comment on this article

Sweeping all before them

Image: Laurence Whiteley

World War I taught the military that the side with air supremacy held a major advantage over an enemy. It was therefore with unease that the Air Ministry realised, following mock air attacks on Coventry and London in 1935, the UK’s air defences were inadequate to repel a bomber force.

Under the direction of the Air Ministry, a committee was formed to determine – and subsequently to make operational – an effective air defence network. Chaired by HT Tizard, the committee included such figures as Appleton, Watson-Watt and Professor FA Linderman (later to become Lord Cherwell, Churchill’s scientific advisor).

Early detection of aircraft by radio beams was discussed and, although opposed by Linderman – who favoured infrared detection – experiments carried out under the direction of Watson-Watt proved this was feasible.

The Air Ministry made £10million available to erect by 1939 a chain of radio direction finding stations to cover the Thames Estuary and the East Coast. Known as Chain Home (CH) and operating from 20 to 50MHz, this was to be the ground detection arm of a network comprising ground command and control and aircraft fitted with air interception (AI) radar.

Taffy Bowen, one of the few pre war radar pioneers – and who was instrumental in the design and installation of CH – turned his research to AI. In anticipation of the research required, and with war with Germany looming, many young physicists had been included on secret lists. One such person was Bernard Lovell (later of Jodrell Bank fame), who was seconded from Manchester University to work with Bowen on AI.

Working at 200MHz, then the highest frequency at which any appreciable power could be generated, the system was cumbersome, unreliable and hard to interpret. Large Yagi aerials reduced the interceptor’s performance and aerial switching introduced noise on the crt display. Ground returns limited the maximum useable range and height, whilst bearing inaccuracies were large due to the antenna’s wide beamwidth. It was realised that many of these problems could be overcome if centimetre wavelengths could be used.

Whilst centimetric klystron oscillators were available, few milliwatts they produced were of little use for radar. Ultimately, klystrons delivering 400W were developed, but, in the mean time, the magnetron had been perfected. A magnetron is basically a thermionic diode whose electron flow is made to follow a spiral path due to the influence of a magnetic field.

But the magnetron of 1920, with its split anode and external tuned circuit, gave little output and was inherently unstable. In 1940, Randall and Boot constructed a magnetron at Birmingham University in which tuned cavities machined into a solid copper anode block overcame the instability problems and gave an output of several hundred watts on a wavelength of 9cm. Later experiments produced 15kW pulses.

Working from the newly formed Telecommunications Research Establishment (TRE) at Malvern, Bowen and Lovell began work on a centimetric AI radar incorporating a 10cm magnetron. Using a spirally scanned dish antenna with common transmit/receive working, they found a high degree of directivity could be obtained with the pencil beam now produced.

By 1941, radar had played a major role in Britain’s defence. Chain Home had been ready in 1939 and, through its use in the summer of 1940, the RAF defeated the Luftwaffe in the ‘Battle of Britain’, causing Germany to postpone its planned invasion.

Going on the offensive

Attention then turned to how radar could be used in an air offensive against Europe. Prior to 1939, aircraft navigated by dead reckoning, but using charts, chronometer and sextant was highly inaccurate and a radar based bombing aid contained within the aircraft was required.

During tests with centimetric AI radar, it had been noticed that towns and coastlines showed up on the display as identifiable ground returns. During further tests over Swanage, the AI scanner was depressed and this allowed the town to be seen clearly. After experimentation, the helical scanner and range/azimuth display were replaced by a fixed angle paraboloid rotating through 360°, with the familiar plan position indicator (PPI) presentation. This allowed a map like view with a 10 mile radius to be seen from an aircraft at 7000ft.

The question of using a magnetron over enemy territory was still in debate as its loss and subsequent discovery would be catastrophic. Eventually, the superior performance of a magnetron swung the argument in its favour. Initial design and manufacture of experimental equipment was carried out by TRE staff aided by a willing group of RAF technicians, but manufacturing experience was now required to carry the design further.

Lovell, by now in charge of the project, codenamed H2S, enlisted the help of EMI in producing a full system to equip a Halifax bomber. Aided by Alan Blumlein of EMI –who had worked with Lovell on AI – the equipment was installed and its operation optimised. But during a night test flight, Halifax V9977 with the only complete H2S system crashed, killing all on board. Five key members of Lovell’s team – including Blumlein – perished. The next day, one of the few identifiable pieces of the H2S was the magnetron.

Despite this loss, many of the problems associated with the manufacture of a serviceable system had been overcome. H2S Mark 1, operating on a wavelength of 10cm, produced a map like presentation of radius 30 miles on the PPI. Height and wind drift could be determined and the display’s orientation could be controlled by the aircraft’s compass. H2S was now given the highest priority and two squadrons of Pathfinders – Lancasters sent ahead of the main group to mark the target – were equipped with the system by December 1942.

H2S MK1 installation

H2S MK1 consisted of 10 major units weighing some 550lb, with ac power from a generator mounted on the Lancaster’s starboard outer engine. The scanner protruded through the central rear underside of the aircraft, protected by a streamlined opaque perspex blister. All units were mounted in quick release trays in the rear fuselage with the transmitter adjacent to the antenna feed to keep rf losses low.

A universal range of W and Pye connectors was used and pin connections standardised. Only male W chassis connectors were made and it was not uncommon to find lethal voltages on exposed pins – a standard installation used 80 such connectors. All electronic functions, with the exception of the crystal detector in the receiver RF unit, were controlled by thermionic valves. Many, like the Sutton klystron in the local oscillator (LO), the trigatron in the modulator and the rhumbartron and the magnetron in the transmitter, were developed for radar and first saw service in AI. In all, more than 70 thermionic devices were used.

H2S operation

The principle behind radar is that a pulse of rf energy will reflect from objects in its path. If the round trip time is measured, distance can be calculated accurately.

In H2S, an rf pulse is generated in the magnetron by a high voltage square wave from the modulator, itself under the control of a master timing pulse produced in the waveform generator. The rf pulse causes a gas filled rhumbartron in the antenna transmit/receive switch to break down, short circuiting the receiver input and protecting the sensitive mixer crystal. The pulse is then propagated away from the aircraft by the antenna.

A sawtooth voltage from the waveform generator – from which the master timing pulse is derived – is fed to a magslip in the scanner assembly, where sin/cos vector voltages are produced to position the trace on the PPI. After the 1us transmitted pulse, the reflected rf energy mixes in the crystal with the LO and the resultant intermediate frequency is amplified and video processed in the receiver unit. At this point, range and heading marks are superimposed.

Coincident with the transmitter firing, a radial time base is produced on the PPI – its rate of progress being equal to 12.4us per mile across the tube face. Video signals fed to the PPI crt grid modulate the trace on a range and bearing corresponding to that from which the echo has been received. As the transmitter fires and the antenna rotates, echoes produce an illuminated map on the face of the PPI. As the first echoes received will always be from the ground directly below, the aircraft’s height can be determined.

Modifications and additions

Lovell was then asked to develop a radar to counter night fighters, which usually attacked from below: the answer was to use H2S itself. There was a hemisphere under the aircraft – with radius equal to the aircraft’s height – from which no radar returns could be received. Should an aircraft enter this area, it could be detected.

To compensate for this anomaly – which would show as a ‘hole’ in the centre of the PPI – the scan was delayed by an amount equal to the aircraft’s height. A second PPI indicator – Unit 182 – was fitted in the wireless operator’s position. This device, with a maximum range of 5 miles, displayed that part of the picture normally suppressed.

Beacon equipment, in the form of the Lucero transponder, was also added. Working on frequencies from 170 to 240MHz and with a range of 100 miles, this allowed beacon and approach information to be displayed in the height tube.

Other modifications followed, including a roll stabilised antenna, automatic frequency control, provision to interface with a bombsight and, in 1943, a 3cm transmitter/receiver, which improved definition and small target detection.

Air to Surface Vessel radar (ASV) working on 1.5m had been available since 1939, but with the ‘Battle of the Atlantic’ reaching its peak in 1942, H2S was modified to be used on Coastal Command aircraft hunting U-boats. Unlike a bombing attack from up to 20,000ft, an anti U-boat strike was at relatively low level (1000ft). A new antenna – which produced a beam, rather than a vertical fan polar diagram –forced U-boats to use the ‘Snorkel’, which allowed them to run on diesel power whilst submerged. Snorkel was too small to be detected using 10cm radar. However, this advantage was lost in 1943 with the introduction of the 3cm ASV MKVII, which could detect Snorkel in all but the heaviest seas.

The secret discovered

On 2 February 1943, the unthinkable happened: whilst probing the wreckage of a Stirling bomber, Luftwaffe technicians discovered the magnetron. Telefunken scientists were amazed: whilst they knew the Allies were developing new radar aids, they themselves still had no more powerful output device than the klystron. Now, they had the magnetron.

Quickly, a complete H2S was rebuilt and its secrets used to develop Germany’s own centimetric panoramic airborne search radar, the FUG224. Another development was the FUG350 (Naxos). At this time, the Allies were jamming German ground radar with strips of aluminium foil dropped in clouds by approaching aircraft. Now Naxos – a 10cm automatic direction finder with a PPI like crt display – was used to home the fighters onto the Allied bombers using their H2S transmissions. But procedural changes in the use of H2S rendered Naxos ineffective after its discovery in a captured Junkers 88 fighter.


At the end of WWII, all H2S models prior to MkIIIG were declared obsolete. MkIV had been developed in 1944 and, with its true motion display, stabilised antenna and automatic height and slant range error correction, incorporated all that was good and eradicated all that was bad in previous marks. But, with mounting pressure on factories, the need for new installation and training techniques and the invasion of Europe underway, its introduction was postponed. Its operational debut was in Malaya in 1948.

By then, new generations of aircraft were being designed to counter a different threat, and, for the first time, aircraft and radar designers worked together. With the Cold War came the V-Bombers – Vulcan, Valiant and Victor – all designed to accommodate a Navigational and Bombing System (NBS) of which H2S MkIX was a part. H2S was last used operationally in 1982, when Vulcan bombers from Ascension Island attacked Port Stanley airfield in the Falklands.

Mick Green

Comment on this article

This material is protected by MA Business copyright See Terms and Conditions. One-off usage is permitted but bulk copying is not. For multiple copies contact the sales team.

What you think about this article:

Add your comments


Your comments/feedback may be edited prior to publishing. Not all entries will be published.
Please view our Terms and Conditions before leaving a comment.

Related Articles

Future horizons

In a rapidly changing world defence companies have to contend with significant ...

Custom MMIC design

Plextek RFI CEO Liam Devlin discusses the technical and commercial ...

Extreme machines

Few companies have the historical pedigree of Curtiss-Wright, formed in 1929 ...