Wheel Rate

[Caito]

I had a doubt the other day.

Basically.. how do you choose wheel rate?

I think I could know how to choose wheel rate ratio front/rear, maybe even roll rate. Let's put some scenario with a non downforce car.

Wheel rate won't change the lateral weight transfer, as that depends on CoG height, weight, wheel base,etc. Will it change how fast the car reaches steady cornering state(if such thing exists, lets suppose)?

Do you use sprung or unsprung weight? In a buggy example, where I have a deep pothole.. I could say i want to go throught it without having too much disturbances on the sprung mass. But a 2G vertical acceleration on a wheel would create a force that depends on the unsprung weight.

Any insight?

Thanks.

Caito.-

[silente]

the easiest way for me (but it was not my invention, of course, this is a well known and i guess much used approach) is to choose a natural frequency for the sprung mass you would like to have in jounce. It is somehow a way to normalise stiffness on corner sprung mass.

Normally it is a compromise between platform movement control and effective roughness absorption. High wheel rate implies higher vertical body accelerations (when hitting a bump, for example) but less body movement compared to wheel movement. On the other side, softer wheel rates insure best control of vertical load variation at contact patch, thus generally insuring better mechanical grip, but the body will move more relative to the wheel.

i don't have experience with buggy vehicles, but i guess other parameters are also very important (like overall wheel travel, and bump stops tuning). So actually i think you could want to end up with a very non linear wheel rate vs wheel travel curve, where it is not sensfully anymore to even talk about wheel rate.

[Jersey Tom]

Depends what you're trying to achieve. Certainly there are trade-offs even for a non-downforce car with respect to both ride and handling. Ultimately I suppose you could go about it either empirically or analytically.

In the former case you could broadly categorize these vehicles and say, "Vehicles with a ratio of sprung to unsprung mass of X tend to have a heave natural frequency of Y," start there and tune around it. In a more analytical approach you could construct anything from a quarter car to full multi-body model of the vehicle, give the system some ride input, and adjust spring rate until you hit some target goal (say, for a pothole of a given shape, I want my peak vertical acceleration experienced by the driver to be less than Z)

[Greg Locock]

The frequency based flat-ride approach was first proposed by Olley, and works fine for a road car (although there are problems with it in practice) but the 'rule' is completely irrelevant for a non-aero race car.

The objective with a non-aero race car is to maximise the lateral and longitudinal forces from both axles at all crucial points in the circuit. In general low wheelrates are the simplest way of extracting maximum force from a given wheel, so the simplest target is to use all of the available wheel travel without hitting the jounce and rebound stops. In practice that is conservative, as it is often advisable to hit the stops a bit, but it is a good place to start. Note that this completely ignores any niceties such as steering and handling, it is solely based on maximising the size of the friction circle. Whether the driver and team can exploit that circle is a separate issue.