At the risk of being too blunt. You guys way over think this stuff.
Possibly. But for me, at least, mainly because it's fun.
Case in point, I don't recall my Koni curves don't have the rears as stiffer than the fronts, but I'll review them again tomorrow. And that would make little sense given the fact the front of the car is in fact heavier than the rear, and that solid axle cars don't want a crapload of rear shock typically.
I'll soon have data for the other two corners (the problem is that the dyno shop didn't bother to make it possible to match specific dyno sheets with specific corners, so all I know is that they independently measured the front struts and did the same for the rears, for a total of four dyno plots, two of which I've supplied here. I know which two of the plots are for the front and which two are for the rear, but I won't know which ones are for left versus right). Given the rather large difference between the front and the rear in terms of the way they're built (one's a strut, the other is a shock), I rather doubt they got them mixed up.
It is what it is.
But more basically, we have a round wheel and I swear folks are constantly trying to reinvent that round wheel. My best friends are engineers. Hell Shelly is, and not a fly a by night one. She's an M.E. Went to some small scrub school called Stanford.
And holds some 60 patents and has started two medical device companies. She's also done things like ground cams for Elgin too. She's well rounded, and I harass her, a lot about being "an engineer" sometimes. One common thread with very technical folks is they get bogged down in minutia.
Yes. But only because it is sometimes important. Yes, sometimes making a small change yields a small change in the result. But sometimes, making a small change yields a
large change in the result. You can experiment and eventually figure out from experimentation where the best result is to be found, but if you have interacting components and make changes to one, how are you going to know that you won't have to readjust once you make changes to the other? Now throw in another couple of variables, and the sheer number of possible combinations becomes overwhelming.
For instance, you can change your shocks and optimize them, but if you change your spring rates, then you have to go through the shock optimization thing all over again. If you're looking for the optimum
combination and can only do that experimentally, then you'll have to go through a pretty large set of combinations before you really find the one that's optimal, and even then you won't
know that it's optimal unless you've tried all possible combinations. It's easier if you can assume that things get progressively worse as you get further away from the optimum point, but even that isn't always the case. Now throw in sway bars, and suddenly things get even more complicated. Now throw in control arm changes. You get the idea.
The alternative, and the one I prefer to use, is to understand the physics behind the suspension, and to create an initial setup on the basis of that understanding. Fine tuning can then proceed from there. If you find that what you experience is substantially different from what the model suggests, then that means your model is wrong in some fundamental way, and thus that your understanding of the physics is similarly wrong (or incomplete) in some fundamental way, and it's then time to start looking at how and why your model is incorrect and fix it. That may require experimentation (it
certainly will if you're using the scientific method to zero in on a correct model).
My suspicion is that you have a working model in your head of how the suspension works. You may have painstakingly built it from experimentation, but you have it, and that's what's important. Many/most of us here don't have the benefit of many years of continuous experimentation from which to build a model in our heads. So physics will have to do.
It's not like our approach is invalid or anything. It is, after all,
precisely the approach the manufacturers themselves use in building the suspension in the first place.
A big one we hear about are suspension frequencies. Fine. But I don't care, in fact I've kicked the crap out of cars setup by numbers. Until a computer drives the car, that stuff is good to know and might help you starting from stratch. But we aren't starting from scratch.
No, but then, if you're modifying something you don't understand, then you're modifying it blind, as it were, and you have to make certain assumptions about your approach (such as that it will converge on a solution).
And when it comes to competition you have to have a car you can drive. That you can lean on. That you can trust. Personally, I am very opposite of the all the discussion here. I drive the cars, stock. I drive them with alignments and tires, etc. And I make a determination about what it is I think the car needs to do better. Is it loose? Tight? Does it roll too much, or just too fast? Etc, etc. It's troubleshooting, and I'd highly encourage folks to do more of it, and more critical thinking about the problem at hand than thinking so much about theory.
But it all goes hand in hand. If it's too loose, what changes do you make in order to fix it? What exactly informs those changes? You have to start somewhere. That's where the theory comes in. It gives you a place to start. If it's anywhere close to being right, then it'll at least put you in a reasonably good starting place, and the rest will be fine tuning. If it's not, then you really haven't lost anything because, after all, you have to start somewhere no matter what.
I know that won't be popular, I've been down this road before. However, my method works. I've done a lot of cars that have won stuff. While I'm a pretty good driver, I'm not head and shoulders above lots of others. It's not a matter of me being able to cover a screwed up setup with driving talent. See also how whatever classes I run I get cars to the top. Swaybars I produce won G-street, A-Street, B-street, and STX. I didn't win SSR, a Porsche did which I don't work with, but I was 2nd and most of the cars up at the top are cars I setup and were running my setups. I also did the the Mustang that just barely missed winning ESP (to a Subaru on E85 that's lighter, narrower, AWD, and makes more torque too), pretty much from top to bottom.
And how did you know what stiffness range to provide with your sway bars? Pure experience? If so, then that's an expensive way to figure it out. Most of us don't do this stuff professionally. We don't have the
opportunity to optimize by trying every possible combination. And seeing how we each have individual needs, we can either take someone else's word for a proper setup meeting our needs, or we can understand the fundamental physics behind the suspension and determine for ourselves what will at least get us in the ballpark of what will meet our needs.
Most people take the former approach, and act on the advice of experts such as yourself. But the problem with that is that even experts have diametrically opposite points of view on much of this. For instance, the fundamental approach you take to setting up a car, even if it's for someone else's needs, is likely to be almost the polar opposite of the approach taken by Terry Fair. Which of you is
right? You both achieve fabulous results in your efforts, so both of you clearly have valid real-world points of view. And yet, they're still diametrically opposed.
If both of you are correct in your approach (meaning, the end result each of you would recommend will meet the needs of the hypothetical singular customer we're talking about here), then that means that there are
at least two optimum points for that customer in the range of possibilities.
The reason people like those here in this thread like to understand the model is that it allows us to find something approaching the optimal point for ourselves. It
empowers us. And it helps that we find it interesting.
Your approach is absolutely a good approach. It works. But it's not the
only valid one.