Howdy all! Another one of my 'technical' threads, pertaining to real life, but also, possibly, having LFS relevance!!

This time, springs! Haven't I already done that? Well, this time it's springs and tyres!

Here we go: This year we've been using crossply (bias-ply for those people across the lake that don't use normal terms) tyres on our car, partly because it came with crossplies, and partly because the loss of grip 'at the limit' is more gradual and novice friendly on them compared to radials.

However, for 2008, we want to change to radials, because they are supposedly quicker, and with a few more miles under my belt I stand more chance of keeping it on the track despite the quicker transition from stick to slip.

To effect the change from crossplies to radials, the car has to be tweaked. More negative camber can be used to generate more grip under more conditions (more accepting of camber curves), and the generally higher sidewall over comparible imperial crossplies means the ride height of the chassis can be lowered towards optimium (more of the 40mm ride height can be from the sidewall height rather than pushrod length).

But radials are also blessed (?) with softer sidewalls (and stiffer tread areas) because of their construction. Which means that the car will have, effectively, softer wheel rates, which won't be a good thing overall.

Bit of data at the moment: Springs 325lb/in (F) and 350lb/in (R). Wheel rates of 110lb/in (F) and 220lb/in (R). Wheel frequencies 140CPM (F), 220CPM (R). Our dampers aren't (currently) adjustable, so they can be ignored for the time being. But we are seriously considering the benefits of adjustable dampers - that's a topic for a future thread though. Overall vehicle mass is 1234lb, with unsprung masses at 47lb (F), 53lb (R), and the weight balance is 40% front (static with driver on board). Preload and bellcrank positions can be ignored, and assumed a constant.

Wheel/tyre sizes:
Front, 13" x 8" wheel. 7.5x21.0/13.0" tyre. Radials would be 195/530R13
Rear, 13" x 10" wheel. 9.2x22.0/13.0" tyre. Radials would be 250/550R13


So the question(s) is this: How much stiffer in terms of wheel rate (ignoring the tyre spring rate) should I go during the change from crossply to radial?

Fire away!

What sort of values are you talking about for the sidewall stiffness of the tyres?
The overall stiffness of springs in series (sidewalls + suspension) can be calculated by analogy with resistors in parallel and using the 'product over sum' equation:

overall stiffness = (stiffness1 * stiffness2) / (stiffness1 + stiffness2)

One of the key features of this equation is that, if the two stiffness values differ by an order of magnitude or more, one of them can usually be ignored. For example:

s1 = 10, s2 = 1. Overall stiffness = (10*1)/(10+1) = 0.909..., which isn't all that different from s2, i.e. the effect of s1 can be ignored and the error is less than 10%.

If the tyre sidewall stiffnesses are significantly higher than the suspension stiffness then you might be able to ignore them in your setup.

I should point out that I speak as an engineer, not a racing driver

as fisher pointed out theres really nothing you can do without any figures on what springrates the cross and radial tyres have
also would help to have those figures in some unit other than mice/hogshead

Thanks both. I shall try and get the info out of Avon - they're at lunch at the moment, so I'll call back in a bit.

As for the units, it's an English car, using imperially dimensioned crossply tyres, in an English series where springs are generally bought in lb/in, and a car made using imperial units. It makes sense to use imperial units for the whole analysis. No more difficult to work with, and no more difficult to visualise (if anything slightly easier as imperials units are based on real world stuff). If I was playing with a metric car I'd use metric units, but I'm not, so I won't.

in that case
what the heck is cmp ? or is it supposed to be cpm as in cycles per minute ?

Yeah, I guess he meant cycles per minute

whoops, the typo monster struck again!

Quote from tristancliffe :

whoops, the typo monster stuck again!

Try smearing him with lard, asking him to exhale and then all of you push at the same time.

Dammit!

Oh, I phoned Avon. The man on the end of the phone didn't have the data for radials to hand, and thought that the data for crossplies had never been found out anyway. But he's going to see what he can find, and I've got to call him in a couple of days. But his two other ideas were to a) stick to crossplies, as we know them a bit, and the car is handling quite well now. The radials are a BIG change, and we might find that we struggle to make them work without trying several sizes/compounds, which we can't afford to do (although I still reckon we can work out what's best theoretically - at least well enough for a lummock like me), or b) to just copy what other people have done with the same car and hope for the best... Not very scientific, and hence not very appealing.

perhaps this is a dumb question, but couldn't you just run higher tire pressure with the radials, and leave the springs alone?

Yes it is
Increasing the tyre pressure would do all kinds of nasty things, like decreasing contact patch size, introduce uneven wear and heating of the tyre surface, etc. Nothing that you really want to mess with just to save yourself some minor spring adjustments.

The real question is, do you even want to adjust spring rates at all? Not only would it mess with the currently not adjustable dampers, but you also have to consider that the radials have a quite different behaviour at the limit. Isn't the credo in LFS' setup adjustment to do one thing at a time to see which effect it has? Changing tyre type, pressure, camber and spring rates seems like a complete overhaul and you won't have any idea if your adjustments made things better or worse.

Quote from tristancliffe :

To effect the change from crossplies to radials, the car has to be tweaked. More negative camber can be used to generate more grip under more conditions (more accepting of camber curves), and the generally higher sidewall over comparible imperial crossplies means the ride height of the chassis can be lowered towards optimium (more of the 40mm ride height can be from the sidewall height rather than pushrod length).

But radials are also blessed (?) with softer sidewalls (and stiffer tread areas) because of their construction. Which means that the car will have, effectively, softer wheel rates, which won't be a good thing overall.



So the question(s) is this: How much stiffer in terms of wheel rate (ignoring the tyre spring rate) should I go during the change from crossply to radial?

Fire away!

Quote from [URL=? :

So if you know the spring rates in your aftermarket spring kit(wheel rate = spring rate/leverage^2), you can calculate the stiffness of your wheel or wheel rate...which is more important because what's happening at the ground is more important than what your spring is doing.

[URL="http://importnut.net/tiretech.htm#caster"]This too[/URL]

Quote from tristancliffe :

However, for 2008, we want to change to radials, because they are supposedly quicker, and with a few more miles under my belt I stand more chance of keeping it on the track despite the quicker transition from stick to slip.

...are bias ply really all that bad?...imo it looks to be a lot of testing/time/money...and what your getting in return ain't all that naff. why not upgrade sompthin else?

Quote from AndroidXP :

Yes it is
Increasing the tyre pressure would do all kinds of nasty things, like decreasing contact patch size, introduce uneven wear and heating of the tyre surface, etc. Nothing that you really want to mess with just to save yourself some minor spring adjustments.

The real question is, do you even want to adjust spring rates at all? Not only would it mess with the currently not adjustable dampers, but you also have to consider that the radials have a quite different behaviour at the limit. Isn't the credo in LFS' setup adjustment to do one thing at a time to see which effect it has? Changing tyre type, pressure, camber and spring rates seems like a complete overhaul and you won't have any idea if your adjustments made things better or worse.

Well, if we went to Radials, then yes we do have to play with spring rates. The overall wheel rate is controlled by the chassis springs and leverage (classical wheel rate) and the spring rate of the tyres. Reduce one, and the other has to increase to compensate. And yes, it's nice to change one thing at a time - the better approach, but in real life club racing time and budget dictate against that! Tyre type and pressure go hand in hand - the manufactuerers give a recommended small range of pressures to work in. The camber is pretty much defined as 3° front and 2° rear (as a base setup, although that can be tweaked). It just leaves the springs. I know it's not a perfect approach, which is why I want to consider as much as possible numerically before committing or not.

Quote from CSU1 :

This too


...are bias ply really all that bad?...imo it looks to be a lot of testing/time/money...and what your getting in return ain't all that naff. why not upgrade sompthin else?

Bias ply aren't that bad, but they DO produce less lateral force than radials, so for a given corner radius they ARE slower. Estimates range from 0.5 - 1.5 seconds per lap slower than radials (assuming a half decent setup).

Castor is easy and doesn't vary too much with tyre type. Our current setting of 4° will be okay for radials.


The thing is this has been done before on various 883s in Monoposto already, and in other series, so I can always fall back on asking/copying what other people have done. But before I do that I'd like to work out, independantly, what we should do, compare that with what people have done, and go from there.

The big question is, I suppose, that although I'm getting quicker compared to the opposition (less than a second slower than cars 12 years newer), we don't really know if we are well setup or badly setup at the moment. So we don't really lose anything by trying radials, having a test session in March to improve that setup further, then seeing where we are at the first race (which I guess is in April, but don't know where yet).

The main factor is that there isn't anywhere else on the car that we can gain ~1 second per lap in one go. Can't get much more power (though we want to increase the crankcase depression, which should help). Can't increase the track. Can't lower the weight. Can't do a great deal on the downforce, though I want to try Gurney's on the front wing, and improve the diffuser a smidgen more too. Each of those will be worth tenths, not whole seconds.

But I have pretty much six months to work out what I need to know to make the car handle 'reasonably' well on radials, and can fine tune it over the season.

OK, for some ballpark figures, radials of those size could be around 800 (F) / 1000 (R) lb/in with the weight of your car on them, with sensible inflation pressures. Lets assume that the sidewalls are 2.5 times stiffer on crossplys (a complete guess I admit but bear with me a moment).

To save explaining things further in text, see the attached spreadsheet, but with the 2.5 guess, you're looking at a 10% reduction in stiffness, less if I guessed too high, more if vice versa. Helpful?

Edit: I nearly forgot, increased load on a tyre makes it stiffer, so with your downforce, the difference will drop with speed, so should be slightly less still. Of course this assumes your suspension is linear, which of course it isn't quite.

Thanks Bob. How much of a guess are those figures though? Vaguely ballpark, or just plucked from the ether?

The reason I ask is because feeding your numbers in, my estimate ends up very close to that used by other 883s on radials, which is nice. Obviously it's not an exact match, as every driver is different, with different strengths and weaknesses and preferences, and every car is slightly different (although for the purposes of this I've assumed the weight and balances are the same), but it's close enough to act as a proof of concept (and proof that the people before me did vaguely the right thing). But it all depends on how accurate you think your numbers are. 10% error? 50% error? 100% error?

bit of a different question but is there any point in approaching the tyres as a soft spring that deals with bumps with a much stiffer actual suspension to adjust the balance of the car ?

Sorry, I don't really understand. Any chance of a rephrase? Tyres are a lot stiffer that my springs.

uuuh ... ignore my blonde moment pls

Wouldn't the whole attitude and handling change, making it relatively pointless to try and compensate the springs for the stiffer or softer tires? I bet the tires themselves make a lot more difference. I bet you could find the whole understeer/oversteer balance to be different, plus of course you'd have to get used to the newness of it all..

Quote from tristancliffe :

Thanks Bob. How much of a guess are those figures though? Vaguely ballpark, or just plucked from the ether?

The 2.5 ratio is an educated guess, I've heard crossplys are a lot stiffer, so I'd guess at least double, but I wouldn't have thought that they would be an order of magnitude different (or even close), as I don't believe that your crossply race slicks would be as stiff as truck radials.

Quote from tristancliffe :

But it all depends on how accurate you think your numbers are. 10% error? 50% error? 100% error?

As for the tyre stiffness numbers, I'd be surprised if they were out by more than 40% or so. They're certainly in the ballpark. I used the formulae I got from Scawen that's coded into VHPA atm to get the stiffnesses, but I think the numbers come out a little low compared to real life. There seems to be no accounting for tyre wall stiffness in the model, yet it's there in LFS; I have asked Scawen about this but he's too busy to look into it atm. It does take into account pressure, load and tyre dimensions (specifically width) quite nicely, although I've seen little real data to make true comparisons, it all seems quite logical and follows what I've read. If the figures are out, it's more than likely they are too soft than too hard.

Edit: I'd agree with Niels though, everything you do goes through the tyres, so substantially changing them is likely to have a larger effect than anything else would. Without experience I'd say it's unlikely you'll be able to adapt for them with some trial and error testing. With the different stiffnesses, deformation and thus contact patch size (and shape) is going to differ, so it could well change the static grip balance.

Yes, I'm sure the balance will change, and I'm still very much a novice anyway, so it being different won't be too hard to get used to again. But the balance can be restored in a test session or two (at least within the broad limits I'm able to detect), so I'm not overly concerned about getting it 'spot on' first time, just within limits.

But compensating for the softer sidewalls is absolutely necessary - the first thing any tyre/chassis person will tell you is that radials need stiffer chassis springs and stronger anti-roll bars to a) keep the overall springing/frequencies under control and b) to keep the camber curves in check with larger lateral loads.

Unless Bob's data turns out to be completely made up, I think it's pretty much confirmed that the previous people were close to the mark, and we just need to start there and try things out. And we need to buy two 25mm roll bars. Or make them.

The list of things to do currently stands at about 40 broad items, some of which are relatively unimportant, and some of which are quite large. They include, in no particular order:

Improve bodywork fasteners
Fix clutch problem (it requires bleeding every race meeting, and we think it's a slave cylinder problem, which means gearbox/engine split).
Replace fuel-ballast with lead-ballast
Replace GRP seat with good foam seat
Fit radials
Consider adjustable dampers
Replace spherical bearings and rod ends
Fit oil cooler
Improve water cooling
Increase crankcase depression to ~10psi gauge
Lengthen diffuser end plates
Fit brake light switch for use in datalogger (cheaper than a pressure sender)
Get cockpit laptimer
Paint helmet (if budget allows)
...

I'm back!

On the same basis as the spring rate choices, which I'm now pretty happy with, and they agree with what other people have done before in the same type of car, I'm looking at anti-roll bars.

I know how to calculate the stiffness of a anti-roll bar, including taking into account effective arm lengths, hollow bars, motion ratios etc...

But I want to know the overall roll resistance of the car with crossplies and radials, soft and stiff springs, & small and large anti-roll bars.

I'm pretty sure I've seen equations that will link spring rates, ARB sizes, CoG heights and lateral accelerations to show overall roll resistance, angle of roll etc etc.

Does anyone here have that(those) equation(s), because I can't for the life of me find them. Any help would be appreciated!

Oh, and even after I play with numbers, I'll probably end up getting the next size up and using that, like everyone else did with radials in these cars.

Ta muchly!

Edit: I suspect it was in Milliken/Milliken's book, which I'll look at when I go home tonight... Can't believe I forgot about it (but if you know the equation(s) still post them in case I don't find the book, or it's not in there).

Such equations are in VHPA, I can dig the code out later tonight if running the program is not sufficient.

I'll have a play with it, and see what it suggests... Ta!

Quote from tristancliffe :

I know how to calculate the stiffness of a anti-roll bar, including taking into account effective arm lengths, hollow bars, motion ratios etc...

Oh and if you don't mind sharing, I'd be interested in seeing that.