roll stiffness distribution & oversteering 1

[mjmghdm]

How can a high rear roll stiffness cause oversteering?

What I know:
1- I can increase understeering by changing the anti-roll bars with a stiffer one in front and a softer one in rear.
2- Higher roll stiffness of an axle leads to higher portion of that axle in lateral load transfer on the cornering.
3- (It might not be related but,) Stiffer springs give lower road holding.
4- Cornering stiffness is affected by the vertical load (I’m not sure but I think reversely).

What I want to know:
If (1) is true, why does it happen? why?
If (2) is true, is this a reason of 1? why?
Does a rigid suspension (not rigid axle) for rear make a highly oversteered vehicle? why?
If (4) is true, Am I right: “I can get a higher cornering stiffness (of axle) on the axle with more equal weight distribution (on its left and right)”

Thank you either you correct my probable mistakes or answer my why questions.

 

[GregLocock]

Read a book (Milliken and Milliken, Race Car Vehicle Dynamics). Draw some diagrams. Do some maths.





Cheers

Greg Locock

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

 

[cibachrome]

You will need to apply your math to a PAIR of tires on the axle. Since a tire is a softening element with load, the action of the load transfer ON THE PAIR will create the under/over -steering situation. Tires are not "springs" when acting in cornering behaviors. They are really velocity based (hence more like dampers). When you sum the net force from the PAIR of tires subject to the transfered load, the answer appears to the worthy student. You need to do this for 2 cases: one where the vehicle's roll gradient is maintained and also for the straight addition of stiffness at front or rear. Just look at the moment balances.

Not sure what is meant by the last part of your last statement. Are you asking if the left and right wheel vertical loads are equal on the same axle having the same tires on it, will you get the highest net cornering stiffness? If the tires are at the same pressure and they are 'conventional' tires, and their plysteer and plyrat values are small, then the answer is YES. Load is the attenuator of cornering stiffness. That's stiffness as in: derivative, not coefficient: as in load normalized force. They are not the same animal. Convenient to spar with but not accurate enough to debate with the Geek Squad.

 

[cibachrome]

That would be load normalized cornering stiffness. Sorry, I didn't have my Red Bull when composing that reply.

 

[hemipanter]

cibachrome,
"Tires are not "springs" when acting in cornering behaviors. They are really velocity based (hence more like dampers)."
Could you please expand that statement a little, just to see if I have understand it right.
Regards
Goran Malmberg

 

[cibachrome]

The spring word comes up because a tire has 'stiffness' which implies a f=kx model. But if you get into a parked car, you can steer or camber the wheels all you want but no lateral force or yawing moments are created. There will be a rigid body moment applied by the steered wheels from a tire scrub torque. Once the tire starts to move (It gets a velocity component) the vehicle receives sideslip forces and turning moments. Thus, the tires operate like they are fluid dampers(infinitely long ones by the way). Now imagine that you were in a railroad car at highway speed with a car running right next to you. If you could reach out an push on the car at its c.g., it would not push back at you to retain its position laterally. Instead, it would drift away. Just like it was connected to the road with pair of shock absorbers with very long rods. If the car were connected by 'springs', it would come back right next to you and travel with you as soon as you stopped pushing. The damper analogy reveals that the vehicle runs in a velocity field, not a displacement field. That means it has no idea where it is, but runs from lateral velocity instincts, hence my damper analogy. You can extend it a bit by placing the attachment points aft of the spindle plane. This now means that there is a trail introduced by the lateral forces. This introduces an understeering moment because of the front steering compliance. The analogy is a bit weak at this point, but hopefully you get the overall point of how vehicle motion is developed. In reality, the driver induces a moment into the steered axle which is reacted by the tire force and moment generation mechanisms to balance this moment. The driver sort of knows where the road is and employs a displacement field constraint set on a feed forward steering torque control law to operate in a desired path.

If you can be comfortable with this explanation, you will be ready to fly an airplane. Last I heard, a plane with no relative air speed is not flying. Wings are dampers, too. If they were springs, you would always not be the first to arrive at the crash site.

This is the basic math model for all vehicle dynamics modeling that we do.