Most Secret War
Page 33
As I read the message, and began to imagine the conditions under which it had been sent, it brought home to me what a pleasant safe job I had compared with those on whom I depended. The poignant humour of having been fired at ‘fortunately with more zeal than accuracy’ and the intention to work on despite the seeming lack of encouragement from our side made me more than ever determined not to let these men down. There was in fact no way of communicating with them directly at the time. I doubt whether anyone in Britain knew who they were, and I am not sure whether they were ever thanked, although I did my best. If any of them is still alive and reads this, I hope that he will now know that their message got through. As will transpire, I was able a few months later to arrange some direct action. The episode illustrates the enormous advantage of my personal position in Intelligence. Having such a vivid contact with those who were risking their lives to get the information on the one hand, and with those at the summit of power, like Churchill, Cherwell, Sinclair and Portal on the other, I was able to ensure that work in the field was appreciated at the top with as little hierarchic attenuation as possible.
Thanks to the Belgians and to Photographic Reconnaissance, we now knew how a typical sector of the Main Belt was organized, and the technical equipment on which it depended. The next advance occurred in September 1942, when Bletchley broke into a new line of Enigma traffic: I had encouraged them to look for anything that might be associated with night-fighter activities, and the hunt succeeded. When the cypher was broken it gave us some of the reports from nightfighter sectors back to their higher command.
Our first important break referred to reports from a Sector 7 on interceptions that had been made on our bombers by nightfighters during a raid on Frankfurt on the night of 8th/9th September. By finding where we expected our bombers to be at the times they had been intercepted, it seemed that Sector 7 must be in southern Belgium. There was mention of three controls, 7A, 7B and 7C, and it seemed that some aircraft had spilled over into Sector 8A, less into 8B, and none into 8C. This would be consistent with assuming that Sector 8 was south of Sector 7, and that the A, B and C sub-units were also in a north/south order. It was also tempting to assume that the three letters referred to a set of three control stations such as we had seen from the pattern on the stolen map. The one other piece of information that I had available was that the Luftwaffe generally had been informed of the existence of a closed area for nightfighting which roughly delineated what we believed to be a zone centred on the old Main Belt. Figure 12 shows this information, and Figure 13 shows the Bomber Command route map, with my pencilled speculation regarding the whereabouts of Sector 7.
Fig. 13. Bomber Command raid tracks for 8/9 September 1942, from which—in conjunction with an Enigma message—it was deduced that Sector 7 of the German nightfighter defences was in south Belgium
As I was turning the information over in my mind I suddenly saw that if you assumed that a Sector covered a 90 kilometre front with radar stations at 30 kilometre intervals in a straight line, and try to fit a set of such 90 kilometre units into the declared nightfighter zone, and if you started at the top and assumed that Sector I was in Schleswig Holstein, the 7th Sector would fall exactly where I had deduced that it should be from the nightfighter interception reports. In a flash, the whole system of the night defences became clear. We could now guess at the position of every other radar station in the Main Belt, and could fill in much of the missing detail, so that the map shown at Figure 12 was now transformed into that at Figure 14. In front of the Main Belt there were the other nightfighter control stations of the Domburg type, extending all the way up the coast of Holland.
Fig. 14. The transformation of Figure 12 as a result of the deduction in Figure 13 and the stolen map at Plate 16. Taking a unit of three radar control stations in a row at 30 kilometre intervals, and assuming that this constituted a Sector and supposing that the Kammhuber Line was numbered consecutively from North to South, Sector 7 fell correctly into position, and the approximate positions of all the remaining 21 control stations could be deduced
I first tried the idea on Charles Frank, who was sitting at the opposite desk at the moment on which it had occurred to me, and I then telephoned Charles Medhurst and arranged an immediate visit to show him what had happened. He said that he thought that Bert Harris of Bomber Command should see it at once, and that he would arrange for me to visit him the following day. I told him that I should be very glad to go, especially since the one gap in our knowledge was the exact characteristics of the German nightfighter radar. We were almost certain that it operated on a wavelength of about 61 centimetres, but we needed to fly in front of a German nightfighter in order to obtain final confirmation. Such a flight would be very dangerous if carried out in an ordinary bomber, and we therefore needed one or two Mosquitoes which would be fast enough to make their escape once they had detected the transmission from any German nightfighter that was about to intercept them. Bomber Command had almost the entire production of Mosquitoes at that stage and our one hope of getting Mosquitoes quickly was to persuade the Commander-in-Chief to release two for this operation. Charles Medhurst told me that it would be up to me to convince the Commander-in-Chief sufficiently to let us have them.
At the same time I showed the work to my old Professor, now Lord Cherwell, and told him I thought it would be helpful if the Prime Minister could minute the Air Staff to stress the importance of finding the German airborne radar wavelengths as the last essential step in our technical knowledge of the whole system, on which we were of course basing our countermeasures. Prime Ministerial pressure was accordingly applied.
In the meantime there was a happy outcome to the problem of giving encouragement to our Belgian agents who had contributed so much. This arose because we ourselves were developing a new bombing system known as ‘Oboe’. This system, developed by A. H. Reeves and F. E. Jones at the Telecommunications Research Establishment, employed a principle which I had sketched in my imaginary minutes of the German conference in December 1940 about the future of radio bombing, where I ascribed it to Dr. Plendl:
‘Dr. Plendl … would add that … newer systems could be developed which did not in fact involve beam flying but depended entirely on distance measurements from two ground stations, either by frequency modulation, phase measurement, or pulses’.
Reeves used to visit me from time to time in London, and told me he would like to develop a bombing system for our own use and tried out on me various of the alternative systems using range measurements from two stations. The attraction was that whereas our directional measurements were not very good, we knew that we could make range measurements very accurately. The idea was to fly an aircraft at a constant range from one station by sending out pulses from the ground which the aircraft would pick up and amplify and then return to the ground station. This ground station would then find the aircraft’s range from the time it took the pulses to return. The path of the aircraft, if it flew at constant range, would thus be part of a circle centred on the ground station, and this circle would have to pass near the target, but slightly short of it, as the bombs went off tangentially after they had been released from the aircraft. All this could be calculated, if the aircraft’s height and speed were known. Moreover, the pilot in the aircraft could be automatically informed by means of signals sent out from the ground station whether he was on track or to one side of it, just as in the German beam systems—but with the great advantage that since there was no beam the Germans could not pick it up as we had been able to do against their bombers.
A second ground station could also determine the range of the aircraft from it, and so instruct the aircraft to drop its bombs when it was at such a range that these should hit the target. The system was therefore a good deal more sophisticated than anything developed by the Germans. It had the same advantage as their Y-system in that the instant of bomb release was determined by a ground observer free from the harassment of a flight over enemy territory, but it also had two disadvantages of
that system in that the traffic-handling capacity was very limited and the aircraft had to transmit a signal which might be homed upon by the enemy. The former disadvantage was not serious in the event, in that a few Oboe aircraft could be used to drop visual markers for the main force; the latter disadvantage was substantially avoided because Oboe was used by Mosquitoes which were too fast for the enemy nightfighters, and because it was later moved to centimetric wavelengths against which the Germans were almost powerless.
The potential disadvantages were given undue weight in some higher quarters—so much so that one scientist wrote:
I regret having to do this but I am sure it is true to say, quite bluntly, that these disquisitions from T.R.E. on OBOE are becoming ridiculous. If they came as inventions from the outside public and not from official sources, they would be rejected without hesitation … I repeat now, even more strongly, that it would be disastrous to permit the protagonists of this fantastic OBOE the chance of causing a sensible and practical system like GEE to share the disrepute into which OBOE—even if raised to Mark 20—will inevitably fall.
If I had the power I would discover the man responsible for this latest OBOE effort and sack him, so that he could no longer waste, not only his time and effort, but ours also by his vain imaginings.
As it turned out, Oboe was the most precise bombing system of the whole war. It was so accurate that we had to look into the question of the geodetic alignment of the Ordnance Survey with the Continent, which effectively hinged on triangulation across the Straits of Dover. F. E. Jones asked me if I could suggest a trial target from which we could get precise information from ground observers on the fall of bombs, so as to be able to see whether our triangulation required any correction. It was then that I saw the chance of showing our Belgian friends that we were making use of their information for by this time they had provided many more reports, one of which said that the headquarters of Sector 7 was in the Novitiate near the town of Florennes.
Fig. 15. The ‘Oboe’ dispositions for bombing Florennes and Essen. The bombing aircraft flies at a constant range from the ground station at Trimingham, taking it on an arc. When it is at the correct range from the ground station at Walmer, the latter sends the order for bomb release. The curved tracks are slightly (but exaggerated in the diagram) to the west of the targets because the bombs continue on a tangent after release
Entirely against the standard policy of not disclosing the target for a raid in advance, I obtained permission on this occasion for signals to be sent out to the Belgian network so that they could have observers in position for the night of the attack which was to be made by a small force of Mosquitoes in December 1942. Each aircraft flew on a constant radius from one Oboe station at Trimingham and the correct instant for bomb release was decided by a controller at a second station at Walmer (Figure 15). In the event, everything was so brilliantly successful that F. E. Jones afterwards said ‘Within 48 hours we had reports of where the bombs had gone—even down to the fact that one had gone into a particular tree. Their distances from the aiming point were given in yards. And when I queried how this could have been done so accurately, R. V. pointed out that the Belgians, risking their lives in no uncertain manner, had actually paced out the distances for us before sending back the information to London.’ And, to the delight of everyone concerned, including the Belgians, one of the bombs had actually hit the building in which the nightfighter headquarters were housed.
We had started tactical countermeasures against the Kammhuber Line in the winter of 1941/42, well before we knew its exact extent or location. The main tactic was to try to fly our bombers in a compact bunch, preferably all through the same nightfighter box in as short a time as possible. While the nightfighter controller was then concentrating on one bomber, many others should be able to get through unscathed. This tactic worked to some extent, but it was difficult to maintain concentration, especially on the homeward route where differences in speed and navigation between individual bombers caused them to be increasingly strung out and spread, and most of our lossses occurred in this phase. Kammhuber’s reaction was to increase the number of nightfighter stations, so that instead of having to pass merely through his original Line our bombers had now to fly through a defence in depth. Figure 16 shows our knowledge of the increased deployment towards the end of 1943, when the original 27 stations of the original Line had now become an area defence involving more than 200 stations.
Fig. 16. General Kammhuber’s reaction to Bomber Command’s concentration tactics, spring 1943. His original Line is expanded to an area defence in depth, with many more radar control stations, each shown as a dot
For locating the new stations as they appeared, we depended mainly on the Resistance movements in the Occupied Countries and on photographic reconnaissance. The stations, with their two Giant Würzburgs, were difficult to conceal, and the Würzburgs were indeed objects of local wonder which strained the descriptive powers of many of our friends in the Low Countries. ‘Inverted Umbrella’ was a typical description, while ‘Miroir Magique’ conveyed an air of mystery. One Würzburg was so much talked about that it became ‘Le Fameux Miroir d’Arsimont’. In areas where we had no direct contact with the Resistance movement, we used to get our bombers to drop homing pigeons in containers which would open after a few hours and release the birds if they had not been found by someone on the ground. Attached to the containers were questionnaires, asking a series of simple questions which, for example, a farm labourer might be able to answer, and which might be helpful to us. My own question was ‘Are there any German radio stations in your neighbourhood with aerials which rotate?’ This feature was an almost certain criterion of a radar station, and we dropped the pigeons wherever we saw a gap in our knowledge. Before the end of 1942 the pigeons had given us the locations of three stations hitherto unknown to us, and more followed during 1943.
Further clues regarding the locations of nightfighter stations could sometimes be derived from Enigma, where the stations were referred to by animal codenames, the first letter of the animal being the same as that of the nearest village or town to the site of the station. ‘Hamster, Biber and Zander’ were respectively Hamstede, Brielle and Zandvoort in the Scheldt Estuary, while ‘Tiger’ was Terschelling (incidentally responsible for the destruction of 150 of our bombers). ‘Waal’ was Wangerooge, ‘Languste’ was Langeoog, and so forth. As a result of this guessing game, and despite my ignorance of German, I came to know almost the entire German menagerie—to the bewilderment of a German forester who took me shooting at the end of the war.
Another tactic that we used against the Kammhuber Line in its original form was the obvious one of flying round its extremities. The outstanding examples were the attacks on Lübeck and Rostock in March and April, 1942. Great damage was done with relatively low losses. Kammhuber’s reaction was to make an area deployment over the whole of Denmark, creating 29 new stations. Thanks to our friends in the Danish Services, and particularly Captain P. Winkel (later Major General and Chief of Defence Staff) and Captain V. Gyth (later Colonel and Chief of the Royal Guard), we knew the positions of all these stations before any was in operation. The Lübeck and Rostock raids had a further effect: Hitler resolved on a policy of retaliation, leading to the V-weapons.
CHAPTER THIRTY-TWO
Lichtenstein
IN BRITAIN WE had decided as early as 1935 that interception at night would very probably require a suitable detector for the nightfighter to enable it to close in on the bomber after the ground control had done its best to put the nightfighter into a favourable position. Such a detector had indeed been the main objective of my pre-war efforts in infra-red, but an airborne radar detector (A.I.—Airborne Interception) had proved more promising. In Germany, too, infra-red had been developed for nightfighter use and a brilliant equipment, the Kiel Gerät, had been produced using the infra-red detecting properties of lead sulphide. But it suffered from the usual infra-red defect of not being able to measure the range o
f the target, and so the Germans also ultimately preferred an airborne radar detector. This equipment, the Lichtenstein Gerät, first came to our notice through a prisoner in April 1941, and it appears to have achieved its first operational success on 9th August 1941 when a Messerschmidt 110 from Leeuwarden shot down one of our bombers with its aid. In the ‘Little Screw’ radio telephony traffic between German controls and their nightfighters there was increasing use during 1942 of the phrase ‘Emil Emil’ which seemed to indicate that a nightfighter had now been brought close enough to the bomber for the latter to be picked up in the nightfighter’s own detector.
Since German radar technique was obviously strong at wavelengths of 50–60 centimetres, and since such short wavelengths would be particularly suitable for nightfighter equipment because of the small overall dimensions, we decided to search particularly in this waveband. And since nightfighters were operating in the Scheldt Estuary, not more than one hundred miles from the Suffolk coast, there was a good chance that we could pick up the nightfighter radar transmissions on listening equipment ground-based in Suffolk. Fairly soon we heard transmissions on a wavelength of about 61 centimetres with a pulse repetition frequency of 3,000 per second, which seemed to come from moving sources. Incidentally, the Giant Würzburg transmissions from Domburg were also heard, on wavelengths around 53 centimetres and with a pulse repetition frequency of 2,000 per second. On 16th May 1942 we even attempted to intercept a nightfighter controlled from Domburg with one of our own Beaufighters controlled from Foreness. Tracks of both nightfighters were plotted by Fighter Command; but, thanks to the skill of both British and German controllers, the attempted combat ended in mutual frustration. The German aircraft refused to be tempted more than about 60 kilometres away from Domburg, so as to remain within range of its Giant Würzburgs.