Every race team needs these numbers. We are yet to work on a race car that didn't
need the calculations. We always learn something. Professional race teams have
the same stuff, but much more sophisticated. ( Maybe some pros need this to get back
to basics.) Yet this is so low cost and straight forward, that owners of road cars
can justify using it.
The benefit is that you get a
baseline set up that works. You save the very high cost of track testing and parts
cost involved in determining a set up by trial and error. Of course, you still need
to test to fine tune your set up.
What the Smithees Weight Transfer Sheet Does.....
This spreadsheet calculates weight transfers, roll resistances and roll angle for a 1G
steady state cornering force. The final figure, "Distribution of Total Weight
Transfer", tells us the calculated balance of the vehicle ie it's tendency
towards oversteer or understeer.
The main use of the spreadsheet is to test the affect of different springs and
anti-roll bars, and compare with standard or current springs and bars. You can also
try changing other factors as well, such as roll centre heights. (For a full
discussion of weight transfer theory see Smithees web site page, www.smithees-racetech.com.au/racetech/theory/wttrans.html.
Although not essential, it helps to read this before doing your weight transfer
At Smithees, we find the spreadsheet helps us to focus on the areas where most benefit
can be gained eg Alfa 105, where we looked to greatly improve the motion ratio on the
front anti-roll bar, a March open wheeler, where we needed to design the front anti-roll
bar a lot softer than original, in conjunction with softer overall springing.
Interpreting the Results....
The equations used in the spread sheet are those recommended by David Gould.
Other writers on vehicle dynamics use slightly different criteria. We assume a stiff
chassis, no friction, constant data and do not take tyre deflection into account.
In real life roll centres might move, spring rates and motion ratios vary, and other data
items vary, so that some guidance is necessary in interpreting results.
In the past, these calculations would only be undertaken by vehicle designers. We
have streamlined the process so that the job can be undertaken by the owner/driver or
amateur racing team.
We find the calculated results work very well. There is some judgement
needed. The final roll angle result should only be taken as a guide to relative
overall stiffness in roll, rather than the angle the car will roll in practise. It's
all about suspension stiffness and it's effects. To help improve accuracy, we have
developed our Smithees "Bounce Test" - a
patented procedure (IP Australia, 2001100047) for determining spring frequency directly,
rather than by calculation.
The accuracy of the results is not as important as the ability you will now have to
compare set ups. What happens if you lower the rear roll centre? If you reduce
spring rates, what anti-roll bar will you need? It helps you understand the
relationships between some of the set up features on your car. The old adage says
"only make one change at a time". A better approach might be "make the
change to the set up meaningfull but progressive - and in the right direction
please!" This often involves more than one change, if you know the correct
relationship between settings for your car.
Suspension Frequency - Key Concept for Race Car Set Up.....
The suspension frequency is a very important starting point for the baseline set up of
your race car, because it is the only real comparison of suspension stiffness. Suspension
frequency, or natural frequency, is a function of wheel rate and unsprung weight.
Increased wheel rate (say a stiffer spring) increases the suspension frequency and
increased unsprung weight decreases suspension frequency. For a full
discussion on spring frequency see Smithees web site page www.smithees-racetech.com.au/performance factors.html
How to Get Started
You need to measure all the data items, indicated by the white boxes in the
spreadsheet. Read help on this measurement process by passing your mouse over the
A unique, patented feature of our methods is the Smithees
"Bounce Test". In minutes, you can determine the suspension
frequencies for your existing set up. Now all racers can get a handle on these important
weight transfer numbers. It would be safe to say that the majority of racers would have
one or more suspension books on their shelves - Smith, Puhn, Valkenburg etc. Yet it
has been too time consuming, or expensive, to measure and do the necessary calculations,
to work out the race car's "vital statistics".
Measuring your roll centres and determining the heights of the unsprung centre of
gravity do require a little work and judgement.
If you have A arm suspension, do a 2D drawing, to scale, of your suspension and measure
the roll centre height from the drawing. Alternatively, Doug Matley has produced a
spread sheet for calculating the roll centre height. (The accuracy required is less
than if we were looking at the design and geometry of the suspension. Then we would
need 3D calcs, and look at the roll centre movement in detail, using Susprog 3D software.)
There is a method of measuring height of the centre of gravity which requires weighing
scales. However, we think positioning major items of weight on a roughly to scale sketch
of the car will give a CG height close enough to use the results for comparison purposes.
In deciding on springs and anti-roll bars for an initial base line set up, it becomes a
question of balancing the suspension frequencies you would like with the overall roll
resistance needed to maintain a reasonable roll angle. Also involved is deciding the
amount of roll resistance we will allocate to the anti-roll bars as opposed to the
springs. Finally we need to make a stab at a reasonable oversteer/understeer balance
for the car - the amount the car will be wedged. To do this, compare the static
front vs rear weight percentage with the total weight transfer (TWT) percentage. If
static percentages and TWT percentages are the same, we say the weight transfer
distribution is neutral. Wedged (understeer) is greater TWT percentage at the
front. Dewedged (oversteer) is less TWT percentage at the front. The latter might
only apply for front wheel drive cars with a heavy front weight bias.
Do not be concerned if this last paragraph involves vehicle dynamics concepts you are not
familiar with. It is my job to help you with all the relevant material for your car,
when I email back to you after receiving your completed data in the spread sheet. My
point at the moment is that we cannot choose springs and anti-roll bars (or shocks for
that matter) in isolation. We must take a "whole of car" approach.
When you have a fitted your new suspension, and done your workshop set up, if it's a
race car, you'll need to test at a race circuit. We'll have a ton of ideas to help
you with that. If it's a road car, you'll simply enjoy a great ride!
Below is a weight transfer sheet for a small sports car. Comparing static weight
percentage ("Calculated Wt Percentage") to the steady state dynamic weight
percentage ("Distrib. Total Wt Trans), 52% and 54%, we can see that balance of the
car is 2% in the direction of understeer. There is a lot to "play"with
when it comes to trying alternatives, and many insights to gain. As an example, this
is a leaf spring live rear axle vehicle ie high rear roll centre and high unsprung weight.
Have a look at the unsprung weight transfer and the weight transfer via the roll
centre, at the rear, 30.6 and 91.56 ie around 122 lb. This is weight transfer that
does not go through the springs (or anti-roll bar if there was one.) Springs or bars
have no affect on it. So even though leaf springs offer such poor roll resistance
(only 28% of sprung weight transfer goes through the rear springs), it still works out OK
at the end of the day, because of the high weight transfer numbers attributable to the
live rear axle itself.