WAR
The two most obvious wartime changes to existing cruisers were radar and increased close-range anti-aircraft weapons. Not so apparent was the introduction of degaussing coils to deal with the new threat of magnetic mines. All of these presented problems because existing cruisers were such tight designs.
Radar
British work on naval radar (RDF) began in 1935, work on a prototype warning radar (Type 79X) operating at metric wavelength (43 MHz) beginning in September 1937. The first operational set, Type 79Y, was installed in the cruiser Sheffield in August 1938, while the first production set (Type 79Z) was installed on board the anti-aircraft cruiser Curlew in August 1939. About 100 such sets were made. Work on a higher-frequency air warning (WA) set (Type 281, 90 MHz) began in December 1939. It was designed to range on against surface as well as air targets, exploiting its surface wave. The Staff Requirement to provide accurate surface ranging to 10nm required a masthead height of at least 110ft. DSD proposed that the first go into a Dido, because ‘for the present war the development of an accurate wireless rangefinder for cruisers is of more importance than for battleships’. Higher frequency made for a smaller, lighter antenna, which was first tested using an army searchlight control set, as Type 280 tested on board the anti-aircraft cruiser Carlisle. It proved possible to use a single antenna with a transmit/receive (T/R) switch, so ships with Type 281 were thus able to make do with a somewhat heavier antenna on their mainmasts, the foremast becoming available for a target-indication radar (Type 293).
These radars were primarily for broad-beam air warning, but given their low frequency they produced surface waves which gave them some surface-warning capability. RAF Coastal Command used much the same technology in its early surface-search (air to surface vessel, ASV) sets, operating at a higher frequency (214 MHz, about 3m wavelength). The surface-ship version of ASV Mk I, using a fixed antenna, was Type 286. Improved versions, including Type 290, had rotating antennas. Although designed (in theory) for destroyers, it was also installed on board cruisers.1 Installation of such radars presented both topweight and electric power problems, as in 1939 many ships had barely enough generator capacity to cover their action loads. Initially radar required separate (but identical) transmitting and receiving aerials, mounted on the two masts. The displays showed only range, the antennas being rotated to look out on each bearing. Given this type of operation, WA required a broad beam and could not be used to track targets. Later radars were given T/R switches which made it possible to use a single antenna for both transmission and reception.
A parallel series of gunnery radars, also with separate transmitting and receiving elements, worked at 600 MHz (50cm). Work began in February 1938, sea trials following (aboard HMS Sardonyx) in June and October 1939. The first set developed was a pompom rangefinder (Type 282) tested in 1940. In 1939 a Staff Requirement was issued for a ranging set which could control main battery fire against aircraft. This Type 283 was installed in a barrage director, the ship’s LA guns being aimed to produce barrages at one of two fixed ranges.2 They were triggered when the range-only Type 283 found that the target was at the right firing range. The first barrage directors were tested on board HMS Charybdis in 1942, and the first four Type 283 went to HMS Berwick. Ideally one barrage director would be fitted for each main battery turret, but that was not always possible. The director was body-trained and handwheel-elevated as in the Mk I pompom director, the eye-shooting operator estimating deflection using a 300kt cartwheel sight. A second operator could train the director for blind-fire. The Type 283 radar office contained an Auto-Barrage Unit (ABU); the turret could be fired either from the director or from this ABU. The director transmitted future bearing and elevation to the turret(s) it controlled.
As completed in 1940, HMS Curacoa shows the two antennas of a Type 279 radar on her topmasts. There was as yet no T/R (transmit/receive) switch, so the radar had to have separate transmitting and receiving antennas (Outfit ATD). The single-antenna Type 79B (for carriers) appeared in February 1941. Type 279 was the production version of the prototype air-warning radar, Type 79 (Type 79Z was installed on board HMS Curlew and on board HMS Suffolk in August 1939; HMS Sheffield had the earlier Type 79Y in October 1938). Nominal maximum range, based on the scale in the display, was 120nm (accuracy 5nm), but actual ranges were shorter. The radar produced both a ground wave clinging to the sea (nominal range 2 to 6nm) and an air wave, and initially it was seen as a dual-purpose air- and surface-warning or ranging set. Thus it had a short-pulse (3 rather than 8 to 30 microsecond) operating mode for gunnery, and it had a special ranging panel (ranging scale 2,000 to 14,000yds, with transmission in 50yd steps). It reverted to air warning only with the advent of the gunnery radars, Types 282 through 285. The antenna proper consisted of four horizontal dipoles, each with a reflecting dipole a fifth of a wavelength behind it. This produced a broad (hence low-gain) beam (84°) which gave good warning but little idea of the direction from which an aircraft was coming. The frequency was low (39 to 42 MHz, depending on version, equivalent to a wavelength of about 7.5m) because it was the best that British radar engineers could do when Type 79/279 was designed. Peak power (Type 279) was 70kW in warning mode. Initially it seemed that the shorter-wave-length Type 281 was much superior. However, in March 1944 the US Naval Attaché in London reported that the Admiralty had a subtler view: there were trade-offs between the two sets. Type 279 offered better cover above 20,000ft and was less liable to be saturated by land echoes. Its performance also varied less on a day-to-day basis. However, Type 281 offered superior medium-height cover and its echoes could be displayed on a PPI, the basis for fighter control and shipboard air defence. Among cruisers, HMS Euryalus seems to have been unique in retaining her Type 279, and in having had it converted to a single-antenna Type 279B in July 1944.
In 1939 the United Kingdom had by far the largest merchant fleet in the world, and the Admiralty had long seen it as both a vital national resource and as a pool from which emergency warships could be drawn. Auxiliary anti-aircraft ships such as Alynbank, shown in 1941, were the mobilisation counterparts of the converted anti-aircraft cruisers, with much the same armament, including a scarce HA director (with the corresponding computer below decks) atop her bridge. She had the same four twin 4in guns, as two (sided) quadruple pompoms. The radar antennas served Type 280, an army set (GL1) developed to serve anti-aircraft batteries ashore (it was first installed on board the anti-aircraft cruiser Carlisle). The antenna developed for Type 280 was later used in the widely-installed Type 281 air-warning radar, whose wavelength was half that of the original Type 79/279. Type 280 operated at 3.6m wavelength (82 MHz) with a peak power of 25kW. This radar was limited to Carlisle and to the converted merchant ships.
In May 1940 work began on low- and high-angle sets (Types 284 and 285 respectively) using the same transmitter and receiver but different antenna arrays. Type 284X was tested on board the battleship Nelson in June 1940, and Type 285 in December 1940. Staff Requirements for longer range and more accurate bearing (by beam-switching) were issued in July 1941. Antennas were sets of directional Yagis (‘fishbones’) or arrays of dipoles along a half-cylinder (‘pig-troughs’). Different ships had different Type 284 arrays on their DCTs, and in some cases Type 285 arrays were used for Type 284.3 Unlike Type 284, the anti-aircraft sets all used arrays of ‘fishbones’. As long as WA procedure was to point the antenna in one direction and range on whatever was there, the gunnery radars could also be used for searching in much the same way.
Designs for ship installation were ordered in February 1939 by Director of Plans, for Didos and Fijis and for the eight projected ‘D’ class anti-aircraft cruisers, but sets would actually be fitted to two Didos per squadron and to all the ‘D’ class conversions. On 30 May 1939 radar installations were approved in principle, the cruisers involved being Suffolk (August-September 1939); Curlew (as soon as possible); and Curacoa, Capetown and Colombo by the end of 1939. At the end of June 1939 WA installations were planned for Suffolk, Curlew, Curacoa, Capetown, Colombo and certain ships (not yet selected) of the Fiji and Dido classes. It was undesirable to stop work on all the Didos, so DNC proposed that the two being built in dockyard, Euryalus and Sirius, be selected. About this time a Naval RDF Panel was formed to decide which ships to fit. Purchase of the first thirty WA sets was approved about this time. The first twenty-four WA sets were allocated at its second meeting on 27 July 1939, the cruisers being six of the first eight Didos, three of the first four Fijis (in both cases not the flagships), Cairo, Calcutta, Coventry, three ‘D’ class AA ships, Canberra and London (if not flagships), Exeter and the first ships of the America and West Indies Squadron (not flagships) coming home for refit. By this time RDF was considered so important that it should always be installed when available. For ‘C’ class cruisers converted to anti-aircraft ships, Plans Division was willing to surrender one or two guns as compensation. Initially it seemed that installation entailed a considerable effort, but by mid-October 1939 it was understood that time in dockyard could be reduced to that required to ship new tripod masts (a week or even less), so that ships in service could be refitted.
As of 18 December 1939, among the first thirty-one WA sets, cruisers were allocated nineteen: four of the five handmade sets (range panels to be supplied later): Curacoa, Cairo, Calcutta and Coventry; four of the five initial production models (range panels to be supplied when available): Glasgow, Fiji, Edinburgh and Belfast; three of four production models with range panels: Naiad, Phoebe and Nigeria; and eight of ten later production sets: two Fijis and six Didos. In February 1940 blanket permission was given for installations, so that cruisers could be fitted with radar whenever they became available. DSD sought approval to fit RDF in all anti-aircraft cruisers, and to all other cruisers except the non-anti-aircraft ‘C’ and ‘D’ classes. That was given in April. In March 1940 cruisers under refit (hence to be fitted with radar) were Belfast and Adventure. Ajax and Exeter were due from the South Atlantic. Perth was going to Australia instead of returning to the UK, hence could not be refitted as had been planned. Drawings were being prepared for installations on board all cruisers, so that work could go ahead as quickly as possible for ships unexpectedly coming to hand. Priority cruisers were Glasgow, Edinburgh, Belfast, Ajax, Exeter and Adventure.
Although the early radars were conceived for air warning, they could certainly detect surface ships. HMS Suffolk is shown on 8 August 1941, two months after she successfully shadowed the German battleship Bismarck. She had been fitted with Type 79Z some time before September 1939, and then with Type 279 in August 1940 (it was replaced by Type 281 in July 1942). Damaged by bombing off Norway on 17 April 1940, she was refitted on the Clyde, 24 May 1940 – 12 February 1941. Improvements included the installation of Type 279 and Type 284 (gunnery) radars; her two quadruple 0.5in machine guns were landed and four single Oerlikons fitted. The gunnery radar is not visible here, but this photograph was presumably censored to eliminate it. In December 1941 a US officer attached to HMS Renown described this operation as the best example of radar shadowing: for thirty-six hours Suffolk managed to keep Bismarck in range at 14,000 to 26,000yds. It seems clear from this report that, until centimetric sets (e.g., Type 273) entered service Type 284 was frequently used for surface search/warning. For search operation, the director trainer (in constant communication with the radar operator below decks by telephone and handset) trained the DCT and thus the Type 284 mounted on it over the arc to be searched. The radar operator had an indicator of DCT train angle, hence could tell the bearing of any echo he saw on his A-scope (showing only range vs. echo strength). If he saw an echo, he reported it by voice pipe to both compass platform (i.e., the command) and to the plot below it (typically visible to the command via a window in the deck). At the same time he turned a switch in the radar office to ‘office controlling’ position. That in turn rang a gong in the DCT and lit up an open-faced indicator. Now the radar operator could control the DCT by turning his bearing indicator (the director trainer matched pointers on the open-faced indicator). A US officer wrote in December 1941 that ‘this system demands great co-operation between DCT and RDF [radar] office and has not been satisfactory’ (in HMS Renown). The radar could be switched between ‘director controlling’ and (radar) ‘office controlling’ modes, the latter being used for surface search. Typically it was switched to office control after an echo from another ship had been obtained by the director office, the Type 284 being kept lined up on the target by its operator below decks. He in turn sent ranges and bearings to the ship’s plotting office as frequently as possible. The plotters used this data to produce a plot of target course and speed. According to the US report, ‘to get bearing accurately requires a lot of experience and co-operation between Director Control Tower and 284. The DCT will have to sweep two or three times for each bearing cut so that the 284 operator can determine the bearing at which the echo was a maximum.’ That is, the operator used direction-finder techniques.
The long-wave radars suffered from reflection off the sea. The shorter the wavelength, the less reflection interfered with searching for objects on or near the surface. In 1940 researchers at Birmingham University found that they could generate useful amounts of power at a much shorter wavelength (10cm), which was well-suited to surface search. The shipboard version of this radar was Type 271 (warning surface, or WS), its two stacked antennas (transmission stacked above reception) housed in a ‘lantern’ (cylindrical radome) immediately above its office (to minimise signal loss in waveguides between antenna and display). The position of the radar depended both on topweight and on topside space available for the radar office. Ideally the radar was atop or adjacent to the compass platform, so that radar information was immediately available to the command. Thus Fiji class cruisers generally had the radar and its office immediately forward of the bridge structure, the DCT occupying the only available space atop that structure. With a larger bridge, a Southampton or Belfast typically had the lantern and office abaft the DCT, looking out over it. Type 271 was initially allocated to convoy escorts. Because it offered better surface performance than the longer-wave Type 284, Type 271 (and later equivalents) were sometimes used for gunnery ranging. Like the WA sets, the early WS sets had only range displays, and were manually rotated. When the set was being used for gunnery, the ship lost situational awareness. That actually happened at the battle of the Barents Sea (in December 1942), when a German destroyer unexpectedly appeared near HMS Sheffield – which promptly sank her. The upshot was that the use of surface-search radars for gunnery ranging was discouraged, although versions of these radars continued to have automatic range transmission to the ship’s fire-control system (suffix M before 1943, PR afterwards). After 1943 the modified Types 271Q and 273Q had automatic rotation and PPI (map-like) displays.
HMS Mauritius shows standard radars (1942), including the important Type 273 surface-search set, in the sixteen-sided ‘lantern’ (replaced by a perspex cylinder beginning in November 1942, to eliminate echoes from the frames of the ‘lantern’) forward of her bridge. On her director is the early version of the Type 284 main battery gunnery radar, with separate transmitting (bottom) and receiving (top) arrays, both of which are ‘pig-troughs’ consisting of a row of vertical dipoles in a long cylindrical reflector. The array at the masthead is for a Type 279 air-warning radar, fitted soon after the ship was completed. Alongside the bridge is a HA director carrying a Type 285 radar to control the ship’s 4in guns. IFF antennas are not visible. Type 273 was the large-ship equivalent to the Type 271 used aboard many smaller units. It was the first British naval microwave radar (operating at about 3,000 MHz, a wavelength of about 10cm), and it offered unprecedented surface-detection capability, particularly of small objects. The catch was that the radar office had to be directly below the radar, to minimise losses in the waveguides connecting it to the radar. The ‘lantern’ was tall because it accommodated two separate antennas, one for transmission and the other for reception. The development of a more powerful radar power source made it possible to move the array away from the radar office, producing Type 272. Type 271 used a pair of ‘cheese’ antennas, but Type 272 used a pair of side-by-side 3ft dishes (paraboloids) offering higher gain. They produced a10° wide beam. The operator turned a crank, the antenna being connected by cable to the office equipment, so the radar could not swing in a complete circle, but only through an arc of about 200° to either side of dead ahead. The same limit applied to the later Type 272. Thus it was impossible to connect Type 273 to a PPI (map-like) display, since it could not rotate continuously. The microwave radars (Types 271, 272 and 273) were given fire-control data transmission capability, since they offered more precise surface ranging than longer-wave (50cm) gunnery sets. They were also considered much better than the long-wave sets against low-flying aircraft. Type 273Q was a complete redesign to accommodate the new strapped-cavity magnetron radar tube, which offered an output of 70kW rather than 5kW. It did not enter service until mid-1943. It is not certain which if any cruisers were given it. Type 284 was a 50cm (600 kHz) gunnery radar, one of a series which included the pompom control radar (Type 282), the barrage director radar (Type 283) and the HA control radar (Type 285); it was the only one of the group not (except in the earliest installations) to use an array of ‘fishbone’ (Yagi) antennas. The long receiving element offered a narrower receiving beam, and later versions produced pairs of beams, which could be switched to find the bearing of the target more precisely. Of this series, the first in service was Type 282, because without radar there was little hope of rapidly providing the range to a pompom director. Type 284 was an adaptation of Type 282, not the other way around, using the same transmitter (initially 25kW, ultimately 150kW) and the same receiver. As of late 1941 (as reported by HMS Renown), the radar could be used either for surface warning (range scale to 48,000yds) or for ranging (24,000yds); it could detect a ‘Town’ class cruiser at 18,000yds and an aircraft (below 15,000ft) at 15,000yds. Bearing accuracy was 1°. Work on fire-control radar began in February 1938, and trials aboard HMS Sardonyx began in June 1939. The Staff Requirement for the radar for a barrage director to counter dive bombers (Type 283) was issued in 1939. Type 282 was first tested in 1940, 200 sets being ordered in April. The prototype Type 284X was tested on board HMS Nelson in June 1940, while work was already underway on common arrays for transmission and reception. The first production Type 284 was installed on board HMS King George V in December 1940, the first Type 285 on board HMS Southampton at the same time. The standard ‘pig trough’ was 21ft x 2ft 6in, but at least at first there were numerous variations. The M and P versions introduced common antennas for transmission and reception, and the M3 and P3 versions introduced beam-switching for greater precision. These suffixes applied to Type 285 as well as 284. Types 282 and 283 used a pair of Yagis, initially one each for transmission and reception, but later both for both functions..
As microwave radar developed, it became possible to locate the office 40ft from the antenna (Type 272). Thus in modified Fijis Type 272 could be placed in front of the DCT barbette, the DCT being raised to clear it. Like Type 271, Type 273 had its antenna in a lantern below which the operator sat. Antennas of the next generation (Types 277 and 293) could be located well away from the radar office. Type 277 was a tiltable dish which could be used either for surface search (with a narrow beam) or tilted for limited height-finding. Type 293 was the first British target-indication radar. Some link was needed between the broad beam of the WA radar and the narrow beam of a gunnery radar, the target-indication radar in effect searching the broad beam. In 1943 the British became interested in Target Indication Units (TIUs) which functioned as track-while-scan memories. A TIU could, in effect, maintain multiple tracks which could be assigned either to different guns or to one gun in sequence as it dealt with nearby targets. After the war the TIUs were associated with Gun Direction Systems (GDS), which were rated in terms of how many targets they could handle. TIU III with its Type 992 fast-scanning radar was a staple of post-war designs. Microwave technology was also applied to gunnery radar, the most important wartime example being Type 274 for main battery fire control.
HMS Enterprise shows standard British mid-war radars in a March 1943 photograph. The large array near the masthead served a Type 281 air-warning radar (the mainmast carried a similar array); it could be distinguished from a Type 279 array by its smaller size, because it operated at about half the wavelength. In fact the antenna (Outfit ATE) was a pair of scaled-down Type 279 antennas side by side, so it had eight rather than four radiating dipoles. Above it is an IFF interrogator for this radar, turning with it. The interrogator below the Type 281 antenna, bracketed to the mast, works with the Type 272 radar in the short ‘lantern’ on the mast, below the crow’s-nest. The main battery director carries the ‘pig-trough’ antenna of the standard Type 284 main battery radar, in this case the version with a common transmitting and receiving antenna (earlier versions had separate antennas on the face of the director). Visible on the face of the bridge are two power twin Oerlikon mounts. Compared to Type 279, Type 281 produced far more power (a short [1.7-microsecond] pulse at 1,000kW and a long [15-microsecond] pulse at 350kW) and had a narrower beam (27°) with higher antenna gain. The original Type 281 could switch beams to find target direction more precisely. The interrogator, Type 243 (IFF Mk III), used a modified ASV Mk II (i.e., Type 286) radar and operated at 179 MHz (with an alternative frequency of 171 MHz if a ship had two interrogators). It superseded a slightly higher-frequency Type 241. There was an associated transponder, to identify the ship to other ships and to aircraft. Type 941 was a Type 243 modified to work with a radar using a PPI (map-like) display, such as the Type 281B fitted to many cruisers. Type 242 was a smaller-ship alternative to Type 241/242 mainly for centimetric surface-search radars. Corresponding transponders were designated in a 250 series, e.g. Type 253 to work with Type 243. The first ship with Type 281 was HMS Dido (1941). The single-antenna Type 281B was introduced late in 1941, but was not aboard many cruisers even at the end of the war and beyond. Type 281BQ (continuous rotation, PPI display) was introduced in 1945, and was an important post-war modernisation item. The arrays were rotated by Selsyn (a form of magnetic motor which followed a remote command) at 2 or 4 rpm, the direction reversing after each complete rotation. Alternatively the operator could point the radar antennas (transmitting and receiving) to a desired bearing by hand. Nominal range was the same 120nm as in Type 279, but performance was much better. Originally, like Type 279, Type 281 had a gunnery ranging function, but it was soon abandoned. In 1941, HMS Renown circulated a radar pamphlet claiming that its Type 281 radar could detect an aircraft flying at 20,000ft at 75nm (35nm for one at 5,000ft) and a cruiser at 15,000yds. Bearing accuracy was given as half a degree. Type 272 used the same superimposed pair of ‘cheeses’ as Type 271, but it had a more powerful transmitter, hence could be up to 40ft (i.e., up a mast) from the radar office. Like Type 273, Type 272 could be used as a gunnery ranging set. That could prove embarrassing. During the battle of the Barents Sea (December 1942), HMS Sheffield used her Type 273 for that purpose, the radar being kept trained on a German cruiser. Because it was not being used to search (manually), the ship’s command missed the approach of a German destroyer – which was very fortunately spotted visually, and destroyed. This practice, apparently common, was then strongly discouraged.
Backing up (and working with) radars were intercept devices and jammers. The first shipboard electronic sensors used by the Royal Navy were radio direction-finders. Work on shipboard HF direction-finding began in 1930, and by the late 1930s many cruisers had remote-control rotatable loops at their fore mastheads. These were not the instantaneous HF/DF devices used by convoy escorts during the Second World War. The British classed interception of enemy signals for exploitation without code-breaking as ‘Y’; initially that meant exploiting enemy communications, but from 1940 on it included exploiting enemy radar emissions (‘noises’). The first effort was directed against German coastal radars in the Pas de Calais which supported attacks on British coastal convoys; later efforts expanded considerably. Beginning about 1941 the FV1 VHF/DF was developed specially for ships engaged in ‘Y’ (it was first tested in the monitor Erebus). In 1942 all cruisers (and many other ships) were ordered fitted with it and with the Type 91 tunable spot jammer operating at 200–600 MHz, the German radar frequencies (Type 91M extended its range down to 90 MHz and provided noise modulation).4 Type 91 was successfully used against German naval radar up to January 1945. Some cruisers assigned to bombardment on D-Day received additional jammers (AC III and ATDV). AC III was probably Airborne Carpet III. A second threat, which appeared in 1943, was radio-controlled missiles. The first jammer produced was Type 650, which was limited by its narrow frequency range. It was soon followed by the broader-band Type 651. It was fitted on board most important warships by the end of the war.
The Type 285 HA gunnery radar is atop this HA director, shown on board HMS Ajax at New York Navy Yard, 16 October 1943. Type 285 began as a range-only radar, and it was developed with beam-switching so that it could track a target in bearing – but unlike its US contemporaries it could not track in elevation, hence could not be used for blind fire (the array did elevate with the sights). It was designed to provide fuse settings, as the associated shells were time-fused. Large HA directors like this one used an array of six Yagis, initially three for transmission and three for reception. Gain doubled when the entire array was used for both purposes, so range increased considerably. The aloft director was part of a larger High Angle Control System (HACS), which employed a below-decks computer called the High Angle Control Table (HACT). The director fed the computer with target range (optical and later radar), elevation (taken from the optical rangefinder), and bearing. There was no attempt to measure target motion using gyros, as in US systems such as Mk 37. The initial HACS I was fitted to all ships up to 1935, after which it was replaced in battleships by HACS III, the sets so released being installed as second sets on board heavy cruisers. HACS I was given improved fuse prediction gear for target speeds of up to 250kts, and ultimately all were to be converted to HACS I*** with maximum enemy speed of 350kts and also with gyro roll correctors (GRUB). The Leander class was fitted with HACS II, which was similarly modified. Compared to HACS I, the director was reshaped to accommodate junction boxes and the rangefinder. Later cruisers prior to HMS Birmingham were given HACS III, which was considerably improved mechanically, including a better means of data transmission. It was reshaped to suit a new 15ft rangefinder and had a slightly thicker shield. Mk III* had an additional rangetaker’s position (III** was completely round and had no additional rangetaker’s position). From Birmingham on, cruisers were given HACS IV, which had provision for roll compensation, cross-levelling (against pitch), and a new means of data transmission (Magslip). In earlier versions each director tower was associated with one calculating table, but in HACS IV any director could be connected to any table. Ultimately the number of targets the ship could engage depended not on how many directors she had but on how many calculating tables (i.e., how much internal volume) she had. Like Mk III**, Mk IV had a completely round shield, but it accommodated the second rangetaker. Mk IV* was designed to control 5.25in guns on Dido class cruisers. Mk V was a high/low angle director of a different type, used on board battleships and carriers. The early post-war standard types were Mks 4 and 4*. The successor director, carrying the Type 275 blind-fire radar, was Mk 6. Directors generally accommodated a control officer, layer, trainer, telephone operator and range taker. Later towers also had provision for a LA rate officer (for when the tower was used for LA control) and space for a second rangetaker, who was needed when duplex height-finders were introduced. The control officer had glasses which moved in step with the two director telescopes; the right-hand glass incorporated a graticule which could be rotated to indicate the apparent course of the target. In later directors the other glass could be used for spotting. Later directors also had a separate telescope (forward area sight, later High Angle Direct Eyeshooting Side [HADES], which was superseded post-war) with a spider’s-web graticule for the control officer, to be used to indicate deflection. Gyro roll correctors stabilised the director telescopes. A contemporary HACS manual advised that, since it took time for the gyros to settle down, they should be kept running when air attack was considered probable.
A new generation of radar direction-finders was developed in 1944–5 for the Pacific: four RU series DF sets were being developed to cover the full radar frequency range, RU4 (2–6 GHz) being the most advanced.5 The war ended before it could enter service, and none of the series survived post-war.
Beyond radars themselves, the great wartime development was the Action Information Organisation (AIO). The Royal Navy had invented and developed plotting as a way of maintaining situational awareness, and in effect the AIO extended the plotting idea to radar. The AIO was not quite equivalent to the US Combat Information Center, because the Royal Navy split it into a number of related spaces, typically (by 1945) a main tactical plot (Operations Room), a Radar Display Room, an Aircraft Direction Room (ADR), a Gunnery Direction Room (GDR) or Target Indication Room (TIR), descended from the earlier ADP, and often a larger-scale bridge or flag plot. This arrangement could be traced partly to pre-war practice, and partly to the accident that British air-warning radars could not be used directly for fighter or gun control, as the narrower-beam US sets could. The British arrangement entailed problems, because there was no way of making sure that the pictures in all the related parts of the AIO matched. Post-war this problem led directly to the development of the Comprehensive Display System, the first (albeit analog) integrated tactical data system. The British form of AIO was installed on board most ships of the Southampton, Belfast, Fiji (not in the four-turret Nigeria) and later classes. Due to the scale of work involved, it was installed on board only two heavy cruisers (Norfolk and HMAS Australia). Sussex and smaller cruisers had only a partial installation.
Achilles shows New Zealand-developed naval radars in this February 1942 photo taken from USS Curtiss at Suva in the Fijis. In February 1939 the British government invited the governments of the technically-advanced Dominions (Australia, Canada, New Zealand, and South Africa) to come to England for briefings on radar. All four produced radars, but only Canada and New Zealand produced naval sets, and the Canadian SW-1C was used only aboard destroyers and corvettes. New Zealand was represented at the British briefings by Ernest Marsden, Secretary of the Department of Scientific and Industrial Research (DSIR). After several months in England, he brought home considerable material, including parts of two television sets (as the basis for receivers) and parts of an ASV Mk I airborne surface-search radar, a predecessor of the British Type 286 naval radar. New Zealand set up a naval radar development organisation, the Radio Development Laboratory (RDL), which in turn set up a group at Canterbury University College in Christchurch (an air and ground radar group was set up in the New Zealand Post Office [responsible for radio] in Wellington). The first experimental gunnery set, probably designated SS-1 (66cm/450 kHz, peak power 5kW) was installed on board HMNZS Achilles at the end of May 1940 (some reports have it installed on board the armed merchant cruiser Monowai in July 1940). Performance was encouraging enough to warrant work on an improved set, which was installed on board Achilles in February – March 1941. It was accurate to within 50yds. From it were developed both a new gunnery set (SWG) and a warning set (SW), both of which were installed on board Achilles in August 1941. They proved successful during a daylight exercise with the liner Aquitania, the radar much outperforming an optical rangefind-er at 15,000 to 20,000yds. The Chief of the New Zealand Naval Staff, Commodore Parry, considered the sets valuable enough to warrant production, and he called for a radar capable of ranging on an aircraft at 12,000yds. Leander was given pre-production SWG and SW sets during her October – November 1941 Auckland refit. Meanwhile, in September, the New Zealand Radio Development Board made the development of ship and aircraft sets its highest priority, and wanted three of each type provided for the ships on the New Zealand station. New Zealand offered radars to the Admiralty for the Eastern Fleet, to make up for shortages in shipments from the United Kingdom. Plans called for producing at least five of each type per month, and in November 1941 an officer arrived from Singapore to ask for thirty of each type (and also to ask the Australians for forty SW sets, mainly for destroyers). The New Zealanders decided to make the SWG set their priority, and by November 1942 fifteen complete SWG had been shipped to Australia. Another nine were sent to Ceylon, one of which was lost to the Japanese when the merchant ship Haranuki was captured. By then sufficient supplies of superior sets were available from the United Kingdom, and eventually the SWG sets went to New Zealand naval batteries ashore. At least eight sets of radar were produced for the Royal New Zealand Navy. SWG operated at slightly lower frequency than SS-1 (73cm/430 kHz) and had an effective range of 7nm (accuracy 50yds). It had two displays, one to give range (using a 30,000yd base with 1,000yd markers) and one to assist in training the director. The antenna was two fourteen-element Yagis, side by side, one for transmission and one for reception, backed by a semi-cylindrical reflector. A later version, on board Leander in 1943, used three Yagis, each backed by its own mesh reflector, presumably one for transmission and two for reception. SW was a 1.5m radar using two Yagis (side by side) atop a lattice pylon, visible between the cruiser’s funnel and her bridge. Some of these sets may have been installed on board Australian ships (HMAS Canberra is sometimes suggested as a candidate). Note that a US report (November 1941) had Achilles fitted, Leander being fitted, and Monowai about to be fitted with an SW set. SWG was described as a 73cm set which might be redesigned to operate at 50cm. According to one account of the radars, in 1941 an improved SW antenna, using four dipoles and a mesh reflector, was proposed, but it appears it was never installed.
This 15 September 1943 enlargement of the mainmast of HMS Birmingham shows the four semi-directional arrays of the FV1 radar direction-finder, the rectangles in which are dark vertical elements indicating the framework and the vertical dipole in each. On the lower right are two hourglass-shaped omni-directional antennas, each canted to the vertical. The four FV1 antennas fed a radar intercept receiver which registered the approximate frequency of the target signal. The two omnis were used to measure its frequency more precisely, so that the associated Type 91 jammer could be tuned properly. Its jamming signal was sent out via FV1 antennas. The process was anything but automatic. During the Scharnhorst engagement jamming failed because it was tuned to the image frequency of the monitor receiver, rather than to the frequency used by the enemy battleship. Fortunately a cruiser shell soon destroyed the German’s radar. There were also successes. In April 1944 HMS Black Prince successfully jammed enemy radar during a battle against three German torpedo boats. Type 91 jammed German shore fire-control radars successfully during the invasions of Normandy and of Southern France. The system was used for the last time in January 1945 (Operation ‘Spellbound’), when HMS Bellona successfully jammed a German Giant Würzberg radar that was being used against her surface strike force. Denied radar coverage, the German convoy lost two merchant ships and an escorting destroyer. In 1942 FV1/Type 91 was ordered installed on board all British battleships and cruisers, but that seems not to have applied to the old ‘C’ and ‘D’ classes except for the modernised Danae and perhaps ORP Dragon. Unfortunately no official register of fittings has come to light. The following list was compiled by Alan Raven based on photography: ‘County’ class: Australia (by late in war), Berwick (on completion of August 1942 refit), Cumberland (not fitted at 1942 refit, not certain about later), Cornwall (not fitted when sunk), Devonshire (fitted by March 1944, maybe earlier), Dorsetshire (not fitted when sunk), Kent (no evidence of fitting as of early 1944), London (on completion of May 1943 refit, on foremast), Norfolk (fitted on completion of refit June 1943, uniquely fore and aft), Shropshire (fitted on each side of the bridge on completion of April 1942 refit, on mainmast as refitted October 1943), Suffolk (on completion of May 1943 refit), Sussex (by April 1945). York and Exeter were sunk before they could be fitted. ‘Town’ class: Birmingham (on completion of late 1943 refit), Glasgow (fitted by July 1943), Liverpool (on fore leg of mainmast at height of after funnel top on completion of large refit, August 1945), Newcastle (not certain), Sheffield (on foremast mid-1942, mainmast array added 1943); Belfast (fitted by 1944), Edinburgh (during large refit January – March 1942). Manchester was lost before she could be fitted. Fiji class: Bermuda (not fitted until August 1945 refit), Gambia (fitted after 1943 refit), Jamaica (upon completion of 1945 refit, not before), Kenya (as completed), Mauritius (not certain), Nigeria (by end of 1943 refit in United States, possibly earlier). Fiji was lost before she could be fitted. Trinidad also probably was not fitted. Improved Fiji: Ceylon (as completed), Newfoundland (as completed), Uganda (apparently not as completed, but had the system as refitted by the end of 1944), Swiftsure (as completed). Dido class: Argonaut (fitted during US refit, November 1943), Cleopatra (by end of US refit November 1943, by end of war aerials moved to higher position just below Type 281 aerial), Dido (not fitted), Euryalus (fitted by 1945, date not known), Charybdis (definitely not fitted as of early 1943), Hermione (not fitted), Phoebe (fitted during US refit June 1943), Sirius (no evidence of fitting as of 1944, may have been fitted later). Improved Dido: as completed. Anti-Aircraft Cruisers: Caledon (upon completion of 1944 refit), Carlisle (upon completion of November 1942 refit), Colombo (upon completion of April 1943 refit). Delhi was almost certainly not fitted, nor were the other ships. Emerald and Enterprise were both fitted when they were refitted in 1943. Ajax was fitted during her 1943 New York refit. Achilles was fitted from the completion of her May 1944 refit. There is no evidence of fitting in any other Leander, but Orion may have been fitted. Hobart was fitted during her big late-war refit, but her sisters were lost before this became an issue. There is no evidence that Aurora or Penelope was fitted, but Penelope may have been. The earliest installations, in Sheffield and Shropshire, used two antennas each, so the ship had to be swung to obtain a bearing. Both ships later received the full four-antenna arrangement.
As modernised, Danae had FV1 antennas on the starfish of her foremast. She had the Type 291 air-warning radar typically installed on board smaller ships, descended from the Type 286 shipboard version of the Coastal Command air to surface vessel (ASV) radar. Type 286 actually used an ASV Mk I set; about 200 were placed in service beginning in mid-1940 (see HMAS Perth for an illustration). An improved Type 286M appeared in January 1941. Both Type 286 and 286M used a fixed antenna array (Outfit ATQ), in which two dipoles pointed along the ship’s centreline transmitted, and two receiving dipoles were canted outward on each side. They were connected to the receiver for alternate pulses, the radar in effect beam-switching. The display was a vertical line, the right- and left-hand echoes being shown to right and left. When the ship was pointed at the target, the echoes on each set of dipoles were the same. The radar could detect targets 50° to either side of dead ahead. Type 286P and 286PQ (and Type 290 and 291) employed a scaled-down Type 281 array with common transmission and reception; Type 291(as here) had a PPI (map-like) display. Type 291 entered service in late 1942; it used a new 100kW transmitter. That represented impressive growth from the original 7kW of Type 286 and its M and P versions and the 50kW of Type 290 (fitted beginning in mid-1941). The rotating-antenna version of Type 286 entered service in February 1941.
HMAS Australia shows the standard radars installed when British cruisers were modernised late in the Second World War. The tilted ‘cheese’ of the Type 293 target-indication radar occupies the masthead. Below it is the tiltable dish of a Type 277 surface-search radar, which could also be used for low-altitude air search or for limited height-finding. The director carries a Type 274 fire-control radar, the centimetric replacement for Type 284, a double ‘cheese’. To the side of the bridge is a Type 285 gun-control radar on a HA director: the intended successor, Type 275, required an entirely new (and heavy) director. Below the compass platform is a twin power Oerlikon on a circular platform. Types 277 and 293 were part of a family of new 10cm radars using a 500kW magnetron and waveguides to carry the signals between radar office and antenna. All were to have PPI displays. Type 277 was intended to replace Type 271 and another member of the family, Type 276, was intended to replace Type 272 (it did so in a few ships, such as HMAS Hobart). Type 293 offered much better air cover, so it largely displaced Type 276 (the latter was reinstated late in 1944 as a temporary substitute when the original Type 293 antenna proved inadequate). Type 277 (Outfit AUK, the ‘great auk’) was stabilised in elevation and could elevate to 40° (277P could elevate to 70). The post-war Type 278, in ‘County’ class guided-missile destroyers, was a modified version remotely controlled in elevation by the ship’s combat direction system.
Type 274, the double ‘cheese’ visible atop the director of HMS Kenya (photographed at Yokosuka on 19 May 1951, during the Korean War), produced a single narrow beam. Note the separate transmitting and receiving ‘cheeses.’ Late in October 1943 the US Navy officer responsible for fire-control radar development, temporarily attached to the Naval Attache’s office in London, reported on British naval fire control sets. He regarded Type 284 as roughly equivalent to the US Mk 3, which was then being superseded by the centimetric Mk 8. He felt that Type 284 offered only two advantages over the US set. First, it provided the trainer with a full view of all the targets in the radar beam, hence did not require co-operation from the range operator in setting the range gate before the trainer could see any signals. He was therefore protected from missing a signal by mis-setting the gate. Second, it provided a direct measure of range rate via a range aiding mechanism, which gave the correct rate when the target pip remained in the range notch even though the knob was not adjusted. Any change in range rate was immediately obvious. The US officer felt that there was no US counterpart to Type 274; the closest was the US Mk 27, just entering production. Both were S-band (10cm) sets using lobe-switching and precision ranging, but the US set could also be used independently, and offered a PPI display. He saw Type 274 as a much-refined Type 284 with greater gain than the US set. It had a separate spotting display, and a separate range marker to indicate where a salvo should be expected to fall. However, its narrow beam did not permit spotting in deflection. The contemporary US main battery radar (Mk 8) scanned its beam back and forth for that reason (but therefore could not be as precise as Type 274). Type 274 was ‘probably the most powerful and precise lobe-switching type of fire-control radar in production today. It is undoubtedly superior to our Mk 8 in almost every respect except the two vital ones of presentation and of deflection spotting. The Type B presentation of the Mk 8 with its ability to present a bird’s eye view of any given area, complete with all targets and shell splashes simultaneously, far outweighs all the advantages which the 274 has.’ He reported that the British were working on a rapid-scan X-band successor roughly equivalent to the US Mk 13, to appear early in 1945, but it never materialised, and no Type number seems to have been assigned. Instead, the Royal Navy developed a broader-beam splash-spotter, initially an adapted army set (Type 930) and then the Type 931 visible under the right side of the Type 274 array. Type 931 was developed in Canada, as part of the wartime integrated Commonwealth weapons development programme, and twelve were made. Post-war the Admiralty ‘anglicised’ it for production as Type 932 (to limit the loss of foreign exchange); it is not clear how many were made (it was tested on board the cruiser Sheffield in 1952).
Anti-Aircraft Improvements
Probably the most important single development was Remote Power Control (RPC), which greatly improved the precision with which a director could control light and heavy anti-aircraft guns. Before the Second World War, directors transmitted their orders to dials at a gun, and the gun crew matched dials or pointers to train and elevate their weapon. But the time lags inherent in such operation became less and less tolerable as aircraft performance improved. The Royal Navy did have a kind of RPC for heavy guns, as used in ships like HMS Nelson, but it was heavy and ill-adapted to light weapons. Magslip, which was much lighter, could move a dial in response to a remote movement, but it sent only low-powered signals. The question was how to amplify those signals sufficiently to drive a gun mount. When the Admiralty Research Laboratory (ARL) produced its first magslip in 1928, it immediately went on to produce a magslip-controlled hydraulic servo, which was tested on board the light cruiser HMS Champion. In effect this servo was the beginning of RPC, although no servo gun mount appeared for a decade. For a naval gun, RPC would both control the gun (in response to fire control calculations) and stabilise it. In 1937-8 ARL produced a hydraulically-driven twin 4in gun.6
Several alternatives were developed: RP 10, RP 40 and RP 50 series (RP 20 and RP 30 were presumably abortive alternatives). Generally the controlling signal (from a magslip) was initially amplified electrically, then either hydraulically or electrically as a final stage. RP 10 and RP 40 were hydraulic; RP 50 was electric.
In 1942 it seemed that the future for close-range control belonged to a new Close Range Predictor, which would supersede the pompom director. The entire predictor, carrying its trainer’s and layer’s sights, was stabilised, and it carried the same Type 282 radar as the pompom director. This director appears in several of the designs described below. It died when Type 282 was superseded by the much more massive Type 262 centimetric radar.
The Royal Navy entered the Second World War with the most powerful close-range anti-aircraft batteries in the world, but unfortunately cruiser designs were so tight that the increases demanded by war experience were difficult to absorb. By 1943 the only major weights left to remove were ships’ boats, aircraft and ‘X’ turret (‘Q’ turret in Didos). British officers noticed that the US fleet eliminated nearly all ships’ boats in favour of boat pools at its bases. For example, by landing her boats a ‘County’ class cruiser could gain one or two quadruple pompoms (with directors) and possibly four twin Oerlikons, plus the desired target-indication radar (Type 293), a ‘Town’ could gain two quadruple pompoms and a three-turret Fiji a quadruple pompom with director and six twin Oerlikons. The 6in cruisers could gain the target-indication set with a compromise aerial (providing surface search) if they surrendered their Type 272 or 273 surface-search sets. All of these figures assumed that the ship had already had her aircraft removed, and that she retained two boats. This February 1943 proposal was rejected because the Royal Navy considered it vital that the cruiser be a self-contained unit; it was unacceptable to lose the strategic mobility which (according to a submission to the Future Building Committee) ‘has always been the pride of the British Navy; that is, the ability to go anywhere and do anything at the shortest notice … [which] more than any other, distinguishes us from the other Services and indeed from any other Navy’.7 Boat cranes were also used for storing, embarking provisions, etc., hence could not lightly be eliminated (they were given up in the Didos as compensation for Type 79Z radar). Moreover, to set up boat pools at specific ports required that which ports the fleet would use, and with what intensity, should be forecast, but the centre of gravity of the war could move.
In May 1943 DNC was asked to find compensation so that ships could be fitted with their ‘ideal’ close-range armament.8 He suggested considering fitting pompoms and Bofors right aft on the quarterdeck, cutting away the after cabin (under that deck) to bring the weight lower and to clear main battery arcs. He was impressed with the new power-operated twin Oerlikons (‘Beehives’) which saved considerable weight. The use of main-battery barrage fire had to be taken into account, because previously main-battery blast on close-range anti-aircraft crews had not been taken into account. Ships should be divided into those with adequate, reasonable and poor anti-aircraft armament, and shown as such in the Pink Lists of operational units. As a typical ‘County’, Kent had two octuple pompoms and seven twin and two single Bofors; ideally she should have twice as many pompoms and thirty-two Oerlikons. Ajax had two quadruple pompoms and four twin and two single Oerlikons; ideally she should have twelve pompom barrels and twenty-four Oerlikons, twice as many. Birmingham had two quadruple pompoms (ideal sixteen barrels) and eight twin and two single Oerlikons (ideal twenty-eight barrels). Kenya had two quadruple pompoms (ideal sixteen barrels) and eight twin Oerlikons (ideal twenty-six barrels). Dido had two quadruple pompoms (ideal twelve barrels) and four twin and two single Oerlikons (ideal twenty barrels).
On 6 October the Board approved removal of ‘X’ turrets from the Ajax, Birmingham, Fiji and Devonshire classes and ‘Q’ turret from five-turret Didos as compensation for increased anti-aircraft armament. By this time it was looking ahead to the war against Japan; while the Japanese might seldom have superior surface gunpower, they would always be able to deliver heavy air attacks against individual ships. Once ‘X’ turret had been removed from a four-turret Fiji, enough weight was available to use the hangar for an AIC. Cabins previously used for that function could be recovered. Removal of a turret was a major shipyard operation, and it was not done in all ships before the end of the war. In 1946, of the heavy cruisers, only Norfolk, Devonshire, Australia and Shropshire had had ‘X’ turret removed. Of the Didos, only Dido and Sirius retained ‘Q’ turret. Of the Fijis, only Gambia and Nigeria retained ‘X’ turret. Of the ‘Towns’, Newcastle retained ‘X’ turret. Only Orion of the three surviving Leanders retained ‘X’ turret.
The first new weapon was the Swiss 20mm Oerlikon, twenty of which the Admiralty bought in 1939. They could be distinguished from guns made in Britain and in the United States by their lack of a shield and by their muzzle collars. Up to about 1942 the main improvements to cruiser close-range batteries were the replacement of quadruple 0.5in guns by Oerlikons. It turned out that a twin unpowered Oerlikon did not weigh very much more than a single. The Royal Navy, but not the US Navy, invested in power-driven twin Oerlikons, which it introduced in 1942. Initially they were visually aimed, but a tachymetric sight was introduced in 1944.
HMS Sussex is shown in modernised form, 4 April 1945. She had just been refitted at Sheerness between June 1944 and March 1945, her ‘X’ turret and her torpedo tubes being removed. She emerged with six octuple pompoms (four added) and four twin and six single Oerlikons. Her radars were modernised, her foremast being cleared by replacing the two-antenna Type 281 with the single-antenna Type 281B. That left space for the Type 293 target-indication radar and the Type 277 surface-search radar (the dish). Her previous major refit had been on the Clyde, November 1940 to August 1942, after having been bombed on 18 September 1940. At that time she was given twin 4in guns in place of her earlier eight single mounts, two octuple pompoms, ten Oerlikons, and radars, including surface search.
After the fall of France, the Dutch gunboat Willem van der Zaan demonstrated both to the Admiralty and to the US Navy her twin Bofors guns on a stabilised Dutch Hazemeyer (later Signaal) mounting, incorporating its own fire-control system. Both navies were very impressed. The US Navy concentrated on the gun, placing it on a simple twin (and later quadruple) mounting with an external director, while the Royal Navy liked the integrated combination the Dutch had developed, perhaps partly because it was better adapted to a ship with limited internal space. A later British account claimed that the Hazemeyer, whose design began in 1936, was a decade ahead of its time. However, it was also very difficult to maintain. The British placed their standard light anti-aircraft rangefinding radar (Type 282) on board the Hazemeyer. The Hazemeyer was placed in production, but it entered service, as the Bofors Mk IV, only in limited numbers, beginning with HMS Whimbrel in November 1942. HMAS Hobart and the anti-aircraft cruisers Caledon and Colombo were armed with this weapon, as were the fast minelayers Apollo and Ariadne. The British equivalent was the STAAG (Stabilised Tachymetric Anti-Aircraft Gun), which the single Mk I was a prototype only, but the twin Mk II entered production after the war to become a standard weapon. The British became interested in an even more self-contained mount carrying its own diesel generator, but this ‘Buster’ (Mk VIII) weighed too much (20 tons) and never entered service. The British also developed a six-barrel power-worked Bofors gun, Mk VII, which entered service post-war. After the war the Bofors company developed a 70-calibre follow-on to their wartime 60-calibre gun, which the Royal Navy planned to make its primary close-range anti-aircraft weapon but it was superseded by the Seacat missile (the L70 did enter British army service). A few British ships were fitted with US quadruple Bofors when refitted in the United States. Single Bofors guns were mounted in adapted Oerlikon power twin Mk V and VC mountings as ‘Boffins’ and were in widespread service by 1945. By that time some British ships, including cruisers, had single US-type Bofors (unpowered ‘army’ mounts). Post-war, many ships had twin Mk 5 and single Mk 7 power-worked Bofors, the Mk 5sin conjunction with off-mount Simple Tachymetric Directors (STDs). The pre-war single pompom was little used during the war, but a power-worked version appeared in some numbers.
In 1940 some ships were given 20-tube 7in rocket launchers (for ‘unrotated projectiles’ [UP]) on turret tops. Each rocket carried aloft a line, at the end of which was an aerial mine suspended from a parachute. In theory an approaching bomber fouled the line and brought the mine down on itself.9
Electric Power
Throughout the 1930s and the Second World War, a hidden pressure on British cruisers was the need for more electric power. Wartime modifications dramatically increased power loads, particularly the installation of radar and additional close-range anti-aircraft guns. Required generator (dynamo) power was based on the night action load and on the average maximum harbour loads. The standard requirement was to provide the action load with one generator out of action. Cruisers typically had four generators (dynamos), any three of which should be able to carry the action load continuously. Since one of the three might go down, the remaining two had to carry the full action load for a short time, giving time to reduce that load. In 1938 standard generators were designed for a 50 per cent overload for eighteen minutes. Unfortunately the turbines driving them were designed for a 10 per cent overload (for two hours). DEE wanted half the generators to be able to carry the full load for fifteen minutes. Both prime mover and dynamo should be able to run at 25 per cent overload for two hours. E-in-C much preferred turbo-generators to diesels because they were easier to overload. There was no interest in US-style emergency diesel generators. In 1941 the total connected load in an Improved Fiji class cruiser was 3,280kW, against an action load of 1,240kW. That was 40 to 60 per cent of the comparable US battle load. The ship’s four 450kW generators added up to only 1,800kW. The US Navy required half the ship’s generators to carry the full battle load, so it would have demanded a total of at least 2,480kW, of which half the generators should produce at least 1,240kW. That would have required six 450kW units, 2,600kW.
Electric loads began to rise when the Royal Navy adopted power-operated turrets. The worst before the Southamptons were the ‘Counties’ (980kW night action load); that fell to 760kW in a Leander and to 660kW in an Arethusa. Day action loads were smaller, because they did not include searchlights: 890kW for Southampton (but 930kW for Manchester), 537kW for Arethusa. Corresponding maximum harbour loads (under General Quarters conditions) were 526kW for Kent (420kW in Norfolk), 670kW for Leander, and 560kW for Arethusa. In 1933 the estimated night action load of the Southamptons was 950kW and harbour load was 770kW because the new ships had more extensive power installations, including their catapults, cranes and even torpedo heaters. The Counties had four 300kW turbo-generators, the Leanders four 225kW, so even three generators did not quite cover the night action load.
The situation was further complicated in July 1938 when DEE proposed a new and more survivable electric-power arrangement (ring main) in major warships. It would run behind armour to the extent possible (entirely behind armour in battleships and carriers). It should be split into independent elements, each with its own generator, the port and starboard sides in different main compartments. A generator should be adjacent to each engine room/boiler room unit. DEE wanted half the generators to be diesels, because they could operate independently of the ship’s boilers. However, E-in-C considered diesels less reliable than steam. They were also bulkier, demanding more maintenance and staff. DNC wanted diesels kept to a minimum – one advance claimed for the Fiji design was their elimination. DNC also pointed out that a ship with unitised machinery had only short steam pipes to the turbo-generators in each unit, which limited their vulnerability. Eliminating diesel generators made it impossible for a cruiser to associate each turbo-generator with a separate machinery unit. The new requirement was applied to both Fiji and Dido classes. Their action loads were 994kW and 967kW respectively. DNC protested that the 50 per cent fifteen-minute standard would require 1,500kW for fifteen minutes in a Fiji, which would involve 10 per cent more weight. The 50 per cent margin seemed to be based on the earlier choice that Admiralty generators should have this sort of overload, not on any particular logic. Fijis (except for the first five) got four 350kW turbo-generators; but the 1938 ships were credited with an 1,100kW action load.
The 1941 Minotaurs were originally given the same installation, supplemented by one 150kW diesel generator, but the turbo-generators were replaced by 450kW units (10 per cent overload for two hours). The late 1941 ships (Tiger, Defence and Superb) were upgraded yet again, with four 500kW units (10 per cent overload for two hours) and one 150kW diesel generator. By this time the ring main was normally split fore and aft to overcome action damage, so DEE wanted each half to carry the full action load, and to do so with one of its two generators out of action. The action load in the Minotaurs was estimated at 1,300kW, but the largest available British turbo-generator was rated at 500kW. His solution, four 300kW turbo-generators and four 300kW diesel generators, could not be accommodated, and neither could an alternative of six 450kW generators, with diesels moved to the ends of the ship for better survivability. Instead the ships were given four 500kW turbo-generators and two 150kW diesels, one forward of and one abaft the machinery box. The diesels were emergency generators, as in the US Navy. Newfoundland already had two such generators, and in February 1942 British policy favoured that practice. Minotaur and Swiftsure had enough weight margin to accept the two diesels, but not Bellerophon, for which heavier boilers had been ordered.
In March 1943 DEE proposed upgrades to cruisers already in service, a plan already having been approved for Liverpool, Sheffield, Birmingham, Newcastle and Belfast. He wanted a separate diesel generator in Didos and Fijis, and the 300kW units in the first Fijis would be upgraded to 350kW. By June 1944 DEE wanted a fifth turbo-generator (200kW) and a diesel generator. Of modern cruisers only the small Arethusa and Aurora lacked space for an additional 250kW diesel generator outside their machinery spaces. A planned Dido upgrade had their 300kW turbo-generators replaced by 350kW units, but in June 1944 it was reported that only three of the four could be replaced in Euryalus, Sirius, Scylla and Cleopatra. In all the others existing turbo-generators either had been or were being upgraded. By July 1944 350kW turbo-generators were on order to replace 300kW units in the ‘County’ and Southampton classes. By October the surviving Southamptons (Sheffield, Birmingham, Glasgow and Newcastle) were to get two additional 200kW turbo-generators in their machinery spaces. Hobart was having her 225kW units replaced by 250kW turbo-generators, plus an additional 200kW unit. Emerald was adding two 75kW diesel generators (Enterprise was already up to standard). Further recommendations were: to fit a 300kW turbo-generator in place of the auxiliary boiler in Sirius, Scylla, Cleopatra, Dido, Phoebe and Argonaut (plus uprating existing 300kW units to 330kW in the last three); to add a 300kW turbo-generator in place of the auxiliary boiler in Bellona, Black Prince, Diadem and Royalist; to fit a 350kW turbo-generator in place of the auxiliary boiler in Bermuda, Ceylon, Jamaica, Gambia, Uganda and Newfoundland; and upgrade existing 300kW units to 330kW and add a 300kW turbogenerator in place of the auxiliary boiler in Kenya, Mauritius and Nigeria.
Norfolk is shown in December 1942 after a refit on the Clyde, 14-23 October 1942. Damage to the ship from near-miss bombs led the DNC to reconsider splinter protection in cruisers.
In the wake of the Norwegian campaign, on 29 April 1940 DNC asked the constructor who had been with the fleet what the future criterion for cruiser protection should be. Most German bombs seemed equivalent to the British 500lb SAP, and they were typically dropped in a dive from 3,000ft, equivalent to a bomb dropped in level flight from 5,000ft: therefore 2½in in one thickness; decks should be 3in thick to provide a margin of safety (to give an equivalent thickness in two layers, another 80lbs – 2in – would have to be laid on top of the existing 80lbs). Ships might have to be bulged or, if the state of work permitted, have their sides moved bodily outwards to maintain stability. Didos nearing completion could be given extra upper deck armour in exchange for two turrets. Those in the war programme could be redesigned altogether, with 100lb NC added over machinery spaces and 80lb fore and aft. Fijis nearing completion could surrender ‘Y’ turret to gain about 200 tons, which might give 80lbs (2in) on the lower deck from the stern to ‘Y turret, plus splinter protection over AA guns and torpedo tubes. DNC ordered calculations to show whether a more radically redesigned Fiji could get the sort of protection he was considering for the war programme Didos.
Controller soon asked whether ships could be protected against near misses from bombs, whose splinters could riddle a ship’s waterline: Southampton reported about 240 such holes, Suffolk also a large number. DNC was more impressed by internal damage: in Norfolk, for example, splinters passed through bulkhead and frame plating (total thickness nearly 2in) to flood spaces 40ft from the explosion. DNC proposed that bulkheads be strengthened so that splinters from a bomb bursting in the middle of a large compartment, e.g. a 48ft engine room, should not pierce either bulkhead at its ends. Experience of bomb damage suggested that 30lbs of mild steel should suffice 27ft from an explosion, so bulkheads should be 1in (40lbs); DNC noted that pre-war experiments (Job 74) suggested that 1½in NC armour was needed. Blast would ruin the watertightness of any bulkhead near an explosion, but bulkheads about 15ft away should stand up well. DNC proposed to take up these issues with builders of Didos and Fijis whose ships were not too far advanced. These ships had insufficient deck armour to keep out 500lb bombs, but the effects of hits could be isolated by strengthening internal bulkheads, both longitudinal and transverse. Surrendering ‘Q’ and ‘Y’ turrets in a Dido would release enough weight for the stronger bulkheads (210 tons). The holes would be covered with 2in NC plates, ‘Y’ mounting being replaced by a 4in star shell gun (100 rounds). Surrendering ‘Y’ turret provide a Fiji with ½in bulkheads (500 tons). Nothing could be done for cruisers already in service or nearly complete.
The constructor assigned to consider extra deck armour in the Fijis assumed that the armour deck should be increased to 3in equivalent. There was already a 50lb (1½in) strength deck, of which ½in counted as protection, so another 2½in was needed on the upper deck at least from the forward boiler room back to the after end of ‘Y’ magazine, with 2in abaft that on the lower deck. Protection over the engine rooms could be 60lbs (1½in) instead of 2in. Forward of the forward engine room the forecastle deck should be 2½in thick to the fore end of ‘A’ magazine. Side armour could be left unchanged. The effect was more on stability than on weight, since most of the weight involved was already present in the armour deck carried lower in the ship. The higher weight would cost 2ft in metacentric height, and on completion Fiji had only 2.27ft altogether. A properly protected ship might displace about 9,500 tons, with 65ft beam. If heavier bulkheads and thicker decks were taken into account, the ship needed about 1,360 tons more armour, and about 12ft more beam to preserve stability.
These calculations convinced DNC that a new cruiser design was needed. In August 1940 he proposed a new design for a 10,000-ton cruiser armed with three triple 6in guns, four twin 4in and four (rather than two) quadruple pompoms, plus torpedoes and aircraft. It would be protected as a Fiji except for a thick upper deck (2in to 1½in NC) worked structurally; magazines might have an additional 1in on decks and ½in on sides. Estimates showed a standard displacement of 10,400 tons; the 80,000shp Fiji powerplant would drive such a ship at 31.25kts in standard condition, or 30.25kts deeply loaded.
As a parallel idea, on 30 July DNC asked for a rough estimate of the displacement and speed of a ship with four 5.25in guns and Dido machinery, with protection amounting to 60lb (1½in) side, 2½in on the structural deck, and boxes over the magazines. He apparently had in mind a new design using the machinery already ordered for the suspended Didos.
Late in May 1940, with invasion threatening, ships which could not be completed within the year were suspended, including five Didos and three Fijis, thus giving DNC the opportunity to redesign them. On 8 August he reported to Controller that work on the suspended ships, particularly the Fijis, had gone so far that scrapping them would be uneconomical. The three Fijis could be given better deck and splinter protection if they were given Dido armament (which would be available if three of the suspended Didos were armed like Scylla, with 4.5in guns).10 Of the Didos, three could have improved splinter protection, while the other two could have more radically improved protection through total redesign. In the event, simpler options were chosen: the Didos and Fijis lost one turret each.
In August 1940 the Chiefs of Staff recommended resuming work on the eight suspended ships, and in October 1940 the Admiralty ordered this. In its view the greatest deficiency in cruiser strength was ships suited to trade protection against surface raiders, which at this time was considered very much a cruiser role. The Fiji design was not altogether adequate, not least because it had insufficient endurance: it was noted, for example, that a Fiji could not escort a convoy from Great Britain to Freetown without fuelling at Gibraltar.
HMS Bellona was typical of the Dido class cruisers completed after having been suspended in 1940. She is shown, newly completed, in November 1943. At this time her close-range armament was three quadruple pompoms, three twin power Oerlikons and four single Oerlikons (only in Bellona). By the end of the war the close-range batteries of these ships had grown: Black Prince had two single Bofors, she and Bellona had six twin power Oerlikons (the others had eight), and she had eight single Oerlikons (six in Diadem, four in Royalist). After the war she and Black Prince were both lent to the Royal New Zealand Navy.
Spartan as fitted, December 1943. Her underwater hull has been omitted because it was identical to that of HMS Scylla. She had a retractable Asdic dome. Note that no boat crane was carried and that the number of boats was reduced from the complement of the Dido class. Some radio antenna rigging has been omitted for clarity. Her sister Bellona was completed with four additional single Oerlikons. HMS Spartan was sunk off Anzio by a German guided missile (glider bomb). (A D Baker III)
The Redesigned Dido class (Black Prince class)
By mid-August it had been decided to eliminate ‘Q’ mounting from the Didos but to retain its magazine as a reserve, i.e. to avoid redesigning the hull, and to increase splinter protection.11 Masts and funnels would be vertical rather than raked. That made it possible to lower the bridge one deck and move it slightly forward. ‘Q’ mounting would be replaced by a 4in star shell gun with ready-use ammunition only.12 As in other British warships, cabins and messing accommodations were rearranged to bring officers and men closer to their action stations. Controller approved this redesign early in September 1940, a Legend for the revised design having been produced on 31 August. Estimated standard displacement was 5,770 tons. The new design incorporated radars. Work was resumed in February 1941.
Internal (rather than external) degaussing, bridge protection and reinforced support for ‘A’ turret were added. In January 1941 a 4in HA/LA gun replaced the 4in star shell gun in ‘Q’ position (it was not ultimately fitted, but that came later). The pompoms were given RPC and shielded (1 ton per mounting). Asdic was added. The ship was also somewhat rearranged internally. In July 1941 the 4in gun was replaced by a third quadruple pompom and an admiral’s bridge added. Three single Oerlikons were added, and two power-operated single 2pdrs replaced the earlier pair of quadruple 0.5in machine guns. The ship was further stiffened. In December 1941 ships were ordered fitted for Arctic service. In April 1942 the 5.25in guns were ordered fitted with RPC. A 50kW diesel generator was added. Seven twin Oerlikons replaced the three single Oerlikons and the two single 2pdrs. The aftermost twin Oerlikon was removed and two blast cabs were ordered fitted to two twin Oerlikons: the ships were completed with three quadruple 2pdr, six twin Mk V Oerlikons, and three double ship mountings for Lewis guns. Estimated standard displacement was 5,956 tons.
The Modified Fijis
The three suspended Fijis were Ceylon, Uganda and Newfoundland (two 1938 ships and one war programme ship). Of the war programme, Bermuda had not been suspended; it turned out that modifications applied to the other three could not be applied to her. The ships were called the Modified Fiji class, the three-turret Fijis (until turrets were removed from the earlier ships), or the Uganda class.
On 19 August DNC, having offered alternative modifications to Controller, asked cruiser designer W G John to calculate what could be done if ‘X’ turret were removed and its main battery control function moved to ‘Y’ turret.13 The alternatives of interest were to add two additional twin 4in and two quadruple pompoms (what additional protection could be added?); to add only the two additional twin 4in guns; or to add no further guns, only more protection. Priorities for added protection were splinter protection to the HA armament and ready-use pompom magazines; splinter protection to torpedo tubes; a deck over the small gap between the steering gear and ‘Y’ shell room; and splinter plating at the side of this gap (splinter plating was taken as 20lb [½in] KN). John found that the first option entailed adding twenty men to an already badly-crowded ship, and another two quadruple pompoms would be difficult to place clear of 6in and 4in blast while also keeping the after HACS in a satisfactory position. Controller chose to add the two twin 4in guns, in tandem on the centreline, one superfiring over the other, using the magazine originally provided for ‘X’ turret.
DNO pointed out that with only two HA computers the six-mount battery could not be used to best advantage. War experience showed that a ‘four-cornered’ system was desirable; provide a ‘three-cornered ship’ using a third AA computer would be almost as good. The mismatch between three AA directors and two computers was particularly bad when radar (Type 285) was fitted to the directors – ideally each HA computer had a dedicated director and radar. Without ‘X’ turret, there was little point in having an elaborate means of divided main-battery control, using the Admiralty Fire Control Clock in the LA computer room (Transmitting Station). Eliminating that auxiliary computer would help make space for the desired third HA computer space. The third HACP would add thirteen ratings, and another twenty-six were needed for radar.
Uganda after her Charleston Navy Yard refit (October 1943 – 14 October 1944) following missile damage at Salerno, 13 September 1943. This was her configuration up to the end of her Pacific service in May 1945. Her port crane was removed. A barrage director was installed in front of her bridge. Close-range armament amounted to three quadruple pompoms (one in place of ‘X’ turret), two quadruple Bofors (US type), and four twin power and eight single Oerlikons. The ship’s radar outfit was modernisd. (Alan Raven)
Weight would be available for splinter protection to the 4in guns, torpedo tube positions, HACS supports, pompoms and their ready-use magazine. Estimated standard displacement was 8,640 tons, not too far from Fiji as completed. DNC convinced Controller that no formal Board approval was needed for this modification to an approved design, and Controller approved it on 26 September.
Newfoundland was one of three Fijis which, like five Didos, were suspended in 1940 and completed to modified designs, exchanging additional close-range weapons for one turret. She was photographed by a blimp of squadron ZP-11 on 19 April 1944.
Uganda in October 1944. She was transferred to the Royal Canadian Navy and later renamed Quebec. (Canadian Navy Historical Office)
By this time war experience had shown that accommodation had to be rearranged so that officers and men could be closer to their action stations; the suspension of the three ships made that possible. DEE wanted the ring main electrical system modified to provide for radar (this for Bermuda as well). Assistant Controller wanted the bridge rearranged, the wheelhouse (below the compass platform) eliminated in favour of a new lower steering position. An alternative conning position should be provided below the existing wheelhouse (itself on the level below the bridge), for use if the bridge were damaged. The front of the bridge should be made square to provide more space.14 However, the square bridge was not included in tracings circulated in December 1940.
When revised drawings were circulated in October 1940 the question of extra light anti-aircraft weapons was raised. ADNO suggested that Oerlikons or additional 0.5in machine guns could be added at the expense of splinter protection. He also wanted RPC for the 4in guns and pompoms, but DNC doubted that there was enough space for RPC if a third HACP was installed. An order of priorities was set: (1) Radar (Types 279, 284, 285), (2) RPC for pompoms, (3) a third HACP, (4) RPC for 4in guns; and (5) Type 282 radar for pompom control. In December, DNO felt compelled to switch the last two items because to take full advantage of pompom RPC the director had to be as accurate as possible, which required the radar. As weight compensation for RPC, DNO suggested eliminating the aftermost of the extra two twin 4in guns (neither mount could fire across the stern, because they were blocked by the remaining after turret). That would save twenty-two ratings (sixteen gun crew and six ammunition-supply party). This compromise was generally acceptable; it was approved by Controller (Rear Admiral Bruce Fraser) on 13 January 1941. All HA weapons would have RPC. The ships also had better splinter protection.15 Two more quadruple 0.5in machine guns were added, for a total of four.
By October 1941 plans for the bridge structure included a Fighter Direction Office (FDO). Ships were to be fitted as flagships, with an admiral’s bridge level below the compass platform. Requirements included ice protection as well as wind and blast protection, and experience of air attacks made it clear that the captain (and others) needed a clear view of the whole sky. The admiral’s bridge level included the char-thouse, plotting office, FDO and Radar Control Office (RCO), the last three adjacent to each other, the plotting office being near the Admiral’s bridge so that the Admiral could keep track of the tactical situation. The RCO had to be on the deck immediately below the compass platform, and it also served as a wireless receiving office.
By the autumn of 1941 the fleet favoured the pompom over the 4in anti-aircraft gun, on the grounds that the main threat to individual ships was clearly the dive bomber. For example, there was serious interest in a destroyer armed with 2pdrs instead of any larger-calibre guns. The extra twin 4in planned was superseded by one quadruple pompom, which weighed about as much. However, the ships retained the three HADTs.
Like the original Fijis, these Ugandas were subject to excessive weight growth. Based on the completed weight of HMS Fiji and calculated deductions and additions, in December 1941 estimated standard displacement was 8,691 tons and estimated deep load was 10,829 tons (half-load, taken as average action condition, was 9,983 tons). At that time armament included two single power-operated 2pdr and four Oerlikons (two on the hangars and two on the after superstructure). The Fiji design was so tight that the Ugandas could not sustain the 25 tons of extra weight involved in the main proposed upgrades (including Type 272 radar on the foremast). Space for the close-range directors was also a serious problem. The delay involved for ships expected to complete during 1942 was not acceptable. In February 1942 some further additions were wanted for both the Ugandas and the follow-on Minotaurs: barrage directors, Arcticisation, spare 4in barrels, Mk VI HA directors instead of Mk IV, and two more Oerlikons. Elimination of the port aircraft crane and lowering the boats, as in the Fijis, would make possible three of these five items (the Mk VI director proved far too heavy, and it was not ready until 1945). The situation was further simplified when it turned out that one of the two desired barrage directors could be eliminated in favour of modifying the HA/LA DCT aft.
Given the decision to remove aircraft and catapult from the Fijis, the same change was ordered for the three Ugandas before they were delivered, so the catapult already installed on board Ceylon was removed during construction. One hangar had a flat installed, to function as an AIO on one level and accommodation on the other (as ordered in October 1943), while the other became a cinema or chapel, as in the Fijis.
In April 1947 the long-term close-range armament planned for Newfoundland, and presumably for the others, was two STAAG, three twin Bofors with STD (Simple Tachymetric Directors, comparable to the US Mk 51), and a number of single Bofors to be approved. Substitution of Mk VI HADTs for the existing Mk IVs was being investigated.
Improved versions of the Fiji class remained in production because to shift to any alternative design would have imposed unacceptable delays. In 1940 there was annual capacity for ten cruisers, four with 6in guns and six Didos. However, a proposed 1940 Supplemental programme included four 8in cruisers. That year no cruisers were ordered, because ASW and mine countermeasures craft had priority. For 1941, the Board considered further Didos unnecessary, so it traded them for additional large cruisers. That left seven such ships, the four 8in cruisers (deferred to autumn 1941 and then to March 1942) and three Improved Fijis: Swiftsure, Bellerophon and Minotaur (the Staff originally wanted all seven ships to be heavy cruisers). These ships had RPC antiaircraft armament and added fuel-oil capacity forward, so endurance increased from 5,130nm to 5,550nm at 20kts and from 6,459nm to 6,900nm at 16kts.
A 1941 Supplemental programme developed in the autumn of 1941 replaced three of the four 1941 heavy cruisers (weapons for only one of which had been ordered) with three more Fijis: Defence, Superb and Tiger (however, all four heavy cruisers remained in the notional building programme). The three Supplemental ships were built to a further modified design (Tiger class).
All seven 1942 cruisers were modified Fijis. Only two were ever named (Blake and Hawke, the latter cancelled).16 Due to her suspension, Bellerophon was completed to this design (as Tiger), which was heavily reworked after the war and therefore is described in a later chapter. Of the other two, Minotaur was transferred to Canada as HMCS Ontario. Tiger was reordered (having been renamed Blake, she was renamed Bellerophon) as a unit of the 1944 Neptune class described in the next chapter. The remaining three ships were suspended in 1945 and completed to the new design, Bellerophon being renamed Tiger and Defence being renamed Lion.
Swiftsure is shown as in August 1945, with emergency close-range anti-aircraft additions (to deal with Kamikazes) in the form of eight Boffins and five single Mk 3 Bofors. All the Oerlikons were apparently removed at this time. In April 1945 the ship had four quadruple pompoms, eight twin power Oerlikons, and six single Oerlikons. She was refitted at Sydney, 29 June–August 1945, the twin power Oerlikons being replaced by Boffins (essentially single Bofors on the same power mounting). The single Oerlikons gave way to single hand-worked Bofors, at some gain in weight. (Alan Raven)
Staff Requirements were issued in September 1941, before the after twin 4in of the Ugandas had been replaced by a pompom. DGD wanted a pair of octuple or quadruple pompoms in tandem instead, with the new close-range predictors (instead of pompom directors). Forward, sided octuple or quadruple pompoms would replace the two HADTs. Close-range predictors would be superimposed immediately abaft the pompoms, and an HADT with Type 285 radar would be superimposed above each close-range predictor (in the position currently occupied by the quadruple pompom). Type 271 radar would be mounted on the foremast clear of the DCT. Oerlikons would be fitted as practicable. The need for better forward AA fire was well understood, and Staff Requirements for a cruiser of this size called for six quadruple or four octuple pompoms.
DNC rejected eliminating the after HADT because it was needed to provide secondary control of the 6in guns, and also because a centreline HA director could be essential in conditions of low readiness and surprise attack (as experience with HMS Warspite had shown). He could offer one octuple or two quadruple pompoms, preferring the former to avoid blast and congestion. It was decided to trade the extra twin 4in aft for two quadruple pompoms (total four, two in tandem aft) with close-range predictors on the after superstructure. Similar predictor/radar combinations would replace the pompom directors on the forward superstructure. The stabilised predictors needed gyro transmitting units in or adjacent to existing gyro rooms. Space was so short that magazine stowage was limited to that for two pompoms, but the ready-use stowage at each mounting would be equivalent to one-third of the usual full pompom outfit. All 0.5in machine guns would be landed, two of them replaced by Oerlikons. The control cabinet would be deleted from ‘Y’ turret, but a fire-control clock could be placed in an enlarged after HACP. The after HADT would be rearranged. Late in 1941 a 150kW diesel-generator was added to the 1941 ships in place of the auxiliary boiler (and 350kW turbo-generators were replaced by 450kW ones). To restore stability, the ships were given a foot more beam (increased to 63ft).17 Requirements for clear lines of sight fore and aft made it nearly impossible to provide the usual galley funnels, so the ships used electric galleys for officers but retained oil-fired galleys for the crew. The planned increase in electric generating capacity more than covered this addition.
In October 1941 DNC compared the new version of a Fiji with the roomier Southampton. The two ships were not too far apart in tonnage (8,650 tons vs 9,100 tons). DNC estimated that to bring the latter fully up to date would require 12,000 tons and would cost a knot. Fiji was handier, and the Southampton armament was now outdated. Her thicker side armour covered a smaller area, and the deck over machinery was thinner. War experience (oil fires in Southampton and Sussex) had shown the box magazine protection of the earlier design was unsatisfactory. Due to her greater oil capacity, Southampton had greater endurance (7,130nm vs 6,450nm at 16kts). Both ships were badly congested (complement of 860 in Southampton, 800 in Fiji), but Southampton was somewhat more habitable.
The ships ordered in the autumn of 1941 (Tiger class) had another foot of beam added (64ft in all). The most visible improvement was the Mk VI HA director with its Type 275 radar, which was adapted for blind fire (the earlier Type 285 radar was essentially range-only and could not track a target sufficiently well). The modified design incorporated new target-indication facilities adapted to engaging unseen targets. As in the Fijis and Ugandas, catapult and aircraft were eliminated in the Minotaur and Tiger classes (this was decided on 15 March 1943). Rearrangement was considered urgent for ships expected to complete beginning in December 1944 (Swiftsure, Minotaur and Superb). Space originally allocated to the hangar and catapult was rearranged and close-range armament increased. The hangar was lowered by 2ft at the fore end and 4ft at the aft end, its top now level with the sea cabin flat. Flats were fitted at mid-height in the former hangar space, the upper part accommodating eight petty officers and fifty-four seamen, the lower part providing eleven cabins as well as recreation space and a chapel. The crane was retained and repositioned on the centreline, as in Fijis. The forecastle was extended over the former flight deck, closing in the waist of the ship. As of October 1943 they were expected to displace 8,864 tons standard (GM 3.05ft); deep load was 11,222 tons (4.8ft). Oil fuel capacity was 1,902 tons.
Anti-aircraft armament could now approach or exceed a cruiser specification set out on 18 February 1943: four twin 4in with four directors, four multiple 40mm and ten twin 20mm. The ships would gain one twin 4in, one quadruple pompom and director, and four twin Oerlikons, for a total of five twin 4in guns (three directors), four quadruple pompoms with directors and twelve twin Oerlikons. Two twin Oerlikons were moved forward from the quarterdeck. There was no space and weight for any more pompoms. The only ship completed to the original Tiger design, Superb, had four quadruple 2pdrs, eight single Bofors, two single 2pdrs and six Oerlikons. A 1947 report from the ship pointed out serious deficiencies, which would have been evident had the ship fought in the Pacific. The 6in battery had only rudimentary control for anti-aircraft fire, the 4in battery still relied on the pre-war antiaircraft computer (the Flyplane later associated with the Mk VI director had not been installed, not yet being ready), and close-range weapons were inadequate both in quantity and in quality. Superb compared unfavourably with the slightly larger USS Cleveland.
At the end of the war the design was reviewed for updating, as they were badly crowded. Superb was too far along to change, leaving a Tiger class of four ships: Tiger, Defence, Blake and Hawke. The main new features were replacement of Mk XXIII by Mk XXIV 6in mounts (with RPC), fitting of ‘ring main’ main service and main suction systems, and making certain after bulkheads solid to the upper deck. The ships had a new ventilation system, an automatic telephone exchange and counter-flooding arrangements. Changes described as modernisation during the building process comprised armament and fire-control upgrades, radar and wireless upgrades, and habitability upgrades such as centralised messing.
At a meeting on 20 September the Staff pressed for quadruple rather than triple torpedo tubes, as in the 1944 cruiser, with power training and remote control (approved January 1946). New self-contained 40mm guns would replace the old multiple pompoms. Eight single power-worked Bofors (Boffins) would replace the twin Oerlikons.18 Existing close-range fire-control systems would be replaced by three of the new MRS to control 40mm mountings.19 The 1947 armament statement for these ships showed three triple Mk XXIV 6in turrets (200 rounds per gun), five twin 4in (250 rounds per gun plus 400 star shell for the ship), four twin STAAG (1,440 rounds per gun), and eight single Mk VII 40mm (1,440 rounds per gun). All guns would be fitted for blind fire. The AIO should be moved below armour and enlarged or moved to the centreline and enlarged, in either case to protect it. Radar offices should be moved below armour. All of this would have been difficult, given the ships’ limited internal volume. New radars should be installed.20 Estimated standard displacement was 9,420 tons, compared to 9,240 tons for Superb. All available weight having been expended, VCNS wondered whether some of what was wanted, such as RPC torpedo tubes and Type 268 stand-by radar, was really needed. In February 1946 at attempt was made to prune back requirements. For example, no place could be found for the YE beacon useful in fighter control.
Superb as fitted upon completion, January 1946. She was the only Tiger class cruiser completed to the original design. As such she had both the new 4in director (Mk 6, with Type 275 radar) and the new radar-only pompom director, one per pompom (still equipped with the wartime Type 282 radar, incapable of blind fire). The Mk 6 directors were removed in 1955 (the port-side director remained briefly, cocooned). She had eight single Bofors Mk 3 (hand-worked), four quadruple pompoms, two single power pompoms, four twin power Oerlikons and two single Oerlikons. She could easily be distinguished from Swiftsure by her two (rather than one) barrage directors forward of her bridge. The single Oerlikons were removed shortly after completion. In 1949 the single 2pdrs were replaced by single Bofors guns and the four single Bofors on the boat deck were removed to allow more boat stowage. About 1952 the two twin Oerlikons abreast the LA director on the bridge structure and abreast the fore funnel were replaced by four single Bofors. By that time the depth-charge rack had been removed. (A D Baker III)
In the end, none of this happened. Hawke was cancelled, and the other three were badly delayed as a shortage of electrical draftsmen made it impossible to supply armament drawings on time. In September 1947 the carriers Eagle and Centaur were given higher priority, and work on the three cruisers was suspended and the ships laid up without armament or fire controls. As DGD saw it in September 1947, the ships might still be worth completing. They could be ready in 1953, whereas the next-generation weapon, which became the Sea Slug missile, would not be ready until 1958 (which was optimistic). If a missile ship were given an anti-aircraft rating of 100, a Tiger would have 40 – but the best wartime ships, Ontario and Superb, would be rated at 20, and older cruisers at no more than 5. All but Didos had ‘extremely low value’ as fighting ships mainly because their anti-aircraft fire-control systems were so abysmal.21 Although plans were being drawn up to modernise the cruiser force, perhaps bringing ships up to a rating of 80, DGD doubted that much would be done. It would probably be less expensive to complete the Tigers with new weapons (to reach a similar rating) because, unlike existing cruisers, they had new hulls and engines. Even as designed they were much better than any other British cruiser.
A similar estimate of AIO efficiency rated Tiger at 90, Superb at 70, an 8in cruiser partly fitted with AIO at 35, and a 6in cruiser partly fitted at 30. Communications ratings were 65 for a Tiger, 55 for Superb, and 40 for existing cruisers. The Tigers would have a considerable advantage over other ships in that they would have the anti-torpedo system which was being developed (however this did not, in fact, ever enter service).
Preliminary studies by DNC suggested that one missile system could be installed in a Tiger hull. The trials cruiser Cumberland had already been allocated for Guided Air Projectile (GAP) trials; she was considered far more suitable for that purpose than a Tiger hull (in fact Sea Slug was tested on board the converted merchant ship Girdle Ness).22 DGD saw little point in wasting a new cruiser on weapon trials when ships of negligible value were available from reserve. Director of Airfields and Carrier Requirements suggested converting the ships into small carriers (DNC had recently studied conversion of 8in cruisers).
DNC favoured competing the three ships as cruisers, as they would be superior to the ships they would replace; on 22 January 1948 First Lord approved: they should be completed as soon as economic conditions made that possible. However, work would be suspended until armament requirements were better known. The outcome is described in a later chapter.