I'm still trying to grasp some of the deeper suspension "tweaks" and have a question on roll centers. From what I understand lowering the car with just springs drops the front roll center significantly farther than the rear. And that dropping the panhard bar (either via the axle side drop backet, or body side drop bracket, or preferebly both you can correct some of that so the roll axis is closer to level)

For a tighter handling car do you want a more level roll axis? or slightly kicked up in the back? Does having the roll center in the rear way higher than the front pull the load off the inside rear tire in corners causing some loss of traction?

 

Thanks for the reply!

It took a second of looking at the diagram you drew to understand the outside dropping and inside lifting. Didn't realize you were talking about the tire specifically. It seems fairly obvious to grasp the concept now.

Are there ways to draw out/formulate where the roll centers/rollaxis is on a car like mine without expensive CAD programs or something? Or is it more of a "beat the crap out of the car hard" and 'feel' what it wants to do?

Is the panhard talk correct? That lowering it's mounting points on the axle/body you can lower the rear roll center?

 

The rear roll center is easy to find. It is the point where the PHB crosses the centerline of the axle.

The front is a bit more complex, basicly you are finding the IC of each side of the suspension, to do that, you use the plane of the A-Arm, and a construction line percondiculare to the upper strut mount. The point where those two planes meet is the IC. From the IC, you then draw a line back to the center of the contact patch of the front tire. The RC will be on that line.

As you can see, the a-arms orientation has a large effect on the IC, and thus the roll center of the front suspension. This is why lowering has such a large effect on the front roll center. You also have to realise, that the front RC is going to migrate as you turn, brake, and accelerate, since as the a-arm pivots, your RC and IC both move. This is part of why our cars like a stiffly sprung suspension, it helps to control the IC/RC migration.

 

Dean,

Did you end up bracing the bracket, back to the axle? I notice that the jegs piece now, uses a brace, much like the UE peice.

 

Time to clear some things up . . .

The roll axis has nothing to do with braking. Neither do the roll centers that define it. Braking (and acceleration) in a straight line involves pitch motion, which is rotation about an axis that's roughly perpendicular to the roll axis. Not roll.

Vehicle motions due to fore/aft forces involve things like anti-squat, anti-rise, anti-dive, anti-lift, and wheel rates due to the spring stiffnesses.

More later (I'm at work).


Norm

 

Mostly right so far. The roll center for most strut suspensions falls about twice as fast as the vehicle ride height at that end drops. For a PHB-defined suspension, it’s about half as fast. So the roll axis of a strut/PHB stick axle car like the 3rd gen F?body gets steeper as the vehicle is lowered. I’d word the idea of relocating the PHB much more strongly, to the effect that the PHB should be made level with the car in its normally loaded condition unless there is a very specific reason to not do so (circle track running comes to mind as a valid exception, but not much else does right off the top of my head).

Since it appears that you’re after better handling behavior and presumably do not want to sacrifice traction, relocating brackets for your LCAs should also be on your list. A lowered TA/PHB stick axle tends toward more understeer and poor traction both off the line and off the corners if the LCA axle end pivots are higher than the chassis side pivots.


Normally, you want the vehicle roll axis to be slightly higher at the rear. Too high makes tuning efforts less effective, as tuning via shifting the roll couple distribution around requires that roll exist in order to work at all. Take it all away via high roll centers, and there goes your tuning.

Anyway, a slightly higher rear RC helps corner entry by transferring relatively more load across the rear tires during the early stages of the cornering maneuver. But since lateral load transfer due to roll and the front bias to the roll couple distribution eventually overwhelms the roll center height effect, the handling tends to tighten slightly as you progress through the turn. There’s more (LOTS more), but the simple point here is that for most driving you don’t want a “loose” condition that stays “loose”, especially if there’s much torque available to push you past the point of no return.

Without considering the effect of lowering the front RC or the effects of the new front springs for a moment, you would also want either a bit more rear spring, a bigger rear bar, or perhaps some combination of a little of each to re-balance your steady-state cornering. This would be a good time to define your usage (road course, auto-x, street, how serious you might be about either of the first two, etc.) as this can affect your parts choices.

However, since the front RC has been lowered, and the new front springs are (presumably) stiffer than OE, the tuning requirements may change a bit from the above. It is possible to mathematically determine most of this stuff, but it still takes experience and/or “drive-the-snot-out-of-it testing” to make much sense out of the numbers or to fine-tune the ballpark that the analysis puts you in.

Normally, I’d refer you to Sam Strano of Stranoparts.com, as he is very knowledgeable when it comes to setting up 3rd and 4th gen F-bodies. But as Sam’s experience is chiefly Stock and Street Prepared autocross, where PHB lowering is not permitted, I think that Jason Swindle of Unbalanced Engineering may be a better resource in this particular instance (full disclosure: UE offers a PHB relocating kit of their own). The only way I know of to directly get in touch with Jason is through FRRAX.com under the username “Unbalanced Engineering”.

BTW, I do agree with the idea of welding these sorts of pieces rather than rely on bolting.

Norm

 

Separating things . . .
to keep the confusion to a minimum . . .


More anti-lift is not necessarily a good thing. Anti-lift works by removing load from the rear tires almost instantaneously, rather than slightly more gradually over the time it takes for one quarter of a pitch cycle (from "at rest" to "nose-down/tail-up") to occur. But at least you don’t have the later cars’ problem with brake hop. Anti-lift in those cars can be positively evil.

Anti-lift's country cousin "Anti-squat" is the more useful feature, though it's entirely possible to have too much of that for some (many?) situations as well.


Ummm, not exactly. Progressive springs get progressively stiffer as additional load is placed on them. As the rears unload, the first bit of chassis rise is less than the next increment, and so on. IOW, braking twice as hard will cause the rear to rise by more than twice as much.


Was hoping I’d see the boldfaced portion.


Lift and squat with the LCAs horizontal are what they are. Somewhere I’ve got a few spreadsheets worked up that compute roll center height, anti-squat and rollsteer for various linkage configurations. With the shorter TA you should have more of the anti’s than stock.

(I don't have those sheets in here, but they use the 3D coordinates of all the various pivot points plus a couple of other dimensions.)


You have dialed in too much rear rebound, and it’s trying to lift the rear tires. This phenomenon takes a little time, as shock behavior is velocity-sensitive, which means that it takes a little time to develop. Let the chassis move around a little more and you won’t be disturbing the contact patch loading (and hence the grip) quite as badly. Locking a wheel can happen rather quickly, much faster than getting it to roll at road speed again. I’m pretty sure that too much rear rebound is adding to too much anti-lift here.


Sounds to me like there’s a rear ride height adjustment going on here, else there would be no need to tinker with the rear sta-bar. That affects the inclination of the LCAs and the anti’s.

The nice thing about the TA/PHB arrangement is that it's possible to tinker with the RCH without disturbing the anti's (one of the reasons why there's a market for TA/PHB conversions for Fox/SN95 Mustangs, I might add).


I’d take you up on that, but CA is a bit of a hike from NJ.

Fair enough.


Norm

 

Wow. So much to take in. I think I'll be reading and re-reading most of this quite a few times to picture it all in my head.

 

Dean and I have had this debate before also. I think it was in the decoupled TA thread and brake-hop thread, in the brake board. I never found mention of it in the Milleken book, the for/aft weight transfer forumals I found, all had to do with the SVSA geometry, and their according IC's, and the affects on the anti's.

Dean, I would love to see a refrence or diagram of some sort, that explaines how the RC's effect for/aft weight transfer.

 

In the specific case of a PHB arrangement, there could be a very slight bit of coupling between pitch and roll due to the asymmetry of the PHB. But that would be a diagonal-ish phenomenon rather than a purely fore/aft one.


Norm

 

Norm, do you care to explain this furhter. Assuming that you have spher-bearings (so you are not taking some sort of slight binding, creating an moment, that we wouldn't normally take into effect), I do not see how a the SVIC's can have any effect on the FVIC's (this is really what we are discussing). Having looked at the millekens drawings, I don't ever see the FVIC's in any of the caclulations for any for/aft dynamics.

 

I've been reading up on suspension stuff for the last couple of weeks so this all makes sense. But for these cars, what is the "best" way to have the suspension? I'm personally trying make my autocross car handle better (it pushes) so this thread is of great interest to me. I'm tired of being bested by miata's...

 

The PHB forces the chassis side bracket to follow an arc, relative to the axle bracket, so the rear end motion is not purely vertical. It's close enough to vertical for most purposes, as long as the PHB is reasonable level. But sooner or later the small vertical component of force at the chassis bracket will act about the roll center. It's a small effect, and it would be really reaching to claim that it would cause any huge longitudinal load transfer phenomenon.

In the literature you'll find lots of unstated assumptions once you start looking for them. Mostly, they're negligible, or mostly so, with any decent design (yet another assumption). But they exist, and they're easier to spot when you think in terms of 3-D arcs instead of "almost straight" lines and flat planes. With the latter in mind, another small one is that the motion of the PHB is not constrained to be within a single plane.


Norm

 

god i would LOVE to drive a car setup by somebody with the same knowlage as you guys. its great that they havent banned you for no reason Dean, you give good info.

 

First, i work on oval track cars, currently cars with a 2 to 4 inch frame height with watts linkage setups and have worked with long PHB setups with 4-5 inch heights on Busch cars and 4-5 inch "normal" length PHB setups like a late model stock car.

I've always found off throttle corner entry has very slight effect by raising the rear RC. We prolly do not make the huge adjustments you are talking about though since we get to cheat with caster split and tire stagger. We do most of our turning in the center where we want the car to rotate and head the other direction and center off, on the throttle is where roll center makes the biggest difference for us.

You confuse me by saying "overwhelms the roll center height effect, the handling tends to tighten slightly as you progress through the turn." Do you mean, off the throttle?

I understand where you are getting at by the overwhelms part because it's because of the spring/bar/shock package not being correct. Through all this, the rear RC is still doing it's job though. It is not because the rear RC is higher that the car might tighten up.

By rasing the rear RC, as the car progesses through a turn and especially when the power is applied the car will not start to tighten up, higher rear roll center removes weight from the inside rear tire while corning and frees/loosens the car up to set for the center of turn. On power this gets drastic. Lowering the rear RC will tighten the car up as you say raising it will, planting the inside rear tire better. When we lower the rear RC without any other changes, our left rear tire temp will climb.


EDIT: After typing all that, and then another post I might understand what you meant anyway. i was just thinking you meant higher over a lower rear RC.

 

Not knowing the first thing about your setup, I'd first ask (not smart assed) if you understand how rear roll center works.

It works off of/with the COG. Roll center is the amount of leverage the car is able to gain on the COG. This length is called many things depending on who you talk to, we call it moment arm. It has little to do with it's height off the ground (at least the amount a car can actually move around.)

If the world was perfect, and under braking the rear end rises 1 inch evenly, the height of the PHB will only move RC 1/2 inch, or half of whatever the chassis moves. This means the COG only has to move 1/2 of what the chassis moves to maintain the same RC ballpark setting. When the car rolls to either side it lengthens the moment arm and would tighten the car up. (Um, duh, this might have been what the other gentleman was talking about, that as the car rolls more it will tighten)

Again, not knowing anything going in, the first place I would look is the front bar being too stiff, effectively laying the outside weight onto the rear tire and overloading it as the bar continues to load the more the car is turned.

Or, in the attempt to get the control arms level (which everyone seems crazy about here) the rear of the car is just too low and it creates an easy spot to dump weight onto. You keep saying "stiffer" rebound which I'm taking you mean slowing it down/adding hold down which makes me think this is also hurting. In the attempt to hold the rear down, you create a hole that the weight just falls into under braking/cornering, overloads the outside rear tire and around it comes. On low speed stuff, I'd try running the rear control arms down in the back anyway. Silverback touched on this in another post, rear steer isn't always a bad thing.

Or, or along either of them lines there is just too much rear brake in the car, adjustable valves only take away brake from where you dial it out of so you are correct, anywhere past 50/50 you lose braking power. Even in a perfect setup, 3-4 clicks will change the attitude of the car while braking, add in if any of the above is true and it just multiplies.

 

We run way softer then that since we run 15 inch tires, more like 15-18 right sides 8-12 on left rear, somewhere in there to get stagger right. Whenever we tighten a car up with roll center we always leave the stagger up where it was just incase. We'll usually scuff tires in our qualifying heats and once they go through a heat cycle with nitrogen in them they wont move much, even with the added heat.

We use a watts that works of a 1/2 inch or so thread down the center.

 

I'll add, I am not a fan of roll-oversteer either. IMHO, it makes the car harder to drive, especialy in transition, but that is just my prefrence. When I have dialed in too much roll-oversteer in, I end up "sawing" the wheel thru corners, and that is no good (this could be, because I am a mediocare driver, at best). With just a bit of roll-understeer dialed in, the car is much easier to drive, as the car rolls you can slowly and smoothly apply a little more steering, to keep the car turning.

(yeah, I know that was a bit off-topic, but the roll-steer talk was brought up.)

 

yes, we run 15 inches of rubber on a legend car. They are asphalt modifieds.

So as the rear of the car goes up in the air, the roll center gets higher and the car wants to spin out. That's what i took from your post. Draw this out and look at it moving mounting points. I also learned that roll center magically becomes a braking tool.


Oh, so let's not discuss it, since it's all the same, now you really lost me.

 

Dean, we are all in agreement on the how the roll centers location in comparison to the CG location effects things when turning. In you last example, you are talking about turning, while braking. The disconnect here, is earlier you stated that the roll centers effect "straight line" braking.

Maybe I am missing the whole point of the conversion. Are we talking about lateral forces, or for/aft forces?

(And you know I am not being defensive or argumentative. )

 

What happened to all DRR's posts?

 

I think that DRR (Dean) has been banned again

 

This is ****ing ridiculous!! I wonder why i keep wasting my time on this site. Deleting pages of information (good technical info, and practical 3rd gen tech) is assenine. What the hell good is this site if they keep deleting the info i need.

 

Yes, as soon as the tech gets brought here (if it is by Dean) it gets deleted.

Dean and I corrisponded by e-mail today, if Dean gives me the go-ahead, I'll post the e-mail up here. It is sort of useless now that all his posts are gone, but it has some further explaination as the what Deans car is doing, and why. Very helpfull info for the discussion that WAS here.

I am sure I am going to get a talking to about this by the mods and admins. But... Come on, deleting one of the FEW threads on this board with real tech, just to try to prove once again, that you can ban Dean, is stupid. We all know that Dean and the mods on this board have issues, we all know why. But, please, we once again, have a trashed thread, that is now USELESS, because the mods chose to delete all of his posts, again.

Ohh well folks, back to your regulare scheduled TGO program, no real technical discussion.

Here is a coming preview.... Yes folks, sportlines lower more than pro-kits, nope, 295's are a bad idea on your stock wheels, and if you only care about looks and not any realy perfomance, intrax and drop-zone springs make you car look slammed, yo.

 

Where are the **** uniforms when you need them? I could see the actions taken as not being completely childish if there was an arguement of some such but thats just rediculous. So much for trying to spark discussion and a transfer of knowledge.

Lama, what other boards (suspension related) do you frequent?

 

Ironically, i dropped this board because Dean was missing for so long. I checked back every so often " I love my KYB's and prokit, it rides great and it rails now" and finally found something interesting. I guess i wont be wasting my time anymore. Is there another board You guys (Dean too) frequent that i can check out?? If you can find somewhere else to go please let me know Laiky@aol.com. I'm done here.

 

I agree with the idea we go to another suspension board where we can get discuss these higher level issues without fear of wasting our time. Just another moron in government we don't need. I wouldn't drop the board, it is the best, and not all the moderator are ***** like the one that trashed this thread and wasted our time, we just need to drop a link or pointer somehow to another board. Who is the AH moderator anyway?

 

I have to admit that I’m more than a little disappointed in finding that Dean’s posts have been arbitrarily removed, as I’d spent some time thinking about it over the weekend.

As near as I can reconstruct it, Dean’s complaint centered around the rear roll center migrating too far upward for the conditions imposed and the (tire) limitations involved. In that respect, it appears to work as a band-aid for a certain amount of “overdriving”. At least that’s what I infer from an autocross background and an interest in road course running. Maybe it’s part of the game in circle track, perhaps for passing maneuvers; but I can’t speak for that. I wonder if Bob Bolles has anything to say about this in either CT or in his own book . . . I’ll have to look.

I don’t doubt that Dean’s observations of what was happening as the static roll center height is relocated are accurate, just that parts of the analysis that followed miss the mark. My take is that (1) braking caused the rear RC to migrate upward (not much question about that), and (2) braking also caused forward load transfer (still OK).

But when it was finally mentioned that this was happening fairly deep into the corner rather than in pure straight-ahead stopping, it becomes apparent that the forward load transfer from the braking wasn’t being backed out of as fast as the lateral load transfer from the cornering was coming in. At some point, the tires’ friction circle (ellipse?) limits are reached, and one end starts to lose grip (and gain slip angle).

It’s not that the higher rear RC or the steeper roll axis pitches more load forward. It’s that the consequence of the higher rear RC forces more lateral load to be transferred across the rear, and it is to be kept in mind that 1g lateral affects tire loading by almost 1.7 times as much as does one longitudinal g (wheelbase/track = 101/60, approximately). The inside rear takes a hit in grip from both effects. Differential type (and condition) also figures into this, as does roll steer.

I still think that progressive springs are the wrong way to go, if you’re more interested in dynamically maintaining the rear RC position than in maintaining as much ride comfort as possible over bumps. Allowing the rear ride height (and its roll center) to rise more than proportionately with respect to the amount of forward load transfer works contrary to that. I think other solutions exist that might even be better than linear rate springs, but whether they meet anybody’s class rule package is a whole ‘nother discussion.


Wardo1 - speaking of roll center height effects vs roll stiffness distribution, these two effects do not increase at the same rate. The roll center height effects peak much sooner, as load carried through the suspension linkage is almost instantaneous. Only the fact that the tire vertical stiffness is finite (1500 lb/in or so) delays the tires’ response at all.

Roll stiffness effects, on the other hand, involve the roll frequency of the sprung mass, which is much slower (unless you’re running something like 4000 lb/in springs in the front and 1500 or so in the rear). The total lateral load transfer (TLLT) is a function of the sprung mass CG height, and is divided into the portion that goes directly through the roll centers without causing roll and the portion that does produce roll. Generally the CG is is around 19” - 22”, and this holds true for a wider variety of cars than most perople might think.

An average roll center height is perhaps 7”, divided something like 2” front/12” rear, or about 15% front/85% rear.

For most RWD cars, TLLT distribution is generally a few percentage points more front-biased than is the amount of front weight.

Roll stiffness distribution is pretty heavily front-biased; call it 82% front/18% rear for purposes of illustration (and because I have those numbers handy from a reasonably similar analysis). With roll fully developed, it’s distributed about 1/3 RCH effects (7/21) vs 2/3 roll stiffness distribution effects. That’s what I was getting at with “overwhelms”, as the roll stiffness term is about double the RCH term. The higher rear RCH relative to the front indicates an initially looser situation, which helps rotation, but as the roll develops you tighten.

Handling “feel” can be considered as that which you can sense but can’t easily identify the ‘why’ of until you think in terms of ‘time’ and ‘frequency’. The first 50 milliseconds or so is almost all RCH effect with its 15/85 distribution. You’d be horribly loose and have to be VERY late into the throttle on corner exit if it stayed that way (I’d rather not think about trail-braking at all if that kind of distribution was all you had to work with). Over the next 200 or more milliseconds as roll continues to develop, the roll stiffness term increases, and that initial 15/85 distribution moves in the more front/less rear direction (which you can read as a lessening of the initial oversteer and eventually becoming increasing understeer). Ultimately, and with the above numbers, you end up with something in the ballpark of 60 front/40 rear. All that in a quarter of a second or so, but with all else equal it’s the difference between a fairly stable “feel” and one that could be described as a bit “nervous”.


Norm

 

Ok guys.

Here is Dean's e-mail.

As for other places. I post on frrax every now and again. You guys could always come hang out on the cascadecrew website, at Cascade Crew Just register.

 

FRRAX member here too, almost from its very beginning. Just so everybody understands, it's primarily oriented toward road-race and auto-x.


Norm

 

FRRAX member here as well. I stayed out of this one since it was above my knowledge but I will be coming back to this thread to read it for the 10th time. Good info and I really appreciate deans e mail being posted, hope it stays.

 

I wouldn't worry too much about its becoming completely unavailable, as there are at least two copies that are not posted to the forum.

Since Dean's writing style and terminology differs quite a bit from mine, I'm still in the process of translating it. Offline.


Norm

 

I didnt save it, but regardless of the disagreement there was still a bunch of good info thats long gone. Nice job. Now it looks like Norm is talking to himself.

Why dont you ban the IP already (the entire block if you have to) if you are so far off the ball and it took you this long to find him this time? Glad to see some things never change, even with new owners.

 

All I've got is the e-mail and a general gist of the rest. Wish I'd thought to grab it all when it first appeared (as I sometimes do). You never know what material will force you to think the hardest, but the stuff that raises the most questions as you read it is right up there whether it's right, wrong, or indifferent.


Norm