The MacPherson Strut


Over the years, there have been innumerable variations on MacPherson’s original design. Some common modifications include:

  • Omitting the anti-roll bar: While MacPherson’s dual-function anti-roll bar is clever, cheap, and elegant, a front anti-roll bar is not always desirable, particularly for lightweight, front-heavy cars that already have fairly stiff front springs. However, if the front anti-roll bar is omitted, some other means must be provided for triangulating the lower control arms. Some automakers resolve this dilemma by replacing the lower arm with a lower wishbone. Others, including early British Mk1 Ford Escorts and the Mk1 and Mk2 Ford Fiesta, use radius rods (leading or trailing links) to locate the lower control arm. One advantage of using radius rods in this way is that they can be designed to allow some fore-aft compliance for better ride quality. It also becomes possible to provide a measure of anti-dive geometry by setting the front mounting point at a different height than the lower ball joint.
  • Single-function anti-roll bar: Having the anti-roll bar do double duty as a radius arm may force designers to accept a spring rate for the bar that is either higher or lower than ideal for optimum ride and handling. Therefore, it’s sometimes desirable to locate the lower control arms with radius rods or use lower wishbones even if the vehicle has a front anti-roll bar. This costs and weighs more, but allows better anti-roll bar geometry and more freedom in selecting the bar’s spring rate.
  • Relocated springs: Mounting the coil spring around the upper part of the strut is simple and tidy, but, as noted above, requires the fender structure to be reinforced to withstand spring loads, resulting in tall, bulky strut towers. An alternative is to relocate the spring to the lower control arm, as in a typical double wishbone suspension. Some manufacturers have used coil springs in this manner, but a few (notably Porsche) have substituted longitudinal torsion bars, typically using the lower wishbone or control arm as a lever. Either way, the primary advantages are better packaging and a lower fender line; struts without high-mounted coil springs are also compatible with body-on-frame construction, which conventional struts are not. Struts with offset springs (or torsion bars) are sometimes called “modified MacPherson struts,” although technically any of these variations could be so described.
  • Double-pivot struts: Patented by BMW in the late seventies and applied to the E23 7-Series and many subsequent BMW cars, a “Doppelgelenk” front suspension locates the strut with a conventional lower control arm triangulated by a short diagonal leading link. The lower control arm is attached to the spindle via a ball joint in the conventional manner. The diagonal link connects to the strut via a second ball joint mounted above and slightly ahead of the first. Together, the link and control arm form a wishbone angled upward at the front to provide anti-dive. More significantly, the additional lower ball joint serves to alter the kingpin inclination and therefore the scrub radius. With two lower ball joints, the kingpin inclination is determined by the line between the upper ball joint and the point where the axes of the lower control arm and leading link intersect (the virtual steer center). The virtual steer center moves as the front wheel turns, so the scrub radius is no longer a constant, increasing as the wheel is steered away from center. The idea is to reduce bump steer in straight-ahead cruising while increasing self-centering action in turns.
Diagram of a modified MacPherson strut with low-mounted coil springs © 2014 Aaron Severson

A so-called “modified” MacPherson strut with low-mounted coil springs. Suspensions like this were used on the Ford Fox platform (used by the Ford Fairmont and the 1979–1993 Ford Mustang) and the third-generation F-body Chevrolet Camaro and Pontiac Firebird. (author diagram)


All of the above variations are fairly straightforward and by now quite common. A significantly more elaborate variation has emerged more recently, driven by the the emergence of powerful FWD sporty cars. The challenge for automakers is to fortify those models to cope with a big infusion of horsepower without sacrificing their commonality with the mundane family sedans and hatchbacks on which they’re based, many of which use MacPherson struts for cost and packaging reasons.

In the early nineties, Toyota unveiled an optional front suspension package called Super Strut for certain sporty FWD and AWD models, including some versions of the AE101 and AE111 Corolla and Sprinter, the Celica and Curren coupes, and the Carina and Corona. (Super Strut was included on some export versions of these cars, but was never offered on any U.S. Toyota.) Put simply, Super Strut was an attempt to approximate the geometric advantages of a double wishbone suspension in a package that could interchange with Toyota’s standard MacPherson strut front suspension.

Super Strut suspension from a Toyota Celica GT-Four © 2005 Kris Carter (used with permission)

The (mostly) complete Super Strut assembly. The connector plate joining the two lower control arms (lower foreground) attaches to the bottom of the knuckle (not shown). The eyelet on the base of the strut is for the drop link that connects the strut to the front anti-roll bar (also absent here). Super Strut is theoretically interchangeable with standard MacPherson struts, although doing so requires replacing the entire front suspension, including spindles and brakes! (Photo © 2005 Kris Carter; used with permission)

On Super Strut cars, the base of each front strut (above the steering knuckle) has a curved extension shaped a bit like an inverted letter “C.” One end of the extension forms the mounting point for the upper ball joint, which is relocated to a point just below and outboard of the base of the strut. The lower end of the strut extension, meanwhile, is connected via a ball joint to a short lever arm (the assist link or “figure-eight” link), which is in turn connects (via another ball joint) to a point at approximately the center of the rear lower control arm (which Toyota calls the camber control arm).

Super Strut lower control arms from a Toyota Celica GT-Four © 2004 Kris Carter (used with permission)

The front and rear control arms of the Toyota Super Strut suspension, illustrating the way the assist link (“figure-eight” link) connects to the center of the rear (camber control) arm. The outer ends of the control arms connect to the front subframe, the rear arm (camber control arm) via a spherical joint. (Photo © 2004 Kris Carter; used with permission)

There is also an additional front lower control arm, a longer, curved arm that Toyota fans have dubbed the “banana bar.” The outer ends of both lower arms are connected via ball joints to a small connector plate on the steering knuckle, allowing the arms to pivot relative to one another as the knuckle turns. The front anti-roll bar, which does not contribute to wheel location, is connected to the strut itself via a drop link with ball joints at each end.

The geometry of this system is quite complicated, but there are five principal effects:

  1. Separating the kingpin axis from the strut. Since the upper ball joint is mounted outboard of the strut itself, Super Strut provides a steering axis is similar to that of a double wishbone suspension, greatly reducing the scrub radius. An interesting side effect is that the strut no longer turns with the knuckle, although the strut does rock fore and aft as the wheel is steered.
  2. Creating a virtual steer center through the use of two lower ball joints. This appears to be analogous to the BMW Doppelgelenk system, although our information does not indicate to what extent Super Strut’s virtual steer center moves as the arms pivot.
  3. Reducing the spindle height by relocating the upper ball joint.
  4. Reducing caster changes as the springs compress or the wheels turn, which also serves to reduce camber changes in tight turns.
  5. Reducing the effective upper control arm length by pivoting the strut extension (via the “figure-eight” assist link) at the center of the rear lower control arm.

The end results are reduced torque steer — an important consideration when putting a lot of power through steered wheels — and significantly more camber gain than a conventional MacPherson strut would permit, improving front-end grip. To take advantage of the new geometry, Toyota specified wider, more aggressive tires and bigger front disc brakes (with two-piston calipers) for most Super Strut applications.

Super Strut suspension installed in a Toyota Celica GT-Four © 2004 Kris Carter (used with permission)

Toyota’s Super Strut suspension as installed in a car, viewed from the rear. Note the way the curved strut extension connects to the assist link and the camber control arm. (Photo © 2004 Kris Carter; used with permission)

The Super Strut suspension was effective, at least as long as it was in good repair, but the system was both heavy and expensive. Its sheer complexity also made it less reliable than a conventional strut. There was a lot to wear out and the components were pricey to replace if they did fail. Toyota discontinued its last Super Strut model around 2006.


In 2004, Renault introduced a loosely comparable system for the Mégane RenaultSport (RS) hot hatch. Like Toyota’s Super Strut, the Renault “double-axis” system separated the steering axis from the strut using a relocated upper ball joint and a broad lower arm with a separate anti-rotation link that allowed the knuckle to turn without turning the strut. This arrangement, now dubbed PerfoHub, is still used on some current Renault Sport models.

RevoKnuckle diagram from a 2009 Ford Focus RS. Copyright © Ford Motor Company

A diagram of RevoKnuckle as used in the 2009 Ford Focus RS. (Ford Motor Company)

In 2009, both Ford and GM introduced their own systems. Ford’s, called RevoKnuckle, was used to allow the turbocharged Focus RS to cope with 305 PS (224 kW) without resorting to all-wheel drive. During the same period, GM introduced its similar HiPer Strut, offered initially on the Opel/Vauxhall Insignia OPC and later on the Astra OPC, Buick Regal GS, and Buick LaCross CXS. RevoKnuckle and HiPer Strut differ from the Renault layout in detail, but are functionally similar.

Diagram of HiPer Strut in the 2010 Buick LaCrosse CXS - X10AR_BU003- Copyright 2014 General Motors LLC. Used with permission, GM Media Archive. (GMMA 19720p2)

The complete HiPer Strut installation, including front subframe, of the 2010 Buick LaCrosse CXS. (Image copyright 2014 General Motors LLC. Used with permission, GM Media Archive.)

Where HiPer Strut, RevoKnuckle, and PerfoHub differ from the earlier Toyota system is that the newer layouts do not attempt to replicate the complex geometry of Super Strut’s figure-eight link and camber control arm. The GM, Ford, and Renault systems do provide some additional camber gain due to their reduced spindle heights, but the manufacturers’ own descriptions generally downplay that point, stressing instead that the primary goals are to minimize torque steer and reduce steering kickback over bumpy roads.

Exploded diagram of HiPer Strut from the 2010 Buick LaCrosse - X10AR_BU002 - Copyright 2014 General Motors LLC. Used with permission, GM Media Archive. (GMMA 19720p2)

An exploded view of HiPer Strut. In case it’s not obvious, “LCA” stands for “lower control arm.” (Image copyright 2014 General Motors LLC. Used with permission, GM Media Archive.)

The new systems are likely cheaper than the Toyota Super Strut layout, although there is still a significant cost and weight penalty that may limit the use of these suspensions to more expensive, more powerful models. It remains to be seen how broadly these layouts will be adopted.


Add a Comment
  1. You can still get some camber change out of a strut.

    Looking at your diagram, when the suspension compresses, the lower control arm will move through an arc that will push the bottom of the knuckle outward.

    The bottom of the strut can’t rotate relative to the knuckle, so its upper mount will have to flex some to accommodate the change in angle.

    Nitpicking aside, it’s always cool to learn the story behind an eponym. When can we learn about Mr Cardan and his double-joint?

    1. MacPherson struts do have some camber change in compression, as do most independent suspensions other than pure trailing arms, but the amount is very small because the effective swing-arm length is quite long. What they don’t provide is camber [i]gain[/i] with body roll, the way a double-wishbone suspension or swing-arm suspension does. As the body leans and the outside spring compresses, the arc of the lower control arm does push the lower edge of the spindle outward (which would tend to create negative camber), but the magnitude of that force is much less than that exerted by the strut as body lean pushes it outward (which tends to create positive camber). The result is always a net camber loss.

      The fact that increasing positive camber is called camber loss and creating negative camber is called camber gain is the kind of thing that gives me a headache.

      Gerolamo Cardano didn’t actually create the DCJ, since he died in the 16th century — the universal joint was named after him because he invented the concept, but the double Cardan joint was a refinement developed many years after his death. He was an interesting guy, though…he and André Citroën would have gotten along well.

  2. Hmmm…true. Suspensions are funny because major differences in ride, handling and character are wrung from tiny geometry tweaks. Through much engineering education, differences in angle that we’re talking about would all be considered “vertical” to a reasonable approximation.

    The funny part about a double-cardan joint is that when there’s there’s one u-joint, it’s usually called a u-joint. When there are two (like the one I blew up on my Jeep recently), it’s called a double-cardan or CV, even though it’s not really a constant-velocity joint. No one ever talks about a single-cardan joint.

    The backwards/upside down one I love is gearing. “I’m going to lower my gears from 3.55:1 to 4.10:1.”

    1. [quote]Suspensions are funny because major differences in ride, handling and character are wrung from tiny geometry tweaks.[/quote]

      This is true, which will tell you something about the horrors of swing-axle suspension — an early Corvair had 16.25° of camber change through its complete suspension travel (about six and a half inches). Holy Snap Oversteer, Batman!

    2. That’s the term because you go down the gears to a lower gear.

    3. Thank you for pointing out the errors on other sites as dating the first MacPherson struts as 1955. My first car in the U.K was a 1951 Mk. 1 Ford Consul follow by a Zephyr and the Zephyr Zodiac also of similar years and fitted with MacPherson struts. My brothers Ford Prefect of similar vintage also had MacPherson struts. It might be interesting to note these struts had a little brass plug that you could remove and top them up with hydraulic oil, very handy as they were prone to leaking profusely.

      1. That sounds like a good feature, honestly! Even today, it’s pretty common for struts to weep a little fluid (especially when brand new) even if they’re pretty healthy.

  3. Nowdays, MacPherson struts are being put on heavier cars, and on heavy 4×4 SUV’s, of which, most off the people who buy them will rarely take them off road, but some will, believing they are true off road vehicles, even when usually called, car based or crossovers.
    So my question or concern is this. Are stuts really strong enough for these uses, usually reserved for double wishbone or solid axles. Has the diameter of the piston rod increased, or is there an increase of the strength of the steel used for today’s struts, to keep up with the increasing demands placed on them.
    I would like to know.

  4. “Strength” is a matter of definition. Are modern MacPherson struts physically robust enough for their intended use, in terms of bending stiffness and so forth? Sure. Do they incorporate shocks beefy enough for heavy-duty off-road use? Rarely. Very few modern vehicles are really set up for that kind of abuse, since less than 5% of them are driven off road at all, even fewer in any serious off-roading; the shocks are usually designed for a civil on-road ride. If you take a Mazda CX-9, say, off road, you’ll probably kill the struts in short order, not because the struts lack bending stiffness, but because the dampers will be overloaded.

    The real limitation MacPherson struts present for severe duty is not the physical strength of the strut itself, but the way the strut transmits its loads into the body. In an SLA suspension with the spring between the control arms, the spring loads are taken by the control arms and transmitted either to the frame (on a body-on-frame vehicle) or, on most modern cars, a front subframe. With MacPherson struts, or with the suspension layout used by old Ramblers, Falcons, and Mustangs (with the coil mounted on the upper arm), the spring loads are transmitted into the structure of the upper fender. (That’s why MacPherson struts are almost always found on unit-body vehicles.) For on-road use, that’s rarely a problem, but for heavy-duty use, it can impose severe, uneven loads on the body structure. You could make a strut strong enough to deal with the loads on the strut itself, but the fender is another matter.

  5. I see that Ford has switched to SLA for the 2011 Explorer as opposed to the MacPherson strut set-up on all their other D platform cars.

    Why do you think they did that? Does SLA provide greater wheel travel or greater durability or some other advantage an Explorer would demand over say, the Flex?

    1. I’ve heard conflicting information about this. A couple of press reports say the new Explorer’s front suspension is an SLA set-up carried over (which I find unlikely) from the outgoing model, but Ford’s own official website says the new model has MacPherson struts, like the other D3 vehicles.

      MacPherson struts present a number of disadvantages for off-road vehicles (scrub radius with fat all-terrain tires, and the way they transmit load to the body structure), but Ford is not positioning the new Explorer as a hardcore off-roader. I would imagine that for Ford, the loss of commonality would probably outweigh any mechanical advantages, especially given that the market for the Explorer is a big question mark — sales of the existing model cratered long ago, and it’s unclear whether the new one will revive them.

      1. Yeah, I noticed the confusion as well but in the end i assumed the press release was right.

        1. First generation Mazda6, whose platform was shared by first-gen Ford Fusion and Ford Edge, had SLA front suspension. The D3 platform has MacPherson strut front suspension. It’s likely the SLA suspension will be phased out with the demise of the first generation Fusion. There was some confusion on this. Even the Mazda website was confused about whether the pre-Skyactive second-gen Mazda6 had SLA or strut front suspension.

          1. As a side note, I happened to learn the etymology of the term SLA suspension, which is not short-long-arm as you might expect. Cadillac applied that term to its first IFS prototype (around 1932), which had what became the early GM double wishbone suspension (with upper arms doing double duty as part of the lever-action shocks). Since that car also had a special aerodynamic body, it was dubbed SLA for Stream Line A, although there was apparently never a Stream Line B or C.

          2. BMW motorcycle’s “TeleLever” front suspension is a form of “modified” MacPherson strut. There are two sliders, one on each side of the wheel, these are sliders only. It uses a separate coil-over-shock that acts on the A-arm.

          3. Thanks for the information. I freely confess I know basically nothing about motorcycles and so I can’t speak intelligently about motorcycle engineering practice.

  6. I’ve sort of followed this progression with bemusement.

    When I was coming of age in the 60s, almost all American cars used unequal A arms in front (Chrysler used a torsion bar variant).

    By the early 70s, after some European and Japanese cars started appearing with struts, references to MacPherson struts appeared in ad copy constantly, it was a badge of sophistication.

    Then in the 80s, manufacturers (and ad copy writers) rediscovered A arms again, and that was the new mark of a sophisticated car.

    Now I’m waiting for solid axles to return (actually I own a Jeep Wrangler, so I am still on solid axles)

  7. Well, it’s not [i]just[/i] a matter of engineering fashion. Now that most cars are unibody, and very rigid, MacPherson struts are a lot more practical than they were 40 years ago. While struts have some downsides, they also have significant advantages. Most A-, B-, and C-segment cars have MacPherson struts for both cost and packaging reasons — it’s cheap, and it doesn’t take up a lot of internal space, important with small transverse-engine cars. Many D-segment cars use them, as well, because it’s an easy way to reduce production costs that most customers don’t perceive or necessarily care about.

    Because SLA or double wishbones are generally more expensive, heavier, and take up more space, they’re now sort of a luxury. Designers have to consider whether the advantages in ride and handling are worth the cost. For a high-end sports car, it might be; for an inexpensive B-segment hatchback, probably not.

    I doubt that [i]front[/i] solid axles will make a comeback — for anything other than off-roaders, the drawbacks are steep. Torsion-beam rear axles, though, are almost as common as MacPherson struts on smaller cars. Even Honda opted for a torsion beam for the Fit and the European Civic.

  8. Had to get my 2 cents in here regarding camber gain with struts. Camber gain can be achieved ,though very limited,if the inner pivot of the control arm is located higher than the ball joint in the static position. As body roll compresses the strut, the ball control arm pushes the wheel outward creating a bit of negative camber which will cancel some of the positive generated by the roll. The down side is a change in track during cornering as well as straight line travel. Raising the inner pivot will also raise the roll center which will affect the handling as well. A friend and I played with this mod on both front and rear drive autocross sedan years ago, and were able to gain some improvement in overall handling and cornering force. The big downer was heavy tire wear due to scrubbing as the track changed. Fun to experiment with though!

    1. That makes sense, and certainly, different manufacturers have found various tricks to mitigate some of the geometry limitations of the classical MacPherson strut (BMW and more recently Ford spring to mind). Of course, as you found, some changes are more suitable to the track than the street!

  9. An observation that may be worth adding is that on cars with MacPherson strut suspensions, the tire tread wears out on the edges much, much sooner than in the middle, even with camber correctly adjusted. On my Audi, tires show virtually none of this, and tires last a good deal longer.

    Another observation that may be worth adding is that in a conventional MacPherson strut using a single ball joint where the lower arm meets the wheel carrier, this ball joint resides on the steering pivot axis, and is thus aligned with the centerline through the strut. However, the strut is generally not perfectly vertical, and the steering pivot line typically insects the ground somewhere under the tire contact patch, as opposed to a point vertically under that ball joint, but also not at the center of the contact patch. Some cars, notably certain BMWs, replace the lower arm with two arms, each with an independent ball joint, the effect of which is to replace the single physical ball joint with a virtual steering pivot point located further outward. The strut has to be angled further from vertical to accommodate this, but not nearly so much as would be required to accomplished the same effect (causing the steering pivot line to intersect the ground plane at the middle of the contact patch) using a single ball joint located on the inner edge of the wheel carrier. The principal advantage is probably not with any handing improvement, but more likely with reduced tire wear and possibly greater life of the ball joint.

    As anyone who studied elementary geometry in high school may recall, a triangle is fully defined by any three parts. If at least one of those three known parts is a side, then the size of the triangle as well as its shape is determined, but even when the three known parts are angles, the shape is fully determined. You will have no difficulty figuring out that the lower arm corresponds to one side of a triangle, and that the chassis is effectively another side of a triangle. The upper mount point for the strut is basically a lazy suzan, and the distance between that point and the pivot joint for the lower arm is another fixed side of the triangle. That gives you two parts for the triangle, which tells you that no other part of the triangle can be fixed, least all the other parts of the triangle also must be fixed, in particular the angles at the ends of the lower arm. But the strut itself joins to the chassis at a fixed angle. As the suspension compresses and the wheel camber changes, the angle of strut body changes relative to the chassis. The upper rod remains at the same angle relative to the chassis, albeit rotating to accommodate steering rotation. It is thus apparent that it is necessary for the strut itself to flex, but none of the articles that I have thus far encountered have made any mention of this fact, which seems to me a defining characteristic of the MacPherson strut. I am interested in other people’s comments on this aspect of the MacPherson strut, and whether anyone has reliable knowledge of just how much flex is required.

    1. R.E. the location point of the top of the strut. The top of the strut is not a fixed point at all. While the strut may in fact flex, or bend to a certain extent, the amount is negligible unless the design loads are exceeded. For instance, fitting larger, sticker tires. Rather, it is located in a rubber mount, both for the purpose of isolating it to reduce NVH trasmission into the body structure, and to allow for the aforementioned camber gain/loss. These mounts also serve as a path for virtually all of the vertical load transmission as well as a portion of the lateral loads. As you might expect, they are highly stressed, and, depending on how well engineered they are, and the quality of the material used, in some vehicles they tend to wear and fail fairly rapidly. As a VW technician, I see this quite often, especially in the older A3 and A4 body Golf/Jetta.

  10. I’m a faithful reader of your blog, but there is a true mistake for this article : the first Ford Vedette was not the first production car with MacPherson struts.
    It’s a common mistake, but the first Vedette used the common SLA arrangement for front suspension.
    You can see a picture of the frame here :

    There is more explanation (in french !) in the message of Vega 770_0 posted 01-12-2009 (21:42:28 pm).
    In fact, the very first was the Ford Consul/Zephyr in 1950. Then, the second generation Ford Vedette was launched with MacPherson suspension in 1954, a few weeks before the takeover of the Ford SAF by Simca. The third one was the German Taunus 17M P2 three years later.
    Surprisingly, as far as I know, the first non-Ford MacPherson car was the Peugeot 404 in 1960…
    Moreover, I’ve seen in some Revue Automobile Suisse catalogs that many japanese brands started to use it in the sixties : Honda already had it on the tiny N360/N600.

    1. Thank you for the correction and the photo — I’ve amended the text accordingly.

  11. what were the earliest Ford models to use these struts in their suspensions (early 1970’s, I think)?

    1. Well, the first Ford product to have MacPherson struts was the Consul. The first Ford marketed in the U.S. with MacPherson struts? A number of European Fords were sold here in very limited numbers by dealers with English Ford franchises, so there were some Cortinas floating around in the mid-sixties. The first to be marketed here in any coherent way, maybe the Capri, followed by the Mk 1 Fiesta.

      The Fox platform used what Ford called modified MacPherson struts (which meant that the coil spring was mounted on the lower control arm, rather than on the strut). I think the first unmodified U.S.-specific car might have been the first-generation, U.S.-market Escort, although I’d have to check.

  12. Tom: I’ve wondered the same thing. In my mind, there needs to be some pivoting or flexing in the strut, otherwise it will bind. I asked some years ago on the AtlasF1 forum, and all I got were blank stares.

  13. Tom, the upper mounting of the strut is not a fixed point. While it is a bearing that allows the strut to rotate for steering purposed, the bearing itself is mounted in a very compliant rubber mount. As the strut compresses and extends and the angles change, the mount has enough range of motion to allow the angles to change without flexing the strut. Having said that, there is a certain amount of flex that does occur, as it must, but strut tubes and pistons are quite robust (in most cases, there have been some notable exceptions)and generally keep the tire located within the intended range of caster and camber.

    In regards to your mention of BMWs use of seperate lower links to create a lower arm, there is indeed a distinct handling benefit to it. The idea is to bring the center of the steering axis closer to the center of the contact patch, thereby reducing the scrub radius. Scrub radius is the arc the contact patch travels through as the wheel is steered left to right. A large scrub radius is undesireable, as it tends to create a “pull” felt at the steering wheel, especially when there is a difference in traction between the left and right wheel. As you noted, there is another problem associated with this is SAI, or steering axis inclination. That is the angle of the steering axis when viewed from the front of the vehicle as it is tilted toward the middle of the car. It has a very profound effect on wheel camber as the wheel is steered from straight ahead, and not in a good way. SAI tends to cancel out caster on the outside wheel in a turn, and multiplies it on the inside. Large SAI’s can even push the outside wheel into a positive camber angle. A good example of this is the Volvo 240 series. Just turn the wheel full lock in one direction and look at the wheels to see what I mean. Typically, you will find that the better handling cars (those that turn in well and remain relatively neutral, or even oversteer slightly) have low SAI’s. VW/Audi found an interesting solution to the problem with the B5 Passat/A6, by using 4 individual links to create 2 wishbones. The result was a nearly vertical SAI. The actual steering axis in this case was not even fixed. Rather, because of the movement of the joints, it moves also in an arc of its own.
    A facinating subject for sure. Hope this clears away some of the fog.

  14. Very informative piece, most of which I already knew, but with a few surprises. The only thing you missed is the inverted strut used on rally cars, where the piston in the strut carries the wheel spindle and balljoint, and the damper end fits into the wheel arch ( to reduce unsprung weight ).

  15. Aaron,

    Two quick notes.
    By some accounts, HiPer Strut is a Saab invention. For instance, Autocar (UK) writes that “This front suspension set-up was designed by Saab engineers for General Motors’ Global Epsilon project.” I’m sure the patent application lists the actual inventors, but I couldn’t find it.

    One of the key elements in scrub radius is wheel offset. It’s also something that amateur suspension gurus almost always get wrong when they “tune” cars by lowering them, fitting bigger wheels, and sometimes even adding wheel spacers. It’s very easy to add an inch or more of offset to a car, or even to take it for negative to positive offset (or vice versa), with dangerous results.

    1. Bernard,

      You make a good point about offset, although the impact of altering the design offset of a car’s wheels is sort of outside the scope of this article. There’s a lot of complexity there that’s really a separate topic.

      I hadn’t seen the item about Saab designing HiPer Strut, although it would make some sense. Saab really got a lot of criticism for the torque steer of some of the more powerful later 900 variations (particularly the Viggen, as I recall), so their engineers may have been giving the issue a lot of consideration.

      As far as patents go, there may not be any specific patents for HiPer Strut per se. There may be for certain specific elements of it, but as the article explains, there’s a lot of prior art in this area. Even with Toyota’s earlier Super Strut, there was no single patent that was recognizably the production layout (I wish there had been because it would have made it easier to figure out the mechanics!), although there was a whole series of patents covering certain elements. One included illustrations of something like a dozen possible variations.

  16. By the way, Colin Chapman didn’t abandon strut rear suspension on the Elan. He abandoned the trailing arm location, using a wide-based lower “wishbone” to control lateral and fore-and-aft location, and rubber doughnuts in the drive-shafts.Some modern aftermarket replacement Elan chassis do use double wishbones instead of the original struts,however.

    1. Thanks for the clarification. Splitting hairs, I would still not call the Elan’s rear suspension a Chapman strut layout — although some contemporary press articles did — because the halfshafts don’t provide lateral location in a way the halfshafts do on the contemporary Jaguar or Corvette Sting Ray independent rear suspensions. I’d just describe it as a MacPherson strut located by a lower wishbone.

  17. I thoroughly enjoyed reading your article about Earle, my great uncle. I never got to meet him, but I do know a few things about him. I didn’t know he had a short temper, but it doesn’t surprise me because he was such a genius and was a Scot. I love knowing stories about him. His father was an inventor, as well. I have some photos of Earle as a young man and in his WWI uniform. I wonder if you would like to see them!

    Thank you for the great article!!


    1. Liz,

      I would love to see pictures of him earlier in his life. If you like, send me a message via the contact form.

      I don’t know that he had a short temper per se, but I got the impression that he was a person of strong opinions and didn’t suffer fools gladly. Admittedly, many of the anecdotes of people who had run-ins with him at Ford were in situations where someone was doing something they technically weren’t supposed to be doing or were just horsing around. For instance, there is an amusing story about some of the designers putting a giant plaster elephant in a wind tunnel where they were supposed to be doing aerodynamic testing, which I imagine was particularly infuriating to him because Ford was renting the wind tunnel by the hour!

  18. I cant help thinking that the passenger area of the Chevrolet Cadet prototype looks remarkably like that of the Vauxhall Wyvern, Velox and Cresta of the mid-fifties(Vauxhall is of course GM owned) .Could that Cadet six cylinder engine have also found its way into the Velox and Cresta, as the size and output seem remarkably similar?

    1. The stylistic resemblance is no great surprise, given the way postwar career advancement typically worked in GM Styling (and for that matter Ford). It wouldn’t have been terribly surprising to have at least a few of the same people have either worked on both or remembered seeing the earlier car, although it may have simply been a reflection of common styling trends.

      As for the engine, I think the resemblance is coincidental. My understanding is that the Velox/Cresta engine was essentially the 1952-vintage 1.5-liter four with two additional cylinders, and I don’t know that there was any relationship with the experimental Cadet engine beyond the broad similarities one saw between various GM OHV engines of similar vintage and displacement. (The Vauxhall six shares the 3 1/8-inch bore of the Victor four, versus the Cadet’s 3 1/16-inch, giving 2,262 vs. 2,173cc.) Vauxhall would have had to buy the Cadet design from Chevrolet — with a markup, in all likelihood — so I don’t think there would have been any particular advantage in using it versus one developed by Vauxhall itself. Vauxhall had been using OHV engines for 20 years and didn’t really need any technical help in that regard. The same logic is how GM ended up with no fewer than FOUR distinctly different 5.7-liter V-8s 15 years later!

      Vauxhall did, of course, offer its own Cadet for a while in the thirties, although there was no relation.

  19. This is easy for me to say 70 years after the fact, but if MacPherson had played up the superior handling of the fully independent suspension, the Cadet could have been the starting point for a sports sedan in the mold of the Rover P6/Triumph 2000/BMW Neue Klasse.

    Of course, it’s easy to see why this didn’t happen. The brief was to design an inexpensive compact, not a car for a market segment that wouldn’t exist for another 20 years. Sports cars were very much a niche market in the U.S. at the time. Earle MacPherson, much less GM management, couldn’t have been expected to think this far out of the box.

    1. I don’t really think the sport sedan in the modern sense was really on anybody’s radar at that time. I think for the most part, even when engineers considered the subject of handling, it was typically prefaced with “ease of,” which isn’t quite the same thing. The standards for maneuverability were somewhat higher in markets like the U.K. or Italy, where narrow, winding roads not originally intended for cars were commonplace, but in the U.S., “handling ease” was generally about stuff like steering effort and how wieldy a car was in tight spaces or for parking. The bigger preoccupation, and what I assume was MacPherson’s principal argument for independent suspension, was getting a smooth ride on what for an American car was a ridiculously short wheelbase.

      Mostly, I don’t think the cultural factors would have been present in the forties or fifties for a sporty sedan to catch on beyond folks like Tom McCahill and a small class of people Detroit generally dismissed as cranks well into the seventies. Some buyers would probably have been pleasantly surprised, I imagine, but even the later American enthusiast crowd would probably have judged it mostly on its suitability for having a big V-8 jammed under the hood.

  20. From what I can find online about the Ford Light Car project which became the Vedette, it seems Maurice Dollfus in 1939 wanted a 4-cylinder engine which Ford US adapted from the German Ford Taunus only for him to later reject it.

    The 4-cylinder Taunus engine in question may have been the 44 hp 1.5 side-valve unit that was due to appear in the Ford Taunus G93A, yet also recall pre-war plans by Ford of Germany (if not Ford UK) to convert the side-valve engine to OHV like on the later post-war 55 hp 1.5 Ford Taunus P1.

    However find it somewhat difficult to believe the 1.5 Side-Valve / OHV unit was in fact the 4-cylinder engine intended for the Ford Light Car project aka Vedette. Probably wrong, though it seems more likely the original 4-cylinder engine for the Light Car was either an all-new design or basically a 4-cylinder version of the Ford Straight-6.

    Would be interested to know what 4-cylinder engine was intended to power the Ford Light Car project.

    1. I don’t know if Dollfus had intended to borrow the Taunus four or something else, but the U.S. development of the Light Car focused relatively little on four-cylinder engines. The main push was for an aluminum I-5 of about 2.5 liters, which had been a pet project of Henry Ford the Elder’s since around 1939. (It was his reaction to the introduction of the Ford six, which he hated on principle.) There were some four-cylinder prototypes of various kinds, some borrowed from Fordson tractors (about which I know very little, although I know there are tractor enthusiasts who probably have more information) and at least one a Jack & Heintz (JAHCO) flat four. I assume these engines were purely for evaluation purposes, which is not at all uncommon for development mules. I have to assume that someone considered and perhaps tried lopping a cylinder off the inline-five and making it into a 2-liter four, but Henry was very resistant to that direction, again on principle rather than for any very good reason.

      1. So the inline-5 was one of the main engines considered for the Light Car project that eventually became the Ford Vedette, being the design likely being a few decades too early for use in passenger cars?

        Do any figures / specs exist for the inline-5 beyond being around 2.5-litre with both SOHC and Flathead layouts being tested? Also was the inline-5 an all-new design or derived from another engine design?

        Never knew Henry hated the Ford Straight-6, yet surprised a Flathead V8-based V6 or V4 was not considered along with a related Slant-4 or even a V6-derived from the Lincoln V12. Then there are the proposals to update the existing / planned engines to OHV and cast in aluminum despite the potential resistance from Henry with a few of the above.

        1. The inline-five was an all-new L-head engine with an aluminum block. It was first conceived I believe around 1939 at the personal insistence of Henry Ford. A number of senior Ford people, including Edsel, had been saying for years that Ford really needed a six to offer in addition to the V-8. The smaller V8-60 engine had never been very successful in the U.S.; it was costlier to build than a six, was too small to have the kind of torque American buyers expected, and wasn’t economical enough to make its deficits worthwhile. However, Henry did NOT want to offer a six. The common story is that his first six-cylinder car, over 30 years earlier, had been a flop and that put him off sixes for life. That wouldn’t be out of character, but I think part of his rationale was that a six would be too much like Chevrolet, not unique or novel enough.

          It’s important to understand that Henry Ford, especially in the last two decades of his life, was a stubborn, obstreperous crank: short-tempered, erratic, mean, and with a penchant for irrational but intractable ideas. He would insist on certain things and strenuously resist others, often on a whim. (When the flathead V-8 was developed, he insisted for some time that it be built without any pumps whatsoever, and it took a lot of wasted time and energy before he grudgingly admitted that wasn’t going to work.) The five was one of his last whims. He worked on it himself at home and hovered over the engineers, telling them to do this or not do that. Very few of the engineers wanted to go with the five, which they thought was impractical. They struggled at length to make it run with anything like acceptable smoothness, and it was not terribly powerful (I think they got around 60 hp out of it).

          As for other derivatives, the prewar “big” flathead V-8 had a bore of 3-1/16 inches (77.8mm) and a longish 3.75-inch (95.3mm) stroke. With its bore spacing, the practical bore limit, which they went to after the war, was 3-3/16 inches (81.0mm), with a stroke of 4 inches (101.6mm). Half of the former would be 1,810cc, half of the latter would be 2,092cc. Either would have been smaller than the V8-60 (which was 2,227cc), less powerful, and probably less economical (thanks in part to the friction of the much longer stroke — the V8-60 had a stroke of only 3.2 inches/81.3mm) while being fairly expensive to build. Henry was categorically opposed to sixes and probably would have balked at a V-6 unless it was his idea. The Zephyr V-12 was not an especially good engine (it was smooth, but its oiling problems dogged it for a long time) and the practical bore limit was 2-7/8 inches (73mm), so the biggest V-6 you could have obtained without lengthening the stroke — which was already 3.75 inches — would have been 2,394cc. That was hardly a great improvement over the V8-60, which would again probably have been more economical and undoubtedly smoother. I can’t see that any of those ideas would have produced worthwhile results.

          As you’ll see, a lot of these prewar engines were absurdly undersquare by postwar standards, which would have been a limitation for postwar development. Part of the reason V-8s took off in the ’50s was that the low-friction, short-stroke OHV engines were more eager to rev and in some cases were more economical than these old undersquare designs. The latter couldn’t grow without adding even more friction and making them more stodgy in response, which ended up making them development dead ends. It wouldn’t have been so simple as just slapping new OHV or even OHC heads on them.

  21. I read somewhere that the vibration of the I-5 was obvious even when it was running on a test bench. For whatever reason the engine mounts were higher on the engine than on most engines.

    The engineers driving the I-5 test mules enjoyed pulling into a gas station, asking the attendant to check the oil, and watching him do a double-take. One of the Ford engineers would respond to the attendant’s “You have five cylinders!” with, “Don’t tell me that, I paid for six!”

    Given Henry’s early insistence that the flathead V-8 not have a water pump, it’s ironic that it ended up with two of them.

    As for mean, it was his idea of a good time to tell two employees that they had the same job and then watch them fight it out.

    The Argentine IKA Torino with the fairly undersquare Willys “Tornado” engine compiled quite a competition history, but it was the exception to the rule, even if the engine was OHC.

    1. IKA eventually made quite a bit of the Tornado engine, but the Argentine version had extensive changes to its porting and manifolds compared to the original Willys engine, which owed its undersquare dimensions to its relationship to the truly ancient Continental “Red Label” flathead six Kaiser Jeep had inherited from Kaiser-Frazier. (I assume that was also why it was essentially breathing through a drinking straw; the Tornado was intended for SUV duty, so off-idle torque was clearly a design priority.)

  22. Following MacPherson’s arrival at Ford, the first production car to feature MacPherson struts was the British-built 1950 Ford Consul and later Zephyr .

  23. When did MacPherson’s original patents expire and which notable UK/European marques were among the first adopters?

    1. The patent term issue is complicated (and would require some digging to sort out the way it worked then compared to how it works now), but I think the 1949 patent (US2660449A, for reference) would have run its course by the late sixties. I’m actually not sure if Toyota paid patent royalties to Ford for the use of MacPherson struts on Toyota products like the Corolla (q.v.), which began in the mid-sixties. By “notable UK/European marques,” I’m assuming you mean “other than Ford.” I don’t know off the top of my head which were first out of the gate (I tend to shy away from trying to argue “firsts” of anything, since it’s often arguable and there’s often some outlier example I may never even have heard of), although certainly struts were ubiquitous by the late seventies.

      1. Indeed am referring to other than Ford.

        The BMW New Class appeared in 1962 in Europe, while the Rootes Arrow appeared from 1966 in the UK although cannot say if both were the first non-Ford adopters of MacPherson struts in the UK/Europe.

        Reputedly an Imp was tested with MacPherson struts at one point which would suggest the patent for both the UK and Europe expired in the early 60s, whereas others would suggest it expired sometime in the mid to late 1960s.

        1. That’s possible (as I said, the question of how long a historical patent term actually lasted is complicated). It’s also possible that BMW et al did actually pay patent royalties or license fees to Ford for use of the design for a few years. Companies DO license outside patents all the time, so it’s not necessarily a categorical impediment (although some companies have historically preferred to avoid it wherever possible; Honda was like that for a while). If a technology is obviously superior to available alternatives, or (as may have been the case with the MacPherson strut) saves enough money to more than offset the cost of royalties or license fees, there may be a compelling business case for doing that. As with other types of royalties, much comes down to how reasonable the rightsholder is inclined to be, which can obviously vary considerably. Unfortunately, these are the kind of details that can be difficult to determine without access to the companies’ actual internal business records.

    2. The Tatra 603 of 1956 has strut front suspension. A single and stiff track control arm rather like many smaller fwd cars of today.

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments may be moderated. Submitting a comment signifies your acceptance of our Comment Policy — please read it first! You must be at least 18 to comment. PLEASE DON'T SUBMIT COPYRIGHTED CONTENT YOU AREN'T AUTHORIZED TO USE!