Friday, July 18, 2014

Towards GOODWOOD, Four: The Green Fields Beyond

(This one is a day late, so I'm throwing in a freebie: my old notes on the British institutional tank design process. Which may or may not be so heavily plagiarised that Zizek would blush, since I originally intended to rewrite them before publishing them. But, hey, blog!)

Oh, the link. The "green fields beyond" is apparently not just a tag that is supposed to bring to mind fake-Faulkner's rendition of "Old Black Joe" in Barton Fink. (For a given value of "supposed.") It is also supposed to remind us of the Battle of Cambrai. I say that it can do both.

 It turns out that "Through the blood and mud to the green fields beyond" is the "unofficial motto" of the Royal Tank Regiment. Someone doesn't make the association that I make. Truth to tell, the story of the Royal Tank Regiment is not exactly full of triumphant breakthroughs onto the green fields beyond. It is mostly about terrific blows of the armoured fist that the Germans contain with sacrifice and courage in a bad cause. That's not necessarily a bad thing, though. The Regiment has achieved its objectives. This is not a "dark future where there is only war."

Warhammer 40K figures, kits and painting By Ryogo Yamane

You knew where that was going, right?

Anyway, the point of putting sponsons on the side of World War I era tanks was that it was deemed necessary to bring a 6-pounder onto the battlefield, and the machines were so cumbersome that hull mountings and turrets were ruled out for technical reasons. 

Which brings me to this little beauty.

Thirty-two tons, 60mm of armour, which was a lot for 1939, 305hp. The thing that stands out is the bizarre armament: a 47mm medium-velocity anti-tank gun in the turret to defend attacking infantry from counter-attacking armour, and a low velocity 75mm howitzer in the hull to suppress enemy infantry and artillery. (Giffard Martel would insist that the main role of the gun is to drop smoke, not kill people, but I am not going to press that argument when I do not have a copy of In the Wake of the Tank to cite from.)

When I say "bizarre," I am being slightly tongue in cheek. The French argument was clearly convincing, as the same configuration appears on the American M3 Grant and early British Churchills. The Wikipedia article on the M3 helpfully points out the T35, as well. The difference here is that the Char1-B was premised around Schneider's new "Naeder hydropneumatic transmission," here explained by a helpful user Gothard on the Axis History Forums back in 2005. (Also, Google Patents.**) The Churchill, similarly, had the Merritt-Brown transmission, which similarly allowed the driver to rotate one track forward while braking on the other to pivot the tank in place and allow it to train a hull-mounted gun. Lacking a similarly reliable transmission for the T-35, the Russians put the main gun in a separate turret. The Americans, meanwhile, do not seem to have taken the need to aim the hull gun on board. Or the need to keep the hull height down.

 Or the need to not have neighbours point and laugh
Tank design and doctrine can be pretty aggressively presentist. The French Army's core doctrine is implicit in the turret on the Char1. An infantry tank needs anti-armour capability in order to support the attack. The United States Army and the British-guys-who-decided-what-tanks-would-look-like (the organisational story is hard to summarise in anything less than booklength. Awful, dull books) agreed. 

Then you turn to Wikipedia and you are told that US Army doctrine called for tanks to be used "to support the infantry in the assault and give them a base of fire. Taking on tanks was the job of purpose-built tank destroyers." This is true. It is, in fact, the basis of an excellent book on American tank destroyers in World War II, Charles Bailey's Faint PraiseUS Army General Officer Commanding Ground Forces Leslie McNair imagined massive German tank break-ins being stopped by massed American anti-tank forces. Since the Germans gave their antitank units a more aggressive sounding name --Panzerjaegern-- and the Germans clearly knew what they were doing, the American "tank destroyer battalion" came into existence --and proved to be rather silly and pointless. 

I will not dilate on the silliness any further except to point out that it falls afoul of the same principle of general utility that stands behind the Main Battle Tank. When you have multiple units for multiple roles, you always have the wrong unit in hand at the wrong time. If it is possible to have a machine that can give the infantry their supporting fire and their antitank defence, that is what you should have. This is, of course, the point that the French had reached by the early 1930s.

How did it go wrong, then? Again, anyone with a smidgeon of historical perspective can answer that question out of the entertainments of the Gay Nineties. For any given armour, there is always a gun that can shoot through it, or a shell that will improve on the gun; and then there is an improved armour. Or, perhaps an armour scheme that makes the existing armour more effective. In any event, ships get every more heavily armed and armoured, and ever larger, and everyone wins but the poor taxpayer, who is eventually called upon to foot the colossal expense of  digging out the necessary drydocks, berths and harbours. World War II tank builders are often discussed in terms of people moving into an undiscovered terrain, when they were actually a little behind hand of the state of the art. There was very little experimentation with improved armours, for example, although Albert Stern's TOG2 project used face-hardened armour, which also appeared on German tanks in North Africa in 1942. As John Stone points out in The Tank Debate, we are now almost a century into a never-ending discussion in which tanks are always getting heavier and lighter, often at the same time.  In fact, since Stone published in 2000, we have managed to have a full iteration of the debate, with the US Army going into Iraq and Afghanistan in Stryker Brigades, and coming out of it proposing a future infantry fighting vehicle heavier than many tanks. 

In 1940, the British were, to all appearances, trying to force the rate of change.  on the field doing everything they could to force the pace of change. The Cruiser Mark 1 A9 was specified as a relatively light, pure "exploitation" tank, and was certainly underweight compared to its German rivals, but the addition of a high velocity 40mm gun, at a time when 47mm medium velocity was state of the art, implied that everyone would have to uparmour and upgun to compete. The British were partway along the curve, with a 14 ton "heavy cruiser" Cruiser Mark II A10 coming in alongside the Mark I, but this was more-or-less accidental. What was not accidental was the inclusion of howitzer-armed variants of the A9 and A10, usually attached to the squadron headquarters troop. Specialised units capable of  high-explosive fire  also appeared in the German armoured force, but the decision to order them from different manufacturers led to significantly different automotive characteristics and the separation of the Panzer III and IV into separate units. 

It's not the place of this posting to sort out the politics of the army's delayed rearmament. Suffice it to say that the recognised Very Important People who emerged in 1938 and who tended to be renforced by defeat in France were tank prophets and production geniuses. This said, the Royal Artillery had no intention of letting sketchy fellows design its equipment, so through all the reorganisations, Ministries of Supply, Director Generals of Prorduction, Tank Boards and Tank Parliaments, there remained the Ordnance Board, headed by Major-General Edward Clarke. So when the postmortem turned to technical matters, Clarke had his say: and his say was to drop the design for the 6 pounder antitank gun on the table, with the intimation that the 17-pounder stood behind it. Tanks were going to get heavier, armour was going to get thicker, and guns were to be needed. The high and mighty took these under advisement, agreed that the 6 pounder in particular ought to be manufactured at some point in the distant future, after the army was reequipped for the decisive invasion battle of spring 1941, and moved on to more important matters.

Now, this is not to paint Clarke in the role of a fearless, truth-telling prophet. I will admit to being fascinated by Clarke, just as I am by Director of Mechanisation A. E. Davidson, and less relevantly to this thread, Colonel Albert Victor Kerrison. Kerrison became superintendent at NPL, while Davidson was the President of the IME in the 1930s, and yet I cannot put more details to the names than that. Clarke just did some back-of-the-envelope calculations, reconciled them with external ballistics, which tends to go for stereotyped calibres, and delivered the largest high velocity weapon that could be installed interchangeably with the 75mm MV and the specs for one that would go into a 25 pounder mounting. The former dictated the fate of both the 6 pounder and the 75mm MV. Either gun could go into a Churchill in either mounting, and a new generation of cruiser tanks designed around the 6 pounder would also take the 75, and there you go. As of the spring of 1940, it might even have been the Char1. The Allies were in the middle of negotiations with the Lima Locomotive Works of Lima, Ohio, to build the Char1 For French, British, and, perhaps, eventually, American use. 

Do, by all means, follow the Wiki link, above, to the article on the Lima works (and from there to the  Baldwin Locomotive Works and Alco works). I am no train buff, so it was only when I followed that links that I learned about "Super Power" steam, the M-10000Winton Engine Company (a GM subsidiary) and a great deal more that provided context for the 1938 Baldwin buyout in which Averill Harriman was concerned. 

In case the dark, tangled motivations that I am implying here are less clear than I think they are, I am circling back on Churchill's 1943 decision to cut domestic British tank production in favour of locomotives. Done at Averill Harriman's behest in the name of the President, it was done for practical and appropriate reasons. The Russians thought that the Sherman was junk, due to its poor flotation and, more importantly, its inability to pivot on its tracks, which meant that it could not dismount from a rail car at a siding without heroically dangerous and destructive adventures. So they refused their 3000 tank 1943 allocation. Since Baldwin had just received a contract for 1250 Shermans.

This was overflow from the huge arsenals around Detroit recently erected by the Federal government and handed over to Chrysler and GM, already with the cynical expectation that they would end up in private hands after the war, with hands greased all around. The American public would put up with that to win the war: but the optics of building those factories and then handing them over during the war in order to save business for Baldwin were awful. Conversely, Baldwin's signature business was a poster child for peacetime expenditure. The country had more pressing needs for its factories in 1943, and the argument for running down the railways in order to create postwar work for reconverted war factories could hardly be more straightforward. Locomotives did go overseas,but no-one likes shipping them. They are large, and they will take a ship under almost instantly if they escape their stays and roll, something that, after all, they are designed to do.*** Moreover, the business was captured by "austerity" makes, just exactly the kind of business for which Harriman didn't buy Baldwin. Baldwin was supposed to be moving into the advanced diesel business, not producing cheap steam locomotives to get Europe moving again.  

Nor was the British selling job unique. Canada cancelled its "Ram," and Australia its "Sentinel
(I suspect that Coombs means the AC3) 

So pretty... Though the design was no doubt stolen from some upper-class British designer, whose confidence was cozened by a low-life servant. Off to Botany Bay with them!

in the face of the expected American cornucopia. As already noticed, however, the American production flood did not occur. Whether the rail disasters of the end of 1943 were decisive or not, Lima, Pullman-Standard, and Ford and Pacific Foundry were pulled out of the Sherman tank programme, followed after completion of their contracts by Baldwin, Marmon-Herrington and even Flint, Michigan's Fisher Tank Arsenal, although in the last case it would seem that this was in order to change over to the M-26.**** As a consequence, Sherman tank deliveries to British forces rose from 200 monthly in January 1944 to a dramatic peak of 1000 units in July of 1944 before falling to 400 in August and 300 in October, with deliveries curtailed entirely at this point. British tank deliveries having held steady at 200/monthly throughout the period. Unfortunately, a look at the example of Leyland Motors shows that efforts to accelerate Cromwell production in the late spring/early summer of 1944 pays off in a sudden surge in production to 225 in August ***** presumably in anticipation of the American shortfall leads to delay in the Comet programme. The original forecast was for the production of 75 Comets in August 1944, a number not, in fact, reached until December 1944. 

  As I have suggested several times, it is the interaction of tactics, industry, politics and engineering that seems decisive to me in the evolution of a complex system like tanks, rather than articulated doctrine, which strikes me as retrospectively applied to developments that have their own logic. (See technical appendix, this number, below.) As John Buckley points out (154--55), there was no shortage of people arguing for big tanks with big guns in the early part of the war. The Challenger's bizarre turret appears to have been lifted directly off the TOG2, although we would have to know more about the Stern gang's efforts to be sure. The problem is that, so long as there was no endogenous logic pressing things forward, theorising could get in the way. So, just as Leslie McNair kept arguing for "tank destroyers" in the United States, so we find in the Spring 1943 General Staff paper a compromise calling for tanks with the 75 MV, 6 pounder antitank --and a 95mm "close support howitzer" along the lines of Martel's old smoke thrower, now rehabilitated with Cherwell's blithe assurance that the newly developed HEAT (HESH, surely?) would make the old bomb thrower as effective an antitank weapon as some expensive old gun.  As Winston Churchill later pointed out, while Mediterranean forces were firm in their preference for the 75mm MV, they had never actually seen any alternatives to it. 

So I want to locate "alternatives" within an industrial locus. That's code for the Rolls-Royce Meteor, of course. I have previously argued --although I am not going to link, because this is late enough already-- that the Nuffield cruisers and their Liberty engine are politically situated in the need to find Nuffield a tank to produce, which meant an engine to produce, which meant the Liberty. With the Meteor, causation goes in a different direction. With a fine design on their hands, Rolls-Royce obviously wanted to capitalise with Sea-Merlins and Land-Merlins and whatnot. Famously, W. A. Robotham got on with the Land-Merlin, drew in underemployed Leyland engineers, and somehow found himself attached to the Challenger project, which I just learned right now. (Given Challenger's suspected links with the TOG2, and thus to "the old gang" of WWI tank developers/industrialists, I suspect that we are seeing deep currents of personal connections within the engineering community emerging unsuspected from the depths here.) Some jinks and dives later, and we come out of the story with Rover manufacturing Rolls-Royce engines for Leyland tanks. The point is that we now have 600hp to play with, and what are we going to do with it? The story, as Buckley tells it, is that the Cromwell is intended to take the "HV 75mm," or 17 pounder. (The Wiki author thinks that it starts as a 6 pounder tank,) In fact, in 1942, preliminary planning for the 17 pounder called for it to go into the Cromwell. It is not clear that Cromwell design ever envisaged taking the 17 pounder, however. Cromwell was the last partly rail-gauge limited British tank, and it should have been fairly obvious that this 28 ton tank would not take a 17 pounder.****** When the engineering problems with this were pointed out, the Comet emerged as the successor project, albeit taking a slightly cut-down version of the 17 pounder, the 77mm HV, and taking time to incorporate slightly heavier armour. 

Now, with Cromwell in production along with a Cromwell variant with a Liberty engine (the Centaur), let us move on to Oliver Lyttleton's January 1944 bombshell: with imports falling, British tank production would have to be cut again, to 5280 in 1944. The Churchill target would be met and exceeded, due mainly to a reluctance to cut work prematurely at Vauxhall, leaving skilled labour unemployed and vulnerable to recruiting by firms in other manufacturing areas or even the military draft. Cromwell numbers, however, would fall, and even then would squeeze out Centaur production. Lyttleton called for a bold step forward. Instead of cutting British production even further and relying on Shermans (a reliance that would have proven mistaken in the fall of 1944), he called for aggressive development of the Cromwell-Comet-Centurion design line (that is, the Meteor family). Buckley's reporting of the key War Cabinet meetings makes it unclear just what the nature of the compromise was, except insofar as it involved cutting Centaur production. If so, the story here is an increase in resource allocation to tank production within a framework of declining total national industrial output. 

But that's speculation. What we do know is that there was an aggressive effort to rehabilitate Shermans on hand into "Fireflies," that is, 17 pounder-armed Sherman tank variants. And this takes me back, one more time, to the founding of the REME, and its early effort to recruit young British engineers into its ranks, an effort stopped stone dead by William Beveridge in 1941. At this early date we are back in the world of "unreliable" British tanks, and the REME was trying to build up its strength in order to do after-production modifications on defective-as-delivered Covenanters, Crusaders and, particularly, Churchills. Beveridge rudely (as was his wont) told the infant tank-and-truck repairing corps of the British army to stop poaching scientific talent and bloody well start training its own technicians. Looked at from the standpoint of class warfare, this was about shifting the army from an engineers-as-officers-as-middle-class-car-tinkerers" model to the "army-of-tradesmen" model, and is the quintessential Beveridge intervention --unwelcome but far-sighted. Look at in manpower terms, it suggests an explanation for the hypertrophied strength of the REME in the summer of 1944. It also comes back to the Firefly, because the conversion of often superannuated Shermans into good-as-new 17 pounders was done by the corps of service tradesmen originally built up to repair the tanks produced in a rush in 1940--2. 

So, on the cusp of GOODWOOD, the situation stands as follows:

--The greater part of British armour is mounted in Sherman tanks, mostly build at government-owned arsenals in Detroit destined for postwar use to make cars for the great American public. But these are wasting assets.
--British armoured divisions mounted in British tanks drive Cromwells, the tank that emerged as the machine-to-put-Meteors in. Some with 75 MV had little prospect of penetrating the armour of Panthers and Tigers. Others, armed with 6 pounders firing improved APSDS rounds, did; but they muddy our story, so we'll leave them out.
--It seems not impossible that such Centaurs as were built, all mounting 95mm howitzers (or AA, or some other specialised equipment) and sent to Normandy in specialist formations, were around, but not at GOODWOOD.
--Many, many British Shermans mounted 17 pounders. Although no better armoured than other Shermans, they had a gun that could kill Panthers and Tigers at battle ranges exceeding 1000 yards, compared with under 500 for the 75mm MV.
--Comet deliveries, which were originally supposed to have begun shortly, have been put off for months in order to maintain Cromwell production.
--Equivalent to the "Firefly" Sherman is, supposedly, the 17 pounder Cromwell variant, the Challenger. Were there Challengers in GOOWOOD? I expect that if there were, people would have told us, but this programme is wrapped with such obscurity that I won't entirely rule it out right now.
--And then there were 17 pounders (and 25 pounders) in an exciting variety of other mountings, suggesting that the Britain that the postwar import car enthusiast knows so well, the nation of garage tinkerers, was just getting started.

Which is as good a place as any to stop and suggest that if I can't wrap up before GOODWOOD, I will try for TOTALIZE instead. 

Technical Appendix

From a treatment of British tank design, some very interesting secondary source bits (cf. Peter Beale, Death by Design: The Fate of British Tank Crews in the Second World War [Thrupp, Stroud, Glos.: Sutton, 1998], 146–67). The Royal Tank Corps spawned a Department of Tank Design and Experiment (“in 1918") under Lt.Col. Philip Johnson. The Department was closed in March 1923. In 1920 a Grade II Directorate under the Director of Artillery, was established with responsibility for overseeing the mechanisation of the army. This presumably [Beale, despite best efforts in this chapter, is not the most lucid of writers] assumed general responsibility for tank design in 1923. In 1927 an independent Directorate of Mechanisation was established, with a Mechanical Warfare Board advising the Master General of Ordnance. (Mechanisation Board until 31 May 1934.) The Board’s ambit of responsibility covered tanks and other vehicles, and a design section responsible for tanks was established at Woolwich Arsenal in close physical and professional proximity with triservice design committees responsible for guns, ammunition, small arms, tanks, and wheeled vehicles. [146]
 In a design, the Tank Design Section drew up a broad specification that went up to the Mechanisation Board for advice on engines and ancillaries. A mock-up was then built and cleared with the Tank Gunnery School, Tank School Instructors, and “other interested War Office branches.” (148) The mild steel prototype was then built at Woolwich, at the Arsenal Royal Carriage Factory, then evaluated at the Farnborough Mechanisation Testing and Evaluation Centre (title uncertain). Manufacture would “normally be put in the hands of a sub-contracted industrial firm.” Up to 1937, though, this meant in effect Vickers’ Carden-Lloyd subsidiary or Woolwich.
 In the second half of 1934, a committee of inquiry into the Design Department at Woolwich including Colonel A. E. Davidson, President of the Institution of Mechanical Engineers, Mr. W. A. Stanier, Chief Mechanical Engineer of London, Midland, & Scottish Railway, and Mr. F. W. Hawkesworth of the Great Western Railway. This committee’s recommendations put the tank design section directly under the Director of Mechanisation via the Mechanised Warfare Board, with a second Deputy Superintendent of Design to take tanks and vehicles under direction. Beale thinks that this was a bad move. (149) There is, however, some evidence of frustrated office politicians, as George Macleod Ross (author of The Business of Tanks) moved on in 1937. Such a climate does not promote reasoned debate and consideration of alternate opinions, as when Macleod Ross condemns the tendency for settled issues to be revisited “the fire hazard of petrol versus diesel” being an example. (150)
 The Ministry of Supply took over tank design on formation in the summer of 1939 (151), with a Director General of Tanks and Transport, Sir Peter Bennett. There was also, of course, a tank design department at Vickers. (151).
 With the war on, tank design devolved on the parent firms, with the Department of Tank Design’s “role will be confined as far as possible that of consultant.” (153), quoting an uncited statement by the Minister of Supply. “Turretry, armour, stowage, and many other factors” were matters for the DTD, which would really prefer to talk about the Centurion.
 The overall problem the Army faced was similar to the one that the Navy and the RAF had already confronted and addressed, that of establishing a community of interests with a group of technically qualified manufacturers and an overseeing apparatus that could enforce the requirements of the army upon those manufacturers. Lord Weir delivered a report in January 1936 calling for just such a group through a munitions supply department through a director responsible to the Secretary of State for War within the War Office covering every phase of supply with the “technical production, inspectional, commercial, and financial experience to enter into the contracts that are necessary.” (156–7). In June 1937 Vice Admiral (E) Sir Harold Brown (Ret.) was made Director General of Munitions with a Directorate of Industrial Planning and a Directorate of Progress [to oversee the progress of contracts] (157). In 1938 the last Master General of Ordnance, Hugh Elles, retired, and the MGO office was moved under the Director General of Munitions Production, and the organisation then passed under the Ministry of Supply in May 1939. Finally in June 1940 the Director of Tank Design was formed, but that Directorate declines to take full responsibility for anything prior to the A41/Centurion –Okay.... Production was under the Chair, Armoured Fighting Vehicle Division, Cdr E. R. Micklem (ret.) a former Vickers director.
 On 29 May 1939 a Tank Board was established to investigate and advise and get the critics “peeing out of the tent.” (161–2)

 On 7 October 1940, Board member General Sir Vyvyan Pope wrote to the ACIGS to summarise what he saw as the ineffectual performance of the Board by giving his summary of its total efforts to that point. On 24 June 1940 it had been directed by the Member of Supply to consider the specification for the A22 (Churchill), consider the views of Sir Albert Stern’s committee, and to consider future matters. It met 3 times in the last week of June on the Churchill, twice in July on Stern’s Committee, twice in the first week of August to deal with Chairman Sir Alexander Ross’ departure for India, and once in September to consider tank gun deficiencies. ACIGS was evidently not impressed with the idea of a chair of the Board being off in India, and he was replaced on 29 December 1940 with Sir James Lithgow, while MajGen G N Macready, the ACIGS and MajGen Stewart, Inspector of the RAC, were brought in as alternate members to replace Brigadier Pratt, who was in the US on a tank procurement team. Beaverbrook was appointed MinSupply in July 1941 and promptly reorganised the Board. It was again reconstituted under Weir on 2 May 1942. On 29 May 1942 W. A. Robotham of Rolls-Royce joined the board, and so did Gen. E. V. Rowcroft, the newly-created Director of Mechanical Engineering. (164)
 Churchill felt the need to bring an even greater number of water-makers into the tent, and on 5 May 1941 he organised a “Tank Parliament” with Beaverbrook, Sir Andrew Duncan, then MinSupply, Sir William Brown (who?), Sir Harold Brown, Sir James Lithgow, MajGeneral Crawford of the MinSupply, and David Margesson in respect of his office as Secretary of State for War, Lt.Gen Sir Robert Haining; VCIGS, MajGen Macready, ACIGS; General Sir Alan Brooke, then CinC Home; LtGen Martel, then Commander RAC; MajGen Pope, Director AFV; and MajGenerals Crocker, Burrows, and Hobart, giving representation of the commanders of 1, 6, 9, and 11 Armoured Divisions, plus Lindemann (165–6). It met a total of 4 times to little result –except to get matters through Battleaxe. (167).
 Going back to industry; i) Macleod Ross, noting how useful it was to have tank design closely coordinated with other design sections, describes how, “as the standard of tank armour rose to 80 mm” manual turret rotation became impractical and at this point reference was made to “Captain Kerrison’s work at the Naval Ordnance Laboratory on anti-aircraft directors for following aerial targets resulting in his experience providing us with hydraulic traversing gear manufactured by Power Mountings Ltd.” (147)
 In 1937 another industrial cross-fertilisation came about when Dr. H. E. (Ted) Merritt joined the firm from David Brown & Co. of Huddersfield, and immediately began work on his regenerative steering system. In early 1939 he left the Directorate for David Brown, where he patented his steering system(!) In other words, Merritt came over from David Brown just at the point when Brown finished with its geared armour planers already noted, and went back with some very valuable intellectual property. (150)
 In the course of 1936–40 Vulcan Foundry worked on the Matilda II, Nuffield Mechanisation and Aero developed the “Cruiser Tanks Mk III, IV, and VI,” Vauxhall developed the Churchill, using work done by Harland and Wolff and the DTD. Other work ws done by the LMS Railway Company, Leyland, and the Birmingham Railway Carriage and Wagon Company.
H. E. Merritt, “The Evolution of a Tank Transmission,” Proc. Inst. Mech. Eng. 154 (1946): 412–28.
This is a discussion of the evolution of Merritt’s regenerative triple differential transmission, showing how the problem developed from theory to reality. The problem is that there is no sufficiently rugged and compact mechanism for providing positive control of the speed of a track. Thus, a tank’s tracks are driven at the same speed. But to turn, a tank must skid the track on the inside arc of the turn. To accomplish this, the drive force on the inner track must be reduced, that on the outer track increased. The theoretical calculation of just how much is very interesting. Without exploring the derivation in any detail, I note that Merritt provides solutions using hyperbolic functions: evidently, the problem is a partial differential equation (I hadn’t thought of this, but it is obviously a more complicated problem than it seems.) Theoretical analysis indicates that for a 500 hp, 50 ton tank, length 16 feet, width 10ft, turning at 30 ft/sec velocity, the net force being exerted back on the tank is no less than 1460 hp, and the motor must drive the outer track at 230hp. Thus, on the one hand heat dissipation becomes a serious issue, while on the other the tank slows rapidly as it turns. (418) Regeneration, in which the final drive unit on the inner sprocket can also function as a generator to drive the outer sprocket, is necessary. (421)
 Merritt reviews a series of possible transmission arrangements, beginning with the basic clutch-and-brake, and leading up to several differential arrangements and ultimately to his triple differential, which was the most easily produced version of a regenerative transmission. He notes the inadequate character of the German tanks, and observes that
>>>>the Merritt-Brown transmission became necessary when the British first began to develop very heavy tanks, specifically, the Matilda II, although it was first applied on the Churchill and later on the Cromwell and later tanks.

The discussion group was large and lively. Lt. Colonel Ewen M’Ewen, REME B.Sc. (E) AIMechE points out that the Germans used a combined double differential and changed speed transmission, an arrangement experimented with in Britain prior to 1940, but only suitable to low speeds and good roads. (421–2)
Lt. Col. W.J.K. Viney REME B.Sc.(E) is a skeptic: how heavy does an AFV have to be before it has to use the Merritt-Brown?
W. Steeds B.Sc. (E) MIMechE catches an error in Merritt’s math; he has been working on the problem since late 1939, and Profesor Hartree has done an even more complete analysis. Major Walter G. Wilson, CMG had designed a simple regenerative system as early as 1926.
Lt. Col. Philip H. Johnston, CBE DSO MIMechE says: the transmissin of the Churchill worked at a time when no other part did. (Merritt contradicts this later.)
Brigadier W. M. Blagden: the Mark VI did fine with a clutch and brake, although with Wilson’s transmission it would have been even better. The Valentine had a similar gear, very well designed, that was adequate for its speed despite having only 10hp/ton. Matilda, not so good; disappointing compared to the performance of its prototype. Crusader had the Wilson, but earlier cruisers had the clutch and brake; going from Cruiser IV to V and VI (Covenanter and Crusader) meant going from 14 tons to 19, and was only possible with the Wilson transmission. In 1941 there was great pressure to adopt Cletrac, although this would have required a considerable redesign from rear to front drive, for example. The simple track differential system used on carriers may seem like a panacaea, but it was only effective combined with a track-setting system, only practical on a small AFV.
Harold Sinclair MIMechE says, out with this comparison of tanks on the basis of bhp/ton, the question is rather, bhp/ton-mph. He also points out just how small are the shafts and gears that take Merritt’s 1430 hp.
Merritt in reply (425) observes that the theory of steering was worked out by M’Ewen, Hartree, Steeds, and also GV Cleave, and M. Barraclough. Also, excellent transmissions by Major Wilson and Mr. Gordon Wilson existed in 1939, but [Merritt’s] was the best compromise between technical and production efficiency. A double-differential with pre-selective gear change had been worked out before the war, but was never put into production.
H.E. Ashfield AIMechE,c hief designer, Tank Gear Division, David Brown and Sons/DAvid Brown Tractors (426–7) points out that a Tiger inspected in 1944 gave the impression that a MB type transmission element had been worked in in mid design, presumably a result of capturing a Churchill.
Mr. David Brown, MIMechE noted that the very heavy tooth loading on the gears required careful control of the production and heat-treating of the special service steels required to give proper reaction to the problems of surging in heavily stressed components.
M’Ewen shows mathematically that Steed’s objections have no merit.
Major G. Lomonosoff, REME says that MB’s theory is great, but his transmission is too complicated. Why not Cletrac?
N.B. Carl Ibbotson, “David Brown Gear and Metrology,” at discusses his experience of porting David Brown’s proprietal FORTRAN programmes over to a new computer system. It was difficult work, for DB FORTRAN embodied immense and highly specific knowledge of gear trigonometry and beyond. He notes that 17 year old David Brown founded his firm in 1860, and over the years there has been a lot of work on gears. is on the Textron website. Textron took over David Brown in October, 1998.
 DB started in 1860, and cut its first machine-cut gear in 1898. DB died in 1903, and his sons Frank and Percy succeeeded. The firm opened its first joint-ventur, Timken-David Brown in the US in 1913. By 1921, DB was the largest worm-gear manufacturer in the world. It bought a speciality steel maker, P. R. Jackson, in 1931. 28 year old David Brown became Managing Director in 1931. It says here that the Merritt-Brown transmission was patented in 1935. In 1936 the firm produced its first DB-Ferguson tractor. In 1939–45 the Aero Division produced 500,000 gears, and was briefly during the Battle of bRitain the only supplier of gears for the Spitfire. The Gearbox Division did 10,000 tank transmissions during the war, and was the parent firm for transmission manufactures. At peak, the firm employed 7000 people. the company was floated in 1993.  

**"concentric berslfserving"!
***Between wartime 8Fs, Austerity 8-2-0s, 10-2-0s, 0-6-0s, "Liberations" and others, I am seeing over 2000 wartime and immediate postwar locomotives, built by Vulcan, Beyer, Peacock, the LMS's Crewe shops and one other "redirected" firm which Coombs does not name. Perhaps Stanier or Ruston-Bucyrus? Coombs does note the urgency of the matter. American and British locomotive production in 1943 was estimated at 4700 at the beginning of 1943, vice demand for 6000. Coombs also includes a very pretty little chart(124) of the Vulcan Foundry order book, showing a transition from a 77%/23% tank/locomotive order book to 17%/83% in 1944.
****Notice that the wiki article has grown rich with research summaries and seems to be converging on the idea that it was the (predictably, in my view) failed diversion into an electrical transmission that delayed the combat debut of a heavy tank. Obviously, various of the more doctrinaire figures within the US Army did press  "doctrine" reasons for not proceeding with the M-26, but their protests look like pro forma statements occurring in discussion channels that ran independently of the engineering developments. 
*****I hate to sound petty, but Coombs' reliance on charts and graphs makes it very hard to pick off the actual numbers and their meanings. I am eyeballing across to get the "200 British tanks/month" delivery rate from Figure 6.2 (124) and the "225 Cromwells produced at Leyland in August 1944" from Figure 5.3 (103). I am not sure if my eyeball is failing me, or if there is some other, nonobvious difference. Perhaps the reported numbers in Figure 5.3 are projected quarterly rates, so we have to divide by 4 to get monthly numbers? Perhaps there was an ever-growing backlogged of delivered-but-not-accepted tanks? 
*******I hope no-one can read my handwriting in the library copy of French's Raising Churchill's Army where I pour scorn on his use of the idea of the "rail gauge" as restricting British tank production. The idea that one might pick up by osmosis (and by "one" I mean "me") is that the argument is that the size of British tanks was restricted by the archaically narrow railways. Whereas in fact the standard British gauge is the same as anywhere else in the world --except Russia. In fact, all rail lines have "loading gauges" based on clearance through cuttings and bridges and under viaducts. The Sherman was rail gauge limited, which is why its tracks are so narrow and its flotation is so poor. Panthers and Tigers were shipped with special, narrow tracks to keep them within continental loading gauges, and if the residual effects of gauge choice are decisive here, Britain ought to have had a huge advantage in home defence, so long as invaders could be persuaded to land in areas served by the Great Western.  My Dad used to insist that if everyone had gone with Brunel's 7ft gauge, we would have condenser locomotives, and steam would never have died. 


  1. A couple of things. Am I right in thinking you mean there are too many support weapon tanks in the fleet at the time of GOODWOOD and not enough tank killers? If so, perhaps the lurker here is the development of the sabot round. When we say gun we really mean projectile, after all. An MBT has to kill the best enemy tank, and then be at least acceptable in the support role.

    Regarding putting the 17pdr on the Cromwell, it wasn't much lighter than Sherman (28 vs 29 tons) or the M10 SPG, 29.5t, of which the UK upgunned 1,100 vehicles with the 17pdr as the Achilles (which Pip Roberts' memoirs says was sometimes used in 11th Armoured as a tank, although only much later).

    On rail gauges, you have this upside down - the UK loading gauge, not the rail gauge, is more restrictive than that on the continent. As a result, even the super-high density, super-low headway southern electrics have never operated a double deck train as are common elsewhere in Europe.

    The SR tried one just before nationalisation, but apparently the passengers didn't like having someone's feet in front of their face. Having been a commuter on the Southern Electrics, I don't believe this - I'd have put up with a lot for twice as many seats and comfort isn't exactly a priority. More likely, the cramped cars took too long to load and unload, and therefore the dwell times didn't permit a headway that permitted keeping the timetable.

  2. There really is no solution to the problem of having a good support weapon and a good antitank weapon on a tank so that it can perform both of its roles. It's telling that the universal solution of 1939 --an antitank gun in a turret and a 75MV in a sponson is now so incomprehensible that tank historians can barely take in the fact that the T-35, M-3 and Churchill all came along in response to the Char1.

    You set that aside, and you get to arguing that either a good high explosive round can so to be a good antitank weapon; either because the enemy won't be so gauche as to put more armour on their tanks, or because of HEAT/HESH/ATGM magic.

    Or you can think outside the box, accept that a tank is not a good artillery weapon, and give the tank a radio that makes it a good artillery weapon.

    The problem with this is much the same as the problems that prevent you from up-armouring in the first place. It implies an infrastructural investment in your army, in engineers and shops and great big trucks. Over the last seventy years, we have seen army after army sidle away from making that commitment in more congenial directions. The Stryker Brigade is the latest of what, four iterations of a version of an American army that could be deployed by air and still fight from a position of technical superiority?

    So we accept that there is an obvious explanation for the "go lighter" side of the Great Tank Debate, which is that light tanks cost less money. Not so much directly, as in terms of the support architecture. But what about when money isn't an object, as in World War II?

    The argument that I am limning here is one about social engineering. Sheppard puts it pretty clearly in 1938. You can either have a deskilled society of middle class managers and assembly line workers, or a trade union based society. The choice is made for you by the kind of total mobilisation army you embrace.

    GOODWOOD is the end of the line for the "middle class" society. We have faced the limits of our infantry replacement pool, and we understand that we are going to solve the limits of our current armoured force with EVEN MOAR armour. Simonds' plan for TOTALISE is fairly obviously an embrace of an REME-centric army, because Kangaroos. But I would also point to the way that the 5.5" is easing out the 25 pounder, improvements in radio, and the Comet that didn't hunt.

    As for the reason for not putting the 17 pounder on the Cromwell, it is simply because there's not enough deck on the Cromwell to punch a large-enough hole through. The mighty 70" turret ring of the Sherman will take it, but the Sherman has such a large turret ring because its deck is up on sponsons over the tracks. This was not a particularly acceptable solution to anyone else because of first flotation, second, height(!), and, third, the fact that this design feature began with the inability to find a proper engine for the Sherman, a problem that no-one else managed to have.

    So we relax the loading gauge (width at height), which is restricted by the limiting dimensions of infrastructure around railways, and accept that British tanks are not going to go by rail anymore, and we get to the 34 ton Comet, a little wider and a little longer than the Cromwell (as opposed to the Challenger, which is just, unsatisfactorily, longer) and you can put a heavier gun in --still not quite the 17 pounder, which goes into the even heavier Centurion.

    See? All of these problems could have been rendered moot if we had just gone with Brunel's 7 foot gauge to start with! (Or we would have just ended up with Tortoises shoooting at Mauses and complaining about exactly the same thing...)

    1. OK. The 6 pounder comes on the scene in early 1942 with a plain AP and a HE ammo nature. The AP round keeps getting upgraded. Sabot comes along in March 1944, well in time to be used in Normandy. That takes the maximum armour penetration from a baseline of 88mm up to 142mm. The US also buys the six-gun, but it doesn't do the ammo upgrades and as a result doesn't get value from the weapon except when units beg APDS rounds off the Brits, which they do whenever they can because they don't want to die.

      The six-gun is installed in a variety of British tanks. As you have discussed, Shermans start to flow into the RAC, to begin with in the Middle East. They bring with them the US 75mm. The combination of US pushing of Sherman, and after-action reports wanting a better direct fire support weapon, leads the UK to accept Sherman and also to introduce the ROQF 75. The important point here is that the ROQF 75 uses the same ammo natures as the US M2/4/6.

      Per Wikipedia, there is never a decent AP round for the M2/4/6 or therefore the ROQF, while there most certainly is for the six gun and the 17. In fact, the M61 AP shells were even delivered without the burster fill, so their ballistics may also have been screwy if the weight in the tail wasn't replaced with something.

      So. HEAT or whatever doesn't turn up in time to be relevant, but sabot certainly does. In fact, run the tape back to Villers Bocage.

      Wittmann kills a Sherman, another gets stuck across the road due to its shitty drivetrain so the rest of the squadron can't gang tackle him. He rips into the RHQ squadron, and then...well, Bill Cotton and a handful of Jackets scrabbling about like untermenschen get a mobility kill and he wanders off leaving his crew. The point here is that the Jackets' AT platoon are loaded for bear or rather tiger with 6 pdr APDS while the CLY tanks have nothing like it. Cotton pulls this off again and again through the day. It happens again in EPSOM - German armour smashes everything until it hits an infantry AT platoon and then it, er, doesn't.

      Question. Why wasn't there a sabot round for the 75? A 75 round weighs about as much as a 6, so the energetics ought to be OK. The Americans never got it for their own 6s - couldn't their industry manage it? In which case, why did ROF not make one as they obviously could? The British were clearly aware they were short on tank killing capability, hence the effort to upgun Shermans and M10s.

      Further question. The requirement for better suppressive fires out of a tank wasn't bullshit - engaging German anti-tank guns was something all armour that fought them needed to do. It turned up as a requirement from Tunisia and AFAIK Sicily - is there some really fascinating deep history of the landscape you're going to tell me about? Or is it more that 1st Army's artillery fire control and forward air control was a bit wank compared to 8th, which after all had learned the hard way?