IFS theory: most predictable rather than "fastest"?

[mikedc]

Anybody have any thoughts about front suspension geometry and roll centers in regards to predictability gains?

Of course the roll-center needs to move as little as possible. But what about past that?


There have been a lot of trees cut down over the issue of CG versus roll center. Conventional thinking seems to indicate that the RC shouldn't be too far below the CG so that the chassis weight has less leverage on the front suspension (in terms of body lean). Some say the RC should be like 3-4 inches off the ground at ride height.


But the longer I think about it, the more it seems like you'd WANT the RC notably lower than the CG, at least if body lean wasn't too big of an issue. My reasoning thinks you'd probably be able to feel/predict what the front end is doing a lot better that way.

So if the CG, sway bars, and track width were already keeping the body lean under control all the way up to the limits of adhesion, then wouldn't it be better to lower the RC a bit more than conventional wisdom indicates?

[Norm Peterson]

The RC is something that you have to balance against all the other things.

Too low, and combined with a tight limit on how much steady-state roll is acceptable is likely to result in ride quality that's not only too firm for comfort, but might also be too firm for optimum overall grip on pavement that's less than smooth.

Too high and you have to make relatively larger changes in spring/sta-bar/shock tuning to accomplish a given change. Those things work only with the Lateral Load Transfer that's resisted through them, and any LLT that goes through the RCs does not. Worst case, it may be impossible to achieve the desired Total LLT Distribution if the RCs are both too high and one is excessively so.

Too high of a front RC tends to be associated with short front view swing arm lengths and rapid camber changes with suspension movement. Too much camber gain is bad for the hard braking case, unless you crutch that problem by limiting front suspension movement with stiff front springs and/or excessive anti-dive. Suspension jacking can occur during cornering, if the RC is high relative to the track width.

You can affect transient handling one way with RCHs, and allow the roll as it develops to change the handling the other way. For example, a nose-heavy vehicle may benefit from some initial "looseness" due to a relatively low front RC and/or a relatively high rear RC to get the rotation started that tightens up as the (usually) much larger front suspension roll stiffness takes over. This gets into front RCH vs rear RCH, which you need to keep in mind anyway.

Actually, there's another approach to looking at this that's probably better as far as understanding chassis roll vs actual suspension movement is concerned (hint: roll does not occur about the 'roll center' except by coincidence, and suspension movement during roll isn't necessarily symmetrical).


Norm

[mikedc]

Thanks for the reply.


Some more background is probably in order here:

I'm trying to come up with an UBER-ideal front geometry for a street cruiser musclecar. Not a track car, not chasing any particular number on a skidpad or stopwatch. Just very, very fun to drive. Basically a tall & floppy suspension with a heavy car and small tires. (At least compared to most handling-oriented modified cars if maybe not compared to a real stock 1960s car.)


I'm thinking that there is only so much spring rate & sway bar I wanna throw at the front end, whether it turns out to truly control all the body roll or not.

The front end mounting point of the spring pressure will be pretty far inwards compared to a lot of cars, so each individual front spring is gonna have less leverage on the body. (Obviously that calls for a stiffer sway bar than otherwise, to restore the body roll control somewhat.)


Given all that, it seems like this is what I want:

-- VERY long control arms

-- moderate-to-aggressive camber gain (with control arms so long, the camber gain would end up being functionally less than normal. Longer arms = less angular movement for a given amount of travel at the wheel.)

-- a pretty low and stable RC



Do ya think all that makes some sense?

[Norm Peterson]

Do you have some target ride frequencies in mind? Might as well start there. Typical cruise speed too, so that the "flat ride" stuff can be looked at.

You don't want to make it TOO soft. A ride that gets some people a little queasy isn't any better than one that's too stiff.

For a given wheel rate, which is what the ride frequencies will set, ride quality actually improves as the spring is moved out toward the ball joint, provided that it can be softened as it moves outward. 100 lb/in at the balljoint feels "softer" than 400 lb/in at the midpoint of the LCA even though the wheel rates and frequencies would be the same. The short version is that the forces input to the chassis at the upper spring seat and LCA chassis side pivot are lower, and generate less NVH because they aren't hitting the chassis quite as hard. If it helps the visualization, picture banging on a flat steel plate first with a tack hammer and then with a 32-oz framing hammer.


Norm