(PhysX 3.3)
Hi! I’ve been trying to create some stiff wheels, but don’t seem to find anyway to disable the suspension. If I lower the suspension compression and droop very low, the constraint start to generate wierd forces forward(local Z-axis).
Everything works fine if I increase the compression and droop, but the handling feels wrong since the wheels are supposed to be stiff.
Any ideas? Thanks in advance!
I confused with what you said about wheels being stiff… The suspension is what supports the weight of the car, and not the wheels from a physics perspective, the wheels are just the link between the drive surface and the suspension. What do you mean by the handling feels wrong ? I know what you mean in general that it doesn’t feel right, but to be more specific what makes it not feel right…to floaty ? too wobbly ?
Exactly, it feels to wobbly and floaty. The thing is a lot of modern vehicles doesn’t have a spring suspension, their drive axle is directly attached to the chassi. (Example: A wheelbarrow)
So the problem is that I want the drivetrain, but no the spring suspension. But if I lower the compression and droop of the spring to about 1cm(0.01m) I get wierd behaviours, like acceleration without any force applied.
Well ofcourse there is force applied but the acceleration comes from the constraint instead of the vehicle update.
I’m not sure that physx vehicles would be the best way to model a wheelbarrow. The model is pretty much a box on springs with the spring forces used as input to the tire force calculation. The model is based on the assumption that the spring force dominates over the force generated from the springiness of the tire. A further assumption is that the tire has a stiffness at least an order of magnitude greater than the suspension spring. This allows us to ignore the tire because it will be almost static on the timescales of the spring. These assumptions are fairly standard in vehicle modelling.
The obvious thing to try is to shorten the suspension travel (maxDroop and maxCompression) and then add a stiff spring. The difficulty is that a stiff spring needs short timesteps, especially when the suspension travel is short. I can’t see this being an efficient or faithful solution.
One thing to think about is that a wheelbarrow typically has a soft tire and you might consider a soft spring as a good approximation of a soft tire. If you really want a stiff tire and are trying to simulate a vehicle more complex than a wheelbarrow then doing this with rigid bodies sounds more sensible because there isn’t really any springiness left in the system.
Cheers,
Gordon