Soft Suspensions

[Mystique Racing]

Yes, Brian "M" you are one good looking dude.

BTW, nice to see you at the track yesterday.

Brian "H" IMHO your stiff suspension theory is all wrong.

[hardingfv32-1]

Then... Why do almost all non-winged cars racing on slicks have suspensions/wheel rates a MINIMUM of 2-3 times stiffer than FV's?

Brian

[Mystique Racing]

Then... Why do almost all non-winged cars racing on slicks have suspensions/wheel rates a MINIMUM of 2-3 times stiffer than FV's?

Brian

Brian, there is no short answer to that question and I have had a long weekend. Lets talk about that next time we see each other at the track.

[Matt King]

You also have to take into consideration the inherent spring rate of the tire when making that comparison. An FF tire for example, has a lot more sidewall height than a FV tire.

[hardingfv32-1]

1) We are very unlikely to ever see each other at the track under the present economic conditions.

2) How do the members of this form benefit from our private discuss? I'm do this to learn things.

Start off simple: FV's run soft because....

FV tires are special? No, we have run with stiff and soft side walls with equal performance over the years. Our bias ply tires are made just like all the other bias ply road racing tires, except for the dimensions. Our side wall dimensions/shape are deferent the FF, but I would doubt the spring rate of the tire is that much different. I can get the spring rate of the FV tire, does anyone have any clue what the spring rate of a FF tire is? Note, that the front FF wheel rate is easily 3-4 times greater than a FV and the rear is a minimum of 3 times stiffer.

What other possibilities are there?

Brian

[Mystique Racing]

Tire spring rates are available on the manufactures web site last time I looked.

[brian]

Mr. Harding, I would think that the vastly superior suspensions of other race cars can accommodate different spring rates. I'm not sure comparing vees to anything made in the last 30 years is relevant.

[hardingfv32-1]

I'm am sure it is very relevant. Vees have camber, CG height, and tire compliance just like any other race car. A modern suspension does some things better, but exactly when does the stiffness issue become a negative in a FV?

Assume average NE dry track conditions.

Brian ( Or do I have to change to Mr. Harding?)

[CitationFV21]

I'm am sure it is very relevant. Vees have camber, CG height, and tire compliance just like any other race car. A modern suspension does some things better, but exactly when does the stiffness issue become a negative in a FV?

Assume average NE dry track conditions.

Brian ( Or do I have to change to Mr. Harding?)

Busy with other non-racing things, so I can't spend too much time, but the CG and more specifically the roll centers might give you your answer.

Our roll centers are fairly high, and a high roll center with a high roll stiffness results in large weight transfer. Since our suspensions are not sophisticated enough to deal with this, and our tires very narrow, everything we know about handling might be wrong....I have heard of stiffer front roll bars reducing understeer becuase the interaction with the zero roll in the back. Not what you would expect..

And I always wondered what happens if we get the GC LOWER than the roll center in the back. Does this lead to reverse jacking?

I think we need some slow motion cornering movies of Vees. The stills I have of my car show too much roll in the front, and a distinct lack of jacking in the back.

Now that we can use height adjusters, wonder if anyone has tried double spring packs in the front, with external swaybars?

Come to think about it, I wonder how these things handle at all.

ChrisZ

[Mystique Racing]

I like the " Mr Harding" it shows respect and solves the double Brian thing. I would suggest that we need to also use " Mr. Macarthy" for Brian "M"

[brian]

I'm too old to appreciate being called MR. Especially from a cute young thing!

Regarding front sway bars, I have run large front bars for a very long time and they do really help front roll and understeer. I ran a stock VW external sway bar on my Caracal back in the 80's. I think it's better than making the rear loose to reduce push. I have developed a hollow chrome moly bar using Sprint car technology that's quite strong and still fits in a h-beam legally.

Many use droop stops on the front to help reduce the amount of total roll. The problem with this approach is that stops have no effect on rate of roll and don't have any effect until the end of travel.

By using a stiffer sway bar and higher rebound control in the shock, you can control both the amount and rate of roll.

Not sure of cg and jacking. If memory serves me, our cg is already below the rear roll center. Cars with tilted engines ala Lazier should have very low cg's but they have other issues. I have noticed that lowering a vee can change the handling and sometime not for the better. Way back when, we all lowered our Lynx's by raising the h-beams and some cars lost their handling in the process.

What really effects how our zero roll cars behave is the location of the rear control arms to the frame. Especially the forward facing arms. Height, width and relative angle, when compared with the axle tube is very critical. I've seen a lot of cars with roll steer and subsequent snap oversteer problems associated with these links. When the first upgrade kits came out for the Lynx's, they had roll steer designed into the new rear sub frame. By lowering the pickup point, I eliminated the roll steer but that exposed the inherent push of a zero roll. That's when i got started playing with the front sway bar.

[robert]

A nice soft front spring allows (requires?) one to run a very stiff anti-roll bar. The very stiff anti-roll bar resists roll by shoving upward on the inside wheel . . . transfers some inside spring rate to the outside. Or . . . as the outside suspension compresses, the anti-roll bar tries to compress the inside as well. That all sounds fine, it reduces roll, but at the expense of the inside wheel's ability to follow the track. The inside wheel can't readily droop when the anti-roll bar is doing its job.

Probably not an enormous issue with zero roll because corner weights are reasonably unchanged . . . However; if compliance is important (and I think it is) then any time (however brief) that the vertical load on any wheel is reduced, total grip suffers. I have no doubt that anti-roll bars reduce compliance at the inside wheel.

At any instant when the inside wheel loses vertical load, the vertical load on the outside front increases. The greater the difference in vertical loads, the less total grip across the front wheels. If you question this, you need to study the nature of tires before you debate chassis setup. EVERYTHING (almost) done to the chassis should be done with the goal of maintaining compliance at each wheel . . . reducing load transfer to the extent possible.

Add a second spring pack, and roll is resisted by the stiffer springs. The outside suspension can't as easily compress, yet the inside suspension now remains free to follow the track.

Additionally, with the added front spring rate, suspension travel is reduced . . . . meaning toe and camber changes are reduced. If you believe lowering the CG is beneficial, a stiffer car can be run with less ride height. That equals a lower CG.

A comment was made suggesting that driver comfort was important . . . perhaps "driver comfort" was describing a predictable car, not one with a soft ride. A comfortable car (to me) would be one in which the driver could complete a race and not have to remain in the car for a period of time before he could trust his wobbly knees to support him. In other words, a car that willingly did as the driver wished.

[robert]

By the way . . .

Everything Brian M. says, or has likely ever said, is valid, and is based on his actual experience. Hard to argue with that . . .

Although roll needs to be addressed, I think stiffer springs might be a better solution than the stiffest bar one can legally install. As I describe in my prior post.

I used to think raising the beam to lower the rest of the car was viable. However, I now feel a lower beam, employing control arms at a greater angle to lower the car is is better. The quite high weight of the beam is lowered, as well as the driver's butt. . . . contributing to a lower CG.

The greater angle on the control arms gives a slightly higher wheel rate for the same actual spring rate, which helps resist roll, and reduce wheel travel.

[jpetillo]

Our roll centers are fairly high, and a high roll center with a high roll stiffness results in large weight transfer.

This is true only during the transition into a turn. Once you're settled, then it should only be the CG height that matters.

I have heard of stiffer front roll bars reducing understeer because the interaction with the zero roll in the back. Not what you would expect.

No, it isn't what I'd expect. I'd like to hear an explanation of that. Any ideas? I haven't measured the CG height of my car because my scale isn't accurate enough. Does anyone have some numbers?

And I always wondered what happens if we get the GC LOWER than the roll center in the back. Does this lead to reverse jacking?

No, not on a zero roll. On a zero roll, all that contributes to jacking is the geometry, the CG height, grip(different on each tire as Robert says) and G-force. However, the case you mentioned wouild try to roll the car in the opposite direction, so the front would be happier.

John

[brian]

Robert, did I read your post wrong or am I confused about a sway bar pushing up on the inside tire? If the car rolls to the outside pushing up on the outside, doesn't it push down on the inside to try to equalize itself? I have always thought that when a car rolls, the sway bar winds up creating a twisting action that uses the opposite reaction on the other side to level the car. Maybe Bruce needs to jump in here and splain it to me.

Every time I have run a stiffer front spring, my cars have gotten nervous and the front end skips over the bumps. When I talk about a softer setup it is still well beyond a street spring rate and not really that soft either.

[jpetillo]

Robert, did I read your post wrong or am I confused about a sway bar pushing up on the inside tire? If the car rolls to the outside pushing up on the outside, doesn't it push down on the inside to try to equalize itself? I have always thought that when a car rolls, the sway bar winds up creating a twisting action that uses the opposite reaction on the other side to level the car. Maybe Bruce needs to jump in here and splain it to me.

Every time I have run a stiffer front spring, my cars have gotten nervous and the front end skips over the bumps. When I talk about a softer setup it is still well beyond a street spring rate and not really that soft either.

Brian M. (it will take some time to think of you as Mac), Robert is right. When the car rolls to the outside, the outside tire goes up (relative to the car) and pushes up on the torsion bar. The torsion bar then pushes up on the inside tire. I always think of it as picking up the inside tire allowing the inside to drop - keeping the car level. John

[robert]

Some comments were made about the CG below the roll center, suggesting the car would try to roll in the opposite direction.

No matter the roll center (axis) location, it is not a mechanical thing. A roll axis is a predicted thing, and often not too accurately predicted. The axis about which a car rolls is real, but it is constantly moving about. Clearly if the roll axis can move, it is not physically "linked" to the car.

The car (CG) does not lean against the roll axis, it leans against the suspension linkages. Forces passing along those links dictate the axis about which the car will roll. A car will not lean (roll) into a turn, no matter the CG to roll axis relationship. Now if you could get the CG below ground level . . . . . . . .

Jacking forces exist no matter suspension design, but the geometry of most swing axle designs make it a problem. Various designs have been developed that pivot the swing axles at a point lower than the VW system. Mercedes and I think some Spitfires employed the design. Such layouts reduce or eliminate actual jacking. Jacking is the actual rise of the car, not simply the presence of a jacking force.

[ external image ]

[hardingfv32-1]

There is a compromise between weight transfer caused by CG height (ride height) and tire compliance, in this narrow example controlled by spring/wheel rates.

With a stiff suspension you will will spend a little less time in contact with the track, but the time in contact will be more productive because the tire is providing more traction do to less weight transfer. You just need to find the right compromise. Why is it that most the other race car find this compromise to be lower and stiffer?

Brian

[brian]

Whle watching F1 last night, Steve Matchet said the teams were struggling with a compromise in handling on straights vs tight sections. The preference for a softer setup in the tight sections, for additional grip he said, was at the expense of the desire for a stiff setup on the long straights at China. I found that gratifying that the best teams in the world are struggling with the same issues as we lowly vee folks.

I get lost quickly on the more technical discussions, like Robert's, because my dyslexia is much more pronounced when reading technical stuff. I do understand that in the "real world", a lot of these discussions are only approximations since so much is going on at the same time it's impossible to explain the infinite variations and interactions. With that said, I do really enjoy these chats and it's nice to go beyond my seat of the pants techniques.

[cendiv37]

A car will not lean (roll) into a turn, no matter the CG to roll axis relationship.

I believe this is incorrect. In fact my father did his PHD analyzing a suspension design that did exactly this. It made the car handle "like a boat with an outboard motor" which leans "into the turn". I will see if I can dig up some pictures of this car (modified '61 Chevy if I remember correctly) showing it clearly leaning "into the turn" with the wheels and tires clearly loaded indicating the direction of the turn is the opposite indicated by the roll of the car on the suspension. The designer of the suspension contended that the car would corner faster and safer because this design "reduced weight transfer" as clearly indicated by the chassis roll. I believe the designer also contended that it actually transferred weight to the inside rather than the outside as clearly "indicated" by the chassis roll.

A simple free body diagram indicates that this is impossible (transferring weight to the inside - without actively moving the chassis weight to the inside), but my dad analyzed the whole "Curve Bank Suspension" design to show how it did in fact accomplish the task of rolling the chassis opposite of what is "normal". In essence this was accomplished by simply moving the roll axis to a position above the CG.

I've also read (long time ago) that cars were at one time designed (I assume race cars) that intentionally had the roll axis AT the CG to eliminate chassis roll altogether. If I remember correctly, this goal was accomplished, but the benefit was minimal because the actual (total) weight transfer was not reduced because (again) it is almost entirely function of the height of the CG and the location of the wheels. CG movement due to suspension movement changes this a small amount unless the CG is well above or below the roll axis. Thus even though the car didn't roll, it still transferred weight and lost overall overall grip accordingly. In fact the cars were considered hard to drive because one of the inputs that most drivers use to sense car performance and grip level achieved had been removed. Chassis roll is/was perceptible to the driver and is typically used by the driver to judge how fast they are/were going. When that input was taken away the car seemed less predictable and to lose control more suddenly.

Things may have changed a LOT since then, but this may actually somewhat answer Brian's original question: why a soft suspension?

Personally, I think the softer the suspension, the better the grip, period. But going too soft slows down the cars responsiveness too much and also results in unwanted ground clearance problems and adverse suspension geometries (big camber changes). So in the end, it's a compromise. Go as soft as possible without getting the tire geometry all messed up and bottoming out the chassis and go as soft as possible without becoming too slow in reacting to the drivers inputs. Obviously each driver will be a bit different in what they want.

Like I suggested to Brian in all seriousness: replace all your shocks with adjustable rigid links and go drive the car. If it's faster, go for it!

another $.02...

[robert]

Bruce, as you suggest in your post . . . . CG height and track width, are the primary factors in load transfer.

So, if a car leaned inward, would it reduce load transfer? It seems well understood that "normal" roll has negligible affect on load transfer, so I doubt load transfer is reduced.

The comfort level of the car you describe would be greater with inward lean . . . fewer spilled drinks because they would tilt with the car. Just as some high speed trains have systems to tilt their cars for passenger comfort.

Preface everything I say with "it is my belief", and don't assume anything I say to be any where close to factual.

My thoughts were more focused on load transfer than roll, yet I clearly spoke of roll . . . oops . . . not the first time I've screwed up. Just because I never saw a car lean inward, I can admit it is possible.
.
.

[CitationFV21]

Found this about roll centers:

http://autospeed.com.au/cms/title_Front ... ticle.html

and this:

http://www.shoptalkforums.com/viewtopic ... 107bd1e419

and then:

http://www.neohio-scca.org/comp_clinic/ ... cs2007.pdf

ChrisZ

[jpetillo]

Wow, you guys have been busy...

Robert, the roll axis is completely mechanical and not a predicted thing. Yes, it can move around, but perfectly predictably. However, without getting into each point, I believe that several things you mentioned were not quite right.

Chris, very nice post. You are absolutely right, the roll axis can be above the CG and the car will absolutely roll the other way - into the turn like a motorcycle. But, no, it will not reduce the weight transfer in any significant way, nor would it add weight to the inside tire. Either way you roll, weight is added to the outside tire by about the same amount.

Like you said, weight transfer is pretty much predicted by the CG, but there can be a small change in the CG location as a result of roll. For a positive roll car (defined as rolling like our vees), the CG moves toward the outside and down. For a negative roll car, it moves toward the outside and up. So, a positive roll is better in this regard, but we're splitting hairs. As long as we keep our cars' roll to few degrees, this effect is negligible.

You did mention the way I look at roll. Roll is fine, except that it messes with the camber. That's why we don't want roll.

Brian M, in you post, was the stiff suspension for the long straights desired because of the excessive downforce?

Guys, please let me know if you disagree. John

[hardingfv32-1]

With in the limits of an acceptable compromise, I want the car as stiff as possible.

AGAIN...

If I am correct in stating that a FV has the lowest, by a large margin, spring/wheel rate of all SCCA slick tired classes... Exactly why is that the case? What makes a FV different in relation to spring/wheel rates?

Brian

[brian]

John, I don't remember why a stiffer suspension is desired for the straights. I think you're right about downforce. I got so hung up hearing about the new "F" duct that I got completely lost.

Brian H., intuitively I think the vee is sprung the way it is exactly because it's so crude and the unsprung weight is so high. I would think a zero roll car would not take to high spring rates very well either. A critical factor in controlling high spring rates is the precision with which the alignment and subsequent movement of the suspension is factored. Neither of which is a stong suit for a vee.