You are viewing a news item in the old website format. There may be display issues in some browser versions.


The Chieftain's Hatch: Centurion III Pt 1.

You will recall that some time ago I covered the US Army’s evaluation of the Sherman Firefly in three parts comparing it to the M4(76) and the M26. Frankly, Firefly didn't fare all that well. The Cruisers didn't do all that well against Sherman in either British or American tests either. Well, a couple of years later, a similar, albeit less rigid, set of tests was conducted, comparing Centurion to M26 and T29. 

These tests were a little more generic, as they were more focused on the tank as a whole rather than, as in the Firefly’s case, the weapon installation effectiveness. I’ve come across three Centurion reports so far, and I’ll go over then over, again, a period of articles. Interestingly, they are not in the sequence you might expect: The first two reports were filed on the Centurion III in the first half of 1950, the third was of the Centurion II in the second half of the year.

This is explained as follows:

The two Cent III reports were divided into gunnery, and automotive. Army Field Forces Board No 2.  [AKA Armored Board – Chieftain] received the Centurion III early in September 1948. The tank received was the first production model to be manufactured and, as a conswequence, it was necessary to proof-fire the main armament at this station. Unfortunately, the initial amount of 20-poinder ammunition received and the relatively short tube life prevented a comprehensive testing program. In addition, shot APCBC is not procurable at present and the gun on hand mounts a counterbalance in lieu of a muzzle brake as the design for the brake is not yet firm. Consequently, only preliminary testing, covered in this report, has been possible to date; further and complete tests have been postponed pending receipt of additional HE and APDS ammunition, an allowance of APCBC ammunition, a muzzle brake, and additional tubes. The interim period will be spent in a complete evaluation of the stabilizer installed in a Centurion II tank mounting a 17-pounder gun for which there is adequate ammunition. This vehicle arrived at Fort Knox in August of 1948, but certain essential components had deteriorated so greatly as the result of arduous trials at Aberdeen Proving Ground as to preclude immediate testing at Fort Knox. It is expected that this tank will be refitted in November of 1948 and that the complete evaluation of the stabilisation system contemplated with the Centurion II will materially reduce the stabilizer testing required with the Centurion III.


The Centurion III as tested by Armored Board. Note the painted-on PT belts.

I shall now start transcibing the results of the tests. This is the first test, that of the fighting compartment of Centurion III. T29 was not incorporated into this first particular part, sorry. Let's see if Centurion manages to redeem British tank design in American eyes which, as you will recall from this and other

SUMMARY OF : TESTS. . Testing of the fighting compartment of the Centurion III has been limited because of the lack of time and the afore-mentioned shortage of ammunition and replacement gun tubes. Tests conducted at this station to date have been of such a nature as to preclude complete evaluation of any or all features and, of course, proper comparison with similar items on the Medium Tank, M26, and the Heavy Tank, T29. Consequently, in no respect can tests of any single feature or phase be regarded as complete; and results, although in many instances strongly indicative, are not necessarily representative of final judgment on stabilization,fire control, and main armament. Result  of limited testing conducted to date are:

 a. Reliability, The Centurion III, which was new when received, has been operated a total of 305 miles. The power traverse system, non-stabilized, has been operated intermittently approximately 9 hours, 40 percent of which was under stand-to or light load. The stabilizer system (which includes power traverse) has been operated intermittently an additional 8 ½ hours, some 25 percent of which cap be considered as stand-to or light load. In addition, 6 specia1 proof rounds, 63 rounds of shell HE and 69, rounds of shot, APDS have been fired, in the main, from a stationary vehJcle. This expenditure of ammunition represents approximately 41 equivalent full charge rounds (EFC) or 69 percent of the estimated life of the presently installed tube. The expected tube life is 60 EFC, which amounts to 120 rounds of shot APDS or l,200 rounds of Shell, HE,and at this figure the expected mean wear is 0.12 inches. The present maximum wear at commencement of rifling, which is 0.087 inches, confirms the predicted short life of the 20-pounder tube. During the life of the gun the average loss in velocity is 17 feet per second per 0.01 inches of wear. During this period no malfunctions or. discrepancies of any significance have been noted despite the fact that the muzzle of the 20-pounder was struck a severe blow by an M26 tank retriever. This accident, incidentally, broke the gun traveling crutch, pushed the open breech block back into the deflector guard, and forced the gun to recoil one inch beyond normal limits without serious consequences.

b. Ease of· maintenance No apparent difficulty has been experienced in handling, disassembling, or assembling those items normally subject to field maintenance by tank crewmen.

c. Fightability.

            1) Observation of Fire. When firing shell HE of comparable muzzle velocities, no material difference has been noted between the Centurion III and the M26 in ability of crew members to observe fire.k However, observation of fire from the Centurion III when firing shot APDS was inferior to that from the M26 when firing shot HVAP-T. Examination of subsequent sub-paragraphs will disclose the reasons for this difference, but it must be remembered that observations apply to the 20-pounder mounting a counterbalance in lieu of a muzzle brake. The presence of a muzzle brake will conceivably alter present findings.

            a) Blast. Test personnel were uniform in their opinion that the blast effect on the commander of a 20-pounder firing APDS was markedly less than that of an 90mm firing HVAP-T. The blast effect under the muzzle of the 20-pounder was much more severe than that under the 90mm muzzle, and as a result, the dust raised by the former was a great hindrance to observation.

            b) Muzzle flash. Muzzle flash with the 20-pounder was greater and it was felt that it interfered considerably with the commander’s observation of tracers. Blinking during the instant of firing did not help materially. This flash was not visible to the gunner.

           c) Tracers. Tracers were always visible to flank observers. However, crew members of the Centurion III experienced great difficulty in spotting tracers because of in-experience, muzzle flash, short time of flight and, above all, dust; whereas HVAP-T tracers in the target area were normally visible to crewment of the M26. With dust dampened, the experienced British gunner was able to pick up about 50 percent of the tracers when firing at ranges in excess of 1,600 yards.

            d) Fumes. Fumes from both shell HE and shot APDS were less noticeable and far less irritating in the Centurion III than in the M26 firing either shell or shot.

            2) Adjustment of Fire. No difference in ease of adjustment of fire was apparent between comparable types of high explosive fired from the 20-pounder and the 90mm. Adjustment of fire with the shot APDS, on the other hand, was more difficult because of the poor observation from the Centurion III as outlined above. However, this did not prevent the Centurion III from acquitting itself favourably in initial and abbreviated ease of adjustment problems. The 20-pounder achieved 3 direct hits and one ricochet hit out of 7 rounds when firing at three 6’ x 6’ olive drab targets at ranges varying from 1,800 yards to 2,800 yards, while the 90mm firing HVAP-T and using an M83C telescope graduated for ABC obtained 2 hits out of six rounds at two similar targets 1,800 and 2,500 yards, and failed to get a hit at 2,800 yards with 5 rounds. Users and observers attributed this favourable showing to the flat trajectory of the discarding sabot ammunition which partially negated ranging errors and deflection errors due to cant. In addition, and because of this flat trajectory, a standard correction of up or down 200 yards (on the range drum) was used in adjusting fire with 20-pounder shot APDS. Again, all were of the opinion that this method of adjustment was feasible and that its use would eliminate the necessity for accurate sensings in terms of displacement between projectile and target since such a correction could be based, normally, upon a mere sensing of short or over. It is interesting to note that after an unsensed (except by flank observers) ricochet hit at 2,800 yards on the first round, the commander of the Centurion violated this arbitrary method of adjustment and elected to correct in terms of displacement and range error. This projectile struck at the base of the target and it required a third round to obtain a hit.

            3) Handling and loading ammunition. No specific trials of this nature have been conducted; however there is no reason to believe that the rate of fire of the 20-pounder will differ appreciably from that of the 90mm. The various loaders used have expressed a preference for the Centurion III, despite the fact that the 20-pounder breech is higher than that of the 90mm, stressing the more ample space provided by the loader and the smooth breech action of the 20-pounder. A total of 65 rounds of ammunition are stowed: 8 rounds are quickly available in ready racks, 20 are in the forward part of the hull to the left of the driver, 2 are in rear bulkhead, and the remainder are in turret floor bins. Access to the front hull bin is possible except when the turret is being traversed counterclockwise approximately from 0-180 degrees, but access to the floor bins is difficult under any circumstances because of the spent case bin.

:Ejection of empty brass. No failures to eject the empty cartridge cases have been encountered. In addition, performance of the spent case bin has been completely satisfactory, which is essential because the turret basket is only a 90 degree segment.

 Gun controls.

            a) Manual Traverse. Operation of manual traverse is far from satisfactory because of the presence of the Autolok which very effectively prevents undesired turret rotation during vehicular movement. As a result, unless operated by a skilled gunner capable of applying a continuous and uniform force, movement is jerky. In addition, at certain turret positions when the tank is pitched rotation in one direction or the other can be accomplished only by repeated blows on the handwheel. The traversing handwheel in some respects is conveniently located. However, its position does not facilitate the continuous effort required for traversing, and clearance between the handwheel and power traverse controller is at a minimum.

            b) Power traverse controller. The vertical spade grip in most respects appears satisfactory although complete evaluation must await further testing. The fine response possible with this system permits accurate and quick laying on stationary point targets and precise tracking of slow moving targets. However, the necessity of continually gripping the controller in order to keep the clutch switch compressed is fatiguing. In addition, the position of the controller requires the gunner to thrust his arm under the manual traverse handwheel and around one side or the other of the vertical handle on that wheel. This handwheel, however, does not rotate when using power traverse.

            c)         Commander’s controller. This controller, located at an angle below and to the right front of the commander, with the grip up, appears to be highly suitable and properly positioned. It is felt that less strain is experienced operating this controller and most users prefer it to the gunner’s vertical spade grip which seems heavier. 

            d)         Elevation handwheel. Operation of this handwheel, located to the left and forward of the gunner, and of the elevation mechanism as a whole is extremely satisfactory. The handwheel mounts a comfortable and substantial horizontal grip with firing button and the position of the hand when elevating seems natural.

             e)         Power elevation controller. Limited observation has disclosed no deficiencies. This controller is positioned head-high and well forward on the gunner’s left and the traverse controller waise high and to his right front. This combination seems to enable the gunner to position himself firmly with brow against sight pad.

             f)         Machine gun mechanical firing gear. This firing control has not been tested to date, but it is a foot pedal with disadvantages inherent in that type of equipment.

             g)         Electrical firing switches. Main armament switches located on both elevation controls appear adequate. In fact, gunners expressed a strong preference for the location of the firing button on the elevation controls. There is no emergency means of firing the main armament, although such a gear is planned.

             6) Fire control equipment.

            a) Gunner’s periscope. The gunner’s periscope, used in conjunction with the range drum, has many advantages when compared with presently installed U.S. sights. The combination is extremely valuable when firing from a moving vehicle as either the commander the gunner, using the range adjuster on the range drum, can continually set in range changes. An experienced gunner, incidentally, can read essential scales with his left eye through a reflector on the left side of the 6-power eyepiece without removing his right eye from the periscope sight. He must, however, release the elevation control and firing button in order to operate the range adjuster. In addition, the periscope appears to lend itself readily to the elevation motions which are of a considerable magnitude and violence in stabilized firing; it has a large and readily available field of vision. The sight-range drum combination is equally valuable when engaging a moving target as it enables the gunner to set in a range and to place a lead line on or in line with the target, eliminating tracking in an empty space on the reticle. On the other hand, it is felt that the lack of a definiteaiming point or cross in this reticle does limit the gunner’s ability to repeat the same laying in stationary firing. The aiming area, so to speak, is an imaginary square, 5 minutes on a side, formed by two pairs of vertical lines. Moreover, as with most mechanical linkages, backlash exists in the rotating range drum and in the pivoted periscope. It is felt that the construction of the range drum could be improved and that, at least, it should be made dust-proof. There is no built-in correction for drift or cant, and the principle of design and basic operation preclude such corrections.

             b)  Cupola episcopes. No limits of vision tests have been run; however, the nine episcopes mounted in the commander cupola would appear to have value during moving vehicle firing.

              c) Vane sight. This sight has not been tested as yet, but appears more practical than the version on the M26.

             d) Clinometer. The clinometer is fastened to the range drum mounting and is so positioned as to make accurate reading difficult. The adjusting nut or lock is unsatisfactory.

              e) Azimuth indicator. The indicator is all-electric and seems to be very satisfactory. It is ideally positioned directly in front of the gunner and is graduated in degrees and minutes.

              f) associated electrical equipment. Illumination devices appear satisfactory. The two bulbs in the azimuth indicator are poorly mounted from a viewpoint of sturdiness.

                7) Stabiliser. Stabiliser tests have not been initiated. Results outlined below are based upon observations during practice and demonstrations.

                a) Warm-up time. It required 36 minutes to place the stabilizer in operation. Thirty minutes of this time are consumed in warming the gyros by turning on the master battery switch which in turn actuates the gyro heaters. In addition, the main motor should be operated 1 minute before staring the auxiliary motor, and it requires 5 minutes for the alternator to warm up.

                  b) Controls. The alternator and metadyne switches to the left and below the gunner are awkwardly located and the switches are so arranged as to encourage mistakes. However, the operation required to stablise is simple and consists of locking the elevation handwheel, which is easily accomplished, and of throwing the stabilizer change-over lever to the right. The procedure to change back to straight power traverse is about the same, unlocking the handwheel and returning the change-over lever to its original position, but the necessity of meshing gears, although a minor problem, consumes slightly more time. The elevation and traverse trimmers appear to be conveniently located.

                 8) Safety devices. The loader’s firing safety switch and loader’s traverse safety switch, depression stop switch and limit switches, and recoil indicator are desirable features. Actually two traverse safety switches appear to be necessary.

                 9) Turret seats and seating arrangement. Seats appear sturdy, are readily adjustable, and no malfunctions have occurred to date. The gunner’s position is confined by such space as is available is efficiently utilized. The gunner does have difficulty reaching his seat and, in addition, the smoke ejector firing buttons are suspended from the turret roof immediately in rear of his head which is a hazard. The commander’s position in the Centurion is exceptionally well planned and arranged in contrast to that of the M26.

                10) Hatches. All turret hatches have functioned normally, although some hatch locks are poorly positioned. The location of the escape hatch in the rear of the turret appears extremely practical.

 I’m going to skip some of the technical stuff but do make the following notations:

With the tank level, it took about 4 pounds of pressure to manually traverse the turret, the average gunner could crank at 19.5mils/second. (A little over two cranks a second). The gun was very well balanced, with average elevation effort being 2 pounds to elevate, and 4 to depress.

                e. Dispersion Firing.  

                 1. Shell, HE.

                (a) Ten consecutive rounds fired at each of the following measured ranges at a vertical, 12’ x 12’ panel with aiming point.


               b) With the cold gun, it usually required two or three rounds to place the center of impact centrally on the target. This difficulty was aggravated by lack of definite knowledge of required correction for drift at various ranges. However, once on the target, all subsequent rounds remained on.

            2) Shot APDS

            a) Ten consecutive rounds fired at each of the following measured ranges at a vertical, 12’ x 12’ panel with aiming point.


           b) The same difficulty noted with high explosive was experienced in locating center of impact of shot APDS centrally on the target. Once the center of impact was located it was possible to shift it to an adjacent piece of armor plate, 55” x 67” at 2,000 yards and in doing so first round hits were obtained on three different occasions.

             c) Note habitual difference between overall size of the 100 percent and the 80 percent group in the table above. This difference points up the observed fact that one or two rounds out of every ten deviated considerably from the center of impact.

             3)  Reasons for dispersion.

            a) Inability to repeat exact lay each time with 6-power periscope as previously discussed.

            b) Low muzzle velocity of shell HE which is 2,000 feet per second. 

            c) Non-uniform discarding of sabot. – 50 percent of carriers were normally found within a 10-yard radius of one another, 30-40 percent within a 50-yard radiud, and 10-20 percent were considerably off in range from the others.

            f. Moving target fire.

             1) Firing results. Gravity type target, 6’ x 6’, moving on a diagonal crossing course varying from 1,550 yards to 1,400 yards measured range at an average speed of 13.5 miles per hour, and engaged by stationary vehicle using power traverse.

             2) Results can be attributed to the fine degree of control permitted by power traverse; excellence of manual elevating mechanism, handwheel and firing button; ability to set range on range drum; and less likelihood of misses due to elevation errors. Results were obtained witrhout benefit of round by round sensings by the crew of the Centurion III. However, a critique was conducted after each target run.

             3) It is interesting to note that an expert gunner firing 90mm HVAP-T from an M26 using power traverse failed to get a hit in 18 rounds, primarily because of elevation errors. When this gunner was shifted to the Centurion III and given the proper sight picture, he achieved 3 hits of 3 rounds.

              4) The lead required by the M26 firing 90mm HVAP-T was 7.5 mils from the forward edge of the target, while the Centurion III firing 20-pounder APDS only required slightly over half that amount of lead.

              g. Stabilized firing. No stabilized firing tests have been conducted. However, after one dry run and during a demonstration, an experienced gunner with an inexperienced crew obtained 2 hits and 3 close shorts with shell HE out of 5 rounds fired. The range varied from 1,300 to 700 yards, tank speed was 15 miles per hour, and dust raised by strikes in the impact area handicapped laying.

              h. Armor penetration with APDS. The first round fired (tube life previously expended: 20EFC or 33 percent) at homogenous plate, 6” x 66” x 67”, at a range of 2,000 yards, inclined at 694 mils (Approximately 39.0 degrees) from vertical and located slightly above gun (angle of site 6.5mil). A complete penetration resulted and the top of an M26 turret backing was dented about ¼ inch and a 2-inch piece chipped from cupola cowling. The second round fired (tube live previously expendsed: 26EFC or 43 percent) at the same plate inclined 803.8 mils (about 45.2 degrees) and at the same angle of site struck within 3 inches of the side edge of the plate, gouding out all but 1 ¼ inches of plate before slewing out the side. The third round fired (tube life previously expended: 40 EFC,or 67 percent) against the same plate, again at 2,000 yards and at about 804 mils inclination, failed to penetrate, bulging the rear surface of the plate slightly. The course of this projectile, which remained in the hole, was at a considerable upward angle. Complete data on these three rounds has not been compiled, nor is the hardness of the plate known. Results appear to substantiate penetration figures cited by the British.

(Note: Earlier in the report, figures are given as follows: “Tested (British) penetration of Shot APDS against homogenous plate at 30 degrees is approximately 8.9 inches at 1,000 yuards, 7.9 inches at 1,500, and 7.1 inches at 2,000 yards)

             i. Center of impact firing. Firing results: 10 rounds shell HE fired at a range of 7,000 yards set up on the range drum, relaying after each round on two aiming stakes in line.

             j. Discarding of sabot. Center of impact of carriers after coming to rest on the ground is between 700 and 800 yards from the tank, and slightly to the right of the gun-target line.

             6.. Conclusions.

            As a result of limited testing Army Field Forces Board No. 2 concludes that:

            a. In respect to the performance of the 20-pounder gun Mark I:

            1) Penetration, firing armor-piercing discarding sabot, is substantially as rated and the gun is superior in this respect to the 90-mm tank gun, M3, firing shot, HVAP-T.

            2) Dispersion, firing armor piercing discarding sabot and high explosive is, by accepted standards, slightly greater than that of the 90mm Tank Gun M3, but that this dispersion is not so great as to impair the effectiveness of the 20-pounder against armoured and unarmoured targets.

            3) The relatively short tube life (estimated at 120 rounds) firing armor-piercing shot is not an excessive price to pay for the improved striking power.

             b. The gun controls provided in the Centurion III are generally superior to those in the Medium Tank M26, exceptions being the manual traversing mechanism and the present lack of an emergency firing gear for the main armament.

             c. The fire control equipment in the Centurion III represents a practical solution to a pressing need and is superior to that in the Medium Tank M26.

             d. The comparatively short time of flight and the flat trajectory achieved with armor-piercing discarding sabot offer substantial advantages respecting deflection errors due to cant, lead required against moving targets and allowing ranging errors against armored targets whether moving or stationary.

             e. The overall characteristics and features of the fighting compartment of the Centurion III, including the 74-inch turret ring, balanced gun, and stabilization in azimuth and elevation, result in a far more fightable turret than that of the Medium Tank M26

             f. The fire power and general fighting efficiency of the Centurion III appear to be superior to that of the Medium Tank, M46, judging from projected improvements to be incorporated in the tlatter tank.

7 Recommendations.

            Army Field Forces Board No. 2 recommends that:   

            a) Based on tests conducted to date the 20-pounder Gun, Mark I, and the turret assembly of the Centurion III be considered superior to like components of Medium Tank M26

             b) Further consideration by development agencies be given the stabilizer, fire control equipment, and main armament of the Centution III


                        H. H. D. Heiberg

                        Colonel, Cavalry.

We'll continue on with the automotive tests next week.

As ever, my Facebook page remains here, my Youtube channel here, and Twitch stream (Every Tuesday, and occasional evenings) is here.

Bob for the Forum!