Suggestions or advice please.

[willeng]

If anyone would like to comment or has any suggestions on the following figures please do so. Please throw bricks if you have to!

Double unequal A arm front suspension-road rally & circuit use.

The suspension has 75mm bump & 50mm droop- static camber
(-2deg).

Camber (-.7deg) at 50mm droop & (-6.05deg) at 75mm bump.
bump.
Outer wheel is (-2deg) static & (-.012deg) with 4deg roll.
Inner wheel is (-2deg) static & (-4.79deg) with 4deg roll.

Static caster is (6deg) & is (5.74deg)at 50mm droop & (6.41deg) at 75mm bump.
Outer wheel is (+.18deg) with 4deg roll.
Inner wheel is (-.17deg) with 4deg roll.

Roll centre static height 63.59mm from ground.
Maximum roll centre lateral offset is a total of 1.98mm through 4deg roll.
Roll centre height change is 3.51mm through 4deg roll.
Roll centre height change for full travel is 7.69mm for chassis & 117mm for ground.

Also i am new to these suspension programs so could anyone explain to me what is the roll centre height(Chassis) measurement?. The program has roll centre height Chassis & Ground for bump & droop, i know what ground is but Chassis?

Also they have a Tramp column, i know what Tramp is with beam axles etc but i am unsure what tramp is when associated with an unequal length double A arm setup, could somebody put some light on this for me?

Thanks

 

[GregLocock]

I'm guessing that roll centre height(Chassis) =roll centre height (ground) +suspension travel, ie it is the RCH relative to the centre of gravity of the sprung body.

There is a school of thought that says that if the RCH to CG relationship stays constant then the handling will be more linear. There is another school of thought that says if the RCH above the contact patch stays constant then the handling will be more linear.

Take your pick.

It's a bit hard to comment on your proposed suspension without knowing the numbers for the rear.

I don't know what they mean by tramp - which program are you using?



Cheers

Greg Locock

 

[willeng]

Greg:

The program is SusProg 3D.

I get what your saying with not knowing until i have the rear numbers. I'm onto that now, basically i was concerned if i can't get the rear roll centre as stable as the front & the effects of having a roll centre on the front that stays laterally centralised & the rear moves.
If i have a rear roll centre that moves laterally should i follow this with the front roll centre.

Or is there a reason why the roll centre should move laterally instead of being centrally stable?
If so what amount of lateral movement is acceptable?

 

[GregLocock]

Roll centre lateral motion is, in my very opinionated opinion, not very important, so long as it is less than 300 mm or so.

Reason I say that is that we are looking at where HORIZONTAL forces are applied to the body - and so far as load transfer is concerned the body doesn't much care where that is laterally, just vertically.

However there is a school of thought that says ideally your roll centre should move out towards the outside wheel, and down, so that as the car is on the brink of overturning the roll centre is coincident with the outer contact patch.

I think I can see their logic, but think they are over emphasising the importance of roll centre location, that is, in order to do that you may have to make some unpleasant compromises elsewhere.

I'd certainly be keen to try and keep the front and rear roll centres, for all practical extents of roll and bounce :

(a) above the ground, or below the ground (individually)

and

(b) with one of them higher than the other at all times

and

(c) preferably, but not necessarily, front lower than rear

- there are some good cars for which (c) is not true.

Probably to be safe they should move the same way laterally, but I don't really subscribe to that.




Cheers

Greg Locock

 

[willeng]

Thanks Greg:

I can see what your saying, ok if i can get the rear roll centre vertically stable & it does move laterally within the specified limit & the front remains the same as above, can you see any unforseen problems that i may be overlooking? Besides stupidity that is!

quote:- there are some good cars for which (c) is not true.
Now, this has been a point of interest for a while, for a front engined, rear wheel drive car, are there any good arguments as to which way the roll axis actually works best?

Also some thoughts on camber gain with an IRS with the proposed front suspension would be good also. Working with live axles has me at a disadvantage with IRS i'm afraid!

Thoughts on what amount & direction in bump of the IRS would help also!

Thanks again!

 

[GregLocock]

Generally the camber gain at the rear should be a little more than at the front. This will tend to give you a little bit more understeer, in roll. You could do the same with toe gain, or even compliance steer.





Cheers

Greg Locock

 

[willeng]

Thanks again Greg,

I have done a search & have read with interest your opinions about roll centres. Just as a matter of interest, if roll centres as you suggest are not that important, why then does the roll axis seem to have an effect on the handling of the cars & also more importantly the roll couple front to rear. The attitude of the car can change dramatically if the roll couple is changed.
I'm not try to be a smart a*se rather than trying to get a clearer picture of things myself.

Thanks

 

[red300zx99x]

Yeh Greg, what the deal. Force based roll centers over geometric roll centers right. Still waiting for a better description of such, better yet you working one out for everyone to enjoy. Def not being a smart ass here, just wanna know some things. Good day mate

 

[GregLocock]

Well, my position on geometric roll centres has changed a bit. Not much, but a bit. They /do/ tell you something useful about weight transfer, particularly at the inititiation of the turn.

But, a force based roll centre seems to be trying to tell you the same thing, and I have read a report that demonstrated that for a given suspension geometry the FBRC and the GRC moved in DIFFERENT directions during a maneuver, and in response to hardpoint changes.

So, it seems to me that since FBRC definitely measures something useful, it must be GRC that is in error, or, more accurately, people assign a meaning to the GRC that does not exist in reality, and that perhaps GRC and FBRC are not simply different facets of the same thing.

In practice I go along with conventional wisdom - I am working on conventional cars, I don't need to invent new theories to make them work, so I give them the GRCs they want, and do the stuff I want a different way.

I suppose what it really comes down to is that if you have a full vehicle model and put it into a turn the load transfer in both directions (across and along the car) can be calculated, and fed into the understeer equation.

Another reason I am suspicious of the practical importance of GRCs is that we used to have an IRS and beam axle version of a car, with the same front suspension.

Now, if the RCH is so important, how come a back end with a RCH of 307 mm is OK, and so is a RCH of (say) 120mm, both with the same front RCH, of say 60mm? By OK I mean, drive them on a smooth tarmac circuit and try and pick the difference. The guys who worked on them could, but I bet only a small proportion of normal people could pick it, and they might prefer the beam axle.

Just because you can change the RCH, and you prefer the resulting car, you should bear in mind that inevitably you have changed other things as well, and they may be the real reason why you like the new geometry (things that often change when you change RCH are toe and camber curves, and antidive).



Cheers

Greg Locock

 

[red300zx99x]

I'm still lost to what the difference is, never seen anything concrete about a force based roll center. This report comparing the 2, is it open to the public? And another thing have you ever seen anything written about what the driver feels in roll?

And the car with the different rears, did they keep the roll couples about the same?

 

[GregLocock]

The SAE defines FBRC. I imagine there are some papers around on it.

I've had no luck finding public copies of the dissertation. Here's the front page, you may be able to get a copy from Aachen

Institut für Kraftfahrwesen der RWTH Aachen

Prof. Dr.-Ing. H. Wallentowitz

Diplomarbeit / Diploma Thesis
CAE investigation of the influence of suspension characteristics on the vehicle roll behavior for passenger cars

Cand.-Ing. Eggord Thomaschky
Matr.-Nr. 188 587

Betreuer: Dipl.-Ing. Peter Holdmann

November 1995

Yes, roll is a very important metric. Not as important as yaw velocity.



Cheers

Greg Locock

 

[willeng]

Greg:

Thinking about your statement on the two cars with different rear roll centre heights, i take it there production cars?
Most production cars haven't got the torsional rigidity or components to show much response in handling when making changes, some do but there are a lot that don't.
Try it on a race car that is sensitive to minor change, with a good chassis minor changes make the difference & the difference in roll centre height from 307mm to 120mm would sure make it interesting to drive.

 

[red300zx99x]

Greg, can you draw a free body diagram comparing a grc and fbrc, in roll would be perfect.

 

[GregLocock]

No, you can.

Cheers

Greg Locock

 

[GregLocock]

Oh willeng, how torsionally stiff are the race cars you are talking about? I really doubt that any open wheeler is stiffer than your typical large production sedan, but I'm willing to learn.



Cheers

Greg Locock

 

[willeng]

Greg,

Some of the monocoque designs are extremely rigid with figures of 28,000 Nm/deg being stated.

Back in the 60's Bruce Mclarens car had around 10,000lbs/deg of torsional strength & they have come a long way since then.

There are a couple of sites you might find interesting to read, i find the results of the tests interesting?

http://www.ces.clemson.edu/~lonny/pubs/journal/sae983051.pdf

http://www.ces.clemson.edu/~lonny/pubs/conference/sae983032.pdf

The force based roll centres subject is interesting & would make an interesting discussion, i'm all ears!