Tire data treatment

[SR-Mike]

I'm in the process of setting up a detailed simulation of our cars performance under pure longitudinal accelerations, but I'm having some issues interpreting the tire data available.

The simulation includes dynamically varying loads on the driven wheels but the majority of tire data I have found is only relevant for a single load. I have looked at normalised tire data but its not been clear to me how exactly to get the propulsive forces from just knowing the normal load and the slip ratio. And does this method take into account the load sensitivity of the tires?

Is it a valid assumption to consider the tyre as having a friction coefficient, which is a function of the slip ratio and then applying a correction factor based on the normal load to account for the load sensitivity of the tires?

I have little in the way of real tire data, but it is the calculation method I am concentrating on just now.

Mike

University of Strathclyde Motorsport
Glasgow

[SR-Mike]

I'm in the process of setting up a detailed simulation of our cars performance under pure longitudinal accelerations, but I'm having some issues interpreting the tire data available.

The simulation includes dynamically varying loads on the driven wheels but the majority of tire data I have found is only relevant for a single load. I have looked at normalised tire data but its not been clear to me how exactly to get the propulsive forces from just knowing the normal load and the slip ratio. And does this method take into account the load sensitivity of the tires?

Is it a valid assumption to consider the tyre as having a friction coefficient, which is a function of the slip ratio and then applying a correction factor based on the normal load to account for the load sensitivity of the tires?

I have little in the way of real tire data, but it is the calculation method I am concentrating on just now.

Mike

University of Strathclyde Motorsport
Glasgow

[Buckingham]

If you can get a good Pacejka model for your tires (and understand it), the calculations will become fairly straightforward.

PS, After doing it myself, I can't see you getting much better than 10% error between actual and simulation without knowing a good relationship between Long. Force, Normal Force, and Slip (desired % slip does vary with normal force). The one I came up with predicts the theoretical car to be 0.2 seconds (~5%) faster than our best tested time. The program assumes an ideal surface, ideal traction control, and ideal clutching, ideal shifting (by ideal shifting I don't mean in shift time, I mean that the driver can shift at the exact perfect RPM each time).