Developing a Proper Suspension Model

[7thgear]

Originally Posted by RBbugBITme

-3 to +3 pure roll, no steering

how much effort would it be to integrate steering and roll into the geometry calculation?


as a car would only roll during steering, so this is rather important.

[ZDan]

I could be wrong, but here is my impression:

I don't think that a *virtual* point "shooting off to infinity" necessarily means bad things happen in the real world. It's just that the mathematical model is running into a discontinuity, probably due to a divisor getting close to zero somewhere.

Looking at the plots where everything goes to hell at -1 ride height, that might be a mathematical artifact and not representative of physical reality. The *actual* roll center height doesn't go from -5ish to +5ish by going to +infinity and flopping to -infinity first.

It's just a limitation of the mathematical model.

I *think*!

[RBbugBITme]

Originally Posted by Calum

@RBbugBITme

What ride height, relative to stock, would you say is best? Or is it just a matter of also preforming the roll center correction when lowering?

Sorry Calum, I would rather not and really can't recommend the best ride height. The intent here is to supply you with enough info to be able to make better educated decisions about your own setup. I have driven this car during a test drive once so I'm not the best guy to talk to about specific setups.

7thgear, if you go to the photobucket page with the image you can blow it up pretty big with good resolution. Yes I'll do a static roll angle and iterate steering angle at two different ride heights when I get home tonight. The point of these camber curves is to isolate your camber change due to chassis roll since you can be at the same roll angle for every degree of steering angle depending on vehicle speed.

[RBbugBITme]

Originally Posted by ZDan

I could be wrong, but here is my impression:

I don't think that a *virtual* point "shooting off to infinity" necessarily means bad things happen in the real world. It's just that the mathematical model is running into a discontinuity, probably due to a divisor getting close to zero somewhere.

Looking at the plots where everything goes to hell at -1 ride height, that might be a mathematical artifact and not representative of physical reality. The *actual* roll center height doesn't go from -5ish to +5ish by going to +infinity and flopping to -infinity first.

It's just a limitation of the mathematical model.

I *think*!

Part of the problem it represents is that your jacking forces will switch from positive to negative (anti-jacking) mid-corner. No one ever wants to design that in on purpose. You either maintain a RC above ground or below and keep it there. The other issue is the wild swing in the moment arm length.

[ZDan]

Originally Posted by RBbugBITme

Part of the problem it represents is that your jacking forces will switch from positive to negative (anti-jacking) mid-corner. No one ever wants to design that in on purpose. You either maintain a RC above ground or below and keep it there.

I don't think it's a problem. You'd never notice the difference between the roll center being 0.1" above the ground vs. 0.1" below the ground. Yeah, there will be an optimal range for RC height (highly dependent on the application), but nothing magical or bad happens when it passes through zero.

The other issue is the wild swing in the moment arm length.

There *is* some wild movement of virtual mathematical points, but if you look at the actual loads going into the suspension mechanisms, you aren't seeing radical changes in actual physical moment arms.

I really think that the crazy curves going to +/-infinity are more indicative of localized regions where the *math* goes unstable and not an indication of actual problem areas.

I would have no qualms about testing an FR-S/BRZ at 0", -.5", -1", -1.5", -2" to see what happens. My guess is that nothing particularly weird happens at -1".

[RBbugBITme]

Originally Posted by ZDan

There *is* some wild movement of virtual mathematical points, but if you look at the actual loads going into the suspension mechanisms, you aren't seeing radical changes in actual physical moment arms.

You've measured this? You can certainly calculate it but I'm not going to since I have no vested interest in this car.

The RC width going to infinity isn't an unstable equation. That is physically when the angle between the top of the strut and the lower control arm change from acute to obtuse, meaning the instant center jumps from one side of the vehicle to the other. You may not notice this in a car full of rubber bushings but no race car ever behaves like this.

The only benefit to the digressive curve of the moment arm is that it is digressive. If it were progressive the car would want to continually roll further and further without a progressive spring rate to match it.

[Racecomp Engineering]

Ryan, can you post the front view of the car at a 1 inch drop? It might clarify what you mean about roll center location at that height...

- Andy

[RBbugBITme]

Yeah, when I get home.

[fika84]

I think another thing to keep in mind here is that this is only taking into account the geometrical roll center http://balancemotorsport.co.uk/suspension-geometry. I haven't personally felt any of the unpredictable behaviors that all of this jacking --> anti-jacking would induce.

What you can absolutely get from the model is the camber gain in roll, heave, (steering if you had it).

I would be much more inclined to see simulations/data with the force based roll center.

Also, did you measure the suspension points on the car? Are you sure they are accurate? I just want to make sure you didn't use the points from the spreadsheet on this thread.

[7thgear]

Originally Posted by fika84

I haven't personally felt any of the unpredictable behaviors that all of this jacking --> anti-jacking would induce.

well, lateral overload or unloading would, leading to the tire giving out sooner rather than later is your immediate result of improper geometry


At the end of the day, we want a chassis that remains relatively stable while the wheels hug the road for dear life, with any change in load intuitive and proportional to a drivers inputs.


[ame="https://www.youtube.com/watch?v=zZQKepbGKiI"]- Robert Kubica Test - Flat out & Max Attack Moments - Test Mc 2015 - - YouTube[/ame]

[fika84]

Originally Posted by 7thgear

well, lateral overload and the tire giving out sooner rather than later is your immediate result of improper geometry


At the end of the day, we want a chassis that remains relatively stable while the wheels hug the road for dear life

I love the way rally cars sound!

I wasn't saying that I don't know what it feels like (though I probably don't).. I was saying that after tracking the car it feels predictable and stable to me at stock ride height.. But I'm not a racecar driver either and may or may not feel it unless I drove it after it was "fixed" and could compare.

[RBbugBITme]

These are based on Toyota/Scion 3D models that have been paid for so I can't share the data points.



I have all the FAP data as well and will try to apply some FAP theory but I'm limited on time, I have a side business developing suspension parts for other cars and I was at this until I think 3am last night.

[fika84]

Sweeeeet.. Subaru wouldn't give me any data when I asked them.. for obvious reasons, but it was worth a shot!

For anyone that wants to play with a model and move the car around to see what the effects are, this is fun http://www.racingaspirations.com/app...try-calculator.

I also wonder what sort of adjustments (for better or worse) the roll center correction kits for this car do.

Thanks for the info Ryan and keep it up!

[7thgear]

Originally Posted by RBbugBITme

These are based on Toyota/Scion 3D models that have been paid for so I can't share the data points.

how much does something like this cost?