[ChrisL]

My rear suspension findings.

I think my approach is slightly different to Ross’s so the figures should also be different in the final wash. My starting point however was to listen very closely to everything people had to say (Ross and Will have put in some serious thought into their solution) but then try and work with what I have. As well as listening to others, I’ve also done some general researched on the rear dynamics of existing production mid-engined cars and some open wheelers. Interestingly, I found quite a lot of info on the Toyota MR2 which is a car I’ve previously owned so I’m familiar with its characteristics. A common thread though that seems to exist amongst all mid-engined cars in that they all seem to run about 0.5deg static toe at the back. The Honda NSX however to name one, has a geometry that will near enough hold that value through the whole of the suspension travel range. The MR2 on the other hand, dynamically increases the toe-in under compression and consequently reduced the toe-in under rebound. This results in the car being very stable through the corner (to the point of understeer) but very flakey under hard braking, swapping ends quite readily.

Given however that I am stuck with a geometry on the GT40 that has the opposite dynamics to the MR2 (decreases toe-in under compression and increases toe-in under rebound) I decided to work with it instead of against it.

My approach was to firstly minimize the amount of dynamic toe. Using Ross’s findings I adjusted rear castor as much as the standard suspension would allow. Secondly, I removed the shock absorbers and raised the suspension to what would be full compression and adjusted for 0deg toe. When I then lowered the suspension to the static position, I was ecstatic to find a toe-in value of 0.6deg. I then adjusted for 0deg static caster and reassembled the car. This setup ensured that I would never get toe-out under any circumstances so that made me feel I had a good starting point.

The remaining dynamic toe however did not necesarily concern me because I figured that, ideally, a car with this type of dynamic toe should exhibit the following rear-end characteristics…

Brake-and-turn.

Will be extremely stable under hard braking since the rear wheels would toe in as the rear lifted. This would allow the driver to brake very late while approaching a corner and continue braking well into that corner (and hopefully even promote a little understeer while braking). At mid corner, the driver would complete the braking maneuver and begin to accelerate out of the corner. If the car is traveling fast enough (and the transaxle has an LSD) then the tendency would be for the car to push (understeer) further. However, at this point, the car would squat, the rear suspension would compress, so reducing the amount of toe-in to the loaded wheel. The harder the car accelerates the less toe-in results and the more the car will turn out of the corner. Balance and steer would then be achievable through the throttle.

High speed through a sweeper.

As the car approaches the corner in a straight line, both rear wheels are evenly loaded so each toe-in value will negate the other. As the car begins to move through the corner, the inner wheel will begin to unload and so will contribute less in terms of rear-wheel steer. Conversely the outer wheel will contribute more toward rear-wheel steer as it begins to load up, so would cause an imbalance between left and right. This however is negated by the fact that the loaded wheel (the outer) is now reducing its toe-in under compression so balance is maintained. This effect is maintained even if the road is irregular (bumpy). When the loaded wheel encounters a bump, the forces between road and tyre increase and the friction values therefore also attempt to increase. If there was no further change, then the loaded wheel (given that it is toeing in) will either tend to kick the car inwards if the friction max was not met, or kick outward if the friction levels were exceeded. However, since the suspension dynamically reduces toe under compression, the effect of the increased forces are also negated, so balance is maintained. Note that this bump effect is also valid while traveling in a straight line.

Oversteer stability.

Simply put, if a driver has pushed the car beyond its limits into a corner to the point at which the back loses traction, in theory the suspension will not compress much further as by definition, the rear end has reached its max lateral force. If the driver could at that point raise the rear of the car, either by lifting off the throttle or in some cases even applying the brake, then the loaded wheel (the outer) will toe-in further and oppose the rear-end slide. Backing off during oversteer is a natural response for an inexperienced driver, so this characteristic should be good.


Well that was the theory. The car was on the track at Sandown with this setup a few weeks ago but unfortunately I developed rear brake problems after a few short laps so my testing was minimal. The signs however were all good. The car was very stable under hard braking and at the long sweeper at the end of the back straight. I pushed the car into oversteer through the esses and managed to recover cleanly by backing off the throttle. The scrub on the rear tyres also seemed to be very even. I didn’t have my temp gauge with me though so I can’t confirm that the tyres were actually loading evenly.

Note that all my measurements are approximate since they were all done with strings and rulers so as I said previously, I’ll confirm my findings and get accurate reading at a later stage.