There still remained an operational requirement for the K-type containers for the Far East. The lid was redesigned to increase its strength and to make it easier to open and close. Special attention was given to improving the tropicalisation – in particular to protect the float cable from corrosion. Samples were sent to India and Australia during April, but it is not known whether they were used operationally.
ROLLER CONVEYORS FOR PACKAGE DESPATCH
The speed and efficiency of despatching packages from an aircraft was vital for successful supply drops. It takes several seconds to manhandle a heavy pannier through a hole in the floor and if the aim is to place, say, half a dozen in the same field from an aircraft flying at perhaps 120 mph, the speed of handling is of paramount importance. The use of a twin-track conveyor to despatch through the side door of a Dakota had become an established technique in the Middle East. The use of the same method of despatch through the jump hole of a Halifax was considered. Station IX designed, and Station XII made, a prototype which was tested successfully from a Halifax at Tempsford on 8 June 1943 and so impressed Grp Capt Batchelor that he recommended its general use. A number of defects were, however, identified and early the following year the conveyor situation for SOE aircraft was reviewed by the ASR Section. It was considered that, although the Halifax conveyor was defective in many respects, with minor modifications it would meet the current requirements for this aircraft. It could also be used with the Liberator either through the jump hole or rear hatch. At this time Stirling bombers, modified for paratroop operation – the Mk IV – became available and although the Halifax conveyor could be used for this aircraft, much greater utilisation of space could be achieved by redesigning it. An operational requirement envisaged that it could carry and despatch twenty 100 lb (45 kg) packages. These conveyors were designed by Shorts, who had built the original aircraft, and tested in collaboration with AFEE in November 1944. The trials were in general very satisfactory but minor problems remained. The ease of moving standard packages within the fuselage needed improvement, and the lashing techniques to prevent movement of the packages in flight had to be designed to facilitate loading and rapid unleashing before the drop. Of major importance now was to evolve an efficient despatching drill. To this end a Stirling fuselage was installed at Station IX. After some experimentation a basic drill was established for demonstration to various interested parties. There followed a full flying demonstration of the conveyor procedure. It was then proposed that a fully loaded conveyor be the subject of final trials but in the event they were held up because of a technical dispute over the effect on the centre of gravity of the aircraft as it despatched its load. A trial was scheduled for 29 April 1945 but no report of this has been found: it probably never took place.
Dropping from the Mosquito
The use of the Mosquito intruder aircraft for supply dropping was under active consideration early in 1944. The bomb racks of the aircraft were not large enough to accommodate any standard containers, but individual cells from the C-type could be adapted specially to fit the existing bomb racks and clearance for their use was obtained. However, the requirement lapsed for European operations, though it was said to be needed in the Far East. The possibility of replacing the long-range wing fuel tanks by supply containers was actively considered. A container having the same shape and dimensions as the wing tank was fabricated in plywood, using the same techniques as those used for the aircraft itself. Prototypes were tested both in drop tests from a balloon and in stressing trials at RAE. An aircraft drop test awaited the availability of an aircraft. In typical SOE style means were found of circumventing this obstacle. It so happened that the Canadian Air Force Mosquito squadron at Ford in Sussex was commanded by Grp Capt Paul Davoud, elder brother of Gordon Davoud at Station IX. Some brotherly collusion then enabled the wing container to be tested successfully in June. However, for some reason production of these containers seems not to have gone ahead.
DETECTION OF CONTAINERS IN THE DARK
It was inevitable that in dropping containers into a relatively small area at night, inaccuracies of the drop or unexpected wind conditions would lead to a scattering of the containers on the ground. The problem of locating them quickly became of major importance. A reception party had no desire to spend hours searching the countryside knowing full well that the noise of the aircraft would have alerted enemy forces.
Concern had already been expressed in late 1942 at the number of parachute containers which had been lost to the enemy because they could not be quickly located in the dark. Station IX was asked to look into the problem urgently. Work towards the solution engaged the Experimental Section continuously from December 1942 to near the end of the war. It was initially an SOE problem since conventional airborne parachute operations were expected to take place mainly in daylight, although later it became an Airborne Forces requirement. The early trials of alternative devices were carried out by Station IX, though later the User Trials Section and the ASR Section played a major role. Five possible solutions were considered: reflective or luminous discs; a ‘birdcall’ whistle operated by compressed air; electric bells; supersonic devices; and a luminous cloud. Reports of early experiments with bells indicated that they could be heard at 350 yards and this gave encouragement for further work.
Initial experiments with luminous discs in December showed that on a dark night a 2 in diameter metal disc coated with radioactive paint was visible at a distance of 100 yd and detectable at 150 yd. This was better than the Country Sections had asked for so smaller discs which were visible at 25– 50 yd were experimented with. The following month a small quantity of 1 in diameter metal discs which could be seen at 25– 50 yd were supplied for trials in the field and in February 1943, 100 belts of luminous discs were supplied for operational purposes. The belts of luminous discs had been tested in Operations Rose and Tulip in March but their success or otherwise is not recorded. What is recorded, however, (see Chapter 6) is the concern of Station IX that the highly radioactive luminous paint was harmful to health if a person was in close contact with it for any length of time. Moreover, flakes of paint could get into the air and become caught in the clothing and be breathed in as dust. It was ruled that the discs should be stored in a lead-lined box and not assembled onto the parachute containers until just prior to loading onto the aircraft. Station IX supplied four radioactive discs per belt and 140 metal plates for use on packages.8
Comparative trials of the whistle, bell and luminous cloud devices were held on a fine clear night with brilliant starlight but no moon on 26 January 1943 with a group of observers from Station IX. In each case the detecting devices were effective at distances up to 300 yd, although with the whistle and bell the directional discrimination fell off with increasing distance.
The third device created a blue-green cloud of chemiluminescent droplets by the mixing at an atomiser nozzle of two solutions, operated by compressed carbon dioxide (CO2). Two versions based on different solution concentrations were developed. In each case the detection device was set off by an inertia switch responding to the impact of the container on landing. In spite of, or perhaps because of, the earlier work on luminous discs, a device based on electric lighting was not considered at this stage since it would not be seen if a container fell behind a hedge or in a ditch. The luminous clouds, using different reagent concentrations were between five and ten feet high, the smaller intense cloud being visible at more than 350 yards. The lifetime of this cloud depended on the size of the CO2 cylinder and the volume of the reagent solutions. It was extended by operating through a modified Horstman gas clock, on for 4 seconds and off for 10 seconds. The containers for the reagents had sufficient for 20– 30 minutes operation, although there was a tendency for the nozzle of the CO2 bottle to freeze up before this time. Development of the luminous cloud device continued through February and March.
Various modified means of operation were tried. The use of only one solution was possible by replacing the hypochlorite
solution with a tube of activated charcoal saturated with chlorine gas which was displaced by the CO2 stream. Morale among the scientists working on the problem ran high and a note made on 2 March commented ‘most of our troubles are over’. But this was not so. Even though the device could be fitted into the chute compartment of the container, it was not easy to ensure that the cloud was released in a vertical direction; mounting the device in gimbals was a possible solution.
At about this time it became clear that a container location device was also an operational requirement for the Army and Airborne Forces, and that there might be a demand for possibly several hundred. It was realised that while the luminous cloud device might be a viable solution for supplying SOE agents with four to six containers per drop, the problems of manufacture on a large scale, and of filling the apparatus with fresh chemicals, made this impractical for major operations. Moreover, it was likely to be an expensive item. At this stage, further development was dropped and attention directed to the bell device and to a revised interest in a lighting device.
SOE was not the only organisation concerned with container location. In particular, the Airborne Forces Research Establishment produced a device which on impact ejected three or four spring-loaded telescopic arms each tipped with an electric light. In some trials of this device at Station IX a number of mechanical problems leading to malfunction were identified. And it did not solve the problem of containers falling into a ditch or behind a hedge. Despite this, Station IX was asked by the Ministry of Supply to develop both the bell and lighting devices.
A programme of top urgency was mounted in the week 26– 31 July to prepare six prototype bell devices for daylight dropping trials in the afternoon of 5 August. These were carried out at Tempsford using sand-filled containers. The devices operated satisfactorily, although some minor modifications were indicated. Because of the wind and background aerodrome noises the audible range of the bells was only 20– 30 yd. Two of the bells were barely audible because on impact the containers had burst and the bells were immersed in sand. That night the trial was moved to farmland near Aylesbury and was combined with trials of reception committee torches. Two more bell devices were dropped. One operated correctly and in the presence of a light wind and little background noise the bell was clearly audible at 150 yd, and just heard at 230 yd. It was judged that at 200 yd it was unlikely to attract unwelcome attention. The other bell rang for a few minutes and stopped. The simple explanation was that the screw holding the hammer had not been fully tightened. The team had devoted another long working day to this problem: they had left Station IX at 10.00 hours and returned after the trials at 03.00 the following morning.
In a further ground trial on the night of 9 August five bell devices were scattered over a distance of 250 yd. A party of five ORs located all the bells in 3½ minutes, a very satisfactory result. The following three weeks were taken up converting the prototype bell devices into pre-production models in collaboration with Station XII and the South Metropolitan Gas Co. who were to undertake the manufacture. Meanwhile the development of the lighting device was held up because the contractor (Bulgin) could not get quick delivery of the necessary switches and bulb holders. To get clearance for these they had to submit a written application to the Inter-Services Communication Committee (no connection with ISRB), which incurred frustrating delay.
The Ministry of Supply Committee on Airborne Requirements met on 1 September with representatives of many interested parties. Widely varying estimates of the needs for container detection devices were bandied about. The internal SOE estimates for 2,200 bell devices and 5,000 lighting devices were trumped by an Army request for 80,000 by the end of 1944. A request was made for the revival of the whistle device, and for consideration of a clockwork version of the bell device. No more work was recommended on radio and supersonic devices, while ARDE’s work on infra-red (IR) methods was to be continued up to the production stage.
Further progress towards the production of the bell device was pursued by Maj Bedford and Capt Ault at Station XII. But it was not until May/June 1944 that it was reported that successful acceptance trials had been carried out on the first pilot production models. Full production models were to incorporate a much more robust bell and use only one battery. Meanwhile, tropical tests had shown that (without the battery) the device stood up well to these conditions. It is not known how many bell devices were produced.
The manufacture of one whistle detection device was undertaken by the Walter Kidde Company in America. When the first sample appeared at Station IX it was plainly grossly over-engineered, with heavy brass cylinders, and was very expensive to produce. J.T. van Riemsdijk, a member of the Fuses and Fine Mechanisms Section, offered to redesign it and produced a device which relied on a capillary tube for the prolongation of the whistle. An order for 37,000 of them was placed with a firm who, unfortunately, made them without the capillary. Hence when they were activated they gave one loud shriek, followed by silence.9
There seems to be little information on the use of container detection devices in the field. After D-Day in the Maquis areas an agent said that ‘they had little difficulty in locating containers’ as they could ‘provide plenty of men as lookouts for ½ to ¾ mile from the dropping grounds. They were extremely useful (sic) particularly when the parachutage was done by the Americans!’ These drops were massive and in daylight. It is not known to what extent location devices were used in other theatres of the Second World War.
Records of work on supersonic methods of detection have not been located though research was being undertaken in February 1943 and field trials to establish the effective range of supersonic signals were carried out the following month.10
HIGH-ALTITUDE DELAYED DROPPING
While the ability to drop supplies by parachute from heights up to 1500 ft were advantageous in many situations in north-west Europe, the real and urgent need to be able to drop from much higher arose with operations to Poland. It is ironic that the Poles, whose Home Army could make greatest use of sabotage materials and weapons were, in fact, the most difficult to supply. The main problem was that a supply drop to Poland involved a 12– 14 hour flight over enemy occupied territory during darkness. The crew of the aircraft would not be at their best after a long cold flight which, to avoid the enemy anti-aircraft fire, had been at an altitude of at least 10,000 ft. They then had a very short time to locate the designated DZ, descend to 500 ft, make their drop and climb back to a safe height for the return journey. Because the aircraft was operating at the extreme limits of its range, the drop would have to proceed faultlessly if the vital fuel needed for the return flight was not to be used up. It was inevitable that under these conditions the payload of stores would be only a fraction of that carried on a shorter sortie in north-west Europe: six or seven tons of fuel were needed to deliver one ton of stores to Poland.
In 1941 the only aircraft available for Polish operations were two long-range Whitley bombers with a range of 850 miles crewed by Poles of 138 Squadron. Later two Halifaxes, with a range of 100 miles more than the Whitley, were made available; three B24 Liberators with a further 100 miles range were added in October 1943. Up to April 1942 only nine successful operations had been made to Poland carrying 48 men but a negligible amount of stores. Later 62 sorties – 41 of them successful – carrying 119 men and 23 tons of stores were flown with the loss of two aircraft. During the summer months the short nights precluded any operations. When they resumed in September 1942 enemy defences had been strengthened and the failure of sorties and the loss of aircraft increased alarmingly. In September twenty-two sorties were flown with the loss of six aircraft – four out of eleven on one night alone. At this point the flight route was changed to a more northerly route over the North Sea, Southern Denmark and the Baltic to reduce the time spent over hostile areas. But this reduced even further the areas of Poland to which drops were possible. It was then decided that, now that much of Italy had been freed by the Allies, the base for Polish op
erations could be moved to Monopoly, near Bari in Southern Italy. This took place between November 1943 and January 1944. By this time the need for a means of dropping stores accurately from a high altitude became a high priority. This would not only reduce the risk of interception and of attack by ground fire during flight but would avoid the extra fuel consumption involved in descending to drop the load and climb again. The need for high-altitude dropping thus emanated mainly from the Middle East and the first successful devices were developed for SOE (Force 133) by Maj G.N. Sanderson RAOC in Cairo. The urgency was heightened by the figures in the period April– July 1944 when out of 315 sorties, only 174 succeeded and eight aircraft were lost. A failure rate of 45 per cent was totally unacceptable and demands for the High Altitude Delayed Dropping Apparatus (HADDA) development intensified.
Full details of the original Middle East (ME) device are not known. It appears that it depended on the ignition of a length of Bickford fuse upon the container’s release from the bomb bay which, after a chosen time calculated to bring the container to the desired height, initiated a switch to release the parachute. Following the preliminary development of the Middle East HADDA, further work in the UK was initiated by SOE early in 1944.
The problem resolved into two main factors. A means was needed either to delay the parachute opening for a given time or until the container reached a predetermined height. Moreover the ballistic inconsistency and indeterminacy of a tumbling container made the aiming of the drop difficult. Unlike a bomb which has fins to stabilise it in its flight and give a reasonably predictable trajectory, a standard container had no such features and its trajectory was difficult to predict. Furthermore, it could reach a velocity far in excess of that specified for the deployment of the parachute. It was, therefore, necessary to use a small parachute to stabilise the fall in the first part of the trajectory and to reduce the terminal velocity to an acceptable level. The main parachute had then to be strong enough to withstand the shock of opening, otherwise the result would be a bursting of the canopy and total destruction of the container and its contents. In some cases the long flight at high altitude to the DZ resulted in iced-up parachute packs which failed to open. The Stirling seemed particularly vulnerable to this problem.
SOE Page 25