Roadholder 364 - April 2018

40 Desmodromics - is it worth the effort? Mark Woodward I’ve found the Desmodromic valve actuation system an interesting subject for quite a while and wondered who 'invented' it, who has tried it and why it is not common these days? Looking at numerous sources of info. (the web!), I’ve found lots of information regarding this subject. Firstly, what is it? The name 'Desmodromic' derives from the Greek words 'desmos' meaning 'controlled' or 'linked' and 'dromos' meaning 'stroke', 'course', 'track'. In the mechanical world it refers to a method of controlling movement in one direction (e.g. opening a valve and then controlling the same movement in reverse i.e. closing the valve). When you think about it, many valves we are familiar with usually use a spring to move a valve in one direction or another with some other method to move it in the 'other' direction (mechanical cam, lever, hydraulics, electric solenoid, etc.). There have been numerous designs and applications of the desmo system over the years. It was used in a few designs, including the 1908 French exotic single cylinder Desmodromic engine for the 1908 'Grand Prix des Voiturettes'. Elie-gratien Michaux-nicolas, who had become head engineer at Peugeot in 1890, designed the 1914 Delage grand prix car using a desmodromic valve-control system with four valves per cylinder in pent- roof combustion chambers. Fiat designed another unusual desmo operating system a few years later. In the early days of the internal combustion engine, metallurgy wasn't anything like as well understood and metals were nowhere near as strong or developed as those of more recent times. Valve springs were often known to break; the early motorists/ motorcyclists were adept at changing valve springs at the roadside as a matter of course. Engine revs were limited by numerous factors, not least of which was the inability of the four-stroke valve actuating mechanisms to withstand high speed operation, resulting in valve 'float' or bounce. One 'work-around' this problem was to fit stronger springs but this was somewhat self-defeating as it simply further increased the stress on the (already delicate) springs and the associated valve gear. Norton attempted an alternative 'fix' by fitting hairspring valve springs which worked to a point, but these springs were somewhat ungainly devices that were difficult to accommodate within the cylinder head and usually had to be left out in the 'elements' to spew oil everywhere and also collect road muck etc., e.g. the International and Manx models. J.L. Norton had designed a motorcycle desmodromic valve operation in the '20s, however, the resulting system was too bulky and complicated and Norton's idea was abandoned. The idea was resurrected at Norton in the '50s for use in the 1957 Model 30 and 40 Manx Nortons. Doug Hele, working at Norton at that time, designed and made two desmodromic heads in 1957. In early tests, the power from the desmo engine was no greater than the 'normal' DOHC motor. Work continued in 1958, and in 1959, a 500cc desmo engine was taken to the TT. Bob McIntyre tried it in TT practice but reverted to a conventional engine for the races and, mostly because of time and resources, Hele and Bert Hopwood didn't pursue its development any further. Bert Hopwood stated that "the desmo programme showed promise but we didn't have the time or resources to recognise its potential. There were too many other things going on at the time". Doug Hele said: "We spent a lot of time making different cams, but we never got more power from the desmo than a standard engine would produce. More time was needed on the engine, but that is one resource that development engineers are always short of." The Norton desmo project died a quiet death. The operation of the desmo system has often been simply described as 'having one cam to open the valve and another to close it'. I have often thought: 'hang-on a minute - it's all very well opening the valve because, as