Soft Suspensions

[hardingfv32-1]

Brian McCarthy states in another post: "MY experience has taught me that a softer, more compliant suspension, is a bit easier to drive and more forgiving. I have converted several cars to a softer set up and the owners have been more consistent and eventually quicker."

I question why this would be the case. It will be said that a soft suspension provided more grip and corner force. I will argue that a soft suspension requires more ride height, higher center of gravity, and thus less cornering power. A good computer model would provide the answer, but FOR THIS DISCUSSION lets call it a wash.

So what aspects of a soft suspension make the car easier to drive? More toe and camber change from the extra suspension movement? Car less nervous? Define nervous... slower reacting?

Brian

[pillowmeto]

One possibility is that the softer suspensions absorb track irregularities better making the car more stable on odd surfaces, like those often found in repaired corners.

On a similar note, I have been working with dampeners lately and have seen a set that came from a car that was very fast at the 2009 Run Offs. The front shock's shims were so soft that the external low speed adjustment had no real effect.

[hardingfv32-1]

1) For this discussion don't worry about odd surface conditions or rain. We spend most of our time in the dry and on smooth tracks.

2) For this discussion we will AGREE that a soft suspension absorbs track irregularities better. Is this all a soft suspension has to offer?

After much research on this subject I have NEVER seen a graph that illustrates suspension stiffness (spring rate?) vs grip (not sure of the term) for a given track surface. I also can not find how such a formula/calculation is plugged into a car dynamics computer model. Just seem odd more isn't available on such an important subject.

Brian

[72jeff]

I don't know where you race....but in the NE our tracks are pretty bumpy (except the freshly paved ones)

[Bill_Bonow]

Brian,

I tend to agree with your theory, but come out to Nelson Ledges. You drive around aiming for the smooth spots. They typically put pylon cones around the suspension damaging potholes.

[Matt King]

This is really just a variation of the age-old debate between soft spring/big anti-roll bars vs. stiff springs/small roll bars that has been going on in NASCAR and other racing series for decades. Factors like track conditions, suspension geometry and specifically, driver preference, all play a role in picking the right setup, so there is no "right" answer. I believe this conversation will generate the same result.

[hardingfv32-1]

OK... here in the West we are talking billiard table smooth tracks. But, in the NE they race F500's and FF near the ground and they do not fly of the tracks.

FOR THIS DISCUSSION ASSUME A SMOOTH TRACK and we are looking for something other than "a soft suspension absorbs track irregularities better".

This is a stiff suspension in general. It does not matter how you get there: springs and/or sway bars, or shocks. We have the same general suspension geometry on most zero roll Vees.

Drivers preference: What would a driver like about a soft suspension? Less vibration in the mirrors??

Brian

[jpetillo]

Brian M. discussed suspension compliance in another thread. (Brian, I believe it was you - sorry if it wasn't.) I was first thinking that compliance would be important to this thread, but Brian H. limiting us to billiard smooth tracks I think nixes that idea. I think if you take even minor track irregularities out of the equation - where suspension compliance probably helps a lot - then there's not much difference to the result of having a stiff or soft suspension. Both will work well on billiard table surfaces. But there still are differences.

Brian H, I'm not sure that a softer suspension would necessarily have a higher ride height given a super smooth surface - no bumps to bottom out on. For a bumpy track or dips in the track, that's different. I suppose that the nose might smack the ground with the softer suspension set at the same ride height.

Okay, that aside, the softer suspension will rise up higher when cornering due to jacking, as was suggested. That will limit cornering force. Brian, I assume that you want to remove droop limiters from this as well.

Another issue is when you transition into a corner and how the car settles, and the time constant for the settling and whether the system is overdamped or underdamped. The softer and stiffer suspensions will do this quite differently, and this is hard to quantify (I don't want to sign up for this calculation). This is a balance between everything - spring, dampers, sway bars, CG, chassis stiffness, front to rear anti-roll, etc..

If we go back to just a long sweeper, perhaps the softer suspension would have a slower response time and that could help the driver deal with the rear swaying out with minor corrections without overcompensating. A stiffer suspension may react too quickly and be hard to control.

For Vees with zero roll resistance, the response time of the whole chassis roll being controlled by the front may affect this. That time constant may be long and a softer suspension may deal with this time scale better. I think this effect could be significant to what people feel.

Brian M., when people were generally happier with the softer suspension, on what part of the course did they feel it was an improvement - shorter corners, sweepers, going into corners, coming out?

Sorry - just thinking out loud, and you could argue about the thinking part. This is a good thread. John

[CitationFV21]

This was posted in another forum:

http://www.youtube.com/watch?v=OgSiK_VarK8

If you listen to the part about the being on and off the power upsetting the car, the softer sprung car is a little more tolerant of driver inputs.

Yes a stiffer car should be faster, up to the point where the stiffness interferes with roll and weight transfer (even a go kart has some flex built into the frame)

One of the problems is, people spring too soft, then tighten up the shocks to compensate. The opposite is also true, they spring too stiff and over overwhelm the shock. Most of the time the shock needs travel to work.....(and not bottom.....)

In a FV, wheel rates are pretty easy to figure out in the rear, a real pain in the front! And with zero roll and torsion leaf front suspension, hard to change and adjust.

Rising rate suspensions are tricker to drive than the reverse. That is why the Citation and D13 are considered so easy to drive. Most Vees are not rising rate, although I am not sure of pull rod cars like the Adams.

So you want to run a car as stiff as necessary and as soft as possible.

This is so much fun to watch: http://www.youtube.com/watch?v=n6aNnZ4q ... re=related


ChrisZ

[hardingfv32-1]

Bottoming under heavy breaking is probably your most common ground clearance tuning issue.

Shocks and suspension response times: Soft shocks do not restrict the suspension's movement as much, so the suspension responds faster. A stiff shock slows the suspension.

Soft springs allow more roll or jacking which means more time required to stabilize. I would estimate that you could set the car to reach the same optimum settings, but a soft suspension is just going to take longer to get there.

Questions:
IF we ASSUME that the cornering performance is perfect at the stabilization point (where the car takes a set), why do I want to waste time getting there? It would seem the more time at the optimum the better.
Would I not be better off with more frequent and smaller in corner corrections? A slow responding car reduces the number of activities I can complete for a given amount of time I have in a particular corner.

Brian

[CitationFV21]

....
Questions:
IF we ASSUME that the cornering performance is perfect at the stabilization point (where the car takes a set), why do I want to waste time getting there? It would seem the more time at the optimum the better.
Would I not be better off with more frequent and smaller in corner corrections? A slow responding car reduces the number of activities I can complete for a given amount of time I have in a particular corner.

Brian

One of the problems is that we cannot quantiy stiff and soft the way we can quantify tire pressure. This is no overall chart for stiffness unless you map one for your particular car. Could I put a 300 lb spring on my Citation? When I was helping with FF in the 80's, we went from factory 180 lb springs to 300 and I think they ended up around 220. I think there was a test in either Racer or SportsCar where they tested 2 Crossle FFs, one with stiffer springs and one with a softer set up. The softer set up was almost as fast, but for a less experienced driver, the difference was smaller.

"Would I not be better off with more frequent and smaller in corner corrections? " That assumes the corrections are smaller. What if they are shorter in duration, but larger in amplitude? Then the driver has to have faster reflexes and be more trained to handle them (I think that is why karters and motorcycle riders make great racers) Imagine of drifting vs a tank slapper. You have to have confidence in the car to push it to it's limit. The driver has to feel in control.

A famous mechanic, Chris Wallach of MRE (Where the D-13 was buildt for a while) took a Skip Barber School around 1980. When he got back to the pits after one session he exclaimed "so that is what trailing throttle oversteer is! (in the LR downhill)....I guess I should listen to a driver who is complaining about the cars handling". Chris could do perfect setups, but theory runs up against reality. Today with data acquisition, we can try to show a driver how to get closer to a car's limits, but we can't make him drive there.

We don't have the time or money to do extensive testing at every track under every condition. So we need to find a good compromise in our setups so we can work on driving the car. If the car is 100% but the driver is 92, it will not be as good as a 98% car and 98% driver.

I think one driver nailed it last year at the Runoffs - he was a close to 100 % in chassis, motor and driver as you could get.

ChrisZ

[robert]

Harding has it pretty much spot on, for the reasons he states. And . . . it probably is more applicable to a V than cars with more suitable suspensions.

Suspension travel on a vee causes unwanted wheel alignment issues. If only for that reason, limiting travel should prove beneficial.

A soft car reacts more slowly to driver inputs . . . the driver must learn and compensate for those delays. For example, the driver of a soft, sluggish car probably turns in earlier, but the car doesn't actually respond immediately. No problem, right? The driver just learns the car. Except there will be times that the driver will likely not have the confidence needed to race wheel to wheel, and common sense will dictate he back off rather than risk a maneuver. The big difference is a car that can be raced, in other words it will go where the driver intends, when the driver asks. Many cars can be driven quickly on a clear track, but are too spooky to race wheel to wheel.

The stiffer a car, the more critical its set up, or the more easily its set up can be dialed out.

I have no doubt that a low , stiff, well dialed in car, will be quicker and more confidently driven. I also have no doubt that a motivated driver can learn, and then drive most any car quickly. Another thing I have little doubt about, many drivers beat them selves up over their driving, when they should be addressing issues with their car.

But . . . where should the longitudinal CG be located?

[cendiv37]

Tell you what Brian, why don't you do an easy experiment? Take it to the limit.

Replace all three shocks with turnbuckles, set your ride heights and cambers the way you want and go out and drive the thing. I suspect you'd find that even your billiard table smooth tracks would suddenly seem incredibly rough with little grip.

I suspect you would finally accept that you can make it too stiff. Now how much softer you should go? Depends...

Why do we need suspension compliance? Simple: to keep the tires in contact with the road as much as possible and to minimize the load changes the tires see so that their grip stays more constant over time which allows the human being at the wheel the time required (about 1/2 to 3/4 sec. typical reaction time???) to react to the grip changes before they cause loss of control. Oh, and it's a lot more comfortable to drive too. Even Jackie Stewart said that being comfortable is extremely important to driving fast. Getting beaten senseless over every tar strip doesn't lead to fast lap times...

Why has the F500 community spent countless hours and $ developing "real" suspensions on cars that were conceived to not allow them? Because they are faster with suspensions than without!

my $.02

[hardingfv32-1]

Bruce... To keep the subject narrowed, I stated from the start that a "soft suspension absorbs track irregularities better". So, beyond that, does a soft suspension offer any other positive traits?

Chris... Answer the question: "IF we ASSUME that the cornering performance is perfect at the stabilization point (where the car takes a set), why do I want to waste time getting there?"

Driver corrections: REGARDLESS of the speed of the driver reflexes is he not better off getting the car set as soon as possible? This would give him more time to respond for a given amount of time spent in a turn. The driver cannot make any judgments until he has crossed the edge.

Brian

[CitationFV21]

...Chris... Answer the question: "IF we ASSUME that the cornering performance is perfect at the stabilization point (where the car takes a set), why do I want to waste time getting there?"...
Brian

That would be correct if we are running a skidpad, (or a certain Carousel..); however, we spend most of our time transitioning from straightline to braking to cornering to accelerating. As I think about it, most drivers don't crash in the middle of a turn (unless they do something like lift or brake) but rather coming into or out of the turn. Which suggests that the transition from straightline to cornering and vice-versa is important. You might want oversteer going in and understeer coming out.....

I am not really disagreeing with you, I said as stiff as necessary, as soft as possible. Once you go past the point where stiffness assures proper geometry, you start taking away from the transition. A car that is perfectly balanced on a skidpad, may be diabloical entering a decreasing radius turn.

Disclaimer - wings, ground effects, active suspensions, etc. none of which have anything to do with FV are a whole different ballgame. F1 cars today are very different from Indy Cars or Stock cars - go watch Hamilton in his go kart days....

ChrisZ

[CitationFV21]

..Driver corrections: REGARDLESS of the speed of the driver reflexes is he not better off getting the car set as soon as possible? This would give him more time to respond for a given amount of time spent in a turn. The driver cannot make any judgments until he has crossed the edge...

Brian

He can make a lot of corrections getting to the edge. The car (and tires) work different depending on which side of the edge you are on. If you are 100% on the edge of handling, there is nothing available for braking or accelerating. If you are under 100%, then you have some choices.

So if you get to the edge sooner, and find you have made a mistake, then there is no room to correct. When I saw Danny Sullivan hit the wall at Indy, I thought, why did he not just steer away? But once you have gone into Turn 1 at NHIS and realize the mistake you made entering the turn is going to bite you exiting the turn, 8 seconds later..., you develop a little more respect....

A better driver gets to the edge and can handle a it better. The premise you put up is correct, but I belive what brought this up was the statement that most drivers find a soft setup easier for them to drive, so they go faster. Not that it is ultimatly the fastest way around a track.

ChrisZ

[hardingfv32-1]

On those rare occasions where my driver goes to the edge and stays there the DA system shows he will be in a steady state suspension geometry condition for almost half the turn. So in general the middle of the turn is about as important as the transitions.

Oscillating from under to over the edge is the only way I know of to insure that you are in fact "on the edge". A closed loop feed back system? How fast I get to the edge has nothing to do with what comes next. But the sooner I get to the edge in a turn the more time the driver has to make the constant corrections that are required to maintain the process of "staying on the edge".

The whole point of this post is to quantify what it is that makes drivers happy with a slow responding car. Do the drivers have something else to do while the car is taking a set? I'm honestly missing what could be so comfortable about a (relatively speaking) sluggish car. I'm up for learning something.

Brian

[brian]

Wow, this really triggered a great conversation. I really liked Bruce's points about compliance and keeping the tires on the ground. Since it makes my head hurt, I'm going to avoid billiard tables and graphs and say that the most important thing in making a amateur driver faster, is confidence. Let's face it guys, we're amateurs; some of us are really old amateurs. That said, the quickest way to improve a driver's confidence is making the car more predicable. It is extremely difficult to find a car's limit if the car reacts unpredicably, or differrently each time it receives inputs. This is all the more true of vees. Our cars are so crude and challenging I consider making a vee predictable the greatest challenge. Don't forget, we don't drive vees, we negoiate with them.

Softening the springs is one way to make the car more predicable. So is increasing toe in. Now I'm not saying to make the car into a balloon and yes, the shocks have to be balanced to the spring rate. But running a stiff nervous car really requires an experienced, and probably, a very young, driver.

I often increase the toe setting for my clients as well. As was mentioned earlier, transisition is one of the most difficult times for a race car and toe in settings help the car in transisition. I like to say that it's better to loose a tenth or so with a greater toe set than to loose 20 seconds to a spin. Believe me when I say, 1/8 toe in at the rear will not kill your times; but it will make the car a lot more driveable and improve your consistency.

A predictable car makes for a better car during the race. As we all know, during a race there's a ton of things going on and the last thing I want to think about is what my car is doing.

[hardingfv32-1]

EXACTLY what is so predictable about soft springs? Remember compliance is NOT part of the thread. It is agreed that "a soft suspension absorbs track irregularities better". Is predictability tied to absorbing track irregularities? After a few laps the bumps are still surprising the driver??

Why would a stiff suspension cause a "car to react unpredictably, or differently each time it receives inputs"?

Brian

[CitationFV21]

...
The whole point of this post is to quantify what it is that makes drivers happy with a slow responding car. Do the drivers have something else to do while the car is taking a set? I'm honestly missing what could be so comfortable about a (relatively speaking) sluggish car. I'm up for learning something.

Brian

I like the other Brian's quote about "we negotiate with them....."

I think the word that keeps coming up is confidence. Softer springs, more toe in, all give the driver more confidence that he is driving the car, not the other way around. Since a stiffer car will resond faster, it works both for and against a driver. If he make the wrong input, the car will do exactly what he wants, without asking. Think of the crashes in Indy Car at the start of the ground effects era. The driver would get slightly sideways, correct, and the car would overreact and put him right into the wall. We lost a few good drivers that way.

Now remember, we are not talking Showroom Stock here. At some point, the car gets so soft that you loose more handling than you gain in confidence.

I have been scanning my Carroll Smith Books and other to find something that might be more technical. Read pgs 10 - 11 in Engineer to Win for some insights. I think that a stiffer car transmits weight transfer faster. The driver is a computer, analyzing all the input, and the brain of an average driver cannot digest all the information in time to make the proper decision. A softer car gives more feedback of what it is doing and what it is about to do, and this information is easier for the driver to work with. This is a driver limitation.

Again, another extreme example, the Indy guys in Texas a few years ago. The cars could go around the track at 5G, but the drivers were passing out.

I have always thought that most FV have too slow steering. it might also be that physically we cannot turn the wheel fast enough to react to a very stiff car.

Chris Z

[hardingfv32-1]

"Since a stiffer car will respond faster, it works both for and against a driver. If he make the wrong input, the car will do exactly what he wants, without asking"

So... in a soft slow reacting car, a wrong driver input is going to be LESS wrong. Or is it going to be the same mistake but just delivered at a later time?

Again, what is the driver doing while he waits for the input to the car to be completed? Feeling comfortable and confident everything is going his way??

A drivers limitations do not start until he has received information from the car to react to. Aren't these activities done in series? Driver input, car response, next driver input, car response and so on.
The driver's input and the car response each have their own place along the time line. They do not overlap.

Brian

[CitationFV21]

Brian,

I know you want to quantify something that might be unquantifyable.

In NASCAR, Indy Car and F1, drivers on the same team might have vastly different setups. You would think on this level there should be no difference.

So to finish my contribution for now, here is a link to a F1 similar discussion:

http://www.f1technical.net/forum/viewto ... 63&start=0

ChrisZ

[jpetillo]

Hmm, it's not clear to me that a softer suspension needs to have a slower reaction time. With bikes, we'd shoot for a spring rate that gave us specific sags with and without the rider, and also have room to not top out. We'd then adjust the rebound and compression damping to give a certain bump and rebound settling times. We'd then make the next level of modification so that the front and rear bounced together. That was our starting point.

If we went to a stiffer spring, we'd play with the suspension setting so that it reacted in about the same time. We'd soften the compression damping so that the bike would move about the same as it would with a softer spring and in a similar time. This would mimic what would happen to the tire when a bump was hit. We'd increase the rebound damping so the tire would rebound in a similar time as for the softer spring. But, what we did for bikes trying to keep similar settling times may have no business for cars.

Brian H., I'll try to shed some light on your question about what happens during settling and settling time. I don't know it if will help, and I'll certainly be telling you something that you already know, but maybe it may bring up some issues that could jog a thought or memory. This will be based on the response of a standard damping oscillator (if I remember it right). If I have a spring of a certain spring constant holding a certain mass on one end and sitting on a platform, and then I instantly move the platform up to a different position faster than the mass can respond, one of three things happens.
- the mass will settle after it oscillates around it's final position - this is under-damped.
- the mass will very slowly settle to the final position - overdamped.
- the mass settle right to its final position as fast as it can without overshooting: critically damped.

So, let's concentrate on the critically damped case, which, let's assume, we'd guess to be our optimum. When you're critically damped, the relaxation (settling) time is 1/(2*pi*freq), where 2*pi*freq = sqrt(k/m), k being the spring constant. That tells us that a critically damped softer spring has a longer critically damped settling time. In general the settling time is 2*m/D, where D is the damping coefficient. When the system is critically damped, the damping rate is D = 2*sqrt(m*k). So, for critical damping the damping rate and the spring constant have a defined relation. This tells us that increasing the damping or the spring constant decreases the settling time. (So, relatively speaking those motorcycle guys were increasing the settling time on compression (underdamping) and decreasing the settling time on rebound (overdamping) when they moved to a stiffer spring.) This definition of settling time is the time it takes for the system to get to within a certain percentage of its final value. So, it may overshoot and oscillate, but will stay within some range.

I'm not sure this is applicable, but it should be resemble the basic (lowest order) operation since it's such a basic theory.

So, now what happens when you initiate a turn? You first turn the wheel to begin the rotation, and then you turn it back towards straight (if your in a four wheel on-the-edge drift maintaining the magic slip angle for maximum cornering force) to keep it in a certain new state. So, there were two actions, one to transition over to the new state, and then a second to keep it in the new state.

So, what's the definition of a stiffer suspension in this thread (not the definition in this post), a higher spring constant that shortens a critical damping time (assuming that the system is critically damped), or a higher damping rate, which gives a shorter settling time in all cases - critical or not? Note that a higher spring constant does not affect the settling time, only the damping rate does that. Also note that the damping rate does not affect the critical time (critically damped), only the spring constant does that. Of course a stiffer spring and a higher damping rate results in a shorted critical time and a shorter settling time.

In all of this, the mass is assumed to be constant.

Brian H., I think this applies, but before we decide how to let the math tell us what might be happening, we need to figure out what our definition of stiffer suspension is. Do you have an idea? Also, do you have any idea what the ratio of the rebound vs. compression damping rates are?

I hope this helps. I think it should be right. Let me know if I've strayed - it's late.

Oops, I see it's a bit too long a thread - sorry!

John

[Mystique Racing]

Ok, one of the two Brian's need to change their names. It's getting me confused.

Sorry to hi- jack this interesting discussion.

[brian]

Sorry about the confusion Scott. Just remember I'm the good looking one