The Stiffness of the Springs - the Suspension Frequency         
How stiff should we make the suspension on our race car?
Smithees "Bounce Test" for Spring Frequency................ 

We choose our springs to control the tyre - to maintain grip, maintain tyre contact with the road.  The spring works together with the shock absorber in this role. In reality, analysis is complex, requiring expensive equipment,   eg the 7-post rig. (see www.insideracingtechnology.com/contentspg.htm ).  So we must choose our preferred stiffness based on experience and testing.   If it's a road car, we also take into account ride quality.

Having determined a reasonable suspension stiffness for grip, relative front to rear stiffness can be modified for balance (understeer or oversteer).   Further stiffening of the suspension may be required to control overall roll, although this will normally be achieved with a combination of anti-roll bars, shocks and springs, rather than the springs alone.   We need to know what the suspension stiffness is, as a first step in calculating roll resistance.  All these considerations are a key part of the Smithees Weight Transfer Worksheet.

Spring Frequency - Compare Stiffness of the Set Up Between Race Cars

The most usefull measure of suspension stiffness is the spring frequency.   It is directly comparable between all race cars.  We can take our experience with one race car and use it to help choose spring stiffness for another.  For instance, we find that small sedans and production sports cars around 1000kgs, without aero devices, will work best around 120 to 130 cycles per minute, running on Dot racing tyres.  This is a lot stiffer than road car springs, where spring frequencies might be around 80 to 100 cpm.  Other considerations might require us to run stiffer again.  But 120 to 130 is what we'd like to run for maximum grip.  If we can run quality racing shocks, we tend towards softer springs and more shock compression.  We can do this because modern racing shock absorbers are much more sensitive to slow speed settings.  We use QA1 motorsport shocks because we can play with the valving to get any shock characteristics we want.

Heavy sedans might run lower spring frequencies, and light open wheelers might run higher spring frequencies.   The controlling factor here is that the heavy vehicle requires greater increase in spring rate to achieve a given spring frequency increase.  For instance, in the graphs below, a 50 lb per in. increase in spring rate for the 1000 lb per wheel vehicle, achieves only about half the spring frequency increase we get with the 50 lb per in. increase for the 300 lb per wheel vehicle.

The spring frequency is the natural frequency of the suspension - if the car was to bounce up and down on it's springs unrestrained by the shocks, it's a measure of how many cycles would it go through in one second (Hz) - or in one minute (cycles per minute, CPM).

The spring frequency is a function of the sprung weight acting at the wheel, and the vertical stiffness, wheel rate or effective stiffness acting at the wheel.

Smithees "Bounce Test" - Measure the Spring Frequency Directly

An important development from Smithees.  IP patent number 2001100047.

The bounce test measures the spring frequency directly, without calculation.
Sometimes the spring frequency cannot be calculated accurately. eg large diameter coil spring working on a poor motion ratio, or leaf spring live rear axle.  With leaf springs, calculation of the spring rate and spring frequency is impractical. 

Smithees "Bounce Test" measures spring frequency quite accurately, in a matter of minutes, and the numbers can be used in the Smithees Weight Transfer Worksheet.  The procedure is a no brainer, but exceptionally valuable in gaining a feel for what's going on. 

The "bounce test" results are a key part of the easy to measure numbers we need to begin a suspension handling project.  (If the car has coil overs or Mcpherson strut suspension, the "bounce test" cannot be done.  But this is OK, because for these suspensions, the spring frequencies will calculate quite accurately from the spring motion ratios.)

Procedure

1. Remove the shocks from the car, and disconnect the anti-roll bar(s) if fitted.

2. Put the car on a level hard surface (eg concrete floor)

3. Seat a driver in the car (we need car at road going weight)

4. Start bouncing the car at one end - push down sharply, let the car bounce, and then push it down again to keep the bouncing motion going continuously.
   Look up at a sweep second hand clock and start counting the number of bounces in one minute.  I usually count for 30 seconds and then double it.
   This number is the "cycles per minute".

5. Re-check a couple of times, to see that you can come up with a repeatable number.  There can be problems.  For instance with independent front suspension, it is just as valid to push down on one guard only, or push at the front, both sides together.  If suspension is binding on one side, you will notice this, and it must be repaired.

6. If the end you are not bouncing is moving around a lot, you will not get a good figure - there is "some cross" over between front and rear.  You may have to put the shocks back in that end your not bouncing.
   
   

Weight Transfer Vs Other Handling Performance Issues

If the suspension and steering is mechanically sound, the greatest gains in performance will come from optimising weight transfer, ahead of any other suspension modifications.