Question on custom suspension

In the motorcycle world, the normal rule of thumb is that the "static rider-aboard sag" should be around 30% of total travel, I see no reason why this should be different for cars, so you might want to get it so that you have 3 inches of compression and 1.5 inches of droop from the nominal ride height (with normal load on board). The consequences of running out of travel on the compression side are a lot worse than of running out of travel on the rebound (droop) side.

I don't think your instant-center-below-ground is a disaster, a lot of OEM front suspensions do that (particularly MacPherson designs), but can you do any of the following and does any of it help:
- Raise the upper ball joint a bit (longer knuckle)
- Shorten the upper A-arm a bit (move chassis-side pivot a little outboard)
- Drop the chassis-side upper A-arm mount a bit (I'd rather raise the upper ball joint, but if you can't ...)

Thanks for the response, Brian. This is for a GT40 replica. The original travel for the front suspension was around 3.5" total- much worse. This is on a ~2200lb midengined car. I physically can't go any more bump or the wheels interfere with the body. (Would actually go through the hood.)

I've attached a few pictures of the rough design. I'm using wilwood pro spindles, (modified mustang II geometry,) chrysler large ball joints, and del-spheres on the inboard end of the arms. Because the upper A arm has to be so much shorter due to the chassis, it is very difficult for me to do anything to the roll center and maintain a reasonable camber curve.

If you're splitting the suspension travel equally between bump and rebound and designing around whatever that gives you, shouldn't that be taken with the driver on board? Or if there's a fair chance of carrying a passenger (or track day instructor or student), perhaps driver weight plus half a passenger?

Since roll center heights and available rebound travel shouldn't matter a whole lot when nobody is in the car driving it, it seems that all you'd need do would be to raise the empty static ride height up by enough to achieve your equal bump/rebound target when you (they) get in.


Norm

Hi Norm- I started with a desired static ride height, track width, and rim size and offset, put the known spindle and ball joint geometry in, and then designed the A-arms and mount points from there. I had to actually modify the frame to get my 2.25" bump clearance through the steering sweep with a 10" wide tire. I purchased coilovers already, and my intended positioning and coilover travel happens to give me roughly equal bump/droop- I didn't go at the design aiming specifically for a 50/50 split. I have not yet calculated spring rates- that will come later, when I have final weights at the corners of the car.

Unless you design 100% anti-dive in the front suspension and 100% anti-lift in the rear suspension you will always find the roll center going down at the front and coming up on the rear under braking, which will balance your lateral load transfer distribution towards the rear of the car, making the inherent tendency for oversteer under braking (due to the load transfer) worse. Whether in "pure" vertical full bump conditions the "lateral" handling of the car will worsen cannot be said, since it depends on what the rear is doing. If for instance your front & rear roll center would drop with a ratio of 1:1 to the wheel travel and you would have equal wheel travels front and rear on full bump the lateral load transfer distribution would be unaffected. If the ratio's are different balance would change. The same is of course valid for different size/rate bumpstops. A negative roll center in full bump is not necessarily a negative thing, for instance a McPherson strut changes roll center height to wheel travel typically with a ratio of around 3. Starting out from a typical static RCH value of 50mm this means that at 50mm compression travel the rollcenter is at -100mm and yet it seems to be fine on many racecars. As always, it is in the balance of things.

Cheers,

dynatune,