The Mark Ortiz Automotive
Reproduction for sale subject to restrictions. Please inquire for details.
Mark Ortiz Automotive is a chassis consulting service primarily serving oval track and road racers. This newsletter is a free service intended to benefit racers and enthusiasts by offering useful insights into chassis engineering and answers to questions. Readers may mail questions to: 155 Wankel Dr., Kannapolis, NC 28083-8200; submit questions by phone at 704-933-8876; or submit questions by e-mail to: firstname.lastname@example.org. Readers are invited to subscribe to this newsletter by e-mail. Just e-mail me and request to be added to the list.
DOUBLE WISHBONE VERSUS STRUT SUSPENSION
Double wishbone suspension systems are said to give better performance than the MacPherson strut type which is universal on most vehicles today....what are the reasons behind this? Which type (independent of driver use/abuse) tends to have a longer lifespan? How is tire life affected by these systems? Finally, was the double wishbone system abandoned solely for reduced manufacturing costs?
I will assume that the reader is familiar with the basic anatomy of these systems.
Nowadays, it is customary to call a strut and lower wishbone suspension a MacPherson strut when it is used at the front of the car, and a Chapman strut when it is at the rear. It should be noted, however, that MacPherson originally conceived his design as one that could be used at either end of the car, and Chapmanís earliest versions did not have a lower wishbone as such; the halfshaft was fixed-length and a radius arm was added. However, Chapman was the first to actually sell cars with strut rear suspension. Later versions did have lower wishbones, and the system in that form was still called a Chapman strut.
While struts are popular and have been for a long time, it isnít really correct to say that theyíve become universal. I donít expect to see them on a Miata or a Corvette any time soon. However, they replaced double wishbone (or short and long arm, or SLA) front suspension on Mustangs and Camaros in the 1980ís, and more recently on Honda Civics.
Strut suspension is space-efficient if thereís room for its height, simple, and gives geometric properties that are inferior to a good double wishbone system but not awful. It integrates well with unit-construction and space-frame cars, but not so well with ladder or backbone frames. It applies its loads to the sprung structure at widely separated points, which tends to reduce the magnitude of the loads.
There are various ways to mitigate the systemís geometric shortcomings, but these involve compromises to the systemís advantages.
The geometric problems are mainly in three areas. The camber recovery in roll is poor. Geometric roll resistance (roll center height) varies a lot with ride height, and so does camber recovery in roll. Steering geometry generally has more front-view steering axis inclination than we would like.
Camber recovery in roll can be improved by inclining the strut more. However, that makes the steering axis inclination problem worse, and also reduces available space between the strut towers.
The ample space between the struts is the reason the system is space-efficient. Vertically, itís not so compact; it tends to be tall. That means it packages well in a sedan, minivan, or SUV, but not so well in a sports car or an open-wheel car. It is particularly attractive for cars that either have the engine far forward, or have it in the rear. The tops of the struts can then anchor very nearly directly to the cowl structure. That makes the system good for front-drive or rear-engine layouts, and not as attractive for front-engine/rear-drive cars. In a front-engine/rear-drive layout, we either have to add bulky structure from the firewall forward to the strut towers or accept poor weight distribution due to the front wheels being too far back.
The strut has sliding elements, and also is subjected to bending loads. This creates friction in the suspension. Making the strut longer helps, or more specifically, increasing the distance between the piston and the shaft bushing helps. However, that exacerbates the height problem, or reduces available travel. With a single-tube damper, a remote reservoir offers a little help on this, but adds cost and complexity. Remote reservoirs are consequently seldom if ever seen on production road cars. Some designs offset the spring centerline from the strut centerline so that the spring force reduces bending loads on the strut.
The strut suspension has geometry similar to a double wishbone system with an extremely long upper arm. That makes the front view force line change slope dramatically as the suspension moves, and makes the front view swing arm length increase as the suspension compresses and decrease as the suspension extends. This implies that geometric anti-roll, or roll center height, decreases a lot on the outside wheel as the car rolls, and so does camber recovery. It also implies that both geometric anti-roll and camber recovery in roll greatly decrease when we lower a strut car for racing.
In contrast, with double wishbones the geometric anti-roll varies less with suspension movement and with ride height setting, and the camber recovery in roll generally improves when the car is lowered for racing.
With either system, making the control arms longer improves geometry. However, packaging considerations will impose limitations on this; long arms take up room. One possibility that has occurred to me, that I have never seen tried, would be to substitute a leading or trailing arm (transverse pivot axis) or a semi-leading or semi-trailing arm (diagonal pivot axis) for the lower control arm in a strut suspension. That would provide a longer front view projected lower control
arm to go with the very long upper one. There is some loss in structural efficiency of the arm itself, compared to a broad-based wishbone, but itís not prohibitively bad. Lots of cars use such arms without a strut, to carry all the suspension loads.
I think somebody should try a semi-leading arm in place of the lower wishbone, anchored near the intersection of the A-pillar door jamb and the toe board, and at a point near the center of the car and further forward. Anti-dive would diminish as the suspension compresses, but it probably would be no worse than an old Chevelle (or a modern upper-division NASCAR vehicle) in that regard. Load paths would be pretty decent. Steering would need to be re-thought, to avoid excessive bump steer, but thatís entirely feasible.
It is also possible to use highly diagonal pivot axes for the arms in an SLA system, creating longer front view projected control arms for both upper and lower Ė again, at some penalty in structural efficiency of the arms.
As for the excessive front-view steering axis inclination usually produced by strut systems, it is possible to make the wheel not steer about the upper strut anchor and the lower ball joint. It is possible to prevent the wheel from steering on the strut, and provide a separate kingpin or equivalent mechanism to define the steering axis instead. One big advantage of this would be to reduce the tendency for the outside wheelís camber to go toward positive with steer. Renault and Nissan have actually implemented this as a performance upgrade on the Clio Cup version of the Clio. The regular Clio has a more conventional strut front end. This has the disadvantage of adding complexity to a design whose simplicity is otherwise one of its main advantages.
As a general rule, double wishbone systems have advantages in terms of tire wear, because they generally provide better camber control. However, this varies depending on the specifics of the particular system. Some years ago, Buick produced some SLA systems with worse camber properties than the average strut system. Bushing compliance and other compliances also influence this a great deal.
Suspension wear properties are likewise highly dependent on the specifics of the particular design, but we can say that the strut system has fewer wear points and should therefore cost less to keep in decent condition over the life of the car. The strut itself generally ceases to provide acceptable damping before it ceases to provide acceptable wheel location.
What does castor on the rear of a race car affect? As there is no steering of the rear I can see no need for it.
In a front suspension, caster (American spelling; the questioner is Australian Ė in the US, castor is bean oil) is the side-view inclination of the steering axis. It relates to the inclination of the upright or spindle, but it is usually measured with turn plates and a caster/camber gauge.
Rear caster is the side view inclination of the rear upright, as measured by some method that is usually specific to the particular car. Usually we cannot define a steering axis for a rear wheel. One common datum is a horizontal surface or pair of points at the bottom of the upright. Measurement is taken with an inclinometer, or angle finder.
The concept of rear caster is generally only applicable to independent rear suspensions, whereas front caster is applicable to all kinds of front ends.
What does rear caster affect? It affects rear bump steer and roll steer. It provides a shortcut to recovering a rear bump steer adjustment, established either by the manufacturer or by the carís crew, when reassembling the system.
Rear caster came into use as a setting specified by race car manufacturers around 1970, when the previously popular reverse wishbone lower arms were commonly replaced by two parallel lateral links, often with roughly parallel radius rods running forward from the upright. In such systems, it is possible to get the bump steer very close to zero, and the lower lateral links very close to parallel in front or rear view, by setting the rear caster to zero, or to the manufacturerís specification.
At the front, caster also affects bump steer. With almost all types of independent suspension, adding caster moves the geometry toward roll oversteer, or toe-in in bump and toe-out in droop. In race cars, there is usually a separate means of adjusting bump steer, often by shimming either the outer tie rod end or the mounting of the steering rack. That way, we can set the caster where we want it from the standpoint of steering feel, camber change with steer, and weight jacking with steer, and also have the bump steer characteristics we desire.
In a rear suspension that does not have adjustable side-view upright inclination, and does have toe control links, it is possible to provide adjustment of bump steer by adjusting the height of one end of the toe control link, as with the tie rod in a front suspension.