TRE Jan 1945 home page

T. R. E. JOURNAL
ISSUED BY THE
TELECOMMUNICATIONS RESEARCH ESTABLISHMENT
MALVERN
January 1945

HISTORICAL NOTES ON H2S
by A.C.B. Lovell Ph.D.

1. INTRODUCTION
Until the spring of 1942 it is doubtful if our night raids did any worthwhile damage to Germany. In fact, it was an unusual event for a sortie to reach even the target area - an official estimate is that only 10% of night sorties succeeded in so doing. The raids were therefore little more than propagandist efforts on behalf of the Allies and had no measurable deleterious effect on the German people or their war effort. The introduction of the first radar aid to navigation - "GEE" - in the spring of 1942 completely changed the situation, however. On the night of March 8/9th 1942 some 300 GEE fitted aircraft raided Essen, and shortly afterwards followed the destruction by fire of large portions of the Baltic ports of Lubeck and Rostock, and in May the first "1000 Bomber Raid" on Cologne. 80% of night sorties now reached the target area, but strategic bombing was the only direct offensive weapon against Germany in view for the winter of 1942/1943, and Bomber Command faced many vital uncertainties - GEE could be jammed, its range only covered a fraction of the German targets and its positional accuracy was still insufficient to prevent an almost useless spread of bombs when conditions were such as to prevent the final visual identification of the target.

Upon this background the two main pivots of Bomber Command's present day Strategic Bombing - "OBOE" and "H2S" were developed.

2. ORIGINS
Historically the development of H2S will be closely associated with the rapid development of centimetric technique which took place in the second half of 1940 and 1941. The ambition to enable an aircraft to be navigated by a self-contained radar device relying on information from ground and town reflections (as distinct from information from ground stations, e.g. GEE) dates from before the present war, however. In the summer of 1939 the Officer in Charge of the airborne group at Bawdsey Research Station flew in a Blenheim equipped with a modification of Mark I 1½ metre A.I. from Martlesham to the West Coast of Wales and was able to present the pilot with an accurate account of the route flown. Also in the winter of 1939/1940 experiments were carried out from St. Athan on 5/10 metres in an Anson in an effort to detect the presence of towns. No serious attention was given to the subject, however, until the critical situation of our night bombing efforts became known late in 1941. Fortunately, by that time, the remarkable development of the centimetre magnetron had enabled considerable power to be obtained on wavelengths which made a realistic technical solution of the problem more probable.

An experimental 9 cm. A.I. system had flown early in 1941 and by the end of the year the first centimetre programme for the R.A.F. had been well launched (A.I. Mark VII). A competitive experimental 9 cm. A.I. system using a different form of scanning and presentation (helical scan as distinct from the spiral scan of A.I. Mark VII) had not fared so well as an A.I., and in November 1941 when the urgency of town detection really became apparent, this equipment was diverted to the new problem. The aircraft (Blenheim V6000) flew with the centimetre beam tilted downwards, rotating at 300 r.p.m., and isolated responses were immediately observed on the range/ azimuth presentation. These critical experiments carried out from Christchurch aerodrome were of the utmost importance in that they demonstrated that a centimetre airborne apparatus could give discrete returns from certain areas of ground as distinct from the general ground returns which had been the bugbear of all A.I.s. There was one uncertainty, however the responses on the tube were comparatively numerous, and it was evident that many objects other than towns were giving responses (e.g. a landing screen near Salisbury gave a particularly strong response, as did the many military camps on Salisbury Plain). Were the signals obtained in separate flights definitely associated with specific ground objects? Before these crucial experiments could be completed the Secretary of State for Air called a meeting on December 23rd 1941; gave instructions for six specific flights to be made immediately to "determine whether the signals obtained in separate flights could be definitely associated with specific ground objects", and for the initiation of the H2S programme.

The results of these historic flights were published in a TRE Report No. 12/106 dated April 23rd 1942. Its first conclusion was that "The H2S Scanning System offers the likelihood of successful target selection and accurate location with some possibility of selective bombing within the target area". A sample of some tube photographs published with this report and obtained in these early flights is shown in Fig.l.

Figure 1. - Some of the first H2S responses obtained with Blenheim V6000 from 7000' altitude. Note that presentation is range-azimuth.

Some of the politicians were so anxious to improve our bombing that they considered the scanning system too complicated and wanted a simpler system. 'The Secretary of State meeting directed that these experiments be done, and a simple split aerial system was also tried in the early part of 1942. The above report summarised the results by stating "The H2S Split Aerial System is unlikely to provide successful bombing of a specified target unless very high navigational accuracy were available by other methods. Successful selection bombing within the target area is most improbable. It may, however, be definitely asserted that the use of this method would minimise the bombing of open country". By this time, however, the effort already-put into the scanning system had presumed that this would be the H2S system and nothing further was heard of the simple split aerial system.

THE DEVELOPMENT OF H2S
The first active steps to put the scanning H2S system into a heavy bomber were taken after the Secretary of State's meeting at the end of 1941. The Blenheim equipment was designed essentially for A.I. purposes and was, in many respects, unsuitable for H2S purposes as they were then envisaged. In particular the scanner installation gave forward looking only and the presentation was range-azimuth, whereas all round looking and plan position presentation were clearly more desirable for H2S. At that time comparatively little was known about the possibilities of housing scanners in heavy bombers and the "heavy" which appeared to offer most alternative positions for the installation was chosen for the initial work. The first contacts with Messrs. Handley Page were made on January 4th 1942 and the experimental installation, of a perspex cupola in the under-turret position was initiated on two Halifaxes. The first of these, Halifax V9977 (Fig.2) landed at Hurn on March 27th 1942.

Figure 2, The first H2S heavy bomber - Halifax V9977, March 1942.

This Halifax was installed with an experimental equipment built in TRE, using a magnetron TR box of Mark VII AI design. It took off for its first flight in the evening of April 16th 1942, but the equipment refused to work due to a hidden switch in the 80v. supply. The next morning's flight was successful, however, and ranges of 4 or 5 miles were obtained on towns from 8000 ft. altitude.

In the meantime work on the design of an H2S equipment for manufacture had proceeded with vigour and the firms chosen were well ahead with development. Many design alternatives (such as high speed versus low speed scanning) were settled by local discussion, but one outstanding question had deep political repercussions. This was the use of the magnetron over enemy territory. This valve is comparatively indestructible, and it was considered that it would be a grave mistake to present the enemy with information on our advanced technique in the centimetre field. An early decision had, in fact, been given that the magnetron must not be used, but a klystron, the principles of which were believed to be known by the enemy and which was in any case destructible. The klystron had only been developed, however, as a local oscillators and the forms then existing could only produce very small power. Urgent development was undertaken to make a klystron suitable for use as a transmitter, and in a few months klystrons capable of giving 5 to 10 kW peak were produced. Many technical difficulties associated with the use of this valve were overcome, and by July 1942 a prototype equipment existed which had interchangeable magnetron and klystron TR heads.

Before this stage was reached, however, there had been many bitter experiences with the magnetron H2S equipment. The results in the Halifax were very much worse than the Blenheim experiments had led one to expect. This was mainly due to the higher altitude of operation and the consequent redesign and repositioning of the scanner which had been necessary to give good coverage and all round looking. The political pressure was so great that it was extremely difficult to settle down to satisfactory experimentation. Moreover, strange though it may seem at this date, the opponents of H2S almost out-numbered its protagonists. Information available in this country was to the effect that the successful British experience on town detection was not being confirmed in the United States; this, added to the British sceptical element, and the unfavourable progress of our own experiments, was mainly responsible for the A.C.A.S. meeting convened on May 19th 1942 to review the whole question. The Air Staff expressed faith in the development of H2S, however, and issued the following directif:-

• 2. (a) That the system should be accurate enough to guarantee that bombs would fall within an industrial or other area selected as a target.
      (b) That the Air Staff would be satisfied in the first instance if the range of the device enabled the aircraft to home on a built-up area from           15 miles at 15,000 feet.
• 3. Subject to there being no delay or interference with the development of the equipment and its introduction into the Service in a form which will fulfill this aim, it was agreed that details in design to enable it to be used as a navigational aid to determine a specific area or target could be incorporated during the later stages of development and operational trial."

Towards the end of the month, however, the outside pressure was still further increased and on July 3rd 1942 the Prime Minister reviewed the position and ordered an all-out effort to be made to fit two squadrons of heavies with H2S by October 1942. Quite apart from the insuperable production and fitting difficulties, the actual development was in a most parlous condition. It is true that a Halifax was now flying with apparatus which could be regarded as a functional prototype; and the replacement for the ill-fated V9977 was also ready, but even when using the forbidden magnetron as a transmitter the results were very unsatisfactory. It was soon evident that the insistence on the use of the klystron would make it impossible to attain even the minimum Air Staff requirements and on July 15th 1942 the Secretary of State ruled that development work on the klystron for H2S should cease, that the two H2S Squadrons should be equipped with magnetrons, but that a decision as to their use over enemy territory would depend on the war situation at the time in question. Efforts to make a satisfactory destructive device for the magnetron were then redoubled, but the results of the most successful trial were that a 10 ft. hole was blown in the side of a JU88; and an expert was able to reconstruct the magnetron dimensions from its fragments. After that, the destructive device was confined to a small detonator so placed as to prohibit the immediate use of the valve should it fall into enemy hands.

This permission to use the magnetron very considerably eased the development difficulties. A programme was evolved to fit 24 Halifaxes and 24 Stirlings by December 31st 1942, and by September 30th 1942 the prototype Halifax fitted with H2S Mark I proceeded to BDU for service trials. Even at this late hour the unbelievers still outnumbered the believers, and during late September we were forced to do some statistical experiments with representation from the sceptics in order to prove that the responses on the PPI presentation corresponded to objects and areas which one expected to see and that if an observer homed on to the most prominent response on his tube it would turn out to be a "worth-while" objective.

4. THE FIRST WINTER (1942-1943)
At that time we were extremely dissatisfied with the H2S equipment and doubted if it could ever be used satisfactorily in operations. The BDU reactions were therefore awaited with some eagerness. When they came they were not altogether unsatisfactory. They found the serviceability hopelessly bad, but thought that H2S would be "valuable to a high extent both as a navigational aid and as an aid to locating targets". This poor serviceability was primarily due to the breakdown of the pulse transformer in the TR unit and frantic and eventually successful efforts were made to improve it.

Meanwhile fitting and training in operation had proceeded as hard as apparatus would allow, and although by the end of 1942 only 50 of the target of 200 equipments had been made, there were 12 H2S Halifaxes at 35 Squadron and 12 H2S Stirlings at 7 Squadron. After a sustained period of training 7 Stirlings and 6 Halifaxes took off for the first H2S operation on the night of January 31st 1943. Hamburg was the target and in spite of very unfavourable weather, 4 Stirlings and 2 Halifaxes reached the target with their H2S serviceable and successfully marked for the main force. This raid was quickly followed by others, using small numbers of H2S planes for marking the target, and on February 9th Headquarters, Bomber Command issued a memorandum which concluded:- "H2S in its present form fully meets Air Staff requirements and has exceeded expectations in that towns have proved easy to identify both by shape and relative positions. In addition to the exceptional value of H2S for identification and bombing of the target, its great navigational value has been proved beyond all doubt. The recognition of islands, coast-lines, estuaries and lakes has been particularly easy. In fact the problem of accurate navigation under almost any weather conditions is solved by H2S when operated by a trained navigator."

The initial excitement that H2S even worked soon passed, and with the increasing numbers of aircraft fitted and with the consequent greater difficulty of maintaining expert servicing, the faults of the equipment came more and more into prominence. The greatest of these was undoubtedly the unsatisfactory picture provided on the 10 mile or "bombing scale". The maintenance of a satisfactory distribution of energy at steep angles in the elevation polar diagram had still not been solved and the close-in picture was, therefore, a hotch potch of gaps, fades and heavy ground returns. These effects were aggravated because, although H2S was designed on the assumption of a 15,000 ft. ceiling as per the Air Staff requirement, the operational height had risen to 20,000 ft. even before H2S operations began. A series of minor improvements in this picture were effected during the Spring of 1943 by modifications to the dipole feed and a major improvement both in elevation coverage and azimuthal definition resulted with the introduction of the waveguide fed scanner in the summer of 1943.

Meanwhile the main production of H2S, as distinct from the original "crash programme" had commenced in May 1943, and this, which included certain minor display improvement, facilities for beacons, rearrangement of controls etc., was known as H2S Mark II.

5. THE SECOND WINTER (1943-1944)
Although the fitting of H2S had fallen well behind the original schedule, a considerable number of operational equipments existed at the beginning of the second winter. For example, in October 1943 H2S sorties represented one-sixth of Bomber Command's total. The serviceability had also improved considerably. In that month, for example, 16% of H2S sorties were unsuccessful due to failure of equipment compared with an average figure of about 30% up to July. There were two outstanding needs, however:-

1. In this manner X band H2S - known as H2S Mark III entered the Service. A new indicator was introduced in the Spring of 1944 to correct the "slant range distortion" at close ranges. This equipment, known as H2S Mark IIIB, is at the time of writing (October 1944) still the backbone of the PFF H2S.
2. Secondly the H2S scanning system was not originally stabilised against the roll of the aircraft. Hence, violent evasive action caused the radar beam either to dip into the ground or lift away from it. The result on the PPI was therefore alternatively either very heavy ground returns or nothing. This considerable extra confusion, introduced at a critical part of the sortie (near the target), proved a very serious deterrent to the effective use of H2S, and urgent development was undertaken in mid 1943 to roll stabilise the H2S scanner. The experimental development was completed in the autumn of 1943 and it was originally hoped to have all the X band H2S equipments roll stabilised by the Spring of 1944. Unfortunately, due to a series of production hold-ups and technical difficulties, this conversion to H2S Mark IIIA is only just being introduced. There has been a similar delay in converting the Main Force to the roll stabilised H2S Mark IIC equipment. The need for and the effect of roll stabilisation is well illustrated in Figs. 3a and 3b.

Figure 3a. With roll stabilisation - Photograph taken with roll stabilised barrel Scanner, aircraft banking at 30°. The target is Leicester, at a range of about 4 miles on the 10 mile range scale, with the marker ring at 10 miles.	Figure 3b. Without roll stabilisation - Photograph taken with ordinary unstabilised Scanner, aircraft banking at 30° to starboard. The target is again Leicester. It is barely visible among ground returns at 8 miles on the 20 mile range scale, with the marker ring at 10 miles.

6. THE THIRD WINTER (1944-1945)
We are now entering the third H2S winter. Almost every heavy which comes off the production line is now fitted with an S band H2S and there are between 250 and 300 X band equipments in service with PFF. That sounds better than it really is, however. The burden of the winter's bombing is going to be carried by equipments on which the experimental development was finished one and a half years ago and on which the design principles were laid down two and a half years ago. At least for the first half of the winter there is not even, going to be much roll stabilisation. It is hoped, however, that small numbers of equipment of modern design will make their appearance in PFF during the next few months. The development of H2S Mark VI (K band) (see page 78) is proceeding satisfactorily and the introduction to the Service of a few aircraft fitted with this equipment should take place early in 1945. This should be followed shortly afterwards by the introduction of larger numbers of a radical new design of X band H2S incorporating new computing and wind finding arrangements (VS. Mark IV - see page 64).

7. CONCLUSIONS
The question "How worth-while has been the effort expended on H2S" is sometimes heard. This note will have shown that H2S was originally conceived as a blind bombing device with comparatively crude ends in view - namely to prevent an aircraft dropping its bombs on open countryside. Before H2S could be used even in small numbers, however, "Oboe" had set completely new standards for bombing accuracies. Developed to its full advantage by a powerful Pathfinder Technique which eliminated one of its inherent disadvantages compared with H2S, and given an unexpectedly long lease of life by the failure of the enemy to jam, Oboe has often overshadowed H2S in the consistency of its lethal blows. The greater part of Germany lies outside Oboe range from this country, however, and for those targets there has been nothing but H2S - for example, during the six winter months, October 1943/March 1944 inclusive, 93% of our strategic bombing effort depended on the navigation and bombing facilities of H2S.

In conclusion it may be remarked that a lot of nonsense is sometimes talked about the supposed gross inaccuracies of H2S. The facts are, that on a town such as Leicester in this country, TRE can get probable errors of half a mile on an aiming point in the town, with the H2S Mark III equipments. BDU has got 400 yards on Blackburn with H2S Mark IIIA. Over Germany the official ORS statistics show (a) that in the 1943/44 winter the 50% zone was about 2 miles (b) that in recent operations the 50% zone with the H2S Mark III equipments is about 1.2 miles. There have been extremely few really bad mistakes and gross errors; these have not necessarily been H2S boobs, but they have unfortunately been given undue publicity.

This note has been confined to the development of H2S as such. Modified H2S equipment has been used for several other highly important applications - in particular for ASV and it is hoped that a future issue of this Journal may include some account of these applications.