Such was the lack of understanding of suspension dynamics, this crude, bent piece of tube ruled the GP roost for almost 20 years, explains Keith Howard
Barring the notable exceptions of Earle MacPherson and, to a much lesser degree, Colin Chapman, the only person whose name has become inextricably linked with a particular form of car suspension is French nobleman Comte Albert de Dion. Actually, he didn’t invent the axle which carries his name: it was conceived by two engineers who worked for him, Messrs Trepardoux and Bouton, who patented it as early as 1893. Except for this inaccurate accreditation, de Dion’s distinction is well deserved.
Acting as a halfway house between live axle rear suspension and the fully independent alternative, the axle which carries his name — although initially disregarded — reigned supreme in the top echelons of motorsport from the mid-1930s through to the late ’50s. Only then did Formula One’s reactionaries finally offer grudging acceptance that fully independent rear suspension was its superior.
Today we think of the live axle, the favoured means of rear suspension prior to the de Dion, as belonging only on off-road vehicles because of its superior wheel articulation and better maintained ground clearance over uneven terrain — attributes which hardly commend it to racing cars. But once belt and chain-drive had been abandoned in favour of the propshaft and differential gear in the early years of the car industry, the live axle became the established standard. And despite its very obvious shortcomings, it remained dominant in racing into the 1930s.
A live axle, for readers unfamiliar with the nomenclature, is one which houses the differential gear and halfshafts in a rigid cross-car assembly to which the wheel hubs are attached at either end. Its failings are legion but, with the physics of suspension design so poorly understood in the early decades of the 20th century, it took some time for these to become widely appreciated.
The first major problem with a live axle is that of high unsprung mass. Suspension designers divide a car’s structure into two distinct components: the sprung mass, which largely comprises the main body of the vehicle, being that part which is isolated from the road by the suspension springs; and the unsprung mass, which comprises those components ‘upstream’ of the road springs (tyres, wheels, hubs, outboard brakes, etc.) that are isolated from the road only by the compliance of the tyre. The ratio of these two masses is of fundamental importance to vehicle dynamics, because as well as influencing vertical acceleration of the sprung mass, and therefore the ride quality, it also — and much more importantly for the race car designer — affects grip. Increasing the sprung to unsprung mass ratio reduces the vertical load variation experienced by the tyre as it traverses the uneven road surface, and squanders less of its grip potential as a result. So minimised unsprung mass is good news.
It goes without saying that the live axle, with its rigid casing and differential and halfshafts on board, is the very antithesis of low unsprung mass. This disadvantage alone counts heavily against it — but the list goes on.
Because a live axle carries the differential, torque reaction becomes a second major problem in powerful cars. Application of engine torque causes the axle to rotate in the opposite direction to the propshaft, winding up the road springs and tending to lift one rear wheel — usually the right. As well as harming traction, this can have disconcerting effects when applying power in a corner, the more so as the car behaves differently according to whether it is turning left or right.
Although having the rear wheels rigidly linked by an axle tube has the advantage of keeping them, nominally, at right angles to the road surface whatever the roll attitude of the body, it also means that disturbance of one wheel — by a bump, pothole or kerb — is immediately translated to the other. Again this does little for rear-end stability. Use of an axle also prescribes the height of the rear roll centre, encouraging jacking under hard cornering, and has potentially undesirable consequences for roll stiffness, which is less than for an equivalent independent arrangement.
De Dion’s rear suspension couldn’t solve all these problems as it was still an axle. But because it was a dead axle rather than a live axle — in other words, it no longer carried the halfshafts and differential (which was now attached to the body of the car) — it did alleviate the torque reaction problem and allow a worthwhile reduction in unsprung mass. Mercedes and Auto Union (after their dalliance with swing axles) both saw the advantage of the arrangement and put it to good use in the years immediately prior to WWII.
It would be no exaggeration to call the de Dion rear axle ubiquitous in grand prix racing of the 1950s: almost everyone used it, although the details differed. In some cars, the bent axle tube was located ahead of the differential, in others behind. Fore-aft location was typically by trailing arms whereas lateral location inspired all manner of solutions — Panhard rod (Auto Union Type D), Watts linkage (Vanwall, Aston Martin DBR4/250), A-frame central link (ERA G-type) — the oddest contrivance of all being the sliding block in a vertical groove. Crude as this sounds (and looks), it was first deployed by technocrats Mercedes-Benz in the W125 and widely copied, by BRM (V16) and Maserati (250F) among others.
Mercedes would later question the supremacy of the de Dion rear axle when it introduced the W196 with independent rear suspension and won the championship with it in 1954 and ’55, but the team’s early departure from racing allowed F1’s traditionalists to carry on as if nothing had changed. Only when the rear-engined Coopers arrived with independent suspension all round, was the de Dion apple-cart finally upended, and a lot else besides.
Too many F1 teams of the era waited for someone else to demonstrate a better way: they didn’t take the risk themselves. The gale of change blowing from Surbiton and Cheshunt, though, would soon sweep them — and the de Dion axle — aside.
Thanks to the Donington Grand Prix Collection (01332 811027) for their help in this feature