The French battleship Richelieu shows her fire controls at Portsmouth in August 1946. She was designed to be armed with a 381mm (15in) main battery and with dual-purpose 152mm (6in) guns. The 152mm (6in) director sits atop the 381mm (15in) one. Richelieu had 14m (46ft) rangefinders (triplex in the forward director, duplex in the turrets, with an 8m/26.2ft duplex unit in the after main battery director). The rangefinder in the director had a stereo element to detect changes of course by the target. As in the previous class, the main-battery director was surmounted by two for the secondary battery, the upper one being for anti-aircraft fire. It was never completed, and it was removed when the ship was refitted in the United States in 1943. The forward main-battery director showed four rather than three windows (with armoured hatches on either side). It had a shorter rangefinder forward of the main director (about mid-length), presumably a spotting glass (this rangefinder is not listed in a book describing the ship, but it is quite visible in a photograph). Similar secondary rangefinders or spotting glasses are evident in the sides of the two secondary-battery directors stacked atop the main-battery director. The secondary rangefinders were apparently removed when the ship was refitted in the United States in 1943 (at the same time the uppermost of the two secondary-battery directors was removed). The director visible alongside the superstructure served the ship’s 100mm (3.9in) anti-aircraft guns. The 152mm (6in) directors visible atop the forward superstructure each carried an 8m (26.2ft) duplex stereo rangefinder, but the after 152mm (6in) director visible carried a 6m (19.7ft) instrument. The secondary turrets and director used single 8m (26.2ft) rangefinders (note the narrower tube). Atop the bridge was a 3m (9.8ft) stereo navigational rangefinder. Just abaft it was a 4m (13ft) stereo rangefinder for the ship’s 100mm (3.9in) tertiary battery, added during construction because the 152mm (6in) guns were insufficient for long-range anti-aircraft fire. The ship was modernised with a mixture of British and US radars, such as the British Type 285 on the secondary director and Type 284 on the main-battery director, the British Type 281B air-search antenna at the foretop, and the US SA-2 radar on the mainmast, SF (surface-search) in the small radome forward, and SG surface-search set on the foremast. Richelieu’s sister ship Jean Bart was completed postwar with a completely different main-battery director similar in shape to a British DCT, carrying a main-battery radar (ABM) on its flat face, with a 14m (46ft) triplex stereo rangefinder. Above the radar were four windows like those of the earlier Richelieu director.
Le Prieur’s combinateur was superseded by a range-keeper produced by the Direction d’Artillerie Navale de Toulon. It became the basis of further development. For the first time in France it separated own and target calculation, using two separate plateaus, each driving its own tachymeters for motion along and across. A feature used again in later systems was to indicate the trial solution by two wires: horizontal and vertical for speeds along and across. A diagonal wire indicated trial inclination. Apparently this was independent of the British work on the AFCT. The method of evaluating a trial solution did not resemble the cross-cut concept in American and British systems. In the 1922 production version the table required a supervisor and three operators (for target, shooter and wind). Although this range-keeper was successful, only eight were made (two by the Direction d’Artillerie Navale in Toulon and six by Breguet in Paris).
Working with the range-keeper were range and bearing plotters. They were the vital means of feedback, and the French also considered them essential to a ship firing at a target that might be visible only intermittently. Readings from the four main rangefinders (masthead and triplex) were averaged mechanically. Simpler receivers showed ranges from the five or seven turret rangefinders and estimates by the fire-control officer. Corrections by the gunnery officer were taken by telephone and entered manually. To allow for limited plotting-paper width, the range plot was periodically recentred.
The first full system, for PC Model 1922, was operated by eighteen men with an officer and a petty officer (chef de poste) supervising them. As in the 1917 version, it separated the target motion (range and deflection) and ballistic elements of the problem, with the target motion in the centre. As before, the supervising officer stood at the graphic plotter for range. He had his aide next to him. To his right (at right angles) was the pair of plateaus (target and own-ship) combined in one casing, each with its own operator. Next to that was the bearing plotter (IVL, indicateur de vitesse laterale), with its operator. Behind it was the deflection operator, with two tachymeter operators and the gyro-compass receiver. As in the earlier system, the ballistic corrections and wind corrections were inserted to one side, in this case behind the plateaus. Spotting corrections and transmission to the guns were all concentrated on the other side of the space, each with its own operator. Errors were registered in both range and bearing (ecartes orientées). Many of the panels and devices needed operators to follow up their indications in order to insert these numbers back into other calculators; the entire PC was hardly integrated as desired. However, its personnel included an electrician, to keep it running, a technician (the derouleur) to keep the plotters running, and a talker at a bank of telephones.
The core of this system was the new separatedelement plateau. It was tested on board the cruiser Pothuau and then installed on board the modernised 23,000-ton battleships. A plateau with the associated integrator (tachymeter) Model 1920–23 was installed for the battleship secondary batteries and on board the new 8000-ton light cruisers.
The first integrated system, the prototype for the later ones, was developed beginning in September 1925 for the new Duquesne-class heavy cruisers. Work required about 6000 man-days; it was completed in 1927. Design was complex partly because some elements had to reverse their positions up to twice a second, yet positions had to be shown precisely. Data were carried internally mainly by rods, and, to avoid inaccuracies as they twisted, it used 600 roller bearings – and 300 gear wheels in differentials to add data. Minimum complement was eight, but normally the PC required two officers and eighteen enlisted personnel. If the tachymeters were not working another six men were needed.
The new design reflected the modernisation of French industry, which was moving towards standardised parts and precision measurements, particularly using standards promoted from 1925 by the Bureau de Normalisation de l’Automobile (BNA). It exploited new American machinery for precision gear-cutting and also newly improved ball bearings. More parts were interchangeable or suited for mass production. Existing commercial synchronous motors no longer seemed precise enough.
The calculating part of the system was progressively simplified, so that in 1930 the new cruiser or destroyer range-keeper required only two operators. It incorporated a parallax corrector and a single dial to enter temperature and pressure corrections. Corrections for drift and for the movement of the firing ship were now automatic. In 1937 a French artillery officer assured students at the French naval war college that the target element of the fire-control systems on board the heavy cruisers were adapted to take target turns into account.
The new cruisers were expected to fight fast, violently manoeuvering targets at ranges as great as thirty kilometres (eighteen and a half miles) (the designers worked to forty kilometres, which meant 100-second time of flight). Maximum target speed was forty knots (twenty metres [twenty-two yards]/second). That was not all: experience had already shown that a heavy cruiser could change course about two degrees per second. The system was therefore designed to take data from an inclinometer, rather than to rely on a deduced constant target inclination.
High-speed combat required that data be input instantly and that results appear instantly. Delays inherent in manual data entry could not be tolerated: even if one turn of a handle represented 500 metres (547 yards), the delay involved in turning it would be noticeable. Past calculators had generated future target position, but in this system it was displayed on a third dial so that future target bearing could be compared directly with present bearing. That did not require any new integrators, but the presentation was unique to the French navy. As in other synthetic systems with separated elements, this one used four integrators (tachymeters): two for range rate (own and target) and two for speed across (own and target). Design was greatly simplified by reducing all range corrections to changes in time of flight; range outputs were reduced to gun elevation and time of flight. A fourth dial showed wind speed and direction, because the longer ranges now desired entailed more wind effect, both because they acted longer on the shell and because the shell attained higher altitudes, where wind was stronger.
As before, the future-target dial showed horizontal and vertical wires for trial speeds along and across the line of fire. The third wire, however, was set by inclinometer (or other observation). The three wires formed a triangle, the chosen solution being its centre. A further innovation making for faster computation was a calculator that could indicate how a result could change if the inputs changed slightly, rather than require resetting.
Associated plotters (for feedback) gave range and bearing.
The first of the new battleship fire-control systems, for the Dunkerque class, was a modified heavy-cruiser system, although it was associated with entirely new directors. Like the cruiser, the battleship required twenty-six personnel in her PC; she also had a four-man computer in each main-battery turret, and three nine-man anti-aircraft control posts for her dual-purpose 130mm (5.2in) guns.
The last pre-war cruisers, the La Galissionière class (7600 tonners) had a new kind of PC designed specifically for dual-purpose fire, although they had single-purpose 155mm (6.1in) guns. (The Richelieu class had the same guns in triple dual-purpose mounts, which may also have been intended for the cruisers). The concept was to transfer the usual salvo firing from surface to air targets, using time-fused projectiles. To do that, calculations had to be completed much faster, the guns ready for instant action. Because anti-aircraft shells were much lighter than anti-ship shells, they entailed different ballistics and corrections were more critical. To accommodate its extra functions, the PC had to be larger than that of a heavy cruiser. The kinetic section was U-shaped, with a desk carrying the plateaus and the target-estimation cross-wires on one arm and the azimuth plotter on the other. The two arms were connected by a long arm carrying range and altitude plotters, with a parallax-inverter between them (parallax was particularly important). Behind this arm was the big block of tachymeters. To the usual pairs of tachymeters for range (own and target) and bearing was added a fifth, for range against an air target. The additional range calculator was needed because range rates against floating targets might be up to twenty metres (twenty-two yards)/second, but against an air target 160 metres (175 yards)/second had to be allowed for (about 315 knots). The usual separated plateau calculators were abandoned in favour of more compact ones easier to integrate into one unit. Because the inclination of the target in the firing plane might change while the shell was in the air, the opportunity was taken to split the target plateau into a plateau for current target inclination and one for future target inclination. In the case of an air target, the second plateau was used for wind at the target. Because it had to be used for air as well as for surface targets, the target plateau (current) had to have two speed scales.
Overall, the difficult requirements imposed by the dual-purpose PC pushed French designers towards greater automation, generally using electrical methods. This included a better method of averaging ranges for plotting. This technique in turn was applied to many existing 1929M plotters at the end of 1936 (as conjugateur 1929M36), the rangefinder data automatically being turned into an average curve rather than the usual series of dots, which had to be traced by eye.
The new PC required twenty-eight personnel, but only four or five of them needed much training (ie, would be required to exercise judgement); the others operated follow-ups or performed other, essentially mechanical tasks. Compared to a heavy-cruiser PC, this one used the same ‘forest’ of rods, but hid them away.
The new generation of fire-control systems changed French naval tactics. The French were very much aware that they were gaining freedom to manoeuvre while firing, to the extent that a 1937 lecturer at their naval war college commented that with such freedom much of the difference between armoured and unarmoured ships was disappearing. Manoeuvre as protection against fire became a theme in lectures at the war college, even though, as the French freely admitted, they knew little of enemies’ fire-control systems. The time scale for manoeuvre depended on the dead time between firing and making spotting corrections, which might be as little as fifteen seconds once the enemy had the range (the idea of changing speed was dismissed because the dead time involved was at least a minute). The idea of manoeuvre as protection against fire also affected ship design. The 1937 lecturer argued that the new Dunkerque- and Richelieu-class battleships were unusually well adapted to such tactics because by concentrating their armament in two superimposed turrets forward they gained unusually wide arcs within which to manoeuvre while firing.
The Mediterranean Fleet battleships conducted their first experiments in long-range firing in 1926, ships opening at extreme range (in one case at 22,000 metres/24,060 yards, in others between 16,000 and 20,000 metres/17,500 and 21,872 yards). Such ranges were far short of the theoretical maximum, but they were set by the limits of the guns and, for ships without directors, by the limited visibility from the turrets (maximum 16,000 metres/17,500 yards). Another surprise was that long range made for a slow fire-control tempo (ie, the entire fire-control cycle) because of the long time of flight. Typical reloading time was forty-seven seconds, and spotting time was fifteen seconds. Thus if time of flight (plus fifteen seconds) was less than reloading time, the ship had to be able to fire salvoes faster than she could reload, so half-salvoes were the rule. However, at a range of 22,000 metres (24,060 yards) time of flight was forty-seven seconds, so the guns had to await spotting to fire. The crossover came at 15,000 metres (thirty-two-second time of flight). This was the logic of the British ladder technique, apparently unknown as yet in the French navy, despite the presence of French officers in the Grand Fleet. At this time the French apparently assumed that they would need four salvoes to begin to hit, which at 22,000 metres (24,060 yards) (interval sixty-one seconds) meant four minutes four seconds. Overall, the rate of fire might fall to a third of that at shorter range, say from eighty to twenty-six shots in ten minutes. Yet speed would be increasingly important in combat. Earlier practice was to expend initial salvoes to correct for line, before correcting for range. With so few salvoes available, that meant wasting too much time. It might be possible to avoid the initial salvoes altogether. In seventeen runs ships usually got line correct on their second or third salvo (three on the first, eight on the second, six on the third). Perhaps salvoes could be fired for line and range. The fleet also found that its dispersion was increasing, in some ships doubling to 600 metres (656 yards) in three years. Director control did not seem to affect performance.
The fleet also tried two concentration shoots (three ships each time), one using centralised control, one using a mixed system of autonomous and paired control. In centralised control, coordinated by short-wave (HF) radio, the first salvoes of the ships (three, three and five rounds) included nine rounds falling within 175 metres (191 yards), and four on target, all within about forty seconds. Firing autonomously, the three ships got seven rounds on target during the first two minutes and thirty seconds of fire.
Despite considerable talk about future ranges of 40,000 metres (43,744 yards), and the design of guns capable of firing to ranges as great as 35,000 metres (38,276 yards) (in response to the French commission on naval lessons of World War I), as late as March 1934 a student thesis accepted by the French naval war college began with the statement that almost nothing had been done to push range beyond 18,000 metres (19,685 yards). Only very recently had tests been conducted with the cruiser Colbert.39 However, shells were redesigned for greater range, the last pre-World War II generation being boattailed with long windshields (approaching half the length of the shell). Maximum range of the 381mm (15in) gun (thirty-five-degree elevation) was 41,700 metres (45,600 yards), compared to 42,260 metres (46,216 yards) for the Italian 15in/50 and 35,550 metres (38,877 yards) (at thirty degrees) for the German 381mm/15in/48.4.40 The last ships to fire extensively pre-war were the Dunkerque-class light battleships; on at least one occasion Dunkerque fired to a range of 41,000 metres.
As late as 1936, the US naval attaché reported that the French preferred to use director fire in train and pointer fire in elevation. Typically salvo bells allowed a six-second interval, during which pointers could fire if their sights were ‘on’. These comments presumably applied to old battleships rebuilt before modern director systems had entered service. In these ships the director was inside the conning tower. In cruisers, however, it was combined with a rangefinder at the masthead, and in destroyers the director was a large structure above the bridge. Ships fired ladders (echelons) with a 183-metre (200-yard) step, typically comprising three steps above rangefinder range (plus a ballistic correction). US observers considered French use of aircraft for spotting undeveloped by their standards. However, they did note that the French conducted offset firing for realism (they thought an air spotter would find it almost impossible to correct fire under such circumstances). Dispersion between turret guns was small, a figure of fifty metres (fifty-five yards) at 28,000 metres (30,621 yards) (presumably for a heavy cruiser) being quoted. Delay coils were introduced in 1941 to reduce salvo spreads.
Pre-war interest in concentration fire was revived. As modernised, the Bretagnes could fire beyond the horizon; three of them firing together could deliver twenty-six tons of shells on target. The French therefore became interested in master-gun tactics, the ship closest to the target (the director) carrying the observer. Ranges would be transmitted by radio every two seconds, alternating with target bearing. All ships would fire together on order, each interpreting the data sent by the directing ship, and using her own computer to work out range and range rate. Each ship would keep track of the position of the directing ship. To this end each of the Bretagnes and Courbets was fitted with a separate central de telecommande forward of the PC for her main battery. Her foremast carried a 2m-(6.5ft-) diameter range dial (marked in hm increments, 0 through 9); a second range dial was on her mainmast. Concentration dials also appeared on board other French warships, down to destroyer size.
Concentration was particularly important for the super-destroyers (contre-torpilleurs), organised in divisions of three; the French hoped that a division could overwhelm a light cruiser. Beginning with the Fantasque class, they had special HF radio circuits to pass range, target course and speed, deflection, and spots (using dedicated antennas strung from the foremast to B gun platform). A separate higher-frequency circuit was provided for a bridge-to-bridge radio telephone, for tactical control of the division. Master ship concentration (directed by the division leader) required that each ship know not only the range to the target but also the range and bearing of other ships from which target data could be taken (to determine parallax). To this end they had pairs of 0.8m (2.6ft) coincidence rangefinders, one for each possible engaged side (they were replaced in 1939 by 1m (3.2ft) stereo units, which could also be used for antiaircraft control).
In the summer of 1936 the French navy began widespread use of K shells, which had coloured dyes to distinguish splashes from different ships firing together. This technique so impressed the Royal Navy that it adopted the French shells in modified form during World War II.
Although the French navy had little chance to demonstrate long-range gunnery during World War II, it did show impressive capabilities in its few battles against the Royal Navy. At Dakar in September 1940 the super-destroyer Le Fantasque claimed hits at 20,600 metres (22,528 yards) (using the ship’s roll to extend the usual maximum range of 20,000 metres/21,872 yards). The cruiser Georges Leygues claimed two hits on the battleship HMS Barham at 16,000 metres (17,500 yards) and she and Montcalm duelled British heavy cruisers at ranges between 24,000 and 27,000 metres (26,250 to 29,530 yards) (the French cruisers were straddled constantly but not hit).