[stretch]

Originally Posted by DuncanG

Here's the link for that pdf - http://www.vehicledynamics-expo.com/...y_3/nowlan.pdf
Hmm, different!

While doing the search I came across many threads on different forums each from a single-post user saying how ****-hot these guys are - .

lol, good find. I see that too now. That doesn't discredit their claims but it sure doesn't make 'em look good, either.

[ace996]

Comments in Red below...

Originally Posted by stretch

Tom, this isn't a competition. -Stretch, I agree...just a healthy difference of opinion with a little debate added... Why are you acting like mathmatical predictions and you results are mutually exclusive? -it appears that "your" mathematical predictions and my results are exclusive... If done properly, they'll agree. - yes, or they should be close...a good point in which to adjust from real-world testing/experience. With that said...run some numbers for stiffer springs on those Ohlins...500/400...and let's see whathappens. If you're dead-set on running soft springs, then the dampers will need to be re-valved.

Nobody has said the RCE struts are overdamped because there is no dyno of them. We can't even begin to discuss them without that as a starting point. Perhaps the reason everyone likes the RCE T2's is because they're the only currently sold coilover setup that isn't overdamped? - I doubt that, it's probably more the function of proper adjustment and the ability to match those adjustments to the different spring choices. Again, mount some stiffer springs on the above plotted struts and the numbers will change...I also have never heard of an Ohlins owner complaining about "overdamping". If I wasn't a driver with a focus on track performance, I'd probably would have stuck with the 500/400 rates or slightly lower...and those who chose the lower rates have also had nothing but good things to say.

You're fine with being fast.-that's my goal I want to know why it's fast. You have data showing you are fast but no data showing why.-Why?...because I've used tried and tested parts that allow my suspension to function at a high level of performance. Why?...because I've tailored my suspension to how I like a suspension to "feel"...that allows me to be more confident and drive faster, even whilst sliding the car at very-entertaining speeds. You showed me the ends but I care about the means.- You've also obtained your ends by going brute force with immensely stiff springs-your opinion...and not a fair description of them....TRY a properly dialed-in suspension with 600/500 or 500/400 springs and you may be quite surprised at how they are not even close to a "solid suspension" (the "solid suspension" approach) and that's not really the topic of the thread. This is not a thread on springs-yes, but the "critical damping" plots are only a result of a given spring rate....so you must take spring-rate into consideration...and sway-bar size is also MANDATORY to consider. Please open your mind to the possibility that the Ohlins appear "overdamped" only because the swaybar rates are too low or not taken into account at all....really, think about it...!!! This thread is for discussion the optimal damping ratio for a given set of springs, nothing else-again...dampers and springs do not work on a car without regard to sways. . Please, let this thread form a useful conclusion or come back with some data that actually contributes towards finding it-your useful conclusion is only one that supports your computational theory....that's why you can't figure out why all the struts are "overdamped"...you have a belief that soft springs and soft sways lead to an "optimum/ultimate" set-up based on what you've read from books and online papers. Critical damping....spring frequency....if your findings contradict what 95% of the competitive drivers of the STI are using then there is a problem or a disconnect with the application of those "guidlines".. Comparisons of lap times on different damper adjustments would be pretty interesting. I'd really love it if you cooperated since you have a good data logger and could contribute.-I'm going to Limerock on Thursday....would it make you feel better if I backed my rear struts to full stiff compression and full stiff rebound and showed datalogging evidence that I'm oversteering throughout the whole track? I've pretty much found "MY" ideal set-up with my car that allows me to drive it as if it were connected...hard-wired if you would...to my intentions. It is phenomenally neutral on throttle, slightly oversteering on trail-brake, and turn-in is as good as my last mid-engined MRSpyder.

How do we know there isn't a better solution yet to be found, perhaps even one that doesn't kill daily drivability?My T2s with 500/400 was just about "my" perfect daily-driver...I bet with 450/400 or 400/350s, they'd be the most amazing street/track compromise available. Other than that, I don't know....but I hear those Ohlins are pretty nice, too....

With a shock dyno, 4banger could predict your ideal damper settings and you could compare the times on those with your current settings. But like I said, maybe the RCE T2's are one of the few that aren't overdamped and that's why they're so fast, nobody knows! Answer THAT question for us!-Fine...you're finally going to pull it out of me...THEY ARE NOT OVERDAMPED!!! I RUN 600 POUND SPRINGS IN THE FRONT AND 500 POUND SPRINGS IN THE REAR!!!!!! HOW COULD THEY BE OVERDAMPED??? And if they were...would I be able to hold lateral G's like I do in the above dataloggs?...even in the wet??? I don't think so.
I had Koni Sports on my old MR2Spyder and when they were turned just a 1/4 turn too much, you knew they were overdamped...you could feel it...no dyno needed. But when a damper is dialed-in...and a car suddenly becomes an extention of your intention...you KNOW you found the setting. That's how I know, Stretch.

I hope he posts it, but 4banger just e-mailed me a document from ChassisSim software. They modeled a Porsche 911 and had new shocks made to the software's specifications. They designed for the parameters of 50% critically damped for low speed bump and just 30% critically damped for everything else, equating to roughly equal bump and rebound. That's OUTRAGEOUSLY soft (less than half of Ohlins, above) yet it dropped 4 seconds off the car's lap time!- Again....load those calculations with 500/400 rates or 600/500 rates and see what those Ohlins come in at on the "critical damping"-o-meter....I bet we'll all be surprised. And again, either you are modeling too soft a spring for the given damper or you need to re-valve the damper for the softer spring.

I really think you can provide some great information, Stretch, but it appears that you get a little hung-up on the details. After all, this all comes down to having a well-handling car...and that's been done. You want to know "why"?...open your mind to other possibilities and other 'benchmark' settings, then work from there.

My best regards,
TomK

[Splash]

Heh, did anyone notice that, in the 911 case, they simulated one track, then got the 4 second reduction at a different one? (perhaps the new track was 4 seconds shorter...)

In my experience, not only can the surface change a lot of this, the average speeds can too...

For example, my car is primarily an AutoX car, and pretty much has to handle neutral-to-loose anywhere from 20-75mph. If I take that car to Mid-Ohio, it would be a twitchy, uncontrollable, beast at the higher speeds. If I adjust the car to handle neutral at Mid-Ohio, I might as well paint it green and yellow, and put John Deere on the side for as much as it would plow on an AutoX course.

I see it all the time for road race folks who bring their cars out to play at an AutoX. They all have massive push across the board.

Note, the surface between the two can be the same, though Mid-Ohio is smoother than a lot of the places we AutoX on, but we do have a smooth one. The biggest differences there are simply the vehicle speed and the sharpness of the corners.

I almost liken tuning the handling as being very similar to sizing a turbo or choosing a camshaft profile. There is no solution that is perfect everywhere in the complete band of operation, you tune for a particular range of that band.

To that effect, if you want street springs/bars, you need street dampers. Even if you don't count the bars (which Stretch doesn't), Ohlins might make theirs overdamped simply because they are using the most open piston they make with the lowest springs they think can get away with and still call it a street/race coilover. They are, after all, a racing damper company...

Heh, plus, it's easier to mask overdamping than underdamping... <grin>

[waktasz]

Based on your comment about the Ground Control/Koni shocks, what would the 'perfect' spring rate be?

Also, did you use the SA or DA Konis?


(wishes I understood wtf those graphs meant)

[slim9300]

This is a very helpful (and entertaining) discussion. Thanks 4banger.

[4banger]

Originally Posted by waktasz

Based on your comment about the Ground Control/Koni shocks, what would the 'perfect' spring rate be?

Also, did you use the SA or DA

They are the Koni single adjustable. I'm trying to add some more data points to the vairous Koni graphs.

It has been a long time since I've looked at the GC stuff. I wouldn't really want to make a recommendation just based off of the full stiff and full soft graphs. I think 400/300 is the softest that you would want to go, but let me sleep on it and try some other spring rates out. I don't remember the max off the top of my head, but more than plenty to run just about any spring you would want.

[ace996]

Originally Posted by 4banger

They are the Koni single adjustable. I'm trying to add some more data points to the vairous Koni graphs.

It has been a long time since I've looked at the GC stuff. I wouldn't really want to make a recommendation just based off of the full stiff and full soft graphs. I think 400/300 is the softest that you would want to go, but let me sleep on it and try some other spring rates out. I don't remember the max off the top of my head, but more than plenty to run just about any spring you would want.

4b, great to hear you'll be tryng other springs. From my experience with Koni, they are quite adept at handling springs from a very wide range. Often times, they can be adjusted to softer than stock to up to 300% stiffer or more.

waktasz,
The "perfect" spring depends on the purpose of the car. Once you have an idea of what type of needs you have, then you can tailor the strut to the application. Thankfully, many of the adjustable struts work with a good range of rates.

Again, 4b...if you're going to try other springs, can you also take into account the effective rates of different swaybars? That'd be great.

Be good,
TomK

[MGizzle]

Wow. I am gone for a month and come back to this thread. Sick.

Anyhow, I think what needs to be mentioned here is that you need to tailor the damper TOWARDS your springs, not the springs towards a damper. Steady state is what needs to be taken care of first and all the great links you guys attached talk about basic calculations NEEDED for creating a good intial car set up. Based on what frequencies you want your car built for and based on what cornering stiffness etc. you want, you decide on your springs. Once you got your springs critical damping is not that hard to figure out. You test a coil with the spring and see how long it takes to settle. Play around a bit and you will see that around 65-75% critical damping you will get the fastest time for the system to get to equilibrium. Once you have this done that is it. You tailor your damper to work for your springs.

I have seen quiet a few matlab programs that calculate the critical damping based on your springs but testing as in the intial graphs is the way to go. I hate programing so I don't pay much atention to the programs.

Anyway, it is very hard to pick a damper/spring combo as a consumer such as us who don't have access to all these tools. Hence, I think using math we can get as close as possible. If you have some sick money to drop buy some damper potentiometers and look it up. I remeber at the optimumG seminar I took Claude Roulle talked about potentiometers as if they were sent by Gods. After using them, I know why.

Now I forgot what I wanted to say. Oh, you can't call a damper overdamped because there are infinite spring choices out there. There are plenty of places that will make you CUSTOM spring rates hence you need to figure out the rates you want first. Than you can start picking dampers based on which one settles fastest or in short, which one is close to 65-75% critical. Don't drop money on expensive dampers like the Ohlins and than just pick some springs you think will be nice. Try and do the basic calcs for frequencies, make sure the front is lower by about 15-20% than the rear and find out what your cornering stiffness is. You can use the understeer coefficient equation to find out if you are oversteering or understeering. A simple constant radius test will tell you this too A LOT MORE ACCURATELY.

I used to bust my behind on trying to figure out what to valve a damper too and never spent time on looking at the basics, such as steady state and spring selection. Once you have the springs that work, making a good damper become significantly less challenging.

Great thread.

[stretch]

Good to have you in the discussion, MGizzle. I agree, damper needs to be tailored to the springs, not the other way around. It sounds like Ace has a beef with the spring rates we're modeling but they're the rates that come with these coilovers. We didn't pick 'em, it's what they ship with. (Switching springs is a different topic, though, so I'm not going to comment on spring rate choices in this thread.)

Anyway, sway bars seem to keep coming up in the discussion. I thought I'd try tackling that subject.

Sways play no part for a two wheel bump, so we only need look at a one wheel bump and body roll.

I have three thoughts here:

1) Is it bad to be underdamped for a one wheel bump? The chassis will remain stable due to the other two or three wheels not hitting a bump. This seems like an opportunity to tune for comfort without a stability penalty.

2) How damped does body roll need to be? In a perfectly steady turn, you'll have a constant amount of body roll and thus no motion to be damped (aside from bumps, which would behave like normal bumps). We only need to worry about damping transitions to and from body roll. I think the driver somewhat self-damps body roll by modulating the steering wheel; furthermore the rate at which this transition happens is limited by the speed in which the driver can turn the wheel. We often hear body roll happens at damper speeds of 3-5in/sec whereas bumps are 5-20in/second, so body roll does load the suspension more slowly. This means any "rocking" caused by the inertia of the body would be small (at least relative to the bounce of being underdamped for two wheel bumps); furthermore it wouldn't take much damping force to minimize it.

3) A sway bar also links the left and right struts. If the sway bar was infinitely huge, we'd need no damping changes because every bump would be a perfect two wheel bump. A sway bar can resist force on one side against the force (spring) on the other side of the car. This gets a little complex and I haven't really thought about this to too much detail, so here it goes. I'm going to think out loud a bit.

Let's assume 400lb/in springs and 400lb/in sways (rate for a ONE wheel bump, double that for body roll). This is somewhat like having TiC Coilovers on Whiteline 24mm sway bars. Let's say we have one inch of compression on a one wheel bump. The total spring rate is 800/in, so we have 800lbs of force wanting to react to this bump, 1/2 of which comes from the sway bar.

The sway bar always has equal but opposite forces on either side of the car, though. So, that 400lb/in is really 400lbs downwards on the bump wheel and 400lbs upwards on the opposite wheel. Thus, the instantaneous rate on both sides of the car has increased by 100% but this will decrease as the opposite wheel reacts.

What will then happen is that the suspension will try to reach equilibrium since we have 400lbs of upwards force on the opposite spring. Depending on the duration of the bump, either the bump wheel will recover of the opposite wheel will start lifting. Once it lifts half an inch, the sway bar will only have 200lbs per side of force (400lb/in x 0.5 inches) which is equilibrium for half an inch of compression on the opposite 400lb/in spring.

So, if you raise one wheel up one inch, you will get a half inch lift of the opposite side when your sway bar equals your spring rate (for a one wheel bump). Total rebound forces thus drop since now the sway bar isn't tensioned as much- we now have a 50% increase in rate instead of the 100% we had instantaneously during the bump. The trade-off is now we'll have movement at two wheels, but as far as damping goes, that's much more predictable.

The inside tire will only lift if it has time to do so. I suspect that on really small, quick bumps the inside tire lift is minimal, although it'll always happen to some degree.

Plotting the numbers for the increase in spring rate due to sways, %cd decreases by exactly 30% for the above example if the inside wheel does NOT have time to react. It's obviously less if the opposite wheel can react because then forces are distributed somewhat among the two dampers.

The question then is: how big a deal is this? The chassis is not going to react much to just one wheel being underdamped because it is stabilized on the other wheels. I would think you would want to be underdamped for a one wheel bump since it seems like an opportunity to do so without upsetting the chassis.

Sway bars have a much higher rate for body roll, double that of one wheel bump, since we have opposite direction motions on both sides of the car. (Basically, the rate is the same but the bar bends twice as far due to the inside unloading at an equal rate as the outside loading.) However, since body roll only happens at very low damper piston speeds, we only need compensate for that there. I've read 3in/sec to 5in/sec as the MAX speed body roll occurs at, depending on how aggressive the car is. Thus we can bump up damping there to damp this.

However, in doing so, we'll overdamp ourselves for a two wheel bump. We thus need to reduce damping forces beyond 3-5/in second to compensate. I am reading recommendations of a damping coefficient of 0.5 and 1 for low speed (the latter for big sways only) and 0.3-0.5 for high speed. I would imagine you would want less high-speed as you get more low-speed so that your overall damping remains in a comfortable range, but I don't have any data to support this. I'm just going by what I've read.

Again going with the example above, a car with a damping coefficient of 1 will drop to roughly 0.6 for body roll. (I'm using my calculator to simulate this.) Again, that's with a sway bar rate of double the spring rate (800lbs/in and 400lb/s, respectively). 0.6 (or 60% critically damped) doesn't sound too bad. I'm not sure how damped this needs to be since generally you enter and exit body roll situations with at least some degree of smoothness even if it's just a limitation of how fast one can turn the steering wheel.

However, since this setup would be 100% critically damped for regular two wheel bumps, it'll have to show very strong digression so that the damper isn't overdamped for bumps (which usually occur beyond 5in/sec). Modeling a 100% critically damped spring (EN4: Dynamics and Vibrations) shows how utterly terrible this would feel; it'd take much too long for a spring to settle since it would rebound at a snail's pace. Again, I'm not sure how low one would need to go, but I'd guess below 50%. Even then I'm not sure you'd have a comfortable solution, but it'd probably be good for a track car.

Maybe we model this ourselves since the above calc doesn't support dynamic damping coefficients.

Another complexity of body roll is that since the chassis is rigid, the front and rear of the car roll together. Even though a car may have a huge front sway bar with a small rear bar, the rear struts can help dampen the motion generated by that front bar.

Crap, I hope at least one person out there could read this post from start to finish.

[slim9300]

Originally Posted by stretch

Good to have you in the discussion, MGizzle. I agree, damper needs to be tailored to the springs, not the other way around. It sounds like Ace has a beef with the spring rates we're modeling but they're the rates that come with these coilovers. We didn't pick 'em, it's what they ship with. (Switching springs is a different topic, though, so I'm not going to comment on spring rate choices in this thread.)

Anyway, sway bars seem to keep coming up in the discussion. I thought I'd try tackling that subject.

Sways play no part for a two wheel bump, so we only need look at a one wheel bump and body roll.

I have three thoughts here:

1) Is it bad to be underdamped for a one wheel bump? The chassis will remain stable due to the other two or three wheels not hitting a bump. This seems like an opportunity to tune for comfort without a stability penalty.

2) How damped does body roll need to be? In a perfectly steady turn, you'll have a constant amount of body roll and thus no motion to be damped (aside from bumps, which would behave like normal bumps). We only need to worry about damping transitions to and from body roll. I think the driver somewhat self-damps body roll by modulating the steering wheel; furthermore the rate at which this transition happens is limited by the speed in which the driver can turn the wheel. We often hear body roll happens at damper speeds of 3-5in/sec whereas bumps are 5-20in/second, so body roll does load the suspension more slowly. This means any "rocking" caused by the inertia of the body would be small (at least relative to the bounce of being underdamped for two wheel bumps); furthermore it wouldn't take much damping force to minimize it.

3) A sway bar also links the left and right struts. If the sway bar was infinitely huge, we'd need no damping changes because every bump would be a perfect two wheel bump. A sway bar can resist force on one side against the force (spring) on the other side of the car. This gets a little complex and I haven't really thought about this to too much detail, so here it goes. I'm going to think out loud a bit.

Let's assume 400lb/in springs and 400lb/in sways (rate for a ONE wheel bump, double that for body roll). This is somewhat like having TiC Coilovers on Whiteline 24mm sway bars. Let's say we have one inch of compression on a one wheel bump. The total spring rate is 400/in, so we have 400lbs of force wanting to react to this bump, 2/3 of which comes from the sway bar.

The sway bar always has equal but opposite forces on either side of the car, though. So, that 400lb/in is really 400lbs downwards on the bump wheel and 400lbs upwards on the opposite wheel. Thus, the instantaneous rate on both sides of the car has increased by 100% but this will decrease as the opposite wheel reacts.

What will then happen is that the suspension will try to reach equilibrium since we have 400lbs of upwards force on the opposite spring. Depending on the duration of the bump, either the bump wheel will recover of the opposite wheel will start lifting. Once it lifts half an inch, the sway bar will only have 200lbs per side of force (400lb/in x 0.5 inches) which is equilibrium for half an inch of compression on the opposite 400lb/in spring.

So, if you raise one wheel up one inch, you will get a half inch lift of the opposite side when your sway bar equals your spring rate (for a one wheel bump). Total rebound forces thus drop since now the sway bar isn't tensioned as much- we now have a 50% increase in rate instead of the 100% we had instantaneously during the bump. The trade-off is now we'll have movement at two wheels, but as far as damping goes, that's much more predictable.

The inside tire will only lift if it has time to do so. I suspect that on really small, quick bumps the inside tire lift is minimal, although it'll always happen to some degree.

Plotting the numbers for the increase in spring rate due to sways, %cd decreases by exactly 30% for the above example if the inside wheel does NOT have time to react. It's obviously less if the opposite wheel can react because then forces are distributed somewhat among the two dampers.

The question then is: how big a deal is this? The chassis is not going to react much to just one wheel being underdamped because it is stabilized on the other wheels. I would think you would want to be underdamped for a one wheel bump since it seems like an opportunity to do so without upsetting the chassis.

Sway bars have a much higher rate for body roll, double that of one wheel bump, since we have opposite direction motions on both sides of the car. (Basically, the rate is the same but the bar bends twice as far due to the inside unloading at an equal rate as the outside loading.) However, since body roll only happens at very low damper piston speeds, we only need compensate for that there. I've read 3in/sec to 5in/sec as the MAX speed body roll occurs at, depending on how aggressive the car is. Thus we can bump up damping there to damp this.

However, in doing so, we'll overdamp ourselves for a two wheel bump. We thus need to reduce damping forces beyond 3-5/in second to compensate. I am reading recommendations of a damping coefficient of 0.5 and 1 for low speed (the latter for big sways only) and 0.3-0.5 for high speed. I would imagine you would want less high-speed as you get more low-speed so that your overall damping remains in a comfortable range, but I don't have any data to support this. I'm just going by what I've read.

Again going with the example above, a car with a damping coefficient of 1 will drop to roughly 0.6 for body roll. (I'm using my calculator to simulate this.) Again, that's with a sway bar rate of double the spring rate (800lbs/in and 400lb/s, respectively). 0.6 (or 60% critically damped) doesn't sound too bad. I'm not sure how damped this needs to be since generally you enter and exit body roll situations with at least some degree of smoothness even if it's just a limitation of how fast one can turn the steering wheel.

However, since this setup would be 100% critically damped for regular two wheel bumps, it'll have to show very strong digression so that the damper isn't overdamped for bumps (which usually occur beyond 5in/sec). Modeling a 100% critically damped spring (EN4: Dynamics and Vibrations) shows how utterly terrible this would feel; it'd take much too long for a spring to settle since it would rebound at a snail's pace. Again, I'm not sure how low one would need to go, but I'd guess below 50%. Even then I'm not sure you'd have a comfortable solution, but it'd probably be good for a track car.

Maybe we model this ourselves since the above calc doesn't support dynamic damping coefficients.

Another complexity of body roll is that since the chassis is rigid, the front and rear of the car roll together. Even though a car may have a huge front sway bar with a small rear bar, the rear struts can help dampen the motion generated by that front bar.

Crap, I hope at least one person out there could read this post from start to finish.

I read it. But I think I need Cliffnotes. Normally you do a little summary paragraph that puts it in terms I can understand, or did I miss it this time?

[4banger]

Originally Posted by stretch

Let's assume 400lb/in springs and 400lb/in sways (rate for a ONE wheel bump, double that for body roll). This is somewhat like having TiC Coilovers on Whiteline 24mm sway bars. Let's say we have one inch of compression on a one wheel bump. The total spring rate is 400/in, so we have 400lbs of force wanting to react to this bump, 2/3 of which comes from the sway bar.

I'm not sure how you calculated the total spring ate of 400lbf/in, can you please elaborate. I think this is sort of like the tire spring rate. Check out Hooke's law - Wikipedia, the free encyclopedia. I think in your example, the total spring rate is 200lbf/in. I don't know why I didn't think of that earlier when Ace asked about adding in the swaybar rate. I'm not going to put words in anyone's mouth, but I think the thought there was that the sway and spring would just add together, which would result in a much lower %cd. I think what really is happening is that spring rate would go way down. I think we need to nail this down first before we continue on.


I have been doing a lot of searching and haven't been able to find much at all about the effects of sways on damping. Maybe it is becuae the rest of the world follows the Dennis Grant theory on only using sways as a tuning device. I have an idea about how to add it to the model, but as is, simple spring mass damper model that I have been usning has been stretched pretty far. I am already adding in the tire spring rate and neglecting the damping effect of the tire. I recently read that since the damping effect of the tire is so low, that neglecting it can have a large effect on the model. Almost all of the papers/examples that I can find leave out this factor. If think I can work this in, but first I have to find a copy of matlab.

[stretch]

Originally Posted by 4banger

I'm not sure how you calculated the total spring ate of 400lbf/in, can you please elaborate. I think this is sort of like the tire spring rate.

Sorry, typo. I meant 800lb/in. I re-wrote my example a few times to make a realistic yet easy-to-follow example. I forgot to change that paragraph, it seems... more than once. I'll edit the post now.

Originally Posted by slim9300

I read it. But I think I need Cliffnotes. Normally you do a little summary paragraph that puts it in terms I can understand, or did I miss it this time?

I deliberately didn't make a conclusion to prevent certain individuals from skipping to it, reading it as a blanket statement, and then derailing this thread by arguing about it out of context. I made some calculations from which I can make an intelligent guess about sway damping, and that guess was in line with what I've read others (with data and experience to verify) recommend.

The short version was that sway bars don't play as big of a role determining damper rates as I originally thought, although they do certainly play a role. With big sway bars it might make sense to go to around 100% critically damped (or said differently, a damping coefficient of 1), but only prior to 5lb/in or lower. With that much low-speed damping, I think you'd want significantly less damping force above 5lb/in to compensate- probably 50% critically damped or lower. Where comfort is a priority I think it'd be OK to ignore sway bar rates when picking the damper valving. This is exactly in line with comments from OptimumG's technical documents, so really there's no new news here.

[4banger]

Originally Posted by stretch

Another complexity of body roll is that since the chassis is rigid, the front and rear of the car roll together. Even though a car may have a huge front sway bar with a small rear bar, the rear struts can help dampen the motion generated by that front bar.

Or maybe the soft bushings and chasis flexing dampens out the rest?

[MGizzle]

Springs hold the SPRUNG mass, hence damping is important. Swaybars are directly conected to the sprung & unsprung masses and move via a given motion ratio. They don't seperate the sprung & unsprung mass like springs do. When you look at a quarter of half model of a car you don't see swaybars in the equation because they don't support the sprung mass.

Anyhow, ideally we don't want swaybars on our cars but tuning becomes a lot more complicated because you have to get your spring/damper RIGHT on. Check out the link below. I have seen their suspension on the Western Austalia University FSAE car and it is sick.

You can lift a corner of this FSAE car with like 80lbs (ride rate) about an inch but if you place your hand on the sprung mass and you try to roll it the car doesn't move AT ALL!!! What this tells you is that you want to be soft in bump because this doesn't upset your car and the other corners. You do want that tire to come down really fast but without lifting the car. Anyhow, with this suspension type, you can hit bumps and affect the sprung mass that much but as soon as you roll the suspension roll stiffness is very, very high, as if you are using enormous anti roll bars. This is perfect. Look at that website and you will notice Citroen was using it in the WRC and they got banned after smoking everyone else.

Anyhow, I agree with Denis Grant that ARB's are used as tuning devices because you can get significant changes quickly and your damper velocities will not change at all. I remember when I tuned dampers with the potentiometers we would disconect the ARB in the back completely and than go full stiff and the wheel velocities werent affected that much when hiting bumps. Velocities during roll changed quiet a bit but still, 85% of damper velocities were under 2in/sec for about 20 laps on a parking lot.

A great test that we could do on our cars would be to buy some cheap string potentiometers and bolt them to our frame & the ground. Push on the roof pilars until you get the car going side to side quiet a bit and than let go. The string poteniometers will tell you how long it took for the car to settle and it will give you the damping curve. You can do the same front to back etc. I know some dudes do this by eye and feel at the track but with our cars it is pretty tough because they are stiff as heck.

Anyhow, check out the link. Sick, sick, sick.

Kinetic Suspension Technology

[stretch]

Originally Posted by 4banger

Or maybe the soft bushings and chasis flexing dampens out the rest?

Bushings, perhaps, but chassis motion is bad because it is undamped. (Or so it is taught.)

Math aside, I will make the observation that even on stock struts and springs, big sway bars don't create much "wobble" in the suspension- it exists but isn't problematic nor distracting even if you're looking for it. The stock struts on a 27mm bar are about as extreme of a situation as there is, and in practice it's not so bad. (I even ran the Strano 32mm bar on stock struts for a while.) Thus, I personally wouldn't change my valving much to compensate for a huge sway bar.

Originally Posted by MGizzle

Springs hold the SPRUNG mass, hence damping is important. Swaybars are directly conected to the sprung & unsprung masses and move via a given motion ratio. They don't seperate the sprung & unsprung mass like springs do. When you look at a quarter of half model of a car you don't see swaybars in the equation because they don't support the sprung mass.

Excellent point, that's a good way to think about it. I'll have to ponder that point of view for a while. I still think sways matter somewhat, though, since since a sway bar changes the load on the main springs and thus indirectly the sprung weight. I think my observations above still hold true, but I'd be curious to hear more of your thoughts.