Open main menu

Radar, Air-to-Surface Vessel, Mark III, or ASV Mk. III for short, was a surface search radar system used by RAF Coastal Command during World War II. It was a slightly converted version of the H2S radar used by RAF Bomber Command, with minor changes to the antenna to make it more useful for the anti-submarine role. It was Coastal Command's primary radar from the spring of 1943 until the end of the war. Several improved versions were introduced, notably the ASV Mark VI radar, which replaced most Mk. IIIs from 1944 onward, and ASV Mark VII radar, which saw only limited use until the post-war era.

ASV Mark III
ASV Mk. III on Wellington MP512.jpg
Wellington XII MP512 was one of the first aircraft to be fitted with ASV Mk. III
Country of originUK
Introduced1943 (1943)
TypeSea-surface search
Frequency3300 ±50 MHz (S-band)
PRF660 pps
Beamwidth~10º horizontal,
~15º vertical
Pulsewidth1 µs
RPM60 rpm
Range1 to 100 mi (1.6–160.9 km)
Diameter28 in (0.71 m)
Azimuth320º
Precision~5º
Power40 kW
Other NamesARI.5119, ARI.5153

Coastal Command's first radar was ASV Mark I, which began experimental use in 1939. A number of minor improvements were made for the Mark II in 1940, but it was not widely available until late 1941. Having realized that the RAF was using some sort of radar to detect their U-boats, in the summer of 1942 the Germans introduced the Metox radar detector to listen for their signals. This gave the submarine a warning of the aircraft approach long before it became visible on the aircraft's radar. The RAF noticed this in early autumn when crews reported that they would detect submarines that would disappear as they approached.

An ASV working in microwave frequencies using the newly invented cavity magnetron had been under development for some time at this point, known as ASVS, but had not matured for various reasons. Robert Hanbury Brown suggested using H2S radar for the ASV role, but this had been rejected by Bomber Command, who wanted all the sets for themselves. Brown continued development with EMI and presented it again in late 1942 when Metox became an issue. Continued interference from Bomber Command led to more delays, and it was not until March 1943 that the first dozen aircraft were operational. Deliveries were rapid after this point, and Mk. II had been largely replaced by the end of the summer.

The Germans had no way to detect the signals from Mark III, which operated in the 10 cm band compared to Mk. II's 1.5 m wavelength. Further confusion was caused by a captured RAF officer who stated they carried a device that could detect the Germans' Metox radar detector. Combined with other anti-submarine technologies introduced around the same time, submarine losses shot up in the late spring of 1943. By the time it was clear what was going on, the German U-boat force was almost completely destroyed and the Battle of the Atlantic was entering its final phase. A microwave detector was introduced in October 1943, Naxos, but it was nowhere near as sensitive as Metox and had little effect on the outcome of the battle.

Contents

DevelopmentEdit

Mark IIEdit

Development of the original ASV systems started in 1937 after the team testing an experimental air-to-air radar noticed odd returns while flying near the shore of the English Channel. They eventually realised these were the docks and cranes at the Harwich docks miles south of them. Shipping also appeared, but the team was unable to test this very well as their Handley Page Heyford was forbidden to fly over water.[1] Further testing was carried out on two Avro Ansons. The system was crude, with a simple dipole antenna being held out the window and swung by hand to find returns.[2]

For a variety of reasons, the 1.5 m wavelength of the radar system worked better over water than land, and the large area and flat vertical sides of the ships made excellent radar targets. After some additional development of suitable antennas, the system was largely ready for production by early 1939.[3] Production quality sets were available in late 1939 and entered operational service in January 1940,[4] becoming the first radar system to be mounted on an aircraft in a combat setting. A somewhat improved version, Mark II, followed in 1941.[5]

These designs had a relatively long minimum range, meaning the submarine targets disappeared from the display just as the aircraft was readying for the attack. This was solved by the Leigh Light, a searchlight that lit up the submarines during the last seconds of the approach. By early 1942 Mark II and the Leigh Light were finally available on large numbers of aircraft. Their effect was dramatic; German U-boats had previously been almost completely safe at night, and could operate out of the Bay of Biscay in spite of it being close to British shores. By the spring of 1942, Biscay was a deathtrap, with aircraft appearing out of nowhere in the middle of the night, dropping bombs and depth charges, and then disappearing again in moments.[6]

The Germans ultimately solved the problem of Mark II with the introduction of the Metox radar detector. This amplified the radar's pulses and played them into the radio operator's headphones. With experience, the operators could tell whether the aircraft was approaching or just flying by. It provided this warning long before the echoes off the submarine became visible on the aircraft's display, allowing the U-boat to dive and escape detection. By the end of 1942, Mark II had been rendered ineffective.[6]

ASVS, original Mark IIIEdit

 
When placed between the poles of a powerful horseshoe magnet, this simple copper block produced many kilowatts of microwave signals, revolutionizing radar.

After the early 1940 invention of the cavity magnetron, which produced microwaves at around 10 cm, all of the British forces began development of radars using these devices. Among these were the Air Ministry teams who had developed both AI and ASV, and now turned their attention to AIS, the S standing for "senitmetric".[7] Tests in April 1941 with early lash-up devices against HMS Sea Lion showed they could detect semi-submerged submarines at several miles range.[8]

In June 1941 a formal application was made to the Director of Communications Development (DCD), at that time run by Robert Watson-Watt, to form a separate group to develop an ASVS. This was, in effect, simply a version of the Mark II with the minimal conversions needed to use the magnetron as the transmitter. However, the system was soon converted to follow the H2S model and adapted a scanner system to provide a 360° plan-position indicator (PPI) on a 9 inches (230 mm) cathode ray tube (CRT) display, and a second range-only display on a 6 inches (150 mm) CRT.[9]

In order to fit the scanning antenna to the Wellington, which had a smaller belly-turret ring than newer designs like the Handley Page Halifax, the reflector was 28 inches (710 mm) wide, compared to the similar units for H2S that were 36 inches (910 mm) wide. With those exceptions, the units were otherwise similar to the H2S Mark I being developed in the same period. Philip Dee noted that the first flight on Wellington T2968 did not take place until December 1941, and it was not until 13 January 1942 that he noted "ASV saw [the small ship] Titlark at 12 miles".[8]

This success led to contracts with Ferranti for production electronics and Metropolitan Vickers (Metrovick) for the scanning antenna system, which would be known as ASV Mark III.[10] Ferranti had a prototype ready by the summer of 1942, although they predicted that the first deliveries would not be available before the spring of 1943.[10]

Testing ASVSEdit

T2968 continued tests until 24 February, and on 7 March 1942 was sent to RAF Ballykelly in Northern Ireland to carry out competitive tests against other ASV developments.[9]

One was the Mark IIA which had a new transmitter that increased broadcast power from 7 to 100 kW. This was found to increase detection range against surfaced submarines to about 14 miles (23 km) and 7 miles (11 km) even when the submarine was semi-submerged with just the conning tower above water. This was about twice the effective range of the original Mark II. However, this also greatly increased the amount of clutter.[11]

A second unit used a similar high-power transmitter that operated on a 50 cm wavelength rather that 1.5 m, however, this was shown to have no advantages over the basic Mark II.[11]

In contrast, the ASVS set showed dramatic improvements. Performance against convoys was 40 miles (64 km) when the aircraft was flying at only 500 feet, in spite of the radar horizon being only 27 nautical miles (50 km; 31 mi) at that altitude. Other aircraft were visible at 10 miles (16 km), and surfaced submarines at 12 miles (19 km). The ASVS was immediately chosen as the new operational requirement, with the 50 cm set also being ordered as a backup. As it became clear the magnetron was going to be successful, the 50 cm system was cancelled.[12]

H2S, new Mark IIIEdit

 
The Mark III's small antenna allowed it to be mounted in a much smaller fairing than H2S. Here it is seen under the nose of a Wellington of No. 458 Squadron RAAF.

Throughout this period, Robert Hanbury Brown was convinced the H2S radar being developed for RAF Bomber Command could also be used for anti-shipping work simply by changing the antenna to one suited to an aircraft flying at 2,000 feet (610 m) rather than 20,000 feet (6.1 km) altitude. He continued working on this project with the primary developers of H2S, EMI.[13]

By late 1942, the problems with Metox were coming to a head, and Ferranti was stating the Mark III would not be available in numbers for some time. In contrast, Brown's H2S-based adaptation was largely complete, and it would be possible to have a small number of hand-built units installed by the end of 1942. This system, working at 10 cm, would be invisible to Metox.[14]

The TRE team in charge of ASVS was not under the control of Dee, and he was happy to point out their problems. On 25 September 1942, at a meeting at the DCD, Dee pointed out that the AI and ASV teams were now both developing separate systems that were, from a signals perspective, almost identical. The only major difference was that ASV had larger displays. He proposed abandoning the Ferranti system and using the H2S-based system.[15]

The meeting took place during a furious debate over the use of the magnetron; the concern was that if an aircraft carrying H2S was shot down, it would fall into German hands and be quickly reverse engineered. Frederick Lindemann was especially vocal against the use of the magnetron in H2S and demanded they use a klystron instead. The klystron was already known to the Germans and so fragile that it would be unlikely to survive any crash.[15]

Such a concern did not exist for ASV, where the magnetron would fall into the water if shot down. This made ASV a much safer choice for deployment of the very few available units. Bomber Harris objected, claiming his bombers would do much more damage to the German U-boat fleet by bombing their pens in France than Coastal Command would by hunting them down at sea.[13] Nevertheless, the meeting ended with Coastal Command being prioritised for the magnetron-based units. On 30 September, Ferranti was ordered to stop work on their design in favour of them building the H2S-based system, also to be known as Mark III.[15]

As if this fight with Bomber Command were not enough, there were problems within Coastal Command as well. This was due to their upset that their original Mark III project had been cancelled by the Air Ministry without even asking Coastal Command's opinion. The fact that the H2S-based system could be available immediately did not seem to impress the higher echelons of the Command. Adding to the confusion, the commander of Coastal Command, Philip Joubert de la Ferté, visited the radar development teams at the Telecommunications Research Establishment (TRE) and told them he did not believe in ASV, which led to demands to see it in action.[13]

Further confusion ensued when the TRE teams suggested fitting the new radar on four-engine airframes. These would provide ample room for the installations and superb range over the North Atlantic. On 8 December 1942, a meeting was called over the topic, but Joubert refused to intercede in favour of the TRE and they were told to continue with the two-engine Wellington.[13]

In serviceEdit

Initial flightsEdit

 
On the Wellington, the unused ventral turret ring was used to mount a retractable version of the Leigh Light that lowered drag during cruise.

The use of the Wellington with Mark III coincided with another change that was being made, the move of the Leigh Light from the wing of the aircraft to a retractable "dustbin" arrangement that extended down through the former belly gun turret ring. This meant the radar scanner could not be placed in that location, as it was on H2S aircraft, and it was instead moved to the nose. This not only blocked scanning to the rear, about 40 degrees, but also meant the nose guns had to be removed, leaving the aircraft with no ability to fire at targets below it, sparking more complaints.[13]

By the end of the year a small number of units were becoming available, and in December 1942 two were sent to No. 30 Maintenance Unit for fitting to Wellington VIII. These began testing at the Coastal Command Development Unit in January.[10] There was little difference between the H2S and ASV except the name. Both included two cathode ray tube (CRT) displays, main 6" scanner display and a smaller 3" "height scope" below it. The latter was used to measure altitude and for use with Eureka radio beacons, and in the case of the ASV, it also became used as a timing system for the illumination of the Leigh Light.[16]

The priority given to Coastal Command lasted only a short time, and on 8 January 1943, priority was once again given to Bomber Command. At this point it became clear that there were not enough fitters to keep the units working, and in addition to local recruits, a class from the recently formed RAF Station Clinton in Ontario, Canada sent another 110 technicians. These technicians first detoured for a short stay in the US where they were able to train on the similar US-designed DMS-1000 unit.[17]

The first operational patrol using the two aircraft was carried out on the night of 1-2 March 1943. The aircraft returned from Biscay without having spotted submarines. However, the aircraft were attacked by German night fighters, and the radar operator was able to give the pilot instructions to successfully evade them. Similar patrols also returned empty-handed until the night of 17 March when H538 spotted a submarine at 9 miles (14 km) but their Leigh Light failed and they could not press the attack. The next night the same aircraft spotted a submarine at 7 miles (11 km) and successfully dropped depth charges on it.[13]

Supplies of the magnetron began to improve at the start of March 1943, and on 12 March it was decided to split the deliveries equally between the two commands. A serious limitation of spare parts then became a problem, but was eventually solved by sending more spares to Bomber Command, to make up for their higher loss rates.[17]

Into serviceEdit

Enough units arrived by the end of March that No. 172 Squadron RAF at RAF Chivenor was able to convert their Wellington XIIs to the Mark III. The squadron was soon pressing multiple attacks every week, and in April the number of sightings in the Bay shot up. Calculations demonstrated that the aircraft were at least sighting every submarine in service at that time.[18]

These were not the only changes taking place. Around the same time as the introduction of Mark III, the first similar US radar units were arriving, built using magnetron technology introduced to them during the Tizard Mission in late 1940. These DMS-1000s were mounted to the Consolidated B-24 Liberator, one of the very few aircraft with enough range to enable it to fly patrols over the entire Mid-Atlantic gap and thereby allow aircraft to provide cover over convoys all the way from Halifax to ports in the UK. A B-24 with DMS-1000 was sent to the UK in January 1942 and used operationally by No. 224 Squadron RAF, where the system was referred to as the ASV Mark IV.[19]

For reasons unknown, the US Army Air Corps decided to cancel development of the DMS-1000 in favour of the Western Electric SCR-517, although it proved to be far less sensitive. The RAF also learned of another unit, the Philco ASG, intended for mounting in US Coast Guard blimps, that was more similar to the original design. They asked that the ASG be used on their Liberator order instead, referring to it as ASV Mark V. In March, a shipment of Liberators with a mix of DMS-1000, SCR-517 and ASG arrived and were put into service in June. These aircraft lacked the Leigh Light and were generally unable to press the attack, but they were nevertheless invaluable for upsetting the U-boats approach and calling in ships to attack them.[19]

The tide turnsEdit

 
Mk. III-equipped Sunderland W4030 of No. 10 Squadron RAAF attacks U-243 in the Bay of Biscay in the summer of 1944.

By May, the U-boats were being attacked continually from the moment they left port to the time they returned. Even if they escaped into the Atlantic, boats were then attacked hundreds of miles from the convoys while they attempted to form up the wolfpacks. This was combined with the arrival of new frigates mounting microwave radars and huff-duff receivers, further hindering U-boat operations. Successfully forming up and pressing on to the convoys proved almost impossible.[20]

Karl Dönitz was convinced this was due to a new detection system but remained baffled to its nature. In a mid-May 1943 report to Hitler, he stated:

Attempting to address the continual attacks in the Bay of Biscay, Dönitz ordered the U-boats to leave port during the day when they could attempt to shoot down the aircraft and day fighter cover could be provided. Coastal Command responded by forming up "Strike Wings" using high-speed aircraft like the Bristol Beaufighter which travelled in small packs and made hit-and-run attacks that overwhelmed the U-boats while also proving difficult for the German fighters to attack as they made a single run and then disappeared at high speed. While the U-boats did manage to shoot down several aircraft, the losses of boats continued to climb.[20]

Another change followed; in June, U-boats were seen leaving port in small fleets of five or more, providing a higher density of anti-aircraft fire to the point where it was dangerous to approach, while also reducing the per-boat chance of detection.[a] The RAF responded by having the aircraft standoff from the mini-fleets and call in destroyers, who could sink the entire fleet with ease. If the U-boats attempted to dive, the aircraft would pounce.[20]

Meanwhile, operations against the convoys, for those boats that managed to evade attack in the Bay, were proving almost impossible. Shipping losses to the U-boats plummeted; in June less shipping was lost than any time since 1941. By the end of the month, a full 30% of the U-boat force at sea had been lost, a catastrophic turn of events. Dönitz was forced to recall the fleet from the North Atlantic, sending them to secondary theatres while some sort of solution was developed.[20]

British lie, German confusionEdit

In late February 1943, German submarine U-333 was attacked by a Mk. III-equipped Wellington. The gunners were already on high alert and managed to shoot the aircraft down, but as it fell it managed to drop charges around the boat. The submarine survived and reported that the Metox gave no warning of the approach and the Leigh Light was not used. The aircraft simply appeared out of the murk and dropped a string of depth charges.[22] On 7 March, U-156 was attacked in a similar fashion, and radioed in that they believed a new radar was being used.[23]

In spite of this early warning of a new system, German efforts were hampered by one of the most effective disinformation campaigns of the war. A Coastal Command captain who had been captured after crashing told a plausible story, apparently entirely of his own creation, that threw the Germans off the scent for months. He stated that they no longer used Mk. II for the initial detection, and instead used a new receiver that listened for the slight leakage of the intermediate frequency used in the Metox's tuner. He claimed that it could detect the Metox at ranges as great as 90 miles (140 km). The radar was now only turned on during the last minutes of the approach to check the range and aid the Leigh Light operation.[20]

At first, the Germans were sceptical of this claim, but a series of experiments soon demonstrated this was indeed possible. This became outright horror when the equipment was installed in an aircraft and demonstrated its ability to detect a Metox at a distance of 70 miles (110 km) while flying at 6,000 feet altitude.[24] The extra 20 miles claimed by the pilot was attributed to the UK's superiority in electronics.[20] On 15 August 1943, a radio message was sent to the entire fleet telling them to turn off their Metox.[25]

From that point, the false information was "treated as gospel",[20] in spite of much evidence to the contrary. This included reports from boats that were attacked while their Metox was turned off, and one report from an enterprising radio operator in U-382 who had been experimenting with a visual display with the Metox and detected signals that were well outside the normal range.[26]

Most surprising of all was the fact that a magnetron had fallen into German hands on only the second night it was used when a Short Stirling carrying H2S was shot down over Rotterdam on the night of 2/3 February 1943.[27] For reasons unknown, the possibility of this system being used for anti-submarine work either never reached the Navy or was dismissed as impossible by Navy engineers.[20]

Attempted countermeasuresEdit

 
The Fliege, named for the fly-like antenna seen in the center of the reflector, greatly improved Naxos. It was connected to a mast on a rotating seal so it could be turned by the operator in the radio room and did not have to be removed when the submarine dove.

Still believing the issue was leakage from Metox, boats returning to port were fit with the Wanze radar detector. This was originally designed to detect signals in the 120 to 150 cm range, but also had the side-effect of having lower signal leakage and had greater sensitivity and range. In spite of the introduction of Wanze, U-boat sinkings continued without pause. On 5 November 1943, the order was sent out to stop using Wanze as well, as they believed it too might be tracked and that might explain its failure.[28] A new version, Wanze G2, reduced signal leakage even more, but at the cost of reducing the system's range. This produced no further improvement in the situation at sea.[29]

Yet another system, Borkum, was introduced in the summer of 1943. Borkum was sensitive between 75 and 300 cm, still outside the range where it might detect the Mk. III. Borkum had significantly less sensitivity than the Wanze, but further reduced leakage to the point that command felt it was safe to use under any circumstances. Once again, the sinkings continued without pause.[29]

For whatever reason, it was not until September 1943 that they finally began to consider the possibility of 10 cm signals. This was about the time the Luftwaffe was introducing the Naxos radar detector to allow their night fighters to track H2S radars. In a hurried conversion, the receiver was adapted to a new antenna and introduced that month. Naxos was a fragile device and its antenna had to be removed in order to dive; the commander of U-625 drowned while he struggled to remove the antenna. On top of these problems, Naxos offered very short-range detection, on the order of 8 kilometres (5.0 mi),[30] so even if it detected the Mk. III it offered very little time to dive to safety.[29]

Several improvements to the Naxos were introduced during 1944, notably the new "Fliege" antenna that did not have to be removed to dive. Additionally, fliege offered not only reception but reasonable directivity, allowing it to provide initial aiming for the anti-aircraft guns. A further improved antenna, "Mücke", added antennas to detect 3 cm signals when an H2S unit working on the frequency had been recovered from an RAF bomber. However, Coastal Command never moved to this frequency on any large scale.[29]

Further efforts to understand the British radars led to dedicated missions with highly instrumented U-406 and U-473, both of which were sunk.[31] In spite of all this effort, Naxos was never a convincing solution to the Mark III problem. It would later become clear that the Battle of the Atlantic was won with the introduction of Mark III.[20]

Improved versionsEdit

IIIAEdit

Shortly after the first IIIs arrived, a minor upgrade was worked into the line, producing the Mark IIIA, or ARI.5153. Although there were a number of minor differences in the equipment, the main difference was the addition of the Lucero system.[16]

Lucero was a transceiver tuned to the 1.5 m-band radio beacons and transponders used for navigation and IFF Mark III.[32] Lucero's 500 W transmitter periodically sent out signals near 176 MHz, or could be switched to the Blind Approach Beacon System (BABS) at 173.5 MHz. When these signals were received by ground-based transponders, the transponder would reply with a short pulse of its own, typically with much greater power. This pulse was picked up by the Lucero receiver, amplified, and sent to the ASV or H2S's height scope. Lucero sent a signal every third pulse of the radar and synchronized to begin at the same time, so that the same time base generator as the normal radar signals could be used.[33]

Two antennas were used, and a motorized switch switched the receiver between them every 4 or 5 signals to produce lobe switching. The switch also turned on a signal inverter on the height scope so that signals from the left-side antenna caused deflection to the left, instead of the normal right side. The result was two "blips" on the height scope, and by comparing their amplitude, the radar operator could determine the direction of the beacon relative to the nose of the aircraft.[33]

As the number of beacons proliferated, a significant problem with spectrum overcrowding emerged. This led to the movement of the Rebecca/Eureka system to the 214 to 234 MHz band, which in turn led to new versions of Lucero that could be used with this system.[32]

IIIBEdit

By the end of 1943 a number of major improvements had been rolled into the H2S systems, most notably more efficient antenna designs, the use of waveguides instead of coaxial cables, roll stabilisation, "north-up" display, and height-corrected displays that showed ground distance instead of slant range. These were of less interest in the ASV role, especially the ground-range modifications which were not necessary due to the low altitudes being flown by the aircraft that meant the slant range was not too different than the ground distance.[34]

As Coastal Command was not interested in adopting many of these changes, the decision was made to introduce the first custom ASV system, Mark IIIB. The major change was to allow the operator to expand the "zero ring" as the aircraft approached the target, keeping the target blip near the outside edge of the display instead of it naturally approaching the centre of the display. This meant the blip was larger on the display which improved the angular resolution, from the original system's ~6° to about 1.7° within the last 1,000 feet (300 m) of the approach.[34]

Other changes were more minor. Before the introduction of the height-range adjustments on the newer H2S, this adjustment was carried out with a simple mechanical calculator called the "height drum". As this was not needed for the ASV role, the range lines used for this calculation were removed from the drum, and replaced by a line with fixed steps indicating 1 mile (1.6 km) ranges that could be used with BABS without having to look over at the drum to estimate the range to the airfield. Finally, the "strobe", a small blip created by the range drum system that was displayed on the height scope, was no longer adjustable and instead fixed at 1 mile range, used to time the use of the Leigh Light.[34]

IIICEdit

 
The well-faired radomes of the Mark IIIC produced less drag than the large antenna sets of the Mark II.

One of Coastal Command's primary aircraft was the Short Sunderland flying boat, which by 1943 was a major part of its fleet. These had been successfully using ASV Mark II, whose antennas were mounted under the wings or on either side of the fuselage. Mark III presented a problem, however, as the nose and belly locations that gave the required all-round view could not be used due to the boat hull of the aircraft. This led to a modified version known as Mark IIIC.[35]

IIIC used two scanners, one under the outer section of each wing. Their rotation was synchronised to a single drive, and the radio signal switched between them during the rotation. To maintain coverage in the important dead-ahead area, the signal did not switch to the port-side (left) scanner until 15° past dead-ahead, so the starboard-side (right) scanner covered 195°, not 180. The signal was provided by a single magnetron, piped to the scanners via a waveguide run through the leading edge of the Sunderland's massive wing.[35]

In tests carried out in April 1944, the IIIC demonstrated greatly improved performance over the Mk. IIIs in Wellington and Halifax, as much as double, although the reasons were never fully determined.[34]

Sea return discriminatorEdit

Large waves have vertical sides that reflect radar efficiently, and this causes false returns on the display. In high sea states this can completely fill the display with noise, rendering the system useless. This led to experiments with a "sea return discriminator" to help filter these out.[36]

The discriminator was simply a high pass filter that muted any low-frequency components of the signal as it exited the amplifiers. This caused a -3 dB reduction in signal below about 40 kHz. In experiments in March 1944, it was reported that the system eliminated wave clutter in medium sea states, and greatly reduced it in high states. Although it also reduced the signal returned from targets, a good operator could adjust the set so it was not adversely affected for tracking.[36]

ReplacementEdit

When Metox was first introduced, the TRE responded with two possible solutions. One, ASV Mark IIA, was a more powerful version of the original Mk. II, but also included an attenuator known as "Vixen". The idea was that the radar operator would mute down the signals as they approached the submarine, thereby hiding the fact that the aircraft was approaching. The second idea was to move to a new frequency, which became the Mk. III. In testing in January 1942, Mark III proved itself the far superior solution and Mk. IIA was dropped.[12]

When Mark III was being introduced, its developers at the TRE felt the Germans would quickly extend the frequency response of Metox so that it would see the new signals and the cycle would repeat. They wanted to be ready for this possibility and began several developments in order to be able to rapidly introduce new models as soon as it became evident this was occurring. As was the case with Mark II, they considered two possible solutions, a more-powerful version of Mark III with an attenuator, and the move to a new frequency. These emerged as Mark VI and Mark VII, respectively.[37]

It was not until October 1943 that RAF crews began noticing the return of the "disappearing contacts" problem, which corresponds to the introduction of Naxos. Given this unexpected delay in countering Mark III, both models were well advanced. Nevertheless, it was not until February 1944 that Mark VI began installations on the Wellingtons.[38]

Mark VIEdit

Two types of attenuator were introduced for the Mark VI effort.[37]

Type 53 consisted of two wire rings, each ​14 wavelengths long, positioned on either side of the waveguide between the magnetron and the antenna. When the rings were rotated so they were parallel to the waveguide, they did not see the signal and did nothing to the propagation. However, when they were rotated to be perpendicular to the waveguide, they began to resonate and gave off a signal that, due to Lenz's law, opposed the original signal, muting it. Unfortunately, these loops also attenuated the received signal, and this was the reason for the move to the 200 kW CV192 magnetron, compared to the original 40 kW version.[37]

An improved attenuator, Type 58, was worked into the system. This added a Sutton tube to the loops, so that they could be switched out of the circuit during the receiver period, allowing the full signal to reach the receiver. With the added power of the new magnetron, units with the Type 58 had significantly improved range.[37]

A further improvement worked into the Mark VI project was the addition of a lock-follow system. It was found that the operators had difficulty reading the extended blips on the display and turning that into an accurate angle to guide the pilots. To address this, the Mark VIA added a lobe switching system with two closely spaced antennas that could measure the slight difference in signals strength between the two and use that to directly guide the motors turning the antenna. Once turned on, the system automatically followed the target with an accuracy far better than the human operators. Ultimately the lock-follow system proved problematic, and it was not available until the U-boat bases in Biscay had been abandoned following D-Day.[38]

Mark VIIEdit

The other solution to the potential microwave Metox detector was to move to a new frequency. This was becoming possible in 1943 as the first magnetrons operating at 3 cm, the X-band became available, already in use for trials during 1943 to upgrade the H2S set further for aerial work. Moving to 3 cm band offered another tremendous advantage — the optical resolution of a radar system varies with the antenna aperture and inversely with the wavelength. In the case of ASV, the 28 inches (710 mm) antenna produced a beam that was about 10° wide, although it was most sensitive near the centre. The signal from a submarine was returned when it was anywhere within the centre section, perhaps 5° on either side, and appeared on the display not as a distinct spot but a 10° wide or greater arc.[39]

This was not a problem on its own, as the operator knew the submarine was located near the centre of the arc. However, any other large objects at the same range would also produce similar arcs, and these might overlap the target's. At long range, these could be miles on either side, and in medium to high sea states, large waves anywhere in the vicinity of the submarine would obscure its return. Moving to 3 cm improved the beam width to about 3°, and thus made the arcs much shorter. In order for waves to obscure the submarine, they had to be much closer to it. This greatly increased the level of sea state that the radar could effectively operate in.[40]

While the advantages of the X-band systems were obvious, it was also clear that Bomber Command was planning on using the same magnetrons. It seemed likely that Coastal Command would once again lose the argument over supply for UK-built units. In the end, Mk. VII was not ordered into production, in favour or similar X-band units that would soon be available from the US. The small number of units produced during the development phase were instead used for air-sea rescue aircraft,[41] where their higher resolution allowed them to detect small lifeboats.

DescriptionEdit

ASV Mark III vs. H2S Mark IIEdit

The original Mark III was largely identical to the H2S Mark II, with the exception being the antenna system. H2S used a 36 inches (910 mm) reflector designed to spread the signal out in a wide vertical angle in order to illuminate the area below the bomber as well as in front of it. The system for ASV modified the design, reducing its width to 28 inches to fit under the nose of the Wellington, and reshaping it to send less energy downward as the aircraft would be flying at low altitude and the area under the bomber was relatively small and did not need to be covered. Another change was to replace the H2S's coaxial cable power feed with a cable that ran to the scanner unit, and then switched to waveguide and feedhorn on the antenna itself. This modification was later applied to H2S Mark IIA as well.[42]

The IIIC installations on the Sunderland had two separate and non-interchangeable antennas, Type 12 and 53. The were fed via a waveguide running through the wing structure, connected to a magnetron in the fuselage. This was combined with an additional switch, Switch Unit 205, which sent the magnetron output alternately to the two scanners as they rotated. The Type 205 consisted of a muting unit similar to the Vixen system, which alternately muted one output and then the other as the loops were rotated.[19]

Physical layoutEdit

The ASV/H2S system consisted of four main components among a total of eleven packages. At the heart of the system was the Waveform Generator Type 26, which was also known more generally as the modulator. This acted as a master clock for the entire system, triggering the output of the magnetron, switching the system from transmit to receive, starting the trace on the CRT display, and other duties. The modulator was connected directly to several of the main components, and even more through a Junction Box.[43]

The radar signal was generated by the 40 kW peak CV64 magnetron that was part of the Transmitter/Receiver unit, TR.3159 or TR.3191 depending on the version. This fed a signal to both the antenna as well as a CV67 klystron. Magnetrons produce slightly different output with every pulse, which makes it difficult to build a receiver that can match this varying signal. The CV67 picked up some of the output pulse and began to resonate at that frequency, providing a steady reference signal for the receiver.[44]

The Transmitter/Receiver was also responsible for the first part of the overall receiver system. A CV43 Sutton tube switched the antenna from the transmitter to receiver side of the system after the pulses were sent. From there it was modulated by a CV101 diode, one of the earliest examples of military-grade solid state electronics and a key element of microwave radars. After the diode, the signal had been reduced in frequency from ~3,300 MHz to a 13.5 MHz intermediate frequency that was then fed back through the aircraft in a coaxial cable to the receiver/amplifier.[44]

The Receiver, T.3515 or T.3516, took the 13.5 MHz intermediate frequency and amplified it to usable levels. The output was sent to the Indicating unit Type 162, which contained the two CRTs. If it was so equipped, the Lucero receiver, TR.3190, was connected to the height display, sitting (electrically) between the receiver and display. Which of these circuits was in use, along with many other controls, was located on the Switch Unit. This also required the use of the Control Unit 411, which timed and powered the scanning system.[44]

Displays and interpretationEdit

The main display on the Mark III was a 6 inches (150 mm) CRT. When the Waveform Generator fired, it triggered a time base generator that pulled the electron beam outward from the centre of the display to the outer edge in the same time as the maximum return from the radar at the current range setting. For instance, when the system was set to its typical 30 miles (48 km) range, the radar signals would take 30 miles / 186,282 miles per second = 0.00016 seconds to travel out to 30 miles and then the same to travel back. So at this setting, the timebase pulled the beam across the face in 0.00032 seconds, or 320 microseconds. The system could be set to scan at 10, 30 or 50 miles, and had a separate mode for long-range Lucero use that displayed signals in the 50 to 100 miles (80 to 161 km) range.[44]

In addition to the time base, a second system rotated the CRT's deflection yoke in synchronicity with the scanner using a magslip. This meant that the line being drawn by the time base was rotating around the screen. When a target returned a signal, it would brighten up the beam. By adjusting the overall brightness of the display, the operator could set it up so that targets appeared as bright patches while the rest of the signal was muted down so it was invisible. They had to continually adjust the system so that it was not muting too much and making real returns invisible as well.[45]

Because the antenna had about a 10° beamwidth, the target did not appear as a single spot on the display, but an extended arc. This was, in theory, over 10° wide as the return might be seen when the antenna was on either side of it, but in practice, the arc tended to be perhaps half that as the signal strength on the edges of the beam was lower. This did not effect the accuracy of the system during the initial approach as the U-boat was somewhere near the middle of the arc, and when it was near the outside of the display this might be a couple of inches wide. However, as the aircraft approached the target the return moved towards the centre of the display where it became progressively smaller, and it was estimated that the average accuracy in heading at close range was only 6°. In later versions this could be addressed by adjusting the unit to push nearby returns out to the edges of the display, using a control originally intended to do the reverse in H2S settings.[9]

In addition to the radar returns, the display also had controls on the switch box to display a "strobe" at a fixed delay. This caused a spot to appear a certain time after the trace began, and as the display rotated, this created a circle on the display. This was used by the operator to make accurate measurements of the range to a selected target, which was displayed on the switch box by rotating the Range Drum. Like H2S, the ASV displays also had the option to display a solid line extending from the middle to the edge that represented the flight path of the aircraft. In H2S use, this feature was used because a second system rotated the entire display so that north was always up, like a map. Coastal Command aircraft lacked this system, likely due to a shortage of Distant Reading compasses that fed this information to the display. This heading-indication line was typically not used in ASV, and the associated Control Unit Type 218 was not carried.[46]

In addition to the main PPI display, there was a secondary 2.5 inches (64 mm) CRT known as the Height Tube. This lacked the system to rotate the display with the antenna, and always drew a line vertically up the display.[42] Receiver signals did not cause the beam to brighten, but instead deflect to the right, causing a blip to appear. A strobe like the one on the PPI could be moved along this display.[46]

As the name implies, the main purpose of the Height Tube was to measure altitude. The operator would move the strobe onto the first major blip, which was caused by signals reflecting off the ground and being picked up in the antenna's sidelobes. This was not as useful in the ASV role, where the low-altitude flights made it easy to measure altitude visually. In ASV, the Height Tube was used primarily with Lucero for beacon tracking.[47]

The separate Switch Unit Type 207 contained most of the controls for range and mode selection. It also included the Range Drum, a simple mechanical calculator. This was the location of the mechanical displays for the range and height strobes, the range being indicated by rotating the entire drum, and the height as an arrow-shaped pointer moving up and down the left side of the display. A radar measures the slant range to a target, not its distance measured over the ground. By reading a series of lines on the Height Drum where one of the lines intersected the tip of the height arrow, the operator could read off the ground distance to the target.[48] This feature was of little use in the ASV role, where the low altitude flights meant the slant range was similar to the ground range, and was later modified to be used primarily with the BABS system.[49]

LuceroEdit

 
Duxford's Sunderland V has bright-yellow Lucero receiver antennas on either side of the nose.

When the Switch Box selected Lucero, the height display was switched off the main signal and connected to the Lucero antennas. There were two receiver antennas, one on either side of the aircraft. A motorised switch rapidly selected between the two antennas. One of the two was also sent through an electrical inverter. When amplified and sent to the display, this caused two blips to appear, one on either side of the vertical baseline. The longer blip was more closely aligned with the transponder on the ground, so by turning toward the longer blip one could navigate the aircraft towards it.[33]

PerformanceEdit

The performance of Coastal Command operations was a significant area of operational research throughout the war, and the Mark III was repeatedly tested both in its own performance as well as relative measures against other radar systems.[41]

In its first major test series, a prototype Mark III was test flown against the high-power Mk. IIA, and a purely experimental system working at 50 cm. The Mk. IIA demonstrated reliable detection a fully surfaced submarine at 14 miles (23 km) at 1500 ft, 11 miles (18 km) at 1000 ft, and 7 miles (11 km) at 500 ft. Against a submarine trimmed down so the deck was closer to the waterline, the ranges were 7 miles at 1500 feet, 6 miles at 1000 feet and 4 miles (6.4 km) at 500 feet. Minimum ranges varied from 3 miles to 1 mile.[11]

In comparison, the prototype Mark III, referred to as 10 cm ASV in the report, turned in much better results. Large convoys could be detected at ranges of up to 40 miles (64 km) while flying at an altitude of 500 feet, which meant the ships were well below the radar horizon and the aircraft was invisible to them. Other aircraft could be reliably seen at a range of 10 miles (16 km) and the operator could make some estimate about their direction of travel. Reliable maximum ranges against a fully surfaced submarine were 12 miles at 500 feet and 10 miles at 250 feet. It was these tests that convinced Coastal Command to move forward with Mark III as their primary system.[12]

In November 1944, similar comparisons were carried out between Mark III and Mark VI, and then compared to earlier tests of the Mark VII from that August. Using Grassholm Island off the coast of Wales as a target, Mk. III provided an average detection distance of 23.5 miles (37.8 km), while Mk. VI's more powerful signals improved this significantly to 38.5 miles (62.0 km), and the Mk. VII's weaker 25 kW demonstrated a maximum around 35 miles (56 km). Mk. III was estimated to detect a U-boat from the side at 22 miles (35 km), improving to 32 miles (51 km) for Mk. VI, and as low as 18 miles (29 km) for Mk. VII. The range against end-on targets was 10.5 miles (16.9 km), 20.5 miles (33.0 km) and 10 miles (16 km), respectively.[50]

NotesEdit

  1. ^ This is the basic reason for convoys, it is easily demonstrated that one large group is much less likely to be detected than the same number of boats travelling separately.[21]

ReferencesEdit

CitationsEdit

  1. ^ Bowen 1998, p. 38.
  2. ^ Smith et al. 1985, p. 359.
  3. ^ Smith et al. 1985, p. 360.
  4. ^ Smith et al. 1985, p. 362.
  5. ^ Smith et al. 1985, p. 363.
  6. ^ a b Smith et al. 1985, p. 368.
  7. ^ Rowe 2015, p. 159.
  8. ^ a b Lovell 1991, p. 157.
  9. ^ a b c Smith et al. 1985, p. 372.
  10. ^ a b c Watts 2018, p. 3-3.
  11. ^ a b c Watts 2018, p. 7-1.
  12. ^ a b c Watts 2018, p. 7-2.
  13. ^ a b c d e f Lovell 1991, p. 159.
  14. ^ Lovell 1991, p. 165.
  15. ^ a b c Lovell 1991, p. 158.
  16. ^ a b Watts 2018, p. 3-4.
  17. ^ a b Campbell 2000, p. 9.
  18. ^ Lovell 1991, p. 163.
  19. ^ a b c Smith et al. 1985, p. 374.
  20. ^ a b c d e f g h i j Lovell 1991, p. 166.
  21. ^ Office of Chief of Naval Operations. Anti-Submarine Warfare in World War II. US Navy.
  22. ^ Gordon 2014, p. 69.
  23. ^ Gordon 2014, p. 70.
  24. ^ Gordon 2014, p. 66.
  25. ^ Blair, Clay (1998). Hitler's U-boat War: The hunted, 1942-1945. Random House. p. 403.
  26. ^ Ratcliff 2006, p. 147.
  27. ^ Hanbury_Brown 1991, p. 311.
  28. ^ NSA, p. 7.
  29. ^ a b c d NSA, p. 8.
  30. ^ Watts 2018, p. 4-1.
  31. ^ NSA, p. 9.
  32. ^ a b Watts 2018, p. 6-1.
  33. ^ a b c Watts 2018, p. 6-3.
  34. ^ a b c d Watts 2018, p. 3-16.
  35. ^ a b Watts 2018, p. 3-15.
  36. ^ a b Watts 2018, p. 3-17.
  37. ^ a b c d Smith et al. 1985, p. 375.
  38. ^ a b Smith et al. 1985, p. 371.
  39. ^ Smith et al. 1985, See images of convoy, p. 377.
  40. ^ Smith et al. 1985, See images of X and K-band systems.
  41. ^ a b Smith et al. 1985, p. 377.
  42. ^ a b Smith et al. 1985, p. 373.
  43. ^ Smith et al. 1985, p. 372-375.
  44. ^ a b c d Smith et al. 1985, p. 372-373.
  45. ^ Watts 2018, p. 3-9.
  46. ^ a b Watts 2018, p. 3-10.
  47. ^ Watts 2018, p. 3-11.
  48. ^ Watts 2018, p. 3-12.
  49. ^ Watts 2018, p. 3-13.
  50. ^ Smith et al. 1985, p. 378.

BibliographyEdit