television was in no way disharmonious — even though he continued to file patents on aspects of stereo reproduction up to 1934, by April of 1933 Blumlein was giving much more of his time to the development of television.

Front page of Blumlein's 'Monumental' Patent No. 394325 which contained an incredible 60 Claims (Click to expand image)

A still from a film made by EMI for testing cinema stereo soundtracks showing Blumlein (left) and colleague "walking and talking" as they cross the stage during the making of stereo demo wax masters

BLUMLEIN AND TELEVISION

By 1934, Schoenberg was convinced that electronic television techniques were technically within reach and had persuaded Louis Sterling, MD of EMI, to allocate in excess of two hundred thousand pounds to the Television research budget; no mean feat in such economically depressed times. Schoenberg recruited more people with academic research experience including J. D. McGee, L. F. Broadway, J. Cairns, F. H. Nicoll, J. L. Pawsey and H. G. Lubszynski.

In all, 23 graduate scientists and engineers, nine PhD's, 32 pre-graduate laboratory assistants, 33 instrument and toolmakers, glass blowers and machinists, eight draughtsmen and nine female assistants were recruited. G. E. Condliffe headed the Research Department Labs, though Blumlein had the same status and a roving brief. Blumlein was asked to consider the whole aspect of the development programme and contribute where he thought best. He had the authority to enter any of the research labs and grapple with current problems.

Most of the EMI team were physicists. Blumlein, however, with his telephony and "light current" background was at this time at the cutting edge of innovative circuit design. He was the group's authority on valve and circuit design applications. From him came the inventive application of negative feedback, artificial delay lines, the theory of stabilisation and filter theory.

Blumlein insisted on a predictable performance of his circuits based on circuit analysis — his approach to linearising amplifier stages, using negative feedback, created a culture of "designability" — i.e. seeking to eliminate the need for post assembly test or preset adjustment of circuits. It is significant that his first patent describing negative feedback was published as early as September 1933. The method had only been recently suggested (in 1932) by Nyquist, and Blumlein was four to six months ahead of a complete analysis published by Black in December of 1933.

TV SYSTEM DEVELOPMENT

Blumlein maintained a complete overview of the whole TV system under development. He had long and involved discussions with McGee in an attempt to unravel the complications of the EMI Emitron camera then under development.

Not being a vacuum physicist, Blumlein became involved in a cross-instructional exercise with McGee, who leamt a lot about electronics from Blumlein, whilst Blumlein himself was schooled by McGee in the process and problems of secondary emission in high voltage cathode ray tubes (CRT). Nevertheless, even with their introduction of "cathode potential stabilisation", much more needed to be done to get the electron beam from the cathode to uniformly hit every spot on the photo-mosaic (the CPS Emitron took another 14 years of development before commercial operation — at the televised 1948 Olympics in London).

Blumlein's first patent in television covered an energy-efficient line-frequency scanning circuit for generating a sawtooth waveform to CRT scanning coils. He also devised a linear frame scan circuit which depended on the anode-grid capacitance in the valve. It became known not as the "Blumlein integrator" but the "Miller Integrator" — wrongly, but simply because John Miller had been the first to point out the existence and effect of inter-grid and grid-anode capacitance in 1919.

Though in this particular instance to be out-famed by Miller, Blumlein is still credited with the much more important cathode follower circuit. This brilliant piece of insight used the cathode resistor of a linear valve stage as the source impedance for the next. Using this configuration achieved a voltage source of very low impedance (=l/gm) without any appreciable phase inversion of the signal.

More famous, however, was the Blumlein "long-tail pair", now so ubiquitous in electronics for impedance matching, phase reversion, signal mixing and common-mode rejection. He first devised this "common-cathode" circuit to solve signal problems on the eight-mile twin video cable used to televise the 1937 coronation procession of King" George VI which wound its way from Hyde Park corner to Alexandra Palace. Blumlein patented the long tail pair circuit in July of 1936.

ALEXANDRA PALACE

Then Blumlein was thrown in at the deep end, in being asked to redesign the transmitter stages for the Alexandra Palace TV transmitter. The output stage itself was a 17kW, 45MHz Marconi device with the modulator attempting to deliver 1400V into the output stage grid — in effect, a non-linear resistive load shunting about 300 picofarads up to 2-5MHz. The cathode of the output stage required 30A at 18V, but the supply had to be insulated from earth and have as low a stray capacitance as possible.

His solution was to employ an isolated d.c. generator using a water-cooled CAT6 triode as a cathode follower and a Blumlein "stand off" constant resistance circuit to eliminate the dynamo's capacitive loading. This exercise has been described as "a really heroic bit of engineering".

Aerial design was next, Blumlein had ideas about "reciprocity" in circuits — thus came the "resonant slot aerial", described as "the most important technical advance in the field of antennas in the 1930s".

But it was in the area of general system design for TV that Blumlein gave his greatest perception and ability. One Sunday in November of 1934 Blumlein met with five of his team at his house to plan the circuits for the EMI TV system. Thus began the development process that was to be embodied in his approach to the signal configuration and waveform structure for the EMI broadcasting principle, which was eventually described in Part 1 of Blumlein's 1938 trilogy of IEE papers on the "405-line sequential" TV system.

The best definition possible at the time was determined by the electron-optics of the camera system and the receiver, and the video performance of the video circuits employed. The signal bandwidth could not be extended indefinitely. What was possible was set by the choice of scanning lines — the bandwidth increased with the square of the number of lines but could not go on indefinitely since each scanning line width was governed by the ability of the HT supply and electron optics to resolve a small spot (pixel or "picture element") on the display tube.

Baird's mechanical system had 243 lines interlaced but this was no good for an electronic system because it limited the potential resolution and required simple divider circuits to produce the scan waveform. Initially, each divider stage was indeed set to three, giving 3 exp5=243 as in the mechanical method but this was far too low for a high definition system. 7 exp3 gave 343 lines whilst 7 exp2x32 gave 441. In the end, the limits set were achieved by the simple expedient of having just one odd divider circuit — thus 3 exp4x5=405.

The bandwidth, including sound, video modulation and synchronisation, was well within bounds at some 3MHz. This approach was tried, and gave sufficiently good results with the experimental Emitron cameras for Schoenberg to give the go-ahead. The 405-line standard lasted some 50 years!

RADAR DEVELOPMENT

Despite EMI's vast experience in high frequency transmission, only Marconi and Metropolitan Vickers had received contracts to pursue radar systems by the time war broke out in 1939. It was later said that, though EMI were an obvious choice for involvement, government scientists were under "some kind of restraint" not to talk to them!

Nevertheless, becoming aware of the secret of radar, EMI set up a private venture and by 1939 had developed its own 60MHz radar system. Much of it stemmed from Blumlein's work on sound location of aircraft which he had started early in 1938.

As the war entered its first stages, the system known as AI (Airborne Interception), initially developed by E. G. Brown and his team at Bawdsey, was being evaluated. The Mark III AI apparatus had a minimum range of 330m — not good enough for night fighters which needed to be as close as 130 metres before there was a chance of visual contact.

Because of EMI's demonstration of the 60MHz radar, they were invited to participate in improving the Mark III AI radar. Blumlein designed an entirely novel transmitter modulator which allowed the 130m criteria to be met. By August 1940 an upgraded version of AI, the Mark IV, was being fitted to the new two seater Beaufighter.

However, AI had to be used in conjunction with the Ground Control Interception (GCI) radar but this was subject to severe delay in getting into service. Likewise, crews had to be trained to use the equipment, and so the complete interception system was not brought into operation until late October 1940.

It was immediately effective (first kill November 7) and subsequently had its effect on German night bombing. By April 1941 over 52 enemy aircraft had been shot down, with a further 88 probables. By May it had risen to 102 confirmed, with some 172 severely damaged.

Later it was decided that the AI system should be installed into single-seater fighters — but this posed a problem. The pilot could not fly the aircraft, engage an enemy and operate the AI all at the same time. Usually it was necessary to set a "strobe pulse onto a target echo". This needed to be automated and Blumlein set to work. Again, he imposed a quick and highly effective solution by creating an automatic pulse-tracking strobe generator. The official report attempts a diffident conclusion — that the automatic AI test which took place at Christchurch in December 1940 "appeared to function satisfactorily in principle" — which was nothing less than a begrudging understatement. Blumlein and his co-workers White (EMI) and Williams (Telecommunications Research Establishment — TRE) were fully aware of the value of their work and patented the system immediately afterwards.

RADAR MAPPING

Reconnaissance photographs revealed that at night very few British bombs were actually falling on target. Only a fifth of the total bomb load ever got within a 5-mile radius of the target, and even on a bright night (full moon) things were little better.

What was needed was a radar-based ground map to identify built up areas. In 1937 E. G. Bowen at TRE Bawdsey had lashed short wave aerials to a Hayford bomber and had flown over Harwich detecting significant differences in the pulse echo from large buildings and open tracts. Likewise, A. P. Rowe had reported that the characteristic curves of the River Severn could been seen from the air by ground reflected short wave. As the war moved beyond its initial phases, Lord Cherwell (Prof. F. A. Lindermann), then scientific advisor to Churchill, independently raised the question of ground targets being detected by radar and opened an investigation. Bowen was later to report that the Air Ministry committee formed to debate matters had stated that it could not be done. He replied on the contrary — "Yes it can — I saw it in 1937!"

From 24 October 1941, centimetric wave AI radar was converted by TRE for ground echoes and flight trials were commenced. In November an adapted Blenheim bomber flew from Christchurch. It was clear that built-up areas could be resolved and EMI were contracted in January of 1942 to develop a radar which could project a plan of the ground being overflown.

The name given to the development programme was "H2S" and involved both TRE and EMI personnel. (Originally the system was called "BN" for Blind Navigation but H2S came about (supposedly) because Lord Cherwell applied the chemical formula for the pungent gas hydrogen sulphide, apparently he thought that the timing of the development was "stinking — because it ought to have been done years earlier". Another theory is that as a ground radar it could seek out home location as well as enemy targets, hence "Home Sweet Home").

MAGNETRONS AND STATE SECURITY

Bernard Lovell, later Sir Bernard Lovell, head of the Jodrell Bank Radio Telescope, was then on Blumlein's team and remarked that the trials were not initially successful. Minimum altitude requirements for bombers could not be met due to the operational ceiling of the method — then, to add to the problem, it was decided at ministerial level that the top secret cavity magnetron (central to ultra-short wave high power operation and virtually indestructible) could not be allowed to fall into enemy hands.

To force the use of the magnetron, and thereby have a hope of solving the other problems, it was necessary to prove that a Varian Klystron (already well known to both sides) would not do. As each problem was painstakingly resolved, time was lost.

Nevertheless, matters progressed and on 7 June 1942 Blumlein, never reluctant to engage with the real world trials of a system, joined two of his colleagues from EMI (Browne and Blythen) and three from TRE on board the adapted Halifax bomber V-Victor 9977, now trialing a magnetron driven advanced prototype H2S. The rotating scanning aerial system was housed in a Perspex cupola where the front gun turret had been.

Lovell goes on to say that the pressure on the EMI/TRE group was enormous with "no respite from day to day — either I or other members were flying but on Sunday the 7th of June I was taking a day off — tired of flying. I returned to Malvera and at about 7 o'clock in the evening I had a telephone call from Defford Airfield to say that the Halifax had not returned".

FATAL PLANE CRASH

Unbeknown to anyone on board the aircraft, which had only been recently serviced, the starboard motor had a broken inlet valve. Failure by the servicing crew to tighten the tappet locking nut led to flash back through the valve and in flight the engine caught fire. The general consensus is that the fire rapidly burnt through the main wing spar shedding the wing section and sending the aircraft into an uncontrollable dive, too rapidly for the pilot or anyone else to get out. All five crew and six passengers were killed.

Lovell reported that the "rest of the night was a nightmare — I was driven by the CO through the winding lanes near Ross-on-Wye searching for the wreckage — it was wartime and no time for emotions and our first duty was to search for the highly secret equipment and collect all the bits and pieces".

The effect on the remaining team was "devastating" but there was no let up in the pressure to get H2S operational. Without Blumlein, the pace could have slowed but it did not. Within three days a second Halifax was airborne with a duplicate system. It must be said, however, that the greater part of Blumlein's work was done, as was the system development.

Watson-Watt, the "father" of radar was later to say that Blumlein was the "key man in H2S". A colleague and engineering assistant, E.A. Newman was in no doubt that "the circuitry for H2S was almost entirely due to Blumlein" with contributions by Blytham and White.

Ultimately, the proposed TRE design was abandoned in favour of Blumlein's EMI design. Churchill wrote on the day of the crash that he was extremely pleased at the satisfactory progress of H2S and wanted an urgent programme of production. This was inaugurated swiftly through a supreme effort by all, seemingly regardless of the devastating loss recently suffered. Though intended for bombers and target identification, the system also proved to be hugely successful against U-Boats.

As for the crash of the Halifax, and Blumlein's death, it will forever be something of a mystery — not least because the loss of all the crew and the technical team with not one survivor implies that the aircraft went down very quickly after losing control, and one can only assume at a very low altitude. If not, why was there no time for anyone to bail out? The question of sabotage has always been raised, but there is no evidence other than that which endorses the "engine failure" conclusion of the 1942 crash enquiry.

RECOLLECTIONS

Alan Dower Blumlein was as inventive and indefatigable as Thomas Edison. Indeed, much of his behaviour had similarities to Edison's life style — though one must hasten to point out that Blumlein's mix of engineering rigour and careful experimentation was some distance from Edison's thoroughly empirical approach to problem solving. However, like Edison, Blumlein would engage in relentless and unceasing laboratory sessions and, exhausted, sleep where he was working.

Similarly, he shadowed Edison in that he had no perception of normal working hours and if an idea occurred outside the usual time for attendance (Sunday evenings or very early in the morning) he would simply go back to the laboratory and get on with things. He even had Edison-like traits in his attitude to other personnel — he kept in daily touch with other laboratory teams and would engage in conversation with the most junior of personnel as equals. He was as capable of putting people at ease in terms of their status as he was in reinforcing his seniority when it mattered.

Blumlein was as versatile in private as he was in his professional life. His interests were extremely broad — he swam (London Otter Club), flew (obtained his private pilot's license in October 1928), shot (pistol), played bridge (of which he had a "passion"), enjoyed the theatre (particularly Bernard Shaw), rode, played golf and liked nothing better than to drive very fast.

MARRIAGE AND LEGACY

Blumlein married his wife Doreen on 22 April 1933. She had met him on one of his periodic returns to Gothic House where he had been a happy pupil and she was now a teacher. They had two sons, Simon and David (1936 and 1938). Her abiding love of ADB was in no way diluted by his frequent muttering at night and sudden departures for the laboratory.

It is impossible to find anyone associated with, or in some way connected with Alan Dower Blumlein who is prepared to speak badly of him, or in any way express envy or criticism. The worst comment is perhaps that he had a "rather loud voice". Yet even this has no overtones of complaint or cynicism. Rather, he is remembered for his "modesty, integrity, unbounded patience, generosity, inventiveness and brilliance".

A later contemporary was to refer to him as "a delightful man — very human — and very good at explaining everything. He had "inexhaustible patience — and — didn't get exasperated if you didn't understand — he would go through a point again and again. He had, it was said, "a great facility for converting quite complicated mathematics into very simple circuit elements".

He seemed to enrich everything he embraced. Colleagues at the Central Research Laboratories of Thorn EMI spoke of his inspirational role on everything and everyone he touched, and had he lived there is no doubt he would have gone on to do so much more. It is a great loss and sadness that he was taken so early in his life.

ACKNOWLEDGEMENTS

Sincere thanks to Mr Simon Blumlein for very valuable discussions. Also Mr Reg Willard at Thorn EMI archives for additional photographic material.