Roll Centre Migration Coordinate System

[Mercator]

I have read a recent debate in the magazine Racetech, about the relative benefits of measuring the migration of the roll centre about a coordinate system attached to the ground plane versus a coordinate system attached to the sprung mass of a car, when designing a suspension system.

In the former, it is suggested that the restriction of the roll centre migration relative to the ground results in a jacking force that varies very little. In the latter, it is suggested that the fixing of the roll centre relative to the sprung mass will result in a roll moment that will vary very little.

There was some debate as to which variance would have the largest effect on the handling of the vehicle and I would like to know if anyone has any input regarding either of the two standpoints. I have always used a ground plane based method in my past projects but this debate has caused me to wonder which variance is the most important when it comes to vertical wheel loads - the jacking force or the roll moment. I am particularly interested in the transient part of cornering.

Any input would be welcome, as I am at the beginning of a new project and have reached a design philosophy crossroads, so to speak

Cheers,

Dan

 

[crseng]

I am a bit thrown back by your post. The coordinate system taken in any problem should not alter the outcome of the problem. This piont is simply a randon piont of referance. I am not sure if in this debate they are comparing the assumption that the CG does not move in one ref and traking the CG movement in the other ref. ??

To your other questions the jacking force or the moment force significants will vary depeding on your overall suspension layout and your tire coeff.

Generally speaking I use the assumption that roll moment will contribute to most of the vertical loads on the tires. And jacking forces should be used to "fine" tune the system.. This works fine in most applications I deal with.

Remember that jacking loads are very dependent on on your tire coeff through time. At low speeds assuming that both tires hold the same lateral turning force the jacking force is zero. Because all the upward jacking force produced by the outside tire load is matched by the inside tire loads downward jacking force. On the other spectrum, at the piont were the inside tire has lost all traction the jacking force is max.

In transition you will have to keep track of roll center, CG and tire patch movement in check in small steps in order to really compare jacking and moment effects. It proves to be a lot of fun. lol..

Good luck Dan

chris

 

[Mercator]

Thanks for the tips Chris. This was just the sort of thing I was looking for.

Sorry if my post was not entirely clear. I was trying to write it quickly and in a different manner to a previous post on the subject, which was removed (although I now know that it was an accidental removal).

What I really meant to ask was: is it more desirable to constrain the movement of the roll centre with respect to the ground or with respect to the sprung mass CoG? I think that you picked this out despite the confusing nature of my post, so thanks very much

Cheers,


Dan

 

[gielhansen]

It seems to me that the roll moment has a much bigger influence on the body roll than the jacking force, since the jacking force is only (a small) one of the different forces that together balance out the roll moment.

in that sense, keeping the RC constant to the chassis will give the most consistant handling I think.

 

[Mercator]

Well, this is encouraging. I had been leaning towards keeping the roll centre migration to a minimum with respect to the sprung mass CoG since I first read about it. I often wonder why this did not occur to me before

Now it seems that we have a 2-0 score in favour of this. If anyone has an argument for the alternative or any further input, even if it is just agreeing with what has already been said, I would appreciate it very much.

Thanks,


Dan

 

[GregLocock]

All the graphs we produce are relative to the body coordinate system, so I guess that's the way we think. This makes sense when you consider that whatever theory of roll axes you are using is discussing how to use these forces to force the body to move.

Cheers

Greg Locock

 

[Mercator]

Well, I think that more or less settles it. 3-0 to the body coordinate system. Thanks everyone. Any further input is more than welcome, though.