Hi all

Another question about a much-discussed subject: roll centre height.

A lot has been said, in books and forum discussions, about keeping the RCH as constant as possible in roll and bump-droop.
But constant to what, to the ground or the chassis?

RCH is normally measured from the ground, but one would expect that by "keeping it constant" they mean: letting the RC move along with the chassis to keep the roll couple constant? Then it would be better defined by "letting the RCH change as much as the chassis height.

Or can anyone explain why the RCH should indeed remain at constant height to the ground, no matter how the chassis moves?

Sorry if the question is confusing... but I have another one:
What is so bad about a horizontally migrating RC in tranverse direction? I can imagine that it causes lifting or jacking effects, but is this really so and/or are there other reasons?

Cheers
Giel

You can think of the roll center as an axis, about which the center of mass moves as forces are applied to it. There are geometric roll centers (derived from drawings as per most suspension books), and force-based roll centers, determined by measuring forces at the wheels (big money approach). Both are meant to get at the same thing, i.e., defining the axis of roll movement.

The distance between the roll center and the center of mass is called a "moment arm", which works like a lever. The longer the lever, the more it will move when force is applied, hence more body roll. The goal is to maintain a constant distance between center of mass and roll center, so the answer to your question is to maintain constant height with respect to the chassis, not the ground.

As for the lateral position of the RC, Carroll Smith explains it as being a feel issue, that handling is more predictable with the RC fixed. I ran across a very insightful article about RC migration, and it made me realize that when geometric roll centers are kept fixed laterally, then force-based roll centers will be at about the same height as geometric roll centers. This explained what Smith perceived in practice. Here's a link to the article:

http://www.auto-ware.com/ubbthreads/showflat.php?Cat=&Number=489&page=0&view=collapsed&sb=5&o=&fpart=1

Damned interesting stuff, I think.

Pete

damned interesting indeed, Pete. I never read an article before that provided such a detailed insight in the theory of roll moments!

Thanks a lot Pete, you've been of great help.

Giel