Originally Posted by
ShaunMac
Lowering the inner camber link point will raise roll center. "Less roll"
Raising the inner camber link point will lower roll center. "More roll"
I think any handling changes you feel from moving the upper camber link around are more associated with "Camber Gain" and not roll center. To really make roll center changes you can feel, you need to raise or lower the arm pivot point on the chassis.
The direction the roll center moves, may actually reverse from this depending on the angle of the lower arm as well. On my 1/8 buggy front suspension, lowering the front inner pivot lowered the front roll center. Also as was mentioned, the roll center does move around quite a bit, especially with body roll. In full size cars, a guy who builds some of the fastest autocross cars told me that when he mods street cars, his goal is to make the roll center stay between the tires in roll. On most strut setups, the roll center moves well outside the car when it leans and this makes tuning very hit or miss.
In my case, I needed more front grip, so I went with lowering the inner front arm mounts. This increased the roll torque as the CG is above the roll center. This makes the outside rear tire have to work harder to resist the increase in roll torque, and the inside front tire gets a little easier time and can help more with cornering. The slight increase in camber gain on the front may have also helped with a tick more front grip, but the camber gain from the caster angle is far greater, so this change was trivial in my case.
If the roll centers are too close to the CG (sounds like a good idea to reduce lean) what you end up with is so little roll torque that you can't tune front rear balance with springs or sway bars, they need roll torque to function at all. But even if you have the roll axis lined up perfectly with the CG axis so there is zero roll torque in the suspension, the vehicle weight will still transfer to the outside tires and even traction roll the car if you have enough grip. The roll torque on the entire car is still the CG above the outer tire contact patches.
As an additional experiment, I raised my rear roll center a fair bit, and the car became very loose. The close roll center to GC cause all the cornering force on the back of the car to be resisted just by the outside rear tire without producing roll torque in the chassis, so the inside front got a great bite as the inside rear was basically floating off the surface. I had adjusted static camber to produice very close to the same camber while cornering. Adding some negative camber to the back did not help much, as it then was riding on just the extreme inside edeg of the inner rear tire. In the end I put the rear camber link back to stock, but if I ever need to make it loose, I have a good tuning tool there. There are a couple choices that should yeild a mor subtle change. I like trying a bigger change first off to make sure I see what it is doing, but I also realise, doing too big of a change can overshoot the ideal setting and give reverse results.
Roll center and camber gain has to be treated together. When I did my roll center change, I looked very carefully on how it made the camber change. My upper arm position coices are quite limited, and I had to make the front upper arm longer as I lowered the inner pivot. This reduced the change to the camber gain curve. I also have 3 caster choices. I have it at max now, upper arm as far back as possible. The resulting camber when I am sliding at full lock is very close to zero. My worn tires are nearly perfectly flat across their full width. The balance is now very good with just a little push to make it easier to hold on a delicat line.