Ah, that's darned close to the question to which I've been searching for an answer for a year: does one ever take the bars into account when choosing the amount of damping. The answer seems to be No, but I'd like to know why not.

As to the difference in feel for different total roll resistances from different combinations of spring and bar, while I, too, probably couldn't detect it, many folks say that it makes much more of a difference on rougher surfaces. This is why I was very surprised by the bars suggested by RalliArt for gravel. Seem too thick to me.

 

You can't really dampen specially for the bars. The bars are reacting to roll, but without having a fluid circuit between the dampers you can't control that motion exactly.

Best bet I guess would be figure the bar stiffness for a single wheel and convert to spring rate. Add it on for the low speed stuff and probably neglect it for high speed.

This all comes down to theoretical though. The ~65% of critical figure is mathematically the best response for a particular set of assumptions. It doesn't really give a crap how the tire actually responds to inputs and I'm pretty sure tires have some time dependent behavior themselves.

Like griceiv said... Do what the voices in the tire tell you to do.

 

Well, bars are half as effective against single-wheel bump as they are against roll, which is why I've wondered why they are never included in the standard equations for calculating damping. But folks rarely seem to mention the 2/3 & 3/2 rule, either, so I started to suspect that it's a given that also is left implicit.

 

in some reading this afternoon, i just ran across this (see bold), which seems to justify most of my thoughts:

Corner entry understeer: car initially points in and then washes out

-Excessive toe-in or toe-out (car is usually ‘darty’)
-Insufficient front droop travel (non droop limited cars only)
-Incorrectly adjusted packers (car rolls on to packers)
-Insufficient front damper bump resistance (similar to roll stiffness example)
-Insufficient front roll stiffness – car may feel like it is pointing in but may actually be falling over onto the outside front tyre due to insufficient front roll stiffness or diagonal load transfer under heavy trail braking. Initial understeer can often be cured by increasing front roll resistance, even though doing so may increase the amount of lateral load transfer.
-Non linear lateral load transfer due to spring and/or bar geometry. Or to non-optimal roll axis inclination


i think my car has a decent amount of front roll...

 

Man that just goes so much against what I experienced.

Not an Evo, BMW 318ti (strut front/semi-trailing arm rear). I tried a no sway bar setup. 680 pound front springs in a 2400 pound car... TONS is spring for roll stiffness. Ride rates were 500/350 f/r. This should have understeered into oblivion. Quite the opposite though, underivable oversteer.

That above saying understeer on entry due to a lack of front roll stiffness doesn't match what I experienced at all. The turn in was mind blowing quick and razor sharp. The back just wanted to keep going straight...

I added a 25mm front sway bar and even still, it will oversteer 99 /100 times. It is far more controllable now though. Next biggest change though, the car is crazy numb on turn in now. Well, compared to how it was before anyway.

Saying understeer due to a lack of front roll stiffness just seems completely wrong to me. In that situation, more stiffness comes from the rear and it overloads the rear first. Same reason a big *** rear bar causes oversteer...

 

But you also have to consider what is happening to your camber and contact patch if the front end is rolling excessively. Uncontrolled (or loosely controlled) rolling of the front end kills your camber, as it is usually accompanied by lots of compression travel on the outer front corner into a non favorable area of the camber curve. All this lead to a loss of grip on that outer front tire and grinding understeer.

As I slowly progress along the road to getting my Evo to handle better, I have come to the conclusion that you have to really control the body roll on this car and keep the suspension operating in a range where you are working with a favorable camber. The car simply turns better and the tires wear better.

Exactly how to manage this with all the pros and cons of the different approaches, well that's the complex part.

 

That was exactly the theory i had running in my head, but until i found that article, i had zero evidence to support it.


I think what i finally understand is that there is a crossover point/point of diminishing returns. go stiffer in front roll is beneficial to a point, but eventually you drop off a cliff and it goes from decreasing understeer to increasing oversteer.

i also think you are exactly right jim, there are probably 100 ways to attack the setup on our cars. I also think that there has been a disconnect between changes in suspension tuning theory and the massive changes that have happened in tire tech seen in z2's/rs3's etc.

i feel like these days the "go to" rates dont necessarily work perfectly anymore.

 

But with the rates that are effective (~10k up front) you are talking like 2-3 degrees of body roll. Going to 16k might drop that to 1.5-2 degrees.

1 degree of static negative camber fixes that issue...

Sure the extra static camber impacts braking and acceleration performance, but as for cornering grip at the limit, it's a wash. Those high rates do make the tire want to "skate" though when they experience shock loads.

 

I am running 14k F and through 3 events so far this year I have had none of the 'hopping' or skating out of the front end of the car. I was just at CMP and even getting more of the rumbles than I intended to on occasion, the front end stayed planted.

 

14K front.... yikes what's out back?
Must be nice to have a track only car. I don't think I could tolerate more that my 10k fronts for DD.

 

There is virtually no difference between front roll and rear roll in a sufficiently stiff chassis. The Evo isn't perfect, but it's stiff enough that your front and rear roll angles are going to be largely the same.

Another key concept to keep in mind is that the amount of weight transferred to the outside tires is only a function of track width, center of gravity, and the mass of the vehicle (assuming no acceleration or braking). Your spring rates and swaybars cannot alter the amount of weight transferred from side to side for a giving cornering force. Well actually they can if they raise or lower the center of gravity, but let's assume the CG is the same for simplicity. And in practice, your CG isn't going to change a whole lot from one set of springs/bars to another.

So if you can't change how much weight is transferred from side to side in a corner (without lowering the car or losing weight) then your only option is to modify how that weight transfer is shared between the front outside tire and the rear outside tire. Likewise, you can't make one end of the car roll more or less than the other end so you're limited to determining the overall roll stiffness and how that roll stiffness is distributed between the front and rear ends.

If you add more roll stiffness to the front end, then the front outside tire will take a larger share of the weight transfer. If you add more roll stiffness to the rear end, then the outside rear tire will take more of that weight transfer. The front end is already more heavily loaded than the rear, so at first glance it makes more sense to get more roll stiffness from the rear end before increasing roll stiffness from the front end.

Increasing the rear roll stiffness will reduce roll in the front, because the front and rear roll are virtually identical.


In practice, it gets more complicated very quickly. In the real world you also must contend with acceleration and braking forces at corner entry and exit, which will induce additional weight transfer to the front or rear of the car.

This is why common wisdom dictates slow-in and fast-out for front-heavy AWD cars. Slow-in minimizes the amount of time that the car must deal with both forward and lateral weight transfer, which is the worst-case scenario for the front outside tire. Fast-out takes advantage of the rearward weight transfer to more evenly distribute the cornering load by forcing some of it to the rear of the car due to acceleration forces.

Also, if you have too much roll stiffness coming from the rear of the car you can compromise power delivery by lifting the rear inside tire too much. If you can't put power down, you can't induce a rearward weight transfer from acceleration, which becomes counter-productive.

There's also a matter of roll axis inclination in all of this, but that's not something that can be changed a whole lot in most classes.

It's also very easy to end up on the bumpstop of the front outside shock, which quickly throws you into a very non-linear suspension region while simultaneously overloading the front outside tire. This is bad news all around, which is why it's important to keep your overall roll stiffness and your front spring rates high enough to stay out of the front outside bumpstop during most turning and braking/turning maneuvers.

 

Why are you saying that this is the worst-case for the outside front tire? It's weight (on the tires) that transfers, not the actual mass (of the car). If you're going to ask the outside front tire to accelerate the mass of the front end towards the inside of the turn, you want to give it some extra weight (for extra grip) to do so. That's why some folks trail-brake in front-heavy cars.

 

Everything in moderation. This works up to a point.

The Autocross to Win guy explains it better than I can in his page here: http://farnorthracing.com/autocross_secrets4.html (Scroll down to tire theory).

The key is to find the right balance and to prevent the car from riding on the bump stop too much. Once you're in to the bump stop the suspension can't absorb bumps properly and you end up losing grip.

Also keep in mind that although we focus primarily on the loading of the outside tires, the inside tires do contribute to the cornering forces as well. Increasing rear roll stiffness will unload the rear inside tire (tripod mode!). Side to side weight transfer for a given corning load isn't changed by your roll stiffness, so the inside tires still have to split the same weight transfer load. But if your rear inside wheel is up in the air (or at least mostly unloaded) then all of the outside load is placed on the front outside tire. That tire won't be in the most ideal camber orientation, but it will still be able to pull the car toward the inside of the turn. And as a bonus, it will be further down that tire load curve, so it will be happier in the process.

In other words, the increased rear roll stiffness induces a diagonal weight transfer to the front inside tire. That, in turn, pushes that tire into the ground and helps it pull your car toward the inside of the turn.

But you're absolutely right: If you were somehow able to unload the front tires enough to compromise their loading, you would sacrifice grip on that end. Everything in moderation.

 

I can't believe that someone just told me to go to DG's website. I wonder if the hidden message to me is still on the Dynamics Calculator page....

 

Sorry if I come off as too simplistic. I try to cover all the basics because I know a lot more people will read this thread than just those of us participating in it.

Also, I'm entirely open to the possibility that I'm completely wrong in all of this. I'm also willing to disregard all of the theory in the world if a certain magic spring, bar, and car setup somehow turns out to be faster than doing things the "right" way.