Technology has become more and more sophisticated over the period in which the car has evolved. In the areas of electronics, metallurgy and fundamentals like engine and suspension design, things are not as they were.
Correct? Well, yes and no.
In fact, if you look carefully at car design, many aspects that appear modern actually date back to within just the first three or four decades of the inception of the car. Double overhead camshafts working multi-valves (such as in the pictured Alfa), double wishbone suspension – even front wheel drive, four wheel drive and removable cylinder heads. However, lubrication technology, electronics and metallurgy have indeed made enormous strides.
But every now and again, I wonder if today’s apparent love of car complexity is fundamentally worthwhile. It’s fashionable to suggest that those car technologies that have come and gone disappeared because they were intrinsically less worthy than those currently available. A Darwinian technology evolution; almost self-evident.
Trouble is, that philosophy often doesn’t stand up to scrutiny. Today’s hybrid petrol/electric vehicles follow those (admittedly few) hybrid vehicles sold in the 1920s. The technology of the Prius, especially in its Power Split Device and electronic motor control, is new and innovative – but the fundamental concept of mixing an internal combustion petrol engine with an electric motor and batteries was being sold long ago.
Steam cars are now regarded as dinosaur technology. But in their ability to burn pretty well any old fuel, develop immense torque and, as a result, use a mechanically simple transmission, I wouldn’t be at all surprised to see one successfully being sold new again. That’s especially the case with the availability of modern electronics to control fuel and water feed rates to a flash steam boiler.
And then there’s suspension. Cars have had pneumatic tyres since they first trundled along - and even the first cars were fitted with suspension. They used leaf springs and solid beam axles – technology now mostly consigned to museums. But it didn’t take long for radical improvements to be made to suspension designs. De Dion rear axles, double wishbones, steel coils, hydraulic dampers – all of these date back 80 or more years. As with those double overhead cams, these ideas are not new.
To take the premise further, you’d also then expect that the fundamentals of suspension were first sorted out long ago. Read any old automotive engineering textbook – say one that dates back 50 or 60 years – and you’ll see some very familiar suspension systems. (And of course, to be fair, you’ll certainly not see some modern designs – one likes torsion beam rear axles and multi-link front suspensions.)
So what are the ‘steam car’ designs of suspension systems – those dismissed these days as being hopelessly outdated... but ones that could well stand being revisited?
The one that I think stands head and shoulders above the rest is a suspension design now universally condemned as having fundamental flaws: swing arm suspension. To picture this approach, think of a solid axle divided into two and pivoted from the inner ends. As the wheels rise and fall, camber angles alter, usually quite dramatically. A variation is to pivot the swing-arms closer towards the other side of the car than the centreline, making for longer arms. This design can be used with driven and non-driven axles. Another variation on the theme is to use semi-leading arms – in that case, the camber change is less pronounced than pure swing-axles, but the castor changes as well.
The grounds for dismissal of swing-arm suspension are these. Firstly, the camber change already mentioned. Secondly, jacking forces where in hard cornering the suspension tries to lift the car, in so doing promoting positive camber change. Hit a big mid-corner hump and the suspension can go positive in camber on both wheels, resulting in loss of lateral adhesion. If the swing axles are on the back, the result is immediate oversteer.
So it seems a pretty open and shut case – swing axles are awful. The trouble
is, that ignores the way that swing axle suspension systems have been used in
Take Professor Ferdinand Porsche, for example. The surname is now synonymous with one of the world’s most respected sporting car brands, and in the 1930s Porsche (the man) was probably the world’s leading car designer. And Porsche used a swing-axle rear suspension design not only on the ‘Beetle’ (which was first designed pre-WWII but only put into full scale production after the war) but also in the 1934 Auto Union racing car. That car raced in the direct equivalent of today’s Formula 1.
And the main racing competition, in the form of the Daimler Benz Mercedes, also used a swing axle rear suspension design...
The Auto Union initially used a transverse leaf spring for its swing-arm rear and then changed to a torsion bar design. This photo shows the supercharged, 16 cylinder 1936-37 Auto Union race car with its swing axle rear suspension. Top speed in normal racing trim was in the order of 295 km/h.
Mercedes Benz moved to a de Dion rear suspension in their 1937 design, and Auto Union stopped using swing axles in 1938.
OK – so they tried swing arms and it didn’t work: both companies moved away from the design. The trouble with that theory is that Mercedes went back to swing axles in their 1955 racing cars! For many years Mercedes also used rear swing axles in their road cars.
Now no-one, no-one at all, would call the Daimler Benz company dullards at engineering, not at any stage in the company’s proud automotive history. Ferdinand Porsche? He’s one of the most respected automotive engineers of all time. And also keep in mind that when Mercedes in the 1950s elected to go back to swing axles, there were plenty of other choices open to them.
So why did these companies and designers use swing axles? What was their engineering rationale?
Criticisms of swing axles can be found in every suspension textbook but it is nearly impossible to find the advantages that caused these very respected automotive designers to fit this type of suspension to the highest performing cars of their day. Some texts somewhat sneeringly say of swing axles that they’re ‘cheap and easy’ but neither criterion seems relevant to the selection of the system for use in very expensive racing and road cars.
The Racing Car – Development and Design (Clutton, C, Posthumous, C & Jenkinson, D, 1956), a highly respected and authoritative book, says of the 1950s Mercedes return to swing axles:
When Daimler-Benz gave up on the swing axle layout for 1937 they were rather reluctant to do so, for it had many theoretical properties which were desirable from the point of view of controlling steering, but in those days they could not see clearly how to overcome the disadvantages. On this new car [ie the 1954 racing car] they made a great advance by mounting each wheel on a half-axle that was pivoted below the differential. Fore-and-aft movement of the ends of these half-axles was controlled by short links, arranged to give a vertical rise and fall of the hub.
Apart from it being hard to understand how short links can give rise to a vertical movement of the hub (unless multiple links are used, the hubs will move backwards and forwards in an arc!), it’s easy to see how the use of a lower pivot point (ie a lower roll centre) would help in reducing jacking forces. But the mention of ‘steering’ is fascinating; is this a reference to deliberately designed-in bump steer? Or simply to lift-off oversteer?
And Porsche and Mercedes are not alone in their use of swing axles. Some cars (notably the infamous Lightburn Zeta) used swing axles for cheapness, but in other cases it’s hard to see how price would have been the governing factor.
Allard, a UK constructor of racing and ‘trials’ cars in the pre-WWII and post-WWII periods, used what I’ve seen universally referred to as a swing-axle front suspension, made by cutting a Ford solid axle in two. But as this diagram shows (it’s taken from Automobile Repair, Vol II, Molloy, E – Ed, published in c. 1950), the cars actually used a semi-leading arm design. (Later the bodywork was extended so the suspension could use a forward radius rod. Presumably it then became a semi trailing arm design.) The semi-leading (or trailing!) design continued to be used for the front suspension even when the cars were using de Dion rear suspension designs – the latter hardly a cheap and nasty system.
The company founded by Soichiro Honda (another of the world’s most respected automotive engineers!) used swing axle rear suspension in the 1960s Honda 1300, one of the first cars made by Honda. The FWD car used a rear suspension system with leaf springs and swing axles pivoted from the opposite sides of the car.
So did these respected engineers and designers use swing axle suspensions for reasons that are no longer relevant? For example, universal joints were insufficiently well developed that keeping the driving wheels moving only vertically put too much strain on the drive joints? (It wouldn’t seem so: the de Dion suspension designs adopted by Mercedes and Auto Union pre-WWII needed two universals per drive shaft.) Or perhaps tyre technology (and width!) altered sufficiently that the camber change of swing axles could no longer be tolerated – it was more of a negative than a positive?
Advantages that I can see of swing-arms include lightness and the use of only one pivot point per side (rear suspension) or two pivot points per side (front steering suspension). Semi-swing arms (ie leading or trailing) can also introduce dynamic castor change (front steering) and toe changes. And yes, without a doubt, the negatives of jacking and (usually large) camber changes remain.
But one thing’s for sure: too many very smart people chose to use swing arms on very high performance cars for it logically to be treated with the disdain it currently enjoys.