Roll couple or roll centre??? HELP!!!!!!

[jonners]

A few basic points before we get to the question.

The point I am querying relates to an old saloon so please remember that the centre of gravity is almost in the stratosphere compared with the single seaters most of you guys talk about on here. As you will see this is an important point.

The car uses MacPherson struts so remember that the method for determining roll centre is different to a single seater.

Also, the scope for redesigning/optimising/relocating suspension mounts is very low given the sort of events the car is used for.

The problem is body roll. At low speed venues such as Prescott the car rolls too much. Needless to say the problem is worse at faster venues.

The car is driven to and from events but even so it is very competetive.

Just throwing ultra stiff springs on is out of the question because the car has to be driven on the road. We are already using an uprated anti roll bar and we have increased the spring rates.

I am not an engineer but from what little I know I wonder whether the excessive roll is caused by too much roll couple.

The bodyshell has been lowered and the roll centre has gone down by a fair amount, but of course this means that the roll couple has increased.

Lowering the bodyshell is a good thing so that will be left where it is but we have worked out a way to raise the roll centre in order to reduce roll couple.

So the question is.....should we keep the roll centre low or should we try to reduce roll couple????????

REMEMBER that the roll couple value here will be much higher than on any pure racing car - I have read on here before people being quite dismissive of the importance of roll couple but I believe that is because it just is not a factor on a single seater.

Other points - photgraphs of the car in action show the body rolling higher at the rear than at the front - but we can't improve the roll axis because the roll centre at the rear is fixed.

What do you all think?? One chassis engineer has already stated his view that low roll centre is best - the lower the better - but won't this just increase roll couple????

But then again if the roll centre is raised almost to the height of the centre of gravity - does this mean the car wouldn't roll at all????

All help/suggestions/wisdom VERY gratefully received....

[Lukin]

Moving the RC height closer to the COG will reduce the roll angle yes. Having the RC at the COG height will mean no chassis roll at all as you pointed out. But it will also change the way the tyre is loaded. For long sweeping corners this can be a problem, but for street circuit's it not always bad, especially as you can run less bar and spring and help your tyre compliance. However, high RC puts biger loads through your suspension components/bushes and will increase tyre wear in most practical cases.

I have to say I don't quite understand what everyone means by roll couple. I've heard a few different definitions so I usually shy away from using that word.

When you say rear RC is fixed, I assume you mean Watt's Link/Panhard Bar? It is changeable if thats the case.

If the rear is rolling more, (assuming that RC/COG ratios for the front and rear are about the same), does the car understeer a bit? What diff do you have?

Increasing your front and rear bars will increase the load transfer and reduce the roll angle. It will also leave you more suseptible to picking up wheels in long radius fast corners (especially with a sedan). That's isn't always a bad thing though. In sedans we usually run a much stiffer front than rear ARB. It's relatively easy to measure with scales.

Depending on your track, tyres etc it's hard to say. When cash is tight making changes to see what happens is hard. Speak to others. Use a little gut instinct I guess.

If you can approximate some of the information required, use the spreadsheet I posted in another thread to quantify everything. It won't say 'yes' or 'no' but help establish equivalencies.

[Cleggie]

At the risk of sounding a bit dim, what is roll couple, perhaps I'm the only one brave enough to show my ignorance, or everyone else knows more than me! Even Lukin, who covers this issue very well, doesn't like using the term....

[graham bahr]

the real problem here is a dual purpose car, the competition and roadgoing requirements conflict, and the car will never be as good at either as it could be.

you cannot use roll bars to effectively replace stiff springs, roll bars are best at fine tuning

[maddogf3]

Originally Posted by Cleggie

At the risk of sounding a bit dim, what is roll couple, perhaps I'm the only one brave enough to show my ignorance, or everyone else knows more than me! Even Lukin, who covers this issue very well, doesn't like using the term....

Suggest you ask your race engineer he knows !!!!!

[maddogf3]

Just a thought and I dont have a lot of experience with saloons but as a very cheap fix for the track you could try limiting the droop I' ve seen it done with an adjustable cable from the body to suspension arm .It might lift wheels but for a couple of quid you have nothing to loose There's still no substitute for stiffer springs

[phoenix]

Originally Posted by graham bahr

the real problem here is a dual purpose car, the competition and roadgoing requirements conflict, and the car will never be as good at either as it could be.

you cannot use roll bars to effectively replace stiff springs, roll bars are best at fine tuning

I'm not sure I agree.

On our car about 50% of the roll resistance is via the springs and 50% via the anti roll bars. Therefore, if we disconnect the bars but wish to have the same level on antiroll stiffness, and to maintain the same from/rear weight transfer balance, we have to increase the front spring rates by 233% and the rear spring rates by 200%.

This changes the wheel frequencies to 193 cpm front/187 cpm rear.

With wheel frequencies this high the tyres give less grip when accelerating, braking, touching kerbs or rumble strips and in bumpy corners - problems which are eliminated by lower wheel frequencies. We find, therefore, that anti roll bars are the best way to increase roll stiffness without compromising the job that the springs primarily have to do.

Of course, such an approach is ideal for a dual purpose car. Stiffer bars can be fitted for racing and removed or disconnected for road use, or correctly designed adjustable anti roll bars would allow for soft settings on the road and firmer on the track.

[graham bahr]

Originally Posted by phoenix

I'm not sure I agree.

On our car about 50% of the roll resistance is via the springs and 50% via the anti roll bars. Therefore, if we disconnect the bars but wish to have the same level on antiroll stiffness, and to maintain the same from/rear weight transfer balance, we have to increase the front spring rates by 233% and the rear spring rates by 200%.

.

the biggest gripe i have with using stiff roll bars and soft springs it that the roll bar will rob you of traction exiting the corners as it tries to pick up the inside wheel, also weigh transfer accross the car means the lightly loaded wheel limits the effectivness of the bar as it is effectivly only the weight of the strut hub assembly thats providing the anti roll effort beyond a certain point

[phoenix]

If you have transferred weight away from an inside rear wheel onto the outside of the car such that the inside real tyre is unloaded and it loses traction, does it matter whether the weight is transferred by a bar or a spring?

[dtype38]

Originally Posted by phoenix

....to maintain the same from/rear weight transfer balance, we have to increase the front spring rates by 233% and the rear spring rates by 200%.
This changes the wheel frequencies to 193 cpm front/187 cpm rear.

Gonna show my ignorance now, but here goes.... when you are mid-corner and the anti-roll bars are at full deflection, wouldn't the effective spring rate at the wheel be the sum of that generated by the coil springs and the anti-roll bar, making the wheel frequency the same as if there was only the stiffer springs fitted?

[maddogf3]

Those wheel frequencies look about right probably too stiff for saloon car and way to stiff for a road car . he doesn' want to use any stiffer springs though .Its a shame because we all know thats the way to go .

[maddogf3]

Originally Posted by phoenix

If you have transferred weight away from an inside rear wheel onto the outside of the car such that the inside real tyre is unloaded and it loses traction, does it matter whether the weight is transferred by a bar or a spring?

having a stiff spring resists the amount of weight transfer in a race car and actually suprisingly gives more traction to the inside wheel I am not going to do maths that makes my brain hurt . and what works in theory dosn't always work in practice .but stiffen the rear up improves traction on a track particularly when suspension movment is limited (Single seater's again ) THR said that in hill climb cars, its the stiff cars that get FTD's and get off the line fastest. one would assume that you would need a lot of weight tranfer to get grip and load rear tyres .

[graham bahr]

Originally Posted by phoenix

If you have transferred weight away from an inside rear wheel onto the outside of the car such that the inside real tyre is unloaded and it loses traction, does it matter whether the weight is transferred by a bar or a spring?

weight transfer is only part of it, you cant do anything about the weight transfer trying to rob you of traction, but stiff roll bars on the driven axle will actively try to lift the lightly loaded wheel further robbing you of traction

[Lukin]

A few things:

1. Why do you want to limit roll anyway? Unless your sedan has DTM style aero (unlikely for a road car which more than likely produces lift) then changing ride heights in pitch and roll probably aren't that big an issue. Does the chassis bottom during braking and turning? That's probably where RC will help the most, allowing you to run lower (which is a godsend for a big sedan).

Or are you wanting better response? Which isn't a bad thing. The problem I find with most racers (or wannabe racers (not saying your in group 2)) is they associate a great handling car with being stiff as a 16 year old in front of the playboy stand. That said, we do that a bit too. Raising the RRC lowers the roll angle and increases the speed of load transfer; basically quickens the car up and reduces US. So that's a valid excuse for reducing roll.

As far as load variance at the wheel, roll isn't always a bad thing. Soft is often better, especially for an undertyred overweight car.

2. Graham Bahr is right. You have to pick one or the other. Good road car = crap race car and vise versa. If a fox goes for two rabbits in a paddock chances are it will get none.

3. Brave man who limits droop on a sedan I would think! Suspension travel is generally your friend when you have 1.5 tonne behind you, crap tyres, crap brakes and a COG 4 foot off the deck (hopefully that last one is slightly off). You could try it, but I imagine the voilent changes in load at the front corners as pitch reduces and roll increases would scare you half to death. For a Formula Ford (which would have 1/3 of the lateral and longitudinal load transfer) it would work, but I'd be suprised if it worked for a big sedan.

4. As for springs v bars; I am not 100% sure about how to pick a ratio, but I am damn sure it varies with situation. As much as I love the Carrol Smith books and think he was a great writer and engineer, I don't think he is always right when he says tune predominantly with bars and bump rubbers.

Dropping the bar off the rear and going for stiffer springs is a lot better than the other way around as maddogf3 said, especially for corner exit.

[Lukin]

Originally Posted by phoenix

I'm not sure I agree.

With wheel frequencies this high the tyres give less grip when accelerating, braking, touching kerbs or rumble strips and in bumpy corners - problems which are eliminated by lower wheel frequencies. We find, therefore, that anti roll bars are the best way to increase roll stiffness without compromising the job that the springs primarily have to do.

What sort of car do you work on Phoenix? Does it have a third spring?

For roll your set, but what about pitch? If you have stiffer springs you can run a lower chassis ride height without bottoming under brakes. It's hard to say one is better than the other.

If you have stiffer spring, you can lower ride height (and COG) and the lowered COG may make up for the loss in tyre compliance due to stiffer springs.

Dropping the springs and adding more bars will mean a higher ride height (unless you run a lot of anti-dive and risk killing front tyres). Also don't forget the effect bumps can have on the ride component of ARB's. You don't want the ARB having too great an effect in the braking and turn-in zones if it's especially bumpy.

And also, don't rule out using different ratio's front to rear. At a track like Catalunya, Phillip Island etc, you might want a relatively stiff front bar for the long corners, so you only need softish front springs. At the rear however, a stiff bar will hurt corner exit balance, so a lower bar and using stiffer springs to hold the car up could be in order.

[phoenix]

Originally Posted by dtype38

Gonna show my ignorance now, but here goes.... when you are mid-corner and the anti-roll bars are at full deflection, wouldn't the effective spring rate at the wheel be the sum of that generated by the coil springs and the anti-roll bar, making the wheel frequency the same as if there was only the stiffer springs fitted?

Wheel frequencies get lower on the ouside wheels when cornering because the load is increased. As anti roll bars add wheel rate in a corner they help to restore the otherwise falling wheel rate. A spring, unless it is progressive, will appear to get softer the more load that is applied. So the two methods do not give the same result.

[phoenix]

Originally Posted by maddogf3

having a stiff spring resists the amount of weight transfer in a race car and actually suprisingly gives more traction

No need for the maths, but can you tell me the theory and give me some technical references please?

[maddogf3]

Roll usualy is acompanied by massive camber changes .... Just the oposite of what you want on race track limit the roll limit camber changes . why have all that suspension movement anyway ?
A lot of saloon racers do have wire droop resistors on them. TVR racers use them as well .
Try it. what have you to loose ? a bit of wire with two ends on from the local chandlery or even B&Q 5 quid max and then you can get one over on your paddock cronies that you are doing something they havn't I am sure it wont be as bad as you are making out

[phoenix]

Originally Posted by Lukin

What sort of car do you work on Phoenix? Does it have a third spring?

For roll your set, but what about pitch? If you have stiffer springs you can run a lower chassis ride height without bottoming under brakes. It's hard to say one is better than the other.

If you have stiffer spring, you can lower ride height (and COG) and the lowered COG may make up for the loss in tyre compliance due to stiffer springs.

Dropping the springs and adding more bars will mean a higher ride height (unless you run a lot of anti-dive and risk killing front tyres). Also don't forget the effect bumps can have on the ride component of ARB's. You don't want the ARB having too great an effect in the braking and turn-in zones if it's especially bumpy.

And also, don't rule out using different ratio's front to rear. At a track like Catalunya, Phillip Island etc, you might want a relatively stiff front bar for the long corners, so you only need softish front springs. At the rear however, a stiff bar will hurt corner exit balance, so a lower bar and using stiffer springs to hold the car up could be in order.

It's a sports/GT car and we don't have a third spring. We are on the minimum ride height already.

Our static weight distribution is 45% front and 55% rear. Under 1G braking this becomes 60% front and 40% rear. Under 1G acceleration (nice if we could get it!) we would have 29% front and 71% rear. Of course you will know that these percentages do not change if we simply increase the spring rates at one end of the car - only moving the c of g would achieve that.

To control potential pitch changes we use anti dive front geometry and a tiny amount of anti squat, as we have more bump travel in the rear of the car and more rear ground clearance and as the car is setup with the rear slightly higher we don't get any worrying changes to the pitch under maximum acceleration.

In terms of setup springs and bars we have just what you advise. A lower front spring rate than rear and a higher front bar rate than rear - it seems to work! As I said previously, to achieve the same balance without bars would not be possible as the front spring rate would have to be higher than the rear.

[maddogf3]

Originally Posted by phoenix

No need for the maths, but can you tell me the theory and give me some technical references please?

NO!
I dont know any technical references to this . Its worked for us and some guy trying to sell us a set up service at Castle Combe also went with this Idea .
In our case Formula Ford, Oulton Park Knicker Brook inside wheel spinning up . In our search for quicker times I had upped the spring rate all around based on Suspension Frequency working to about 130 /140 actual spring rate was if I remember was about 450 front and about 650 rear .That was quite stiff result no more wheel spin and another drop in lap times .

[maddogf3]

Originally Posted by dtype38

Gonna show my ignorance now, but here goes.... when you are mid-corner and the anti-roll bars are at full deflection, wouldn't the effective spring rate at the wheel be the sum of that generated by the coil springs and the anti-roll bar, making the wheel frequency the same as if there was only the stiffer springs fitted?

Yes I do believe it would . Wheel frequency is as far as I know just measured statically to give a usable spring rate to start with and then Anti-roll bars worked out to give adequate roll stiffness at either end .I personally used frequency as a theoretical measure to see just how stiff and what rate springs to aim at in the first place Best Regards Maddog

[norman-normal]

I don't understand a lot of these turms; and I suspect the users have the same problem. But I have a lot of hands on experiance, so here goes:
(1)You mentiond "anti roll bar" you should have two of them.
(2)You are picking up an inside rear wheel on tight turns because of the weight transfers lingering effect due to braking deep into the turn. A much stiffer ft sway bar will eliminate much of the plunging of the outside ft, whitch is the cause of the rear pickup. Very stiff springs will make your wheels airborne over the slightest surface irregularity. Stiff springs are the crudest method of improving handling and should be used with caution.
Idon't know what kind of car you have, I assume it is a ft engine/r wheel drive and I am unaware of the regulations you race under, so I can't offer more help.

[dtype38]

Originally Posted by norman-normal

You are picking up an inside rear wheel on tight turns because of the weight transfers lingering effect due to braking deep into the turn. A much stiffer ft sway bar will eliminate much of the plunging of the outside ft, whitch is the cause of the rear pickup. Very stiff springs will make your wheels airborne over the slightest surface irregularity. Stiff springs are the crudest method of improving handling and should be used with caution.

Interesting analysis... I've never actually suffered from wheel pickup, but when I used a sway bar to limit excessive dive at the front outer corner on turn-in I just got terminal understeer (push). I found that just going to stiff springs instead cured the problem.

As for references about where these idea's come from... I'm using Fred Phun's book on How to make you car handle. From my reading of it, he basically says to go as stiff as you possibly can on the springs without suffering wheel skip under braking or acceleration. Also, go as light as you possibly can on the damping without getting excessive bounce or rebound. Then, when you've got that sorted, if you're rubbing your wheel arches in the middle of max lateral g corners, then add a very light sway bar. Personally, I've done the first and second bit, and didn't need to add the sway bar! Only down side is that I now have a very stiff car that handles great in the dry and aweful in the wet

[Cleggie]

Norm, I don't agree with that last comment about stiffer springs bieng the crudest method, or perhaps all the people that we raced against in FF1600 and ARP F3 were way out on thier own settings, (and I'm sure they weren't). Whenever we had a handling problem, we just stiffened the car and our problems were solved, (except for the car starting to slowly break up). My race engineer does all this, I drive it and report back as accurately as I can, so, sometimes I don't understand how he's worked it out, but the easiest and cheapest methods always seem to work,and its not a long road back if you've gone too far, (as we occasionaly found out). We had 900lbs springs on an 89 FF1600 Reyard, and won races in the wet like that, when everyone else was much softer, typical springs on this car would be about 600lbs. There's a lot of good imput on this thread by people who know a lot more than me on the technology, but as a driver I'd stiffen the car up with springs and test it, bet you'd go quicker....

[Cleggie]

one more thing, Norm... you say that stiff springs would make you airborne in the bumps. You race in the US, right? where, I think, the track surface is much more irregular than in the UK, so, in context, I can understand your comment.