Joe Weinstein

Hi all.
If someone is corner-balancing, adjusting their spring loads
to optimize *all-around* cornering, how do they choose their
target numbers?
I can see that if the car's center-of-gravity (CG) is exactly
in the center of the four wheels, the weights should be equal,
RF=RF=RR=LR. If the CG is somewhere along the center axis, then
LF=RF and and LR=RR, and the front and rears would be in
proportion to the distance to the CG. I believe that in either
of these cases, the cross-weights should be equal (LF+RR =
RF+LR), making for an easy check. In the real world, however,
the CG is offset laterally and longitudinally. In this case
LF+RR != RF+LR. So how do we know what spring to tweak, to
achieve what values?
We have 4 variables (LF,RF,LR,RR). We have these known
equations:
1 - the sum of the values equals the car weight.
2 - the CG is fixed, so we know
2.1 - that LF+RF = X * LR+RR (where X is a fixed fraction)
2.2 - that LF+LR = Y * RF+RR (where Y is a fixed fraction)

Do we know any more? That's not enough, because we need
4 equations to solve for 4 unknowns. Clearly, as soon as
the GC is within one of the four triangles formed by any
three of the wheels, the car could be 'balanced' to have
zero weight on the farthest wheel. If we started there,
and started dialing weight back into that wheel, when would
we quit?
Unless someone can save me with another equation, we're
at a guessing game. I'm *guessing* we would want to minimize
the difference between LF+RR and RF+LR. So we would want to
do the first tweaking with the car loaded as driven, including
driver weight. But where to start? I assume one would soften
the heaviest corner or add load to the lightest. But what do
you look for to know when you're done?
And then, let's say the GC shifts because the gas tank
goes from empty to full? The change in cross-weights will
not be huge, but is there any information to be had in
observing the changed corner weights, to determine if the
setting we have is in fact optimal?
What if we temporarily put weights in the car, just to shift the
GC to the center line? Then we can use the simple LF+RR=RF+LR.
Do our settings then have any real relevance when we remove the
artificial loads?

thanks,
Joe Weinstein

Bryan Watts

What exactly are you asking?

Tim Pruitt

Joe,

I do not think there is such a thing as the perfect corner balance that fit's everyones driving style. You have to find out what works best for you. Fill the car with gas. Put some weight in the driver seat that matches your weight. Balance the car as close as you can to 50%/50% and go drive it. Then feel what happens under braking, under acceleration in corners, when car is full or gas or near empty. Then you will have more information the next time you corner balance the car.

If the course has higher speed right hand corners you may add more weight to the right rear when balancing. If the car needs more grip during intial breaking you may add more weight to the front and vice versa.

Joe Weinstein

Thanks Tim. I am guessing there's no perfect balance, but there has to be
some rough initial numerical/proportional target. What do you mean by close to 50/50?
What if your car is naturally 35% weight in front, and 65% rear, and 60% left, 40%
right, what weights should your corners be?
thanks for the discussion...
Joe

Joe Weinstein

And I understand that for a given course and driver you might want to shade
the weights to favor making one side more 50/50 if that's the outside for
the longest/most important corners, but for a car that has to go left and
right just as often, what should the base numbers be (such as for the
35/65 60/40 weight distribution car)?
Joe

Cory M

Get the front and back weight distribution (%), lets call them F & B. Then get the left and right side weights (lbs or kg), lets call the L & R.

Now adjust until left front = LF, left rear = LB, right front = RF, right rear = RB.

Bryan Watts

http://tech.rennlist.com/911/pdf/corningbalancing.PDF

Joe Weinstein

Hi Cory. Thanks. Somehow I was thinking it was more difficult.
Let's say a car weighs 3000 lbs, and it's a 35/65 front/rear 60/40 left/right,
so the front wheels take 1050 lbs, and the rears take 1950 lbs. The left side
wheels take 1800, and the rights take 1200.

So it seems the wheels weights should be (neutrally)

LF=630, RF=420, LR=1170, RR=780

So in the general case, the cross-weights are irrelevant.
In this case the 'ideal neutral' cross-weight is 180 (LF+RR - RF+LR)

So, to meander this thread,...

We have no problem putting stiffer springs on the end of the
car that's heavier, but what about side-to-side, and in the
case above, shouldn't such a car have a different strength
of spring for each corner?
Joe

Nordschleife

errr.....

adjust the corner balances to minimise the difference between the weights of the two diagonals.

EG

Before
LF 600 400 RF
LR 400 600 RR
LF+RR = 1200
RF+LR = 800
Diff = 400

After
LF 445 473 RF
LR 671 695 RR
LF+RR = 1140
RF+LR = 1144
Diff = 4

These are example weights chosen to make the method obvious. Most of us use a spreadsheet to calculate this. Its fairly straight forward to to calculate target values for the mechanic to work to in order to achieve the desired result.

R+C

2BWise

Not necessarily a different spring for each corner, but a different pre-load. On a traditional production car setup...Spring/Strut attaching to the upright it would be the same as adjusting the spring perch height. Unless you've got inboard mounted shocks and springs where you'd use preload and push/pull rod length to adjust corner weights.

Joe Weinstein

Thanks Nordschliefe. Your numbers don't quite illustrate
corner-balancing however, because your After numbers
added 244 lbs to the weight of the car

However, i am going to ponder whether the optimal neutral
balance coincides with the minimum cross-weight difference.
Joe

Joe Weinstein

Hi 2BWise. I understand that typically the preload is used to do these adjustments,
but even if we get the corner-weights as we want, for the car example I gave,
why would we want one rear corner that carries 1170 lbs to have the same spring
rate as the other rear corner that carries 780 lbs? Even if the load is balanced at
rest, why would we want the motion of the suspension to be different? As I ponder
your answers, I am answering myself. In a corner, whatever weight transfer there is,
is balanced, such that 100 lbs more load on the outside corner will mean 100 lbs less
on the inside, so independent of the static corner weights, the in-motion transfers are
the exactly compensating, so the spring rates should be the same, side-to-side and
front-to-back.

thanks all for the thoughts!
Joe

2BWise

Forgive if you have already considered this, but with your reply I think of different question myself. For the situation above I assume you're talking static loads, if thats the case I would say that you want the equal spring because otherwise the car would respond differently left vs right if the spring rates were not equal.

I assume you're talking about setting the car with different preloads where suspension travel will differ slightly. This tends to be one of the compromises , but as long as the kinematics don't have any major bump steer characteristics you'd be alright.

100 lbs from inside to outside yes, but where does it go is the important part. Does 80 lbs go to the front with 20 to the rear or 20 lbs to the front and 80 to the rear. In this case the left to right spring rates should be equal but front to rear does matter.

analogmike

Diagonal cross weights are only perfect on a mid seat formula car (or my '73 with a ton of ballast on the right!)

Do the math, it's not that hard.

Here is an article I wrote ages ago with the formulas and some 911 specific info.

http://www.pelicanparts.com/techarti...nment_tips.htm

Adam@Autometrics

Basically, you can only transfer weight across diagonals, so no matter where the perches are, you will not be able to change the front/rear bias or the left/right bias.
LF = Front%*Left%*Total weight
RR = Rear%*Right%*Total weight
and so on...

If you do just percentages, you'll notice the diagonals will not be equal if you have a car that is biased significantly in both directions. It will be 49.6/50.4 or very similar.

This is a starting point, and many will make changes for various reasons. Corner balancing is pretty subjective...