Suspension Design

Hi all, I have been messing with a string computer all day, trying to suss out suspension mounting (inboard) points for my mid engine machine.
I have arrived at the point where I am fairly happy with the figures I am getting at the moment but would just like to check if I am chasing the right tails so to speak!
So I have been concentrating on the static roll centre and currently have a set up that is 4" above ground (this is for the rear of a mid engined car) giving the car 1" bump and 1" droop the RC moves about 1/4" up and down, then with 2" bump and 2" droop the RC moves another 1/2", so in all 3/4" up and down over a 4" travel, sound ok or should it be better than that??
Is the Static Roll Centre top of the list or should I concentrate more on keeping the wheels upright in roll, with the current set up I will have to run some Neg camber which I would expect anyway.
Any other ideas, pointers or tips to consider that I may have missed???
Cheers Chris
Chris, can I mention my 20 cents worth.

From high speed applying the brakes hard on and the nose makes a dip a few things happen .

First you and your competitor are trying to out-brake each other. Right now you want the front wheels to have the best grip possible. To achieve this them 2 must be flat on the road. You don't want them to ride on the edge of the casing.

As I said this is just my opinion.

All depends on what you want to use it for.

Braking is ~5% of a race track lap...

Ideally you want tyres to be upright wrt the ground as the car rolls.
Chis I am assuming its a gt40

My opinion I think your r/c will end up lower than that, the 40 is a low car with a low CG between 260-300mm, to higher R/C may premote jacking.

If you can build adjustability into R/C it may come in handy later on.

On some off the sus programs you can state the end result and it will work out the locations.
I use Susprog3

Hi Guys, yes it is a 917 not a 40 but i guess about the same CG etc.I had planned to run the rear at 5 1/2 inch and front at 4 1/2" ride hight, so thought 4" to be somewhere handy (for the rear and circa 1 1/2 at the front), I have read Stanifoth thanks Cliff, and have messed around with a string computer so maybe I will have a fiddle some more with my string computer and try and get the RC lower esp at the rear and then plan a compromise to also keep the wheels as upright as poss in roll.
A couple of questions, How much Neg Camber do you guys with "Fat" Tyres run??
And what sort of roll angles are you guys working to?

I have not done the sup mounts yet (will wait till chassis is finished) so have plenty of scope to fiddle with it, although as with all things it will be a compromise between RC and wheels remaining upright in roll.

Thanks for your input.
regards Chris
You are right about the roll center being a compromise. Essentially the whole suspension is a dynamic assembly with your parameters changing as you droop and bump.
I am assuming you will run your lower wishbone in the rear at level for ride height, and depending on the length, and inner mounting position the upper link, will be lower at the chassis than the upright mount. The inner mount and angle of the upper link will set your roll center, and the length of the link will determine camber gain which, depending on your tire width, should not be that great.
Space limitations also play a big role in where you place the inner mounts. I would think you will not need a lot of camber gain as these type chassis typically don't roll a lot so you may find that your upper link needs to be longer than will practically fit so there is your compromise.
As far as initial camber, depending on tire width, I would not run greater than 1.5 deg. neg. and probably would set it at less if your upper link is relatively short or you have a lot of suspension travel.
I set my chassis at 140" of virtual swingarm in the rear, and I am using 14" wide wheels and the camber gain is more than I wanted, but as you said the compromise was the roll center location.
Just my 2c
Sounds like you will get it sorted.
Hi Phil,

Yes, had planned on keeping the rear lower arms parallel to the ground and the top links slope down towards the gearbox, when you speak of camber gain I assume you mean when the wheel is in bump the neg camber then heads towards pos camber (as am extreme example), views on how much is acceptable??
I think when the wheels arrive and I get them shod I will make up a scale mock up and see where we are.
Thanks for the imput all
regards Chris
Camber gain will be negative to compensate for body roll, as the chassis rolls towards the outside of the turn the tire and wheel will camber negative to compensate.
Hi Phil,

Yes, had planned on keeping the rear lower arms parallel to the ground and the top links slope down towards the gearbox, when you speak of camber gain I assume you mean when the wheel is in bump the neg camber then heads towards pos camber (as am extreme example), views on how much is acceptable??
I think when the wheels arrive and I get them shod I will make up a scale mock up and see where we are.
Thanks for the imput all
regards Chris


Camber gain

The rule of thumb is you are looking for 2 deg of neg if you had 2 deg of body role.
some static neg may have to be added to compinsate for dynamics such as tyre deflection
ect ect.
You will work that out when you do tyre temps.
If it were me I would under shot it than over shot it.
Its harder to get rid of it than putting on static.

If you are making a front upright and a rose joint is used for the lower ball joint you can use a spacer between it and the upright,this alows you to lower the chassis and space the arm to maintain a level liower arm.
Just a thought

Tyre data is really hard to get hold of but Avon provide some for their racing tyres:

From this you can clearly see the tyre with 4 degrees of camber in it is generating the largest side force when at max slip angle, so for the highest cornering force the tyre should have 4 degrees camber on the outside tyres when at max roll, so what would the static camber be?
Obviously though tuning to any single parameter in the suspesion to its most optimal and ignoring all others will not give the overall best solution.

I spent months working on susprog an longer reading books to come to the conclusion the more I learn the less I know, which is not an argument for not studying the problem in as analytical manner as you can manage just be prepared for things to get complex. And if they've not got really complex then study more.

Trevor Booth

Lifetime Supporter
The AVON charts are only for THOSE tyres. the ratio of camber to slip angle to side force is DIFFERENT for every tire. Carcass construction, radial or bias , soft wall , stiff wall, tread compound, tread thickness, wheel load all play a role.
I have been playing with suspension for 45 years and still learning and it aint just that simple as looking up a dynamic tyre performance chart. Susprog and all those types of programmes are excellent for telling you the kinematics of the wheels BUT THEY WILL NOT TELL YOU HOW THE CAR WILL HANDLE. That part is the black art born out of experience and experimentation, ask anybody involved in any form of motor racing and they will tell you the same thing. You need to know a lot more about the vehicle and its intended use to even begin to think about suspension geometry.

Mike Pass

Hi Chris,
I would guess that you will need an estimate of where the centre of gravity is front and rear on your car as this will have a bearing on the roll moment which is making the car roll. The stiffness of the springs plus the antiroll (sway) bar will oppose this roll moment. The amount of roll you get will be where these two are equal and opposite. Once you have an estimate of the degree of roll you have an idea of how much camber gain you need to get by difference in the ratio of the top and bottom arms. I would guess that a 917 would not roll much as you will have a very low car with I guess a flat configuration engine which will mean also that the c of g is very near the roll centre and this will give a small roll moment as whatever force produced from the tyre grip will act over a very short lever distance. This would tend towards wishbone lengths which are closer in length which will give less camber change in roll.

You need to know what your end use will be and therefore the tyres and tyre sizes and this will also tell you how much compliance you want on your suspension. Road or racing?

Alan Staniforth used to run in the same hillclimbs as us some years ago and was very good at telling people almost enough. He liked people to pay him to set their cars up so giving everything away was not good for business! His book is the same in that it tells you quite a lot but does not give you the full story. I always used to seek out dyed in the wool old time race mechanics who built their own cars and had picked up experience in set up the hard way.

In handling one of the things rarely mentioned is the relation ship between the front and rear roll centres which can be seen by drawing a line between the two. If the line changes angle a lot then the car will have grip etc. but will feel unsettled and you can't see why as the tyres are working and flat to the road etc. Keeping the roll centres at a sensible height is good but so is controlling the movement as much as possible. This will help to ensure that the relationship of roll centre movement front to rear is stable which helps to give car friendly handling characteristics and a feeling of stabilility and trustworthyness with no sudden uneasy feelings. The slope of this line from front to rear for a mid engined car would be for the line to be slightly higher at the rear. If this line suddenly tilts the other way with suspension movement then the car will feel nasty to drive.

A reasonable range of adjustment for static camber and a box with a range of springs and an adjustable antiroll (sway) bar will allow you to get a good useable setup. I have only had two cars where the pickups were so bad that the only way forward was to relocate them. I also used the "string computer" to get the best compromise pick up points - it works.

PS Don't forget to check for bump steer. Start suspension design with the steering rack and uprights and work on from that.

Thanks all, interesting....! I plan on using the car on roads and track but not an all out racer. Mike, thanks for the info, most informative and the sort of info I can understand and make use of (I am the average layman without a computer based degree!!)
As said before, you can read Staniforth, understand most of it but at the end of the day it does not tell you what to do, so string computer and full scale mock up is the route I think I shall take, and then see how we go once up and running!!
Thanks all, Chris

I think your basic assumptions are about right.

As Phil mentioned, you'll want to project your camber gain using your string computer with the lower wishbones parallel to the horizontal in static, then roll on and compression and muck around with the upper wishbones inboard location points and length in order to get enough camber gain to compensate for roll. Net, you might want to aim for maintaining 1 degree negative on the rears both in static mode and with full lateral loading/roll and compression. In other words, the camber gain should roughly counteract the roll effect such that the wheel maintains 1 degree of negative throughout the full cycle of loading and unloading. Easy to say, tough to do.

One basic concept is that the fatter the tire you're running, the more sensitive it will be to camber change. If your car is primarily a road car then might be best not to get too extreme on the tire widths - plus a really wide tire is a bitch if you get caught in a rain storm on the road. So, if you're planning on using really wide tires then the measurements and angle on that top wishbone becomes increasingly critical.

One thing I'd recommend is building a scale model. You can do your string computer work, get your measurements, then start with a basic mock up in wood and square tube steel. This will help you refine the dynamic changes and confirm your measurements are in the right ball park before spending a whole lot of time fabbing up and finishing the suspension components.

Good luck!