[4.0] Microwave Radar At War (1)

v2.0.6 / chapter 4 of 12 / 01 jan 15 / greg goebel

* Although the US ramped up preparations for war through 1941, the country was still caught off guard when war came in December. Once America was involved in the fighting, radar went to the top of the priority list. By 1942, new radars were coming into service on both sides of the Atlantic.

antennas on jeep carrier USS COWPENS



* On the morning of 7 December 1941, strike aircraft launched by a carrier task force of the Imperial Japanese Navy (IJN) attacked the American navy base at Pearl Harbor in Hawaii, causing extensive damage and many casualties. America was now at war with Japan. Although American President Franklin Roosevelt had no mandate to declare war on Germany as well, Adolf Hitler conveniently declared war on the United States on 11 December.

The attack on Pearl Harbor had an immediate effect on American radar efforts. The US Army had deployed five new SCR-270 longwave sets around the island of Oahu, and in fact two officers, Navy Lieutenant Commander William E.G. Taylor and Army Major Kenneth P. Bergquist, were trying to build up an "Information Center", something like a British filter room, to collect and integrate the information provided by the radars.

Taylor had been in the RAF during the Battle of Britain, flying with the RAF "American Eagle" squadron of expatriate Yanks, and had been impressed with British fighter control. He had acquired a good knowledge of it, and so the US Navy offered him a commission in the summer of 1941 to come back and help organize similar American efforts. Taylor ran into Bergquist in Hawaii and found a kindred spirit, but their superiors gave them little support and they had nothing close to an operational system.

As a result, when one of the SCR-270s picked up the Japanese strike force flying to the island, the information fell into a hole. The operator reported the intruders, but unfortunately a formation of Boeing B-17 bombers was flying in that morning from the US mainland, and they lacked IFF, so the warning was disregarded. In fact, the bombers arrived just in time for many of them to be attacked by Japanese fighters and destroyed.

There was also an SCR-270 at Iba Air Field in the Philippines, and it was able to pick up flights of Japanese aircraft as they attacked Clark Field and other targets on 8 December. The radar was bombed and destroyed.

* The radar blunders had a valuable silver lining, since the radars had functioned perfectly, the problem being that nobody had been prepared to make use of them. The military tends to be conservative and slow to pick up new technologies, but the lesson was painfully obvious, and the brass immediately wanted radars as fast as they could get their hands on them.

The Navy had the longwave CXAM radar, which was fitted to all the aircraft carriers, and the Army had the longwave SCR-268 and SCR-270 radars. On the outbreak of war, the Signal Corps quickly modified the SCR-268 for low-altitude warning by adding a PPI display, with the result designated "SCR-516".

For the moment, the US had no operational airborne radars of their own, but the NRL, no doubt holding their nose at having to use British gear, put ASV Mark II into production as "ASE" or, in Army service, the "SCR-521". The British had already put ASV Mark II on their Consolidated Catalina flying-boat patrol aircraft, and so it was straightforward to mount it on US Navy Catalinas as well, making the type the first US aircraft to carry radar in operational service.

Since ASE was too big to fit on smaller aircraft, the NRL built their own longwave set, the 58 cm (515 MHz) "ASB", originally "XAT", which was fitted to the Grumman TBF Avenger torpedo bomber. Like ASE, it featured a Yagi antenna fitted under each wing, skewed 7.5 degrees from the centerline, and presumably used lobe-switching. Peak power was 200 kW and pulse width was 2 microseconds. It was the first operational US carrier-based aircraft to be fitted with radar. The ASB was very widely used, with 26,000 units built.

The Americans also began to dive into the more abstract considerations of just how radar could be used in combat. The British had demonstrated remarkable foresight in developing the filter-room system well before the shooting started. The US military services were basically at square one on the issue, and were having to catch up as fast as they could while they were trying to react to one military disaster after the next.

* As the Japanese tide swept south to the north coast of New Guinea and the Solomon Islands chain, Australia mobilized against the threat of invasion. The Australians had created a "Radiophysics Lab" in November 1939 to work on radar and sent experts to England to learn about British radar technology.

The Australians used the British 1.5 meter (200 MHz) LW air-warning radar, with installations in place to protect Sydney only a few days after Pearl Harbor. LW sets were set up around Darwin on the north coast of Australia in early 1942 in response to Japanese air raids, and helped make further raids very expensive to the Japanese. The Australians produced their own version of the LW and it saw extensive service in the South Pacific.

The Yanks also used the British 1.5 meter (200 MHz) LW radar, building it themselves as the "SCR-602". Later on in the war the Americans would develop their own portable lightweight search radar, the "AN/TPS-1", which was designed by Bell Labs and built by Western Electric. It operated in the band around 24 centimeters (1.25 GHz), featured a PPI, and broke down into ten components for transport. It was a popular set, being adopted by the British and other US allies, and remained in service into the 1950s.

AN/TPS-1 radar

The Aussies also got their hands on some American SCR-268 longwave gun-laying radars early in the war. Since anti-aircraft guns were not available at the time, the SCR-268s were resourcefully adapted to air warning with some minor modifications as "Modified Air Warning Devices (MAWDs)".

* Better radars were on the way. The Rad Lab, whose long-term future hadn't always seemed assured, was now operating at full throttle. The Navy, the Army Signal Corps, and the USAAF (the Air Corps had been superseded by the "Air Forces" in June 1941) opened liaison offices there, a sign of improving cooperation, while the Rad Lab set up a Transition Office to work with industry to help move Rad Lab prototypes into assembly-line production. The lab expanded its facilities on the MIT campus, and new researchers were recruited.

Lee DuBridge reorganized the laboratory into ten divisions, and Army military police arrived to provide security. The Rad Lab would continue to grow for the rest of the conflict. By the end of the war, the organization would employ 4,000 people, have field stations all over the world, and have an "air force" of 95 aircraft.



* After the Battle of Britain, RAF Bomber Command began to ramp up night attacks against German cities. Unfortunately, although Bomber Command reported grand results from the raids, an independent analysis based on daylight air reconnaissance performed in the summer of 1940 showed that half the bombs fell on open country. Only one bomb in ten actually hit the intended target.

Radio electronics promised some relief. The British developed a radio navigation system called "Gee", mentioned earlier, and then a second long range navigation scheme known as "Oboe", with both discussed in more detail later. Gee and Oboe were limited in range to a line of sight to the transmitters; a bomber carrying its own, self-contained night targeting system would not be limited in range to a UK-based transmitter. Taffy Bowen had noticed during his early AI experiments before the war that the radar returns from fields, cities, and other areas had distinctly different appearances. He had suggested development of targeting radar, but the matter was forgotten in the chaos.

The idea resurfaced in 1941. Philip Dee's group had got a 10 cm (3 GHz) AI flying in a Blenheim in March of that year. The experimental set was known as "AIS" in reference to its S-band operation. During tests of the AIS, Dee's team rediscovered that radar reflections could reveal different types of terrain.

In October 1941, Dee attended a meeting of the RAF Bomber Command where the night targeting issue was discussed. After the meeting, on 1 November 1941 Dee performed an experiment in which he used an AIS radar mounted on a Blenheim to scan the ground. He was able to pick up the outline of a town 55 kilometers (35 miles) away. The brass were impressed, and on the first day of 1942, the TRE set up a team under Bernard Lovell to develop an S-band airborne targeting radar, based on AIS. The new targeting radar was designed to fit in a blister on the belly of a bomber, where the antenna would rotate to scan the terrain and feed the reflections to a PPI display, producing a map of sorts of the land below the bomber.

The targeting radar was originally designated "BN (Blind Navigation)", but quickly became "H2S". This acronym remains somewhat mysterious, with different sources claiming it meant "Height to Slope"; or, with a little rearrangement, "Home Sweet Home". The "S" might have also had some connection to "S-band", but it is plausible the acronym was deliberately obscure and misleading as a security measure.

H2S performed its first experimental flight on 23 April, with the radar mounted in a Handley-Page Halifax bomber. There was much still to be done. For example, in order to display as a uniform a "map" of the terrain as possible, the radar had to have low sensitivity or "gain" for targets directly underneath the bomber, with the gain increasing with the angle of the radar away from vertical. This scheme would become known as "cosecant-squared" scanning, after the mathematical function that defined the change in gain, and in was implemented by using a special antenna whose curvature provided the proper focus for the beam.

H2S was the TRE's top priority, and Lovell's team had use of the brilliant Alan Blumlein and other top EMI engineers, but there were snags. Intelligence reports had revealed the Germans had stationed a company of paratroopers near Cherbourg, across the channel, suggesting the enemy might be planning to raid the TRE. On 25 May, the entire organization moved out in another mad, infuriating fire drill from Swanage to Malvern College, about 160 kilometers (100 miles) to the north. Fortunately, this would prove to be the last move.

As if that weren't bad enough, then an outright disaster occurred. On 7 June 1942, the Halifax performing H2S tests crashed, killing everyone on board and destroying the prototype H2S. One of the dead was Alan Blumlein, and his loss was a major blow to the program.

Furthermore, Churchill's science adviser Lord Cherwell, previously Professor Frederick Lindemann, wanted the design team to build H2S around the klystron instead of the magnetron. Lord Cherwell was opinionated, obstinate, contrary, something like Churchill himself but without quite as many redeeming features. Most people who had to deal with Lord Cherwell regarded him, with some justification, as an obstructionist who tried to create problems instead of working at how to overcome them. He was not always wrong by any means -- he was one of the first in Allied technical leadership to realize the potential of nuclear bomb research -- but was opinionated, tactless, and almost always exasperating.

Lord Cherwell did not want the secret of the magnetron to fall into German hands. Once the Germans understood it, they would not only try to duplicate it, but could quickly develop countermeasures against it. The klystron wasn't as powerful as the magnetron, but it could be much more easily destroyed in an emergency than a magnetron. A magnetron's copper core could survive even large self-destruct charges. One of the best stories about the origins of the designation H2S was that Cherwell had declared: "It stinks!" -- so the design team named it after hydrogen sulfide, and invented the alternate designations as a cover.

The H2S design team did not believe the klystron could do the job, and in fact tests of an H2S built with klystrons instead of the cavity magnetron showed a drop in output power by a factor of 20 to 30. The H2S team also protested that it would take the Germans two years to develop a centimetric radar once the cavity magnetron fell into their hands, and that there was no reason to believe they weren't working on the technology already. The first concern would prove correct; the second would fortunately be proven wrong, though given the widespread parallel development of the cavity magnetron, in hindsight it wasn't an unreasonable assumption.

Despite all the problems, on 3 July 1942 Churchill held a meeting with brass and the H2S group, where he shocked the radar designers by demanding the delivery of 200 H2S sets by 15 October 1942. Bomber Command had to have H2S. The H2S design team was under extreme pressure, but they were given priority on resources. The pressure also gave them an excellent argument to convince Lord Cherwell that the klystron-based H2S program be finally dropped.

Despite the extraordinary efforts of the TRE, there was no way to meet the 15 October deadline. By 1 January 1943, however, twelve Short Stirling and twelve Halifax bombers had been fitted with H2S. On the night of 30 January 1943, thirteen "Pathfinder" bombers, which dropped incendiaries or flares on a target to "mark" it for other bombers following in the bomber "stream", took off to give H2S its introduction to combat by marking the German city of Hamburg for a strike. Seven of the Pathfinders had to turn back, but six marked the target successfully, which was hit by a hundred Lancasters.

The Germans did not know about H2S at the time. Unfortunately, on 2 February 1943, a Pathfinder was shot down near Rotterdam, and the Germans noticed the unusual gear in its wreckage. The British had been clever with electronics, and the Germans were careful to look for anything out of the ordinary in RAF aircraft forced down in the Reich. Elements of the H2S set were recovered, and German engineers began to work on the "Rotterdam Geraet (Rotterdam Device)", as they called it.

Bomber Command didn't use H2S in a big way until that summer. On the night of 24 July 1943, the RAF began Operation GOMORRAH, a large-scale systematic attack on Hamburg. With the target marked by Pathfinders using H2S, RAF bombers hit the city with high explosive and incendiary bombs. They returned on the 25th and the 27th, with the USAAF performing two daylight attacks in between the three RAF raids. Large parts of the city were burned to the ground by a terrifying cyclone of fire. About 45,000 people were killed.



* Meanwhile, the British had been working hard on countermeasures against Doenitz's U-boats through 1941, and had been getting results. Merchant vessel sinkings by U-boats decreased, and German submariners judged that the First Happy Time was over.

As attacks on Atlantic convoys grew less rewarding, in January 1942 Doenitz shifted the focus of submarine operations to the American East Coast in what he codenamed Operation PAUKENSCHLAG (DRUM ROLL). American antisubmarine defenses were almost nonexistent, and the U-boats had another party, sometimes sinking merchantmen in sight of major ports. The U-boat crews called it the "Second Happy Time".

The British had proven that ASV radar was an important weapon against the U-boats, and now the Americans went through the same painful learning curve. The longwave ASV Mark II / ASE fitted to Catalinas and ASB fitted to Avengers were useful, but not completely adequate. Development of centimetric ASV now became the Rad Lab's top priority. The initial ASV was a fast modification of the Western Electric SCR-520 AI, the "SCR-517", which was installed on USAAF Liberators for ocean patrol beginning in the spring of 1942.

With increasing numbers of air patrols and more convoy escorts, the US East Coast became unsafe for Doenitz's U-boats. They transferred operations to the Caribbean, but by July 1942 rising American vigilance there forced Doenitz to give up PAUKENSCHLAG completely. The Americans had gone through hard lessons in their attempts to deal with the U-boats, but were finally beginning to get on top of the learning curve.

* The British had the benefit of more experience in antisubmarine warfare than the Americans and were correspondingly better organized, though they had difficulties of their own. In June 1942, the British had managed to compensate for the long minimum range of ASV Mark II by strapping a high intensity searchlight known as the "Leigh Light" under the wing of patrol aircraft. When a patrol aircraft flew closer to a U-boat than the minimum range of the radar and lost radar contact, the Leigh Light was switched on, trapping the submarine in its glare. The number of U-boat kills by aircraft rose dramatically, and TRE work on centimetric ASV dropped in priority.

The success was short-lived. In August, the Germans deployed a radar detector designated the "R600A", called "Metox" after the French factory that built it. The Metox swept wavelengths from 3.8 to 0.9 meters (79 MHz to 79 MHz), using an antenna named the "Cross of Biscay" that was mounted on the hull of the submarine. The antenna could be quickly removed and stowed before the U-boat dived. Metox provided a warning that allowed a submarine to dive and escape, and negated the British advantage. Most U-boats were fitted with Metox by September 1942, and ASV Mark II was reduced to a system for alerting a U-boat that a patrol aircraft was around. Making a submarine dive was all for the good, but it was better to kill the damned thing.

The TRE effort to develop centimetric ASV went back up the priority queue. Lovell's team was able to perform a few minor modifications to H2S to produce "ASV Mark III", which was a great advance over ASV Mark II. However, there was no way to produce both H2S and ASV Mark III in the short run, and Bomber Command insisted that H2S had priority.

That meant asking the Americans for help. Winston Churchill wrote Franklin Roosevelt to ask him to "consider the allocations of some thirty Liberators with centimetre A.S.V. equipment from the supplies which I understand are now available in the United States." Helping the British was high on Roosevelt's agenda, and by January 1943, Liberators with SCR-517 centimetric ASV, as well as centimetric ASV sets to be installed in other aircraft, were being sent across the Atlantic. They weren't arriving fast enough. The U-boats had regained the edge in the Battle of the Atlantic. In the first 20 days of March 1943, the Atlantic convoys lost 95 vessels, against the destruction of only twelve U-boats.

* Naval planners despaired, but the edge the U-boats possessed was actually very thin. More Allied escort vessels and patrol aircraft were available every month, with processes and tactics undergoing continual refinement. Codebreaking played a major role, as well as other improved technologies. One simple new weapon, the "Hedgehog", tossed out a pattern of contact-fused rocket-boosted bombs ahead of a ship, allowing an escort vessel to attack a U-boat without passing over it and losing sonar contact.

A particularly interesting new antisubmarine detector was the "sonobuoy", which was a buoy containing hydrophones and a radio transmitter. An aircraft could drop a sonobuoy to help track a submerged U-boat. Professor P.M.S. Blackett had come up with the idea in 1940, the initial notion being that they would be dropped by escort vessels to spot U-boats trailing a convoy. RCA developed the device for the British, but it didn't make it into the development queue.

Just before Pearl Harbor, the US Navy had demonstrated another sensor scheme, the "magnetic anomaly detector (MAD)", which could sense the slight variation in the Earth's magnetic field that occurred when an aircraft flew over a U-boat or other large submerged metal object. MAD was carried by US Navy blimps on antisubmarine patrols and much was expected of it, but its use turned out to be something of an art form. It had limited sensitivity and range, and tended to have a high rate of false alarms, since it could pick up old wrecks and other metal objects, or be spoofed by natural variations in the Earth's magnetic field.

The RCA sonobuoy seemed like it could help with the problem, with a blimp using sonobuoys to listen for the presence of a U-boat and then using MAD to close in on it. Field experiments were performed with prototype technology in March 1942, with a US Navy blimp hunting a Navy submarine.

Of course, if sonobuoys could be dropped from blimps, they could be dropped from heavier-than-air aircraft as well. An aircraft test drop was performed by a Douglas B-18 Bolo bomber in July 1942. The USAAF became enthusiastic about the sonobuoy, and ordered over 6,400 of them in 1942. The "AN/CRT-1A" sonobuoy was in formal service by 1943, initially carried by USAAF Liberator ocean patrol aircraft out of Newfoundland and the UK. The US Navy hadn't followed through on the blimp experiments, but the USAAF experience convinced the Navy that the sonobuoy was a good idea, and the Navy bought almost 60,000 AN/CRT-1As during the war.

The AN/CRT-1A had a detection range of about 5.5 kilometers (3 nautical miles) and a radio range of about 9 kilometers (5 nautical miles), could transmit on one of six radio channels, and could operate for about 6 hours. It was a passive unit, only fitted with an omnidirectional hydrophone, and precise targeting had to be performed by MAD. Since a MAD contact only occurred when the carrier aircraft passed over a target, a special retrorocket was designed by the Jet Propulsion Laboratory at the California Institute of Technology to allow depth charges to be dropped behind the patrol aircraft's flight path.

Yet another useful tool was "high-frequency direction finding" or "Huff-Duff", which was carried by escort vessels to pin down U-boat radio transmissions. The history of Huff-Duff, incidentally, is extremely obscure. It appears, and would be consistent with other Allied electronics efforts, that the British and Americans built their own different versions. The British version seems to have owed a great deal to Robert Watson-Watt's thunderstorm-location system of the 1920s. The American version was actually French-designed, implemented by French researchers who had fled to the US after the fall of France in the spring of 1940. The researchers initially insisted that the Huff-Duff antenna be mounted above the radar on a destroyer's mast, but eventually destroyers featured separate masts for Huff-Duff and radar.

* Of course, to get back to the main topic, centimetric radar was a big help in the war against the U-boats. The early SCR-517 centimetric ASV radar had a number of obvious limitations, particularly in that it only swept forward of the aircraft, limiting the radar's field of view over the ocean, and lacked a PPI display, meaning that it could only really observe one target at a time.

It was accordingly followed by the improved Philco-built "ASG" or "AN/APS-2", known as "George" to its users, which scanned full-circle, displaying echoes on a PPI display, and had a longer range of 24 kilometers (13 nautical miles). Reliability and operator training were poor at first, but as spring came it helped the Allies gained the upper hand in the Battle of the Atlantic.

The USAAF built their own version of the ASG, the "SCR-717", which differed in using a "B-scope" display and not a PPI display. A B-scope display gives a view somewhat like that of some early video games, with the operator staring down on top of a rectangular radar map of the area ahead of the radar, with the target tracked left or right and forward or back on the map. The SCR-717 supplanted the SCR-517 in USAAF service.

The next step after ASG was to build a 3 cm (10 GHz) X-band ASV, which emerged as the "ASD", or simply "Dog", and was produced as the "AN/APS-3" by Philco. It could be mounted in a pod under the wing of an Avenger torpedo-bomber, and proved much more effective than the Avenger's earlier longwave ASB radar.

Centimetric ASV had greater maximum and shorter minimum range than longwave ASV, was much more accurate, and Metox couldn't detect it. U-boats were hit without warning on the surface at night and in low visibility. Escort vessels were also fitted with centimetric ASV, allowing them to hunt down U-boats at night. Doenitz and his senior officers were baffled, suspecting at first that the sinkings of their submarines were due to the work of spies. They were further confused when a RAF Coastal Command prisoner told his captors that RAF planes were homing in on emissions from Metox, leading to an order for the removal of Metox from all U-boats.

That exercise was an act of desperation, and its only real effect was to make U-boat crews, a justifiably jumpy group to begin with, distrustful of all electronic equipment. In May, the Allies sank 38 U-boats, and lost only 34 ships in the North Atlantic. Doenitz finally had to call off the campaign. He found out about centimetric radar later in 1943, but it was too late. The U-boats operated on the margins of the conflict for the rest of the war.



* While centimetric ASV had been the Rad Lab's priority effort, radars for other platforms were not ignored. Bell Labs had developed a longwave fire-control radar, initially known as the "CXAS" and then the "FA" or "Mark 1", and then the improved longwave "FB" or "Mark 2". These radars were basically just rangefinders.

The Mark 1 was built in small numbers. The Mark 2 was on the drawing board when the magnetron came along. Since the design of the Mark 2 was modular, the design team found it straightforward to adapt it to centimetric wavelengths. This exercise produced a 40 cm (750 MHz) surface fire-control radar, the "Mark 3" or "FC", with horizontal lobe switching to give it a horizontal targeting capability; as well as a 40 cm (750 MHz) antiaircraft fire-control radar, the "Mark 4" or "FD", which added vertical lobe switching. Both these radars were in production by late 1941. They were eventually linked to the vessel's gun gyroscopic stabilization systems to improve targeting in rough seas.

The Mark 4 proved to have difficulty determining the altitude of low-flying aircraft, due to reflections from the surface of the water. This led in 1943 to the "Mark 12", which was essentially an improved Mark 4, coupled to a smaller X-band height-finding radar, the "Mark 22", mounted alongside. The Mark 22 had some conceptual similarities to the British AMES Type 13 CMH, but many differences as well. The Mark 22 had an "orange slice" antenna, in the form of a curved, narrow, elliptical grid mounted with the long axis vertical, to focus the radar beam in a narrow horizontal fan. As with its British equivalent, the antenna nodded to sweep the sky for targets. A land-based version of the Mark 22 was also built, known as the "AN/APS-10" or "Little Abner", named after the hillbilly in Al Capp's popular comic strip who liked to sit in his rocking chair. The British also used the AN/APS-10 as the "AMES Type 60".

Mark 12 radar & Mark 22 height-finder

* The next level of sophistication involved addition of "electronic steering", in which adjusting the relative phase of the waveforms supplied to array elements shifted the direction of the beam without mechanically moving the radar. A radar using electronic steering is known as a "phased array radar".

Bell Labs used this approach to build a new X-band surface fire control radar, the "Mark 8" or "FH". The Mark 8 had a very unusual appearance. It was based on an antenna element known as a "polyrod", which was a pipelike microwave waveguide with a polystyrene plug in the end. The Mark 8 featured an array of 42 polyrods, organized as 14 rows of three. Signal phase to each triplet of polyrods was controlled by mechanically switching electronic delay elements into the output signal path. If the signals were sent in phase to all the triplets, the beam went straight out forward. If the signals were delayed from one end of the row to the other, the beam was diverted in the direction of the delay.

Mark 8 radar

The Mark 8 provided increased accuracy, with a beam width of 2 degrees that could be swept over a 30 degree arc, and a 0.4 microsecond pulse width to provide tight range accuracy. Of course it did not provide height information. A high peak output power of up to 20 kW gave it excellent range, and it also featured a plan-type display that made it much easier to locate and pinpoint multiple targets in the radar's field of view.

Following tests of the prototype "CXBA", the Mark 8 was put into production by Western Electric in October 1942. While phased array technology predated the war and had already been implemented in certain German longwave radars, the Mark 8 was the first microwave phased-array radar. A derivative of the Mark 8 with an auxiliary height-finding radar, designated the "Mark 14", was introduced late in the war.

The fire-control radars were mounted on top of turreted "directors" that could be rotated towards the direction indicated by long-range search radars. The radars would be linked to the guns through a "ballistic computer", or analog tracking system, fitted in the interior of the ship. The Navy learned to use their surface fire-control radars to indirectly pinpoint ground targets during the island-fighting campaigns. Prominent land features were identified from aerial reconnaissance photographs, the features were ranged with radar, and the ranges were used to obtain a position fix for the ship. It was then relatively simple to map out ground targets for fire. The scheme worked in day or night, and all but the worst weather.

* The US Navy stayed with the long-wavelength search radars search radars such as the SK through the war, but such devices had their blind spots. However, the Rad Lab's work with the experimental S-band centimetric radar on the destroyer SEMMES went very well, and the NRL turned it into an operational set, the Raytheon "SG". The first operational SG set was installed on the cruiser USS AUGUSTA in April 1942, and was in action by the fall of 1942.

The SG featured a rotating elliptical parabolic antenna, producing a beam 5 degrees wide and 15 degrees high, along with a PPI scope. The SG was gyroscopically stabilized to ensure level scanning in rough seas. It gave naval radar operators a neat electronic map of their surroundings, and its usefulness was so obvious that it was an instant hit, successful beyond the expectations of its designers. A warship captain could use SG to see 360 degrees all around his ship in the dark and in the fog, identifying shorelines, tracking other ships in a convoy, and spotting intruders such as enemy submarines.

SG did have a limitation relative to the longwave CXAM. For one, it was really optimized for surface search and was not good at height finding. Another S-band set, the "SM", of which more is said later, was developed and fitted on aircraft carriers as a complement to the SK, where it was used for height-finding and close-range fighter direction.

Another limitation of the SG was that it did not have a switch to change the PRF for sorting out ghost echoes. This would lead to an unusual naval action off the Aleutians on the dark hours of the morning of 26 July 1943, when a US Navy surface force spent about a half hour bombarding empty ocean. Intelligence had reported that there was a Japanese force in the area, and atmospheric conditions had allowed ghost echoes to be returned from local islands at unusually long range, probably from a phenomenon known as "ducting" in which the radar beam bounced back and forth between a low and a high atmospheric layer.

Given that US Navy crews had good reason to fear getting into a shootout with the Imperial Japanese Navy at night -- night actions were an IJN specialty and the Japanese were murderously skilled at them -- it was no great surprise that they opened up on the targets even though the returns were intermittent. Some of the radar operators got suspicious when they realized that the echoes didn't seem to get any stronger as they closed range, and that the returns from the impacts of American shells were stronger than those of the presumed targets. It is unclear if the switch was ever added to the SG, but it appears that it was a feature with some later sets.

A lighter variant of the SG designated "SF" was built for destroyers, and a similar coastal defense set designated the "SCR-582", with a PPI and a 1.2 meter (4 foot) dish, was put into production and used for harbor defense. The SCR-582 proved very successful during the American campaign in North Africa, and was also updated in the field to act as an air-defense radar. Although the SF program had been focused on design of a microwave radar that could be carried on the smallest vessels, the SF still turned out to be too big, and so further work was done on an even lighter S-band radar that emerged as the "SO". It was mounted on torpedo boats, landing craft, and other small vessels in a thimble radome.

* It took some time for the US Navy to learn how to make use of radar. Not only was the technology new and unfamiliar, but many US Navy officers had a peculiar complacency, believing in staggering obliviousness to the evidence that the IJN was a pushover.

The Japanese weren't informed of this fact. On the night of 8 August 1942, a Japanese surface force ambushed an American-Australian force supporting the invasion of Guadalcanal in the Solomons. Although the Allied force had radars, even an SG, and the Japanese had no radars, the well-trained Japanese hit the Allied force near Savo Island, sending four cruisers to the bottom and chewing up a fifth with no serious harm to themselves. Only their timidity prevented them from remaining into the day to finish off the landing force.

Longwave airborne warning radars like CXAM (and, on shore, the SCR-270) did help in further actions off Guadalcanal, though the lack of both experience and IFF was a great hindrance. In the meantime, the US Navy learned how to make use of radars and incorporate them into training exercises.

They put this knowledge into use off of Cape Esperance on Guadalcanal on the night of 11 October 1942, when a US surface force sank a Japanese cruiser and three destroyers, for the loss of one American destroyer. This was a more significant win than the simple tradeoff in kills would have suggested, since the tradition of Japanese invincibility in night naval actions had finally been broken.

The US Navy continued to refine their skills with radar, reaching a peak of sorts on the night of 24 October 1944, when a US Navy surface force jumped a Japanese fleet that was trying to sneak through the Surigao Strait to attack the American amphibious force landing on the Philippine island of Leyte. The US Navy all but annihilated the Japanese force at little loss to themselves.

Of course, US aircraft carriers made heavy use of radar, becoming something like floating radar centers, with a half-dozen different types of radars. The early "radar plot room" gradually evolved into something like a floating version of the RAF filter room, the "Combat Information Center (CIC)", also known as an "Aircraft Direction Room" to the British.

The CICs evolved individually, with different carrier crews devising their own schemes, some using horizontal tables to track the battle action, some using vertical transparent screens on which staffers wrote in mirrored writing to track air actions. By the end of the war, a CIC could have 50 staffers, with the ship or fleet commander residing over the battle at the center of a web of radars and communications links.

US INDEPENDENCE Combat Information Center

* Bell Labs also built a series of radars for submarines that helped America's own wolf packs to locate targets and avoid escorts. The crude 2.45 meter (122 MHz) "XAS" radar was initially tested in the spring of 1941, leading by the end of the year to the production "SD" radar. The SD provided little more than a early warning capability using a fixed horizontal bar antenna mounted on the submarine's HF radio mast. The SD could not give direction information, had a range of only 16 kilometers (10 miles), and a distance accuracy of only a kilometer (3,280 feet).

Submarine commanders distrusted the SD, since it wasn't all that useful and they worried that the Japanese would home in on it with direction-finders. In fact, the Japanese were astonishingly inept at antisubmarine warfare, and never developed a competent defense against American "pig boats". Whatever the misgivings of the American crews, by mid-1942 most US Navy submarines were fitted with it.

By that time, the S-band "SJ" surface-search set was being introduced, the initial prototype having been completed in December 1941. It used an elliptical scanning dish, which was solid at first but then built as a mesh, mounted on its own mast, connected with a shear pin that allowed the antenna to pivot away if struck by surface debris or a shock wave from a depth charge. The SJ produced a beam about 9 degrees wide and 29 degrees high, and could pick up an enemy destroyer from 10 kilometers (6 miles) away, with a range accuracy of 25 meters (80 feet). There were reliability problems at first, unsurprising given that the radar system had to work under difficult field conditions in a salt-water environment, but it was still a very useful piece of gear, and was further improved by the addition of a PPI display in 1943.

The PPI was a particular plus, allowing aggressive submarine commanders to infiltrate Japanese convoys at night and in bad weather, and then turn about, firing torpedoes at intervals to create destruction and chaos. Since the SJ could switch PRF, they even used it with a telegraph key to perform secure line-of-sight communications when hunting in "wolf packs". US Navy submariners became believers in radar.

SV & SJ submarine radars

The SD was retained in service for a time to provide air warning, but it was eventually replaced by the "SV" air-warning set, an up-to-date microwave set that operated at 8 centimeters / 3.75 GHz instead of 10 centimeters, presumably to ensure there were no conflicts with the SJ radar system.