[FONT="]
Originally Posted by SoundGuyDave
[/FONT]
[FONT="]I'm not sure exactly what you're fishing for here... The goal ABSOLUTELY IS maximum grip. Anything that compromises that leads directly to a lack of control and precision.[/FONT]
[FONT="]I'm not sure I buy that.
Consider the possibility that when the suspension loses grip, it does so in an entirely predictable and controllable manner. That it does so is not necessarily a consequence of maximizing the grip, but rather is a consequence of the loss of traction occurring in the same way every time, in a gradual manner such that the driver's inputs can control it.[/FONT]
[FONT="]Agreed. It goes back to the point where you need to be off the bump-stops for the suspension to function properly and provide that predictable, controllable breakaway.
Suppose, for instance, that you've set up the car (which has a natural weight bias towards the front) to oversteer a bit, and are using a square tire setup. By definition, this means that you have not maximized the total grip, because the rear tires will lose adhesion before the front tires have reached their maximum grip, when the natural maximum grip would have the car understeer a bit due to its front weight bias. Does that mean that the car is, therefore, less controllable than it would be if it were set up to understeer? Not as long as the oversteer is constrained, predictable, and sufficiently gradual.
[/FONT]
[FONT="]I would argue that you HAVE maximized the total net grip, as you have become grip-limited at one end. Balance is equally important, but plays directly into that whole concept. I guess I took that as a “given.” Controllability is a natural side-effect of proper setup. Short form: Start working on the end with the problem, and maximize the grip potential there. Next, work the other end, and bring the car back into balance. At that point you have “maximized” the grip, specifically because you cannot create additional mechanical grip at the limited end. [/FONT]
[FONT="]In my specific case, I’ve found myself grip-limited at the front. I’ve got nearly the whole catalog thrown at it, and I’m pretty sure I’ve gotten as much out of it as I can without going to aero or an SLA. I then brought the rear up so that I get a very mild overall understeer on entry and mid-corner phases, which transitions to mild oversteer on corner exit under throttle. A couple weekends ago, I had four different drivers run the car at Road America, and the universal qualitative assessment was that the car was very predictable, and very easy to drive. The quantitative assessment yielded less than ½ second in lap times from three of the drivers, with the fourth being only one second off. Peak LatG in all cases was between 1.2 and 1.4, and the lap time differences were all in the approach and exit phases of the corner.
Put another way: controllability can be considered to be a combination of the range of inputs to the range of outputs, combined with the precision of that mapping. Responsiveness is the speed with which the car responds to changes in the inputs or the conditions. Grip is just a single variable out of many, and I don't (yet, at least) see how maximizing it guarantees anything about controllability or responsiveness.
[/FONT]
[FONT="]I would define controllability more in terms of linear input to linear output, specifically in terms of predictability. In other words, “If I do this…. Then that happens.” I will agree completely with your definition of responsiveness. Looking more closely at the grip portion of the argument, however… The easiest way to change one grip variable is to substitute environmental conditions. Take exactly the same car, with the same setup, and now change from a dry track to a wet track, or ice/snow. With reduced grip, the car will NOT respond nearly as well as it will on dry pavement. I’m speaking strictly from a responsiveness perspective. When you turn the wheel, the car no longer immediately begins to describe an arc, as the front tires lack the grip to begin the turn in rapid fashion. Your control inputs necessarily have to become both smoother and slower versus dry pavement, which we could call the “maximum grip” model. This mimics the higher-CG/softer-spring model to a certain extent.[/FONT]
[FONT="]
[/FONT]
[FONT="]If you look at a car with a high polar moment, the time taken before the suspension "sets" is exaggerated, and the inside tire unloads more. In what way is that "controlled" or "precise?" Now, taken to the other extreme, a car with a very low CG, but essentially compromised suspension angles, the camber curve goes positive VERY quickly and you wind up loading only a fraction of the contact patch. Also "not controlled" and "imprecise." This is particularly true if the suspension travel is compromised to the point where you land on the bump stop and your spring rate goes infinite in a sudden way (exaggerated, but not by much). Again no control, and no precision. [/FONT]
[FONT="]Right.
But how much (as a percentage of the original) of a reduction in polar moment do you get by lowering the car each inch of lowering?[/FONT]
[FONT="]I can’t quantify that. I’m also not advocating massive change. Below is a pic of the car sitting in the paddock, and we can all agree that it’s
hardly slammed into the weeds. Lower than stock? Certainly. A
lot lower? Not really. More than an inch, though, probably closer to two inches. Again, not extreme.[/FONT]
[FONT="]
[/FONT]
[FONT="]Two points: First, getting it to do what you want it to do is all a function of grip. If the car is a horrid, pushing pig, you simply cannot just toss the car into a corner and sort out where you want to go at the drop of a hat, you need to plan FAR in advance exactly what you want to do, and then execute perfectly. [/FONT]
[FONT="]But isn't that more the result of
relative grip, i.e. the relationship between front grip and rear, as well as the responsiveness of the suspension itself?[/FONT]
[FONT="]Of course it is. It’s also a function of having sufficient
overall (maximum) grip to be able to get the car to turn in the first place. Relative grip (balance) is a tuning factor applied once one end has reached the maximum limit.
[/FONT]
[FONT="]If you have simply insane amounts of mechanical grip (lowest "reasonable" CG, minimally compromised suspension angles), you can put the car ANYWHERE on track (or the street in an emergency) that you want.[/FONT]
[FONT="]Sure, but that's true as long as you're well within the limits of the grip you have, no? For instance, I found my car very easy to put where I wanted it when driving on the track even though it is supposedly not going to be that precise at the limits.[/FONT]
[FONT="]I think we’re getting into semantics here. Any car works well within the limits of grip available. All discussions here revolve around pushing the grip limits, though. If you (or anybody) are satisfied with the available limits, then there’s no reason to worry about altering the car from stock. If you drive like a human being on the street, yet still achieve 80% use of the available grip (Citroen 2CV, for example), the car will appear to handle just fine. If you need to push it, though, say in an evasive maneuver to avoid a ball-chasing child, you can/will rapidly exceed the net available grip, and depart controlled motion. If, on the other hand, your normal commute only uses 3% of the available grip (Ferrari F430), that same evasive maneuver (at the same speeds) can be handled with aplomb.
[/FONT]
[FONT="]Second, maximizing grip does NOT necessarily mean a compromise in either reliability or street manners. Yes, the usual formula of stiffer springs and dampers to suit will make the car ride a bit more "firmly," but with proper damper and curve selection, that does NOT have to translate to "harsh" in any way shape or form. Take a drive in an M3 some time to see what I mean. Very taught, very responsive, very precise, but not at all harsh. [/FONT]
[FONT="]I haven't driven an M3, but I have driven a 3-series (335i). I found the ride to be near the limits of the hardness I was willing to tolerate in a daily driver. It wasn't harsh. It
was hard.[/FONT]
[FONT="]If this is the case, then perhaps a dual-purpose vehicle isn’t the right choice for you. Keep your DD stock, on the soft suspension, and get a second, dedicated track-toy. Remember that the needs of the street and the needs of the track are generally at polar opposites. The instant you try to make the car do both will result in compromise. The street car becomes hard/harsh, and the track toy becomes less than it could be.
[/FONT]
[FONT="]Why hamstring yourself? Look at your desired target, and then design/mod to reach that goal. If it takes aftermarket bars, so be it. All a bar does is add to the wheel rate in roll, without affecting the rate in pitch. [/FONT]
[FONT="]Well, I'm presuming that I'll want to retain the stock balance at the limit. But that may not prove to be the case.[/FONT]
[FONT="]Very possibly. [/FONT]
[FONT="]The higher the polar moment, the longer it takes to set the suspension, and the less precise things are. Yes, if you drop the car to lower the CG, you will have compromised your suspension angles, as well as shortened the available stroke under bump, but those are things easily fixed. [/FONT]
[FONT="]How do you fix these things without compromising NVH?[/FONT]
[FONT="]Seeing as the LARGEST contributor to suspension-based NVH are the bushings selected, perhaps leave those stock. I was specifically thinking of the various control arm relocation kits and/or extended-length ball joints along with shortened damper housings to correct for the changes that resulted from dropping the ride height. None of them will affect NVH, with the possible exception of the damper curves. See below:
[/FONT]
[FONT="]Hike the spring rates, shorten the damper housings, even relocate the roll center via ball joints or arm mounts if needed. Tame the now "stiffer" suspension with proper damper settings, and it can still be quite comfortable if not taken to extremes. I guess I'm just saying not to be afraid of dropping the car by an inch, inch-and-a-half, or even two. Even doubling the spring rates won't wind up with a harsh ride assuming proper damper curves, which are available with the better dampers out there. [/FONT]
[FONT="]Well, I've experienced what driving an S197 Mustang with roughly double the spring rates and good dampers (KW V3) feels like. It's about as high as I'd want to go.[/FONT]
[FONT="]Again, perhaps a “track toy” is in your future.
The only question in my mind is with respect to lowering the car. Won't stiffening the springs result in greater responsiveness all by itself? If so, then just how much of a gain in responsiveness am I going to get by lowering the car, say, an inch instead of the half inch I intend to start with?
Since my plan is to go with coilovers, ride height is something I'll be able to adjust to my liking. I have practical considerations such as ground clearance which will severely limit the amount I'm going to be willing to lower the car. Remember: it's a daily driver first.
[/FONT]
[FONT="]Again, perhaps a “track toy” is in your future. Only you will be able to qualify the limit on NVH you’re willing to tolerate, and only you can determine exactly how much compromise you’re willing to deal with in terms of viability as a street car versus capability as a track car. In both cases, though, the goal is to run as soft a spring as you can while still keeping the car off the bumpstops. If you need 2” of bump travel on the track, but 4” on the street, then you will have to compromise somewhere, either in spring rate or in CG to maintain the necessary clearance on the street.[/FONT]
[FONT="]I think this is becoming one of those "internet myth" things, where people believe that geometry trumps all. It doesn't. It's a part of the equation, nothing more. I don't think that we could easily quantify the value of CG vs geometry, other than saying that some combination of those two, plus the alignment, can deliver the maximum possible grip the tires are capable of delivering. Let's face it, as long as the car is off the bump stops, you have a working suspension. Dial that in, and you're good. [/FONT]
[FONT="]Well, okay, but "good geometry" means that the available grip changes minimally as the suspension moves, right? Which is to say, doesn't that
alone act as a major contributor to controllability and predictability?[/FONT]
[FONT="]Not necessarily. “Good geometry” just means that you’re in the part of the curve where minimal angle changes occur with a given movement distance. Think about camber curve with a strut car as an example… If we assume that the LCA being parallel to the ground defines the “zero” point of the suspension, small movements in bump or rebound will yield a minimal change in camber measured at the wheel. As you increase the size of the movement, however, the camber change begins to accelerate. The further you get from “zero,” the faster things go to hell. Now, consider a car with “perfect geometry,” i.e., the suspension is at “zero” at rest, yet has a sky-high CG. This results in a very large polar moment, which we also know acts as a lever arm on the suspension. With any given lateral load, that lever arm will cause a larger motion in the suspension than that of the same vehicle with a lower CG under the same lateral load. I guess if you wanted to quantify, you could look at the total camber delta comparing the two. The lowered-CG vehicle might start off in a worse portion of the camber curve (thus “compromised angles”), but the total change may be less than a car starting off in the perfect area but with a longer lever arm. In a Lat-G model, total grip at that tire is pretty well defined as contact patch management. If you do a numeric model, make sure you set the static camber angle the same though. It’s not unreasonable to assume that if you lower the CG, you will at least align the car properly when done. If the model does prove out that the lowered CG provides a more consistent contact patch, wouldn’t that necessarily be a more controllable and predictable car?
And how can I change the understeer/oversteer balance without compromising grip at one end or the other?
[/FONT]
[FONT="]Again, balance is a given assumption once you have maximized one end or the other.
[/FONT]
[FONT="]Purely an experiential viewpoint. The more you push the car (approaching the limits of grip more closely), you'll start to notice all the ills that have been mentioned. Nose-dive/ass-hike under braking, wallowing and numb turn-in, a decided tendency towards understeer will all become apparent to you. You are, however taking the proper approach: DO NOT MOD until you can identify the first thing that is holding you back from going even faster. I know you're not looking to push or be competitive, but if an increase in speed isn't the goal, then why worry about ANY mods? Just drive what you have. Anyway: once you've identified the first thing that is positively holding you back, research the cause, research the solution, then implement. Then push harder, until another issue comes to the forefront. Lather-rinse-repeat. [/FONT]
[FONT="]Yep, that's exactly the approach I'm going to take. Furthermore, my changes are going to be, as much as I can reasonably make them, single-variable changes. Springs and dampers first, with minimal changes to ride height. Then (if necessary) sway bar changes. Then ride height changes. Camber will be set based on tire wear patterns.
Given that track events are at least a month apart from each other, this is going to take a long time, and I worry about not being able to remember how the car felt the previous time when testing a change I've made in response to it.[/FONT]
[FONT="]That’s another reason not to mod until you’ve reached the point of reasonable consistency as a driver. Given long-interval testing, you need to develop some metric to judge each change, and that’s impossible if the driver is all over the place. That said, if you reach the point where
that one thing is pissing you off, go ahead and fix it, and screw the metrics. Once you get to the point where you’re experimenting, hunting for more grip (balance is easy, assuming reasonable adjustability in the hardware), then metrics become more important. They could be as simple as rough raw lap times, or as focused as multiple-zone tire temperatures. When you get into the testing/experimentation phase, though, be sure to keep a logbook on the car’s setup and environmental conditions. You’ll also want to do your testing at a single track, for sake of consistency.
[/FONT]
[FONT="]A few points to consider: At stock ride heights, the suspension geometry is just about as good as it's going to get. [/FONT]
[FONT="]Stock ride height for which car, though? The Boss 302 has a different ride height in the front than the GT does, for instance. My suspicion is that, for handling purposes, the Boss has a better ride height in front, a better geometry versus polar moment compromise.[/FONT]
[FONT="]I think that’s a valid assumption, particularly given that they all have different spring rates.
[/FONT]
[FONT="]ANY change to ride height will have a corresponding negative impact on the geometry. Now, look at pictures of various S197 race cars that stay on strut suspensions, like Grand Am GS cars, and note how they ALL have been dropped, to lower the CG. What conclusion can you draw from that? Hint: they weren't dropped to look good...[/FONT]
[FONT="]Well, of course. But then, they don't care about ride hardness, or even harshness for that matter, nor do they
really care about ease of control. They're able to optimize for a single variable: maximum grip, even if it comes at the expense of some ease of control, since they're going to be driven by professional drivers who have a
very high degree of precision control capability, along with very fast reflexes.[/FONT]
[FONT="]Correct, NVH means nothing. “Ease of control,” though? Hardly. The car MUST be “easy to control.” IF the car is exhibiting excessive understeer or oversteer (balance is off), then the driver has to back off to keep the car on track, out of the tire walls, and off other competitors. Once he backs off, he’s slow, and loses. The “maximum grip” model by definition means that the car is capable of being pushed, and pushed HARD prior to displaying any handling ills. All the solid bushings, spherical bearings, and solid mounts that are swapped into the suspension are there to preserve the maximum grip and precision, allowing the driver to eke out 99.99% of the net available grip. The fast cars are also very predictable at the limit. I don’t care HOW fast your reflexes are, snap oversteer on an unpredictable basis cannot be caught. Bushing compliance/bind is one way to get snap oversteer when you’re pushing right up to the limit.
[/FONT]
[FONT="]If you want to see what the car can do stock, and what might just be the first (second, and possibly third) issue to pop up, get an instructor or advanced-level driver to take the car out with you in the passenger seat, and see what happens when they push the pace a bit. You may find that the "not objectionable" body roll has suddenly become objectionable. Or it might be the face-plant under braking, or the command of "full right rudder" to get the car to turn-in that becomes the glaringly obvious limiting factor in what the car can do, not the driver. Or, perhaps, it won't. THAT is all up to you. If you do have a "guest" drive the car and find the limits, DO NOT MOD, but push yourself as a driver to find the SAME limits, and THEN mod. [/FONT]
[FONT="]That is an insanely good idea. I think I'll do precisely that. But yeah, I agree that I shouldn't bother to mod the car until I've reached those same limits, because modification of the car can't be done properly until I know from personal experience how it's deficient in
my eyes, how it doesn't feel the way I want it to or respond to my inputs the way I want it to. Can't determine those things without supplying the inputs myself. [/FONT]
[FONT="]
[/FONT]
[FONT="]The bottom line is that it's still about maximizing the grip available at the tire, and to do that will require some compromise in BOTH CG height as well as suspension geometry. Perfect geometry and sky-high CG will yield a numb, unresponsive, imprecise understeering mess. Perfect CG and a lack of suspension travel will yield a very responsive but uncontrollable mess. The sweet spot is somewhere in the middle, and most likely compromised on both sides of the equation. [/FONT]
[FONT="]Sure, but realistically, we're talking about lowering the car an inch at most. Daily driving considerations prevent me from lowering it any more than that.
So how much of a benefit will I really get from lowering the car by an inch instead of, say, a half inch? Is that
really enough to make a "night and day" difference in how responsive the car feels?[/FONT]
[FONT="]Obviously, I can’t answer that, as there are far too many variables, some of them qualitative. Again, perhaps a “track toy” is in your future. We’re speaking of pushing the car, the driver, and the suspension system right up to the limits, and then optimizing from there. It may well be that the required hardware and settings for that optimization may exceed your threshold of acceptability for daily-driver duty.[/FONT]
Here is my reference chart for the sliding scale between street machine and track toy:
[FONT="] ........Trips over it’s feet braking................... Rear end still wags................................ Hand of God braking[/FONT]
[FONT="] ........Numb responsiveness ......................... OK, but not great ...................................Crisp responsiveness[/FONT]
[FONT="] ........Rolls like the Queen Mary ...................Still a boat, but a smaller one................ FLAT through the corners![/FONT]
[FONT="]DAILY DRIVER[/FONT]
ß[FONT="]---------------------------------------------------------------------------------------------------------[/FONT]
à[FONT="]TRACK TOY[/FONT]
[FONT="] .......Comfortable ride .................................. Kinda hard/rough ................................. I’m peeing blood![/FONT]
[FONT="] .......Clears speed bumps.............................. Find a new way to work ....................... Can’t get out of the driveway[/FONT]
[FONT="] .......Quiet .....................................................Gotta turn up the radio ..........................What? I have hearing damage now…[/FONT]
).
