Ride Height and Roll Center

[David Inc.]

There's a general rule that has come up (wisely), that it is a less than wonderful idea to lower a 911 below the point where the front control arms go horizontal at rest as any lower and the roll center really starts to drop from that point and the handling characteristics start to get ugly. Any drop below that point basically requires spindle modifications as well as rear trailing arm pickup points to avoid that ugly handling.

Bill Verburg has very helpfully posted the below in many posts, but I was wondering if anyone had a ride height X for the front suspension (vertical distance from torsion bar center to wheel center), at which point spindle adjustments need to be considered.

I know this information is out there buried in a number of threads, but I thought it would be nice to plunk it down in its own thread and call it a day.

I want to point out my own selfish interests here, as my weekend/track car has definitely proceeded a little bit below horizontal control arms and I have to say the handling impact is noticeable. Even with 22/28 torsion bars it still feels to roll nearly as badly or worse than it did with stock bars at Euro height. This is something to pay attention to if you want your car to drive well!

Even better, if someone has a chart of ride height v. static roll center that would be the coolest.

Originally Posted by Bill Verburg

Only US 78-81 SC were set at the factory at the raised height
here's a survey of the specs


height is the difference between a & b

[Bill Verburg]

Originally Posted by David Inc.

There's a general rule that has come up (wisely), that it is a less than wonderful idea to lower a 911 below the point where the front control arms go horizontal at rest as any lower and the roll center really starts to drop from that point and the handling characteristics start to get ugly. Any drop below that point basically requires spindle modifications as well as rear trailing arm pickup points to avoid that ugly handling.

Bill Verburg has very helpfully posted the below in many posts, but I was wondering if anyone had a ride height X for the front suspension (vertical distance from torsion bar center to wheel center), at which point spindle adjustments need to be considered.

I know this information is out there buried in a number of threads, but I thought it would be nice to plunk it down in its own thread and call it a day.

I want to point out my own selfish interests here, as my weekend/track car has definitely proceeded a little bit below horizontal control arms and I have to say the handling impact is noticeable. Even with 22/28 torsion bars it still feels to roll nearly as badly or worse than it did with stock bars at Euro height. This is something to pay attention to if you want your car to drive well!

Even better, if someone has a chart of ride height v. static roll center that would be the coolest.

As the kinematic roll center drops the length of the lever arm causing roll increases, leading to(surprise) more roll

I'm sure the exact measures are out there somewhere but I don't have them,


Fortunately, it's fairly easy to estimate where the roll center is from knowing where the instant centers are as these define where the roll center is

here A1 and A2 are the lever arm lengths


to find the instant centers just follow the A-arm extended back into the chassis

you will find the roll center at the intersection of the vertical dropped through the CoM axis and the line from the IC to the middle of the tire /wheel, The roll center moves around a lot when the car is cornering, so just get the static case right and all else will follow

details aside, the more down angle on the A-arm the higher the roll center(this is good)

so rule of thumb, if the A-arm static angle is approaching or heaven forbid beyond down, then the spindles need to be raised

now this alone doesn't cure bump steer, as bump aka roll steer is caused by the difference in arc described by the ball joint and the steering knuckle as the move up and down

The difference in arcs causes the toe to change w/compression/rebound which is the steering effect

[David Inc.]

Thanks as always for your input on this. I went looking some more and found this: https://www.racingaspirations.com/apps/macpherson-geometry-calculator/

This can be used to roughly mimic the dimensions of the 911 front end and calculate the roll centers. I think I saw you reference this calculator in a post some years back which was what got me looking for it. I saw from playing around with it and guesstimating dimensions that the delta for roll center to ride height was not 1:1 nor linear. Seemed like it was a 2.5:1 relationship that got worse with decreasing ride height (not linear but close), such that a 20mm ride height drop lead to a 50mm decrease in roll center, so body roll increased way faster than the camber curve even accounting for the lower center of gravity.

I'd need to go out and physically measure the car to try again for real numbers, but it's pretty shocking in its severity.

At least with bump steer some rack spacers help as the ride lowers, but looking at this it's no wonder my car still rolls so much even on 22/28 t-bars. I think I'm going to end up coming back up some as a result until I have time to re-do my suspension and raise the spindles (my wife and I are having our first baby in April so now is not the time to tell my wife I'm starting a new project!).

[Otter74]

Great link - thanks for sharing it.

[PeteKz]

Bill, as always, thanks for sharing your knowledge. I haven't thought much about MacPherson strut dynamics because very few of the other cars I've owned had that suspension, so thanks for explaining it again.

I first saw suspension diagrams like that in a book I read about 40 years ago by Fred Puhn, "How to Make your Car Handle," published by HP Books. It's still in print. I highly recommend it. And it even has a picture of a racing Porsche on the front cover! Gotta add that to your library.

https://www.amazon.com/How-Make-Your-Car-Handle/dp/0912656468/ref=sr_1_1?crid=129M3639HE6JQ&keywords=how+to+make+your+car+handle+by+fred+puhn&qid=1672871054&sprefix=how+to+make+your+car+handle%2Caps%2C157&sr=8-1

Something that appears to me from studying that diagram again: If you increase the negative camber (ride height remaining constant), it also raises the roll center a bit, and reduces the lever arm which reduces the roll couple. Maybe that's part of why some people like more negative camber.

[Bill Verburg]

Originally Posted by PeteKz

Bill, as always, thanks for sharing your knowledge. I haven't thought much about MacPherson strut dynamics because very few of the other cars I've owned had that suspension, so thanks for explaining it again.

.......

Something that appears to me from studying that diagram again: If you increase the negative camber (ride height remaining constant), it also raises the roll center a bit, and reduces the lever arm which reduces the roll couple. Maybe that's part of why some people like more negative camber.

Only to the extent that the wheel center line moves outboard, same w/ a lower ET wheel or wider track as w/ spacers. This will cause only a very small change in roll center height. This is another reason why siff springs are desirable for a performance car, as a major cause of roll center height change is from the compression of the loaded side which changes the geometry there adversely.

The largest and most easily changed variable will be the A-arm angle, which can only be changed(easily) by raising the spindles or changing ide height.

A problem w/ really short tires is the extra visual room between the tire and lip when the car is raised enough to affect the a arm angle, one reason why people like a little bit taller tire, but a taller tire has its own issues mainly, torque cost

[stownsen914]

Strut suspensions can be tough because they are a performance compromise out of the box. It only gets worse the farther you take it from the normal range of motion. If it were just roll you needed to reduce after lowering the car, you could crank the sway bar tighter. Unfortunately it’s not that simple.

When you lower the front of a 911 too far, in addition to the roll center problem, you also get:
- camber gain (additional negative camber) in bump goes to essentially zero
- scrub (side to side movement of the tire as it travels up and down) goes up substantially
- the above also causes more bump steer. Correctable to some extent..

Basically it’s all bad. On a track car, it may outweigh the benefits of lowering the car.

Another thing - roll centers being too high or low (debatable what constitutes ideal), even aside from the effects mentioned above, can cause ill or vague handling.

As for measurements to find the ideal height, don’t forget that they may be dependent on wheel/tire diameter, depending on where you are measuring.

One way to keep suspension closer to its intended range of motion is to run 15” rubber, but that’s not as cool to some 🙂.

[David Inc.]

Yeah it's a rough situation for sure. I think I was just shocked to see at how quickly the body roll issue is exacerbated by lowering the ride height. I love the lowered look but it's just not worth it if the car is going to be tracked.

[PeteKz]

With the flat-6 engine, the CG on these cars is very low as-is. I will not lower mine below "Euro" specs for the street, just because of hitting speed bumps, parking blocks, etc. At some point the "street racer" look became dated and goofy. With all the Japanese "Tuner cars" out there now, I think we passed that point some years ago. It looks better at OE Euro ride height. (now dodging the rotten tomatoes..)