Desmodromically Speaking

Griffith Borgeson has a high reputation as an enquiring historian happy to investigate the smallest details of high-speed internal combustion engine development in painstaking detail. He is fortunate in having as a platform the American Automobile Quarterly, a publication willing to devote the considerable space required for his findings to be properly expounded. Renowned for his books on the designs of Ettore Bugatti and on the twin-cam engine, Borgeson’s writing is irresistible, and MOTOR SPORT has devoted much attention to it, even crossing swords at times with the author.

The latest Borgeson manifestation is a remarkable expose of incorrect technical descriptions and illustrations relating to the forerunners of all twin-cam engines, those of the racing Peugeots which became famous for this engineering breakthrough and for winning convincingly the 1912 and 1913 French Grands Prix and the 1913 Coupe de L’Auto.

With multi-valve engines now so much in evidence in production-car circles, albeit for reasons rather different from those which induced Ernest Henry to use a four-valve-per-cylinder configuration when laying out the design of the invincible pre-war racing Peugeots, Borgeson’s latest technical paper, in the October 1986 ( Vol XXIX, No 3) issue of AQ, is interesting.

Incidentally, Borgeson quotes the arrival of the twin-cam valve-gear in 1912 for the Peugeot L76 GP engine as a “towering monument in the evolution of the high-performance internal combustion engine, although it might be more correct to say, as Laurence Pomeroy Jnr did, that this derived from a combination of twin-camshafts, multiple-valves and effective inclination of those valves in a pent-roof cylinder head.

But that is by the way. The significance of Borgeson’s newest findings will be apparent when it is revealed that he says he is exposing a technical (and journalistic) misconception, “deeply and authoritatively rooted for more than four decades”, which has “culminated in recent years in a nexus of total confusion and incoherence in the milieu of automotive history”. Phew!

What started Borgeson on this exposure of erroneous information was his desire to protect Arthur-Leon Michelat, designer of the 1914 S-type GP Delage, from the inference that Peugeot beat him, by two years, to the innovation of reliable desmodromic valve-gear — that is, a mechanism in which the valves are positively closed mechanically, and not by springs.

Borgeson’s argument starts with the statement that while works drawings of the revolutionary 1912 twin-cam Peugeot GP engine were never published, clues about its valve-gear details were contained in an article, published, after the Peugeot GP victory, in La Vie Automobile by the greatest of the motoring writers of those times, Charles Faroux, in 1912.

At the start of his description, Faroux told his readers that any real “speed-secrets” would be withheld, so according to Borgeson his prose was “somewhat opaque”. It seems remarkable, if Peugeot was anxious to keep details of its racing engine secret, that it released any information about it. Yet Faroux was able to tell of the new twin-camshafts, so why distort details of a far less important aspect of the engine? In describing the tappets, he referred to the eccentric on the interior of which the cam acted, and to rollers in the tips of the cams.

Borgeson thinks that when the more journalistically-minded W F Bradley wanted to deal quickly with the victorious new Peugeots in British and American journals, he revamped Faroux’s article. However, when the equally-successful 1913 3-litre racing Peugeots had been sold by the works, it became possible to examine their valve-gear. It was discovered that they used an L-shaped tappet, not a D-shaped ring cam-follower, and that the roller-tipped cams, called “implausible” by Borgeson, did not exist, despite what Motor Age said in 1914, in an article by Bradley.

From this, and the war-time Peugeot V8 aero-engine, the inference is that the 1912 GP cars had the same, or similar, cam-followers. Indeed, Borgeson used the ingenious but quite possible explanation that, by “eccentric”, Faroux simply meant “off-centre” — the Peugeot L-tappets being kept in contact with the cams by off-centre springs working in little pistons.

The size of the cam-covers of the 1912 Peugeot GP engine do suggest that large D-type followers were not used for it, as Borgeson points out. Unfortunately, the Faroux description led to many drawings appearing of the valve-gear on this famous Peugeot (The Automobile in America being the first offender in 1915), assuming this power unit to have had the L-tappets of the later pre-war racing Peugeots. No-one thought to check this after the war and, as Borgeson says, the last of the 7.6-litre cars had disappeared by 1927. Now we may never know the truth.

However, when Pomeroy wrote his WW2 articles for The Motor called “Milestones of Speed”, the artist L C Cresswell, having no 1912 GP Peugeot to examine, depicted the D-type cam-followers in his drawings, and Pom was very reticent in his description of the engine. This mistake (if such it was) was repeated in Pomeroy’s epic book The Grand Prix Car, along with the Cresswell drawings. This seems to have led Cecil Glutton and his fellow-authors, in their 1956 book “The Racing Car”, to repeat the fallacies and to imply desmodromic valve-actuation on the 1912 GP Peugeots (although the word “desmodromic” is not used).

Following this, Court’s Peugeot Profile, the Parisian L’Automobiliste and the American Sportscar Graphic used the Cresswell or similar drawings. The inference of desmodromics became stronger, and was categorically endorsed in expert Paul Yvelin’s racing history of Peugeot. Meanwhile, in a later book, Pomeroy (with no explanation) now said the engine concerned had finger-type cam followers!

Eleven years after Yvelin’s book appeared, Edward Eves reassessed the 1912 Peugeot for Autocar, describing its desmodromic valve gear with a drawing to show how it worked. After protests in AQ and MOTOR SPORT, what Borgeson calls a “grudging, partial correction” was published, with a hasty new drawing showing springs for closing the valves!

All this exasperated Borgeson because the desmodromic 1914 GP Delage did have a D-type tappet-follower. There was apparently about as much secrecy over this engine when it appeared as there had been over the Peugeot two years earlier. However, Griffith Borgeson now justified his commendable research. For when he learnt that Stuart Murdoch owned the sole surviving S-type 1914 GP Deluge, in Australia, he obtained engineering drawings by Jack Nelson of its valve-gear. “the first accurate ones the world has seen” — although he says photographs in the USA journal The Automobile in 1916 showed the same things.

Borgeson also persuaded the director of the Briggs Cunningham Automotive Museum, John Burgess, to undertake the very difficult job of stripping down the engine of their 1913 3-litre Coupe de L’Auto Peugeot. Drawings of this now show conclusively that it has the L-type tappets, with off-set keep-springs housed in cups or pistons, as, incidentally, shown in Hugh Rose’s drawings made for Louis Coatalen, who copied these Peugeots for his 1914 Sunbeam engines.

Alas, this was too late for CresswelI, who in his drawing put the cups on the valve stems themselves (a refinement which Henry rather naturally adopted for his 1919 racing Ballot engines).

So what Griffith Borgeson is saying in his masterly discourse, is that all Press references to D-type or stirrup cam-followers on the 1912 GP Peugeot are probably invalid, but that the true nature of its valve-gear may never be established. And that the official drawings of the 1914 GP Deluge engine are also largely invalid (having been faked for security reasons), but that Michelat’s Delage engine had the first successful desmodromic valve-gear on a high-performance engine — even if the best it did in the 1914 French GP was 8th, with two retirements. Pomeroy’s verdict of “positive unreliability” is regarded by Bergeson as unfounded.

The inaccuracies in The Grand Prix Car, have caused Borgeson to inscribe in his copy: “The user of this very great book must learn to sort its content of honest error, guesswork, and fantasy fiction from that of hard fact”. Which those now buying this three-volume work at £520 may well ponder . . .

I strongly recommend study of Borgeson’s erudite findings. I have wondered myself about some of the drawings of that advanced Peugeot engine, being unable to connect desmodromics with descriptions of apparent adjustable tappets. Bergeson tells us that even the Peugeot historian Yvelin has now abandoned this claim; but no reasons are given.

I suggest, however, that there is a possible explanation why the stirrup-follower was thought to have existed in the 1912 Peugeot. When Henry sat down at his virgin drawing-board to set out his revolutionary racing engine, valves and valve springs were a source of trouble, which is partly why he duplicated his valves. Could it not be that he also sought a means of eliminating valve springs, and drew a D-shape tappet that would permit desmodromic valve closing, but had insufficient time to complete the scheme before the Grand Prix?

As with Watergate and other political leaks, these drawings may have been got to journalists, and used as a means of distracting attention from the twin overhead camshafts. Faced with what he saw as positive valve closing, Faroux would try to conceal this other original feature of Henry’s creative brain, and his “opaque prose”, in deference to Peugeot’s call for discretion, would be excusable.

Thus the drawing of a stirrup tappet may have leaked out, to be copied and to mislead historians for nearly 70 years! If my assumption is correct, might the credit for introducing effective positive valve-operation for a racing engine be fairly assessed as 30:70, Henry: Michelat? WB