trailing arm roll center

Regarding an early VW style trailing arm front susepension:

Where does the roll center go when the body rolls?

Does it remain at ground level?

If the roll center changes height during roll, it would seem an important factor in lateral load transfer.

If this has been addressed or answered here in the past, I have not found it . . .

Thanks

 

[GregLocock]

I don't know, but I'd be very careful about assuming it starts at zero in a real corner. The reason is that compliance of the arms, body and pivots will make a significant difference to the instantaneous centre of rotation.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

Greg,

I first need to know how the roll center is found, and would like an explanation that even I could understand.

I am assuming that during roll, the instant center rises for one wheel, and drops for the other wheel. If tire thrusts remain equal, I guess the roll center would remain where it was.

But the outside wheel will generate more lateral thrust in a corner, so I'd think the roll center would be more strongly influenced by the outer wheel.

If that is the case, body roll raises the instant center for the outside wheel . . . right? . . . so will the roll center rise with body roll?

I'll appreciate any help offered.

Thanks, roofus

 

[GregLocock]

Draw a free body diagram and figure out at what height lateral forces at the cp act on the body.

or draw the arc of the wheel's motion and find the ic. Connect the ic back to the cp and the rch is at the intersectoin with the centreline

A few seconds thought will reveal that in both cases the predicted rch is 0.


Neither of the methods above necessarily results in the same RCH for both wheels.


Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

Greg,

In the case of the trailing link, VW style suspension, the wheel travels vertically (instead of an arc)relative to the car. The instant center will be at an infinite distance, right?

With the frame level, the RC will be at ground level, but when the frame rolls in a corner, the RC will be above ground for one wheel, below ground for the other.

I'll guess the car won't rip itself apart by rolling about both RC's. So, will the roll center wander towards the RC described by the wheel producing the most lateral thrust?

Everything I've run across, diagrams only a level car, and states the RC is at ground level. It seems to my highly untrained mind that it does not remain there when cornering, and I am seeking information to confirm or deny my thoughts.

I think I'd like to understand your first sentence, so I'll hit the library and make an effort.

Thanks for your replies . . . roofus

 

[bikesrfast]

To my knowledge and very old memory, the RC is at the center in the middle of the 2 torsion beams and it generally stays there during roll as the inside and outside do equally opposites and there isn't camber change thru wheel travel. I guess with an anti roll bar the practicle roll center must migrate some towards the direction of the contact patch of the outside tyre.

Please feel free to point me to diagram that says different.

When I was a young 'un, I wanted to take a ball joint front end, cut the middle of the top tube and angle the centers down to the bottom tube ('V' shape). Make new top arms to suit. This would have given me camber gain thru roll and a lower roll center but with the advantages of the trailing arms. I never did.

 

[bikesrfast]

http://i88.photobucket.com/albums/k180/DawieCoetzee/RTLS02.jpg

Found this, may help you, didnt help me because I dont agree that the front end has the RC there.

But I've been wrong before, lol!

 

[bikesrfast]

http://www.lawrence-tune.co.uk/lawrence_link_2.html

and then I found this!

 

[GregLocock]

The car does not roll about the roll centre, in theory, or in practice.

The roll centre height is defined by the SAE in terms of the forces into the body (the second method I suggested). As such it has no particular relationship to the roll axis of the body.

The Lawrence is interesting. By moving the ic inboard from infinity, but still keeping it at ground level, you do get a nicely decoupled suspension. Bear in mind that modern tires have totally different requirements for camber than racing crossplies from the sixties.

I didn't really understand the sketch. A picture may be worth a thousand words, but a bit of context might help!





Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

I'd say for the RC to stay on centerline requires equal suspension travel on each side, which could easily not be the case. If the inside wheel hit a droop stop, it seems to me the inside contact patch becomes the RC. At that time the inside suspension will no longer extend, but with additional lateral acceleration, it is possible for the outer supension to further compress.


The RC is a product of lateral acceleration, right? Lateral acceleration will transfer vertical load to the outside wheel, and the outside wheel will produce more thrust than the inside wheel. It just seems to me at such times the outer wheel has more influence. Rather than splitting the difference between the predicted left and right side roll centers, I'd think the RC would rise above ground level with roll.

roofus

 

[patprimmer]

Not my area of expertise, but as I understand it, the roll centre is not a product of lateral acceleration. It is the point where the chassis applies the lateral acceleration.

G forces apply through the centre of gravity and lateral forces from the chassis apply through the roll centre. The relative difference in height produces roll or lean in on a corner.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[murpia]

"G forces apply through the centre of gravity and lateral forces from the chassis apply through the roll centre. The relative difference in height produces roll or lean in on a corner."

More realistically, lateral forces apply through the four suspension instant centres in proportion to the tyre grip (for independent suspension front and rear).

Regards, Ian

 

[hemipanter]

Maybe this is a bit of topic, so the webbmaster may do something about it.

However, the roll centre issue has been up to discussion many times and it appear to me that the term is not "upp to date". The word "roll centre" tells that there is a centre about which the car turn its rolling action. And after older theories this centre is a product of the forceline intersection.

Now, the forceline intersection is not a hinger function, the geometry allow the chassis to roll about a wider area of spots. So, if we put 100p one side of the car that side will go down, and if we lift of 100p the oposit side this side will rise, and create a roll motion having its centre in the centre of the car.

Then we have different loading during cornering each side.
Each side is also having its own forceline angle as a function of the tilt of the chassis. The % of side loading times the forceline angle is creating a horizontal lift or lowering of the of the chassis. So, we usually have one roll, one lift, and one lowering situation of the chassis, together creating a unic curve of motion.

We must subtract the force creating "jacking" over the forcelines from the total weight transfer to get whats left for roll.

This is my way of looking at Rc.

Regards
Goran

 

[GregLocock]

FWIW my way of looking at it is to forget the name 'roll centre height' and just call it RCH. That way there is a lot less baggage associated with it.

The location of the roll axis for the sprung body in a given manouevre is not stationary in any coordinate system I can think of (ie body centric or suspension centric or world centric) for most suspensions. It would be interesting to create a suspension with a fixed roll axis, relative to the body at least. Not very hard to atleast approximate it with live axles.

Worse than that, isn't the roll axis an instantaneous axis of rotation in reality?







Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.