Tyre / Camber issue

[danowat]

Hi,

Been playing around trying to make a set for the FXO at Aston GP, and I am having an issue with the tyres / camber.

The problem is as follows, no matter how much camber I put on the wheels (up to a point obviously), the outside third of the outside wheel on any given corner is having a more force than it should, I would ideally like the force to be as flat as possible throughout the turn, this wasnt so much of a problem with the previous patch.

I think that the issue might lie with the tyres, it seems as though the sidewall is deforming far to much, I know I have said it before, but I truely think that the sidewalls are far too weak, this leads, on road tyres anyway, to having trouble setting camber for even tyre wear, and to some degree pressures, as the tyres are getting warm on the inner and outer thirds, but sometimes cool in the middle (which generally means low pressure), it seems that no amount of camber adjustment or tyre pressure setting can give good results, it just seems as though the sidewalls have no strengh at all and just "blob" deform.

The following pictures are of an FXO, using the default RACE_S set, I thoughts?.

Dan,

[Blowtus]

I find the tyres much easier to make a set for, myself. it doesn't look as 'balanced' when you have the side sections heavily loaded, but it still grips fine and makes for more balanced tyre wear / heat. Does seem a little over the top, but this was discussed before...

[danowat]

Yeah, I know it was discussed before, and I am not just talking about the "look" of the tyres, I am talking about the effect that the sidewall has on setting up the camber and tyre pressures, it just seems that the sidewall is too easy to "push over" and causes the problems I am having with odd tyre temps/wear

Dan,

[Tweaker]

I know what you mean here danowat, I've thought about this myself.

Whenever I put HIGH camber for my wheels, it always seems to have the same affect of sidewall deformation. In a sense, if I had like -4 camber on the front, the inside should be touching mostly I would think. However it is always the outside edge that gets pulled underneath.... Isn't camber one way to get rid of that too?? That is one principle of camber as far as I know, and LFS is doing it quite wierd I think.

Not only that, but I have noticed that some people are setting up their cars with some incredibly high amounts of camber. Take the XR GT for example. People are using -3 on the front and upwards or near -4 on the rear. This is A LOT of camber... yet that makes the car faster in cornering apparently. :really: I really do think this is incorrect, since there is usually a limit for how far some cambers are put on cars in racing. Too little of camber won't perform well, and too much camber won't perform well. Except LFS seems to be different... somehow.

But seriously, -4 cambers being functional, and fast? Ehem...

[danowat]

I think the excessive camber is coming from the fact that the tyre sidewall is deforming so much that you need serious amounts of camber to get the tyres to run "flat" throughout a corner.

This causes a multitude of problems for me, I just think that the sidewall has little or no strength at all, the elastic deformation of the sidewall almost makes it seem as it is not reinforced at all and it is just rubber.

Dan,

[AndroidXP]

Well, 4° isn't all that much when you think about it, or is it? Compared to what was good in pre S times it's extreme, yes, but who says the camber settings were *realistic* back then?

I guess we'd really need a mechanic here who could answer these questions with actual values used in racing, everything else will just be speculation of how we think it should(n't) be. :shrug:

[danowat]

Yes, very true, we need someone who knows more about tyres/camber than I do, at the moment I am basing my conclusions on the default RACE_S set, which I am assuming has "realistic" camber and tyre pressure settings, depending on weather this is indeed the case would make a big difference in whats right/realistic and what isn't.

I am all ears.

EDIT : Interesting piece here I found about Toyo tyres

" Camber Settings
Toyo motor sport radials will provide optimum cornering grip with camber settings between negative 3 and 6 degrees on steer axles. Where adjustment is possible, negative camber on drive axles will also improve cornering grip. Optimum camber angles will provide optimum cornering grip, and therefore will be an important contributing factor for improving lap times.

NOTE: On some vehicles a lack of negative camber can over-heat the outside tread shoulder causing grip levels to drop-off suddenly and in some cases cause blistering of the outside tread shoulder. Large heavy, front engine cars are most susceptible to this type of tyre damage.
CAMBER SETTINGS & INFLATION PRESSURES

Where camber adjustment is limited or class rules limit the amount of negative camber allowed, increasing cold inflation pressures is necessary to prevent the outside tread shoulder overheating. This also applies to vehicles that are used on the street as well as the track, where the camber settings are a compromise for this dual purpose. Where this is the case, adding 3 - 6psi to recommended cold and hot inflation pressures usually prevents the outside tread shoulder over-heating. "

Dan,

[Tweaker]

The only race cars I know of that have some wild cambers for the wheels is the V8 Supercars. But even then, the cambers are not all that wild on the rear.

http://www.lightmoves.com/news/v8supercar3.jpg

http://www.macktrucks.com.au/Content/Image/News/V8%20Supercar.jpg
http://www.ford-team-mittelland.ch/Motorsport/V8%20Superstar/V8%20Supercar%20Jason%20Bright2.jpg

But even for a slick, it doesn't deform as much as LFS I think :)

[AndroidXP]

@danowat: very interesting read. Seems not too different from the current situation we have in LFS :)

@Tweaker: yup, those cambers are really extreme... look something like 8° to me.

[ORION]

In LFS it depends a lot on the air pressure in the tyres, and because lower = faster, you need extreme camper to get the inside of the tyre in the ground ;)

[Gentlefoot]

Trade offs here too. Too much neg camber on the front of a front wheel drive will result in less traction as you unload the suspension out of a corner or from a standing start. Similarly, too much neg leads to poor braking grip.

But I think there is no hard and fast rule. Different suspension geometry reacts differently to loads which result in camber angles in bump and droop varying. So I don't see how anyone could say that static camber should be between x and Y because it varies from car to car, setup to setup. Other factors that would influence your choice of camber include the level of grip, corner radius, downforce, tyre choice etc.

Best approach is to test and feel I reckon.

[boxer]

There's something not quite right with the games tyre model imo. We took the BF1 around BL1 last night and no matter the camber/pressure settings the inside of the tyre was burning after a couple of laps - the tyre itself remained cold.

This shouldn't, and certainly doesn't happen to tyres designed for formula-one cars at least.

We finally managed to find a solution, fiddling with the tyre pressures and using high-compound tires. I still think there is something not quite right in the game concerning this..

[danowat]

Best approach is to test and feel I reckon.

There in lies my problem, I am testing and feeling, get it set up reasonably well, then I watch a reply with the F9 tyre display on, and its clear that on high load corners the outside third of the tyre is taking a huge bashing because the sidewall is collapsing, and to cure it the camber needs increasing to values were the inside of the tyre takes a bashing.

I understand that setting up a car is a balancing act, but I can't help but feel that the tyre sidewalls give up the ghost far to easily.

Dan,

[Gentlefoot]

What tyre pressures you using Dan? I have noticed that best tyre pressures in LFS are way too low. This would definately contribute to poor sidewall strength.

[danowat]

32 PSI on the fronts, which should be enough IMO.

Dan,

[Gentlefoot]

You're dead right there, probably even a bit higher than IRL for short runs anyway.

Is the problem occuring on all corners at Aston?

I always tend to set up for the two or three most important corners on a circuit.

Oh and I'm also slower than you so what do I know!

[danowat]

Its more of a general thing, rather than a specific corner or track, I first noticed it while setting up for KY Nat a few weeks ago, then I posted about the tyres deforming to much, now I believe its just the sidewalls that are giving the problem, I wonder if they have anymore strength in the sim then the rest of the tyre has, they just seem to stretch and pull all over the place.

Dan,

[Gentlefoot]

I wonder if other factors are contributing. If you're setup is generating good mechanical grip then this will put more force on the tyre sidewalls.

But it sounds like you have hit a problem with the tyres. Maybe they are modelled on Korean remoulds.

[Shotglass]

we had the discussion before in your thread about the tyres deforming too much ... most there agreed with me that it seems like the sidewalls strech instead of just bending ... which is wrong

[bal00]

When comparing the static camber of the roadgoing cars you also have to keep in mind the camber curves of the specific suspension layout. A car with a twist beam/trailing arms setup like the XFG will always need lots of static camber. Same goes for MacPherson struts on the XRG.

[Bob Smith]

I find I need to run maximum negative camber (both front and rear) on the UF1, even when the set is as stiff as low as I can get it. But I suppose it still suffers from a lack or ARBs.

[Chaos]

isn't the "problem" that in lfs you don't see many really stiff sets... I think that IRL the sets are pretty stiff...
or i didn't catch the real meaning of this thread, skimmed through only, cause i got a lot of work on my hands now (race weekend preparation and testing)

[Hahni]

we had the discussion before in your thread about the tyres deforming too much ... most there agreed with me that it seems like the sidewalls strech instead of just bending ... which is wrong

Maybe there is a difference between the visible tyredeformation and the deformation actually calculatet by the engine. To display real bend you need a lot of polygons, and I guess the tyre has in radial direction only two or three intersections. Could be that the bending looks like strecht, caused by lack of polygons

[danowat]

I am going beyond the visual representation of the tyre, and using the force bars and tyre temps to try and figure out what is happening to the tyre on high load corners.

My main concern is the amount of camber I need to run to get a decent "flat" tyre through a corner, and how quickly that amount of camber ruins the tyres.

Dan,

[Shotglass]

Maybe there is a difference between the visible tyredeformation and the deformation actually calculatet by the engine. To display real bend you need a lot of polygons, and I guess the tyre has in radial direction only two or three intersections. Could be that the bending looks like strecht, caused by lack of polygons

doubtful because of serveral reasons
a) most improtantly even if that is the case the sidewalls still strech a lot ... too much if you ask me ... even considering they might be bent normally and straighten out when they strech
b) it wouldnt fit into the general style of lfs being a wysiwyg or rather a wygiwys (what you get is what you see) game with a very close realtion between graphics and physcis
c) the performenace impact of having more sampling spots on the sidewall would be a lot higher than the impact of drawing the appropriate polygons

[Hahni]

doubtful because of serveral reasons
a) most improtantly even if that is the case the sidewalls still strech a lot ... too much if you ask me ... even considering they might be bent normally and straighten out when they strech
b) it wouldnt fit into the general style of lfs being a wysiwyg or rather a wygiwys (what you get is what you see) game with a very close realtion between graphics and physcis
c) the performenace impact of having more sampling spots on the sidewall would be a lot higher than the impact of drawing the appropriate polygons

Yes, right.

Just wanted to point out, that the visualisation has other restrictions to

[bal00]

My main concern is the amount of camber I need to run to get a decent "flat" tyre through a corner, and how quickly that amount of camber ruins the tyres.

Dan,

Well, who's telling you that this is the case in real life. If the tires of F1's were "flat" through a corner, the outside would never get warm.

[danowat]

Not totally flat, but at least flat enough to be able to "zero" the tyre temps across the 3 sections.

Dan,

[bal00]

Well, you want to even out the temps across the tires for a race, but how much camber or roll stiffness you need for that depends on the track. Braking and accelerating heats up the inside, fast cornering heats up the outside. When you use a one-size-fits-all set on a slow track, you'll cook the insides, if you use it on a track with lots of fast sweepers like Aston, you'll cook the outsides.

Plus, the FXO suffers from bad dynamic camber curves due to the MacPherson/trailing arm suspension. If the body rolls 4° and the tire rolls 1° in a corner, you need 5° of static camber in the rear to make up for that, because trailing arms do not at any negative camber when being compressed.
The MacPherson setup in the front adds some neg. camber during compression, but not a whole lot, so you still need lots of static camber.

[Ball Bearing Turbo]

I seriously cannot imagine having an even load across a tire that is holding a car on the road at around a G (or more). Even more so considering that most curves are not flat themselves. Perhaps I am full of dung, but cornering a car on it's limit, into an incline: how could this possibly NOT load the outside of the tire more without some SERIOUS camber? No matter how you think of it - if you have any reasonable amount of mechanical grip at the contact patch, pushing the tire to the outside is going to load the outside of the tire more than the rest, and the tire will "fold" as the contact patch desperately tries to retain it's composure while untold amounts of force are being applied to the tire. Would think you would need more than a few degrees of camber to compensate for such an extreme situation!! It's (camber) not meant to totally alleviate that, just make it better than without.

Fantastically, imagine a car sitting at a right angle to the ground, on a wall. (say it's being held against the wall at another 1g by a super powerful vacuum!) There's your 1G (both vectors). How much camber will you need to make the load across the tire surface even? :pillepall

[tristancliffe]

Bear in mind we can see the real time temperatures of the tyre, so we can set them up for crazy camber (for hotlapping). In real life they can only measure the temps in the pits after a run and get a very vague idea of tyre temps, unless they have rare and expensive in wheel temperature sensors. Also, we don't have to worry about knackering a set of tyres, or the cost of doing so, as we get new ones each time. And finally, we can run silly tyre pressures because it's not a real tyre, and isn't affected by manufacturing recommendations, nor risking blowouts etc. Basically we can have silly setups because it's not real life, and doesn't (and can't) have all the variables associated with real life.

[Glenn67]

No matter how you think of it - if you have any reasonable amount of mechanical grip at the contact patch, pushing the tire to the outside is going to load the outside of the tire more than the rest, and the tire will "fold" as the contact patch desperately tries to retain it's composure while untold amounts of force are being applied to the tire.

I totally agree with you, and have read all the threads regarding this, and find it perplexing how much discusion is going on about how wrong it is :shrug:
In my mind the forces view fits perfectly with what I'd expect to see happen :x maybe it just because its so different from the previous patches that people are having trouble coming to "grips" with it :scratchch :D
What is the bar representing in the forces view anyway (maybe that is adding to some of the confusion) is it the vertical component of force only or is it the total of all combined forces acting on that portion of the tyre? (i.e. longitudinal, lateral, vertical load, etc)

To get even tyre wear and the right temp profile accross the surface of the tyre on a race setup (i.e. with more camber than road suspension) wouldn't that mean that at some phases of the curciut (i.e. max cornering g force) that the outside of the tyre would be optimumly loaded more to compensate for the rest of the time (i.e. straight sections) where the oustside is practically unloaded...

And as to stretching side walls, why do people think its impossible for them to stretch and therefore it is wrong? That's why they put steel belts in radials to keep the shape of the treaded section of the tyre, I've never heard of steel belting in the side walls :shy: and any other material while it will reduce the strech in the sidewalls it will not eliminate it entirely... when you drive normally in LFS (i.e. as you would at street legal speeds in the city) you see quite normal looking deformation, it's only right at the very edge of the grip limit (i.e. forces are measuring in excess of 1g and peaking 1.2 - 1.4g for road normal tyres) that the tyres appear to be stretching and in my mind that seems normal?

What I would like to know (just curious :D) is if temperature affects how much the tyre deformation is in LFS, I know it effects grip but it should also affect deformation and the elasticity of the side walls.

[danowat]

My road car tyres, Continental Sport Contact II's (225/40/18) have reinforced sidewalls.

Dan,

[Gunn]

My road car tyres, Continental Sport Contact II's (225/40/18) have reinforced sidewalls.

Dan,A 40 series tyre has very little sidewall at all compared to a conventional road tyre, so perhaps not a good example to compare to the average tyre (60-75).

[danowat]

No, but I was just making the point that sidewalls are sometimes reinforced.

I don't know what the profile is on the FXO, but I would'nt imagine its much more than 40.

Dan,

[Gunn]

I would say it 60.

[danowat]

I would be extremely suprised if a 2lt turbo charged saloon would have 60 profile tyres IRL, a Subura Impreza has 45's, and the Astra VXR has 35's.

Dan,

[Gunn]

I would be extremely suprised if a 2lt turbo charged saloon would have 60 profile tyres IRL, a Subura Impreza has 45's, and the Astra VXR has 35's.

Dan,Be surprised then.

[Shotglass]

A 40 series tyre has very little sidewall at all compared to a conventional road tyre, so perhaps not a good example to compare to the average tyre (60-75).

all radial tyres have threads running through the sidewall (usually kevlar afaik) which dont strech much at all ... at least not half as much as they do in lfs

[Bob Smith]

The FXO has quite wide tyres so the profile is lower, only 35% IIRC (I'm at work atm so can't really check). The info is all sitting there in GRC... ;)

[Shotglass]

concerning sidewalls streching
ive quickly made 2 pics with an lx4 that show that the sidewalls behave rather weird atm
one of them shows two red lines obtained by drawing lines along the undistorted upper and the distorted lower sidewall and then stacking them to a single line so you can approximately see how much the sidwalls strech
the other one is a flat tyre (afaik modelled by a tyre with 0 pressure atm) with a sidewalls that streches way too much

[Tweaker]

ROFL!! that second picture (http://www.lfsforum.net/attachment.php?attachmentid=9718&d=1147483020) is some insane amount of stretching!!!

That is definitely not right too. The tire would fall off in that situation. The stretching is odd, I don't think the concept of the flexing is all that great right now. It should be pulled underneath, but the shape of the wheel should still retain its 'sidewall' somewhat.

I guess you could best describe it as "pinch and pull" or something.

The last picture I attached is somethiing I found online. This is of course an exagerration of tire flex, but it gives examples for the different sidewall structures. The econo tires on the UF1000 for example, they should be flexing a lot. But we have road_supers which need to flex differently. Performance tires are not developed to flex so much.

And these are some pics I found that someone posted on another forum. This looks BAD

http://img131.imageshack.us/img131/1079/lfss2alpha05t4hd.jpg

[Tweaker]

Two other pics here.

The old Maserati (beautful car I might add), with some classic tires that would probalby not even flex as much as any tire in LFS... but it still flexes quite a bit. Newer tires, like in the other pics, pretty much don't flex much at all.

Seriously, I think this effect is just there to show off the feature, regardless if it looks wrong.

[KiDCoDEa]

I see many people judging lfs physics, based almost solely on a lowpoly visual feedback interpretation of tyrecode (not refering just to this exact thread). Imagine doing the exact same for all other sims... rock solid sims those are.:scratchch

sometimes i wonder if its really nice to implement cutting edge crap, to be honest and "debuggy", providing analysis tools ingame, or if solid frozen cylinders like everyone else does without any kind of diagrams or force feedback vectors and bla bla would be better. kinda like, its a ferrari badge, now imagine all the rest. " we used real data and our sim is used by nasa". hide all debuggy stuff, call it hardcore mode and instant belief factor increases...

With this i dont mean the flex anim cant be tweaked, in fact it was tweaked by request for the bf1patch, and it obviously worked since many ppl only now noticed lfs had visually flexible tyres (which it had for so long already).

judge the visual feedback freely and contructively suggest with pics and vids how it can improve, but please, dont waste everyones (i guess) time, by typing physics essays based on how 7 polys (2+3+2) bend over and look.

[Shotglass]

I don't think the concept of the flexing is all that great right now.

really hard to judge which part of the model goes bonkers with the little information we have about how the model works (none actually, might be different for you 2)
but i guess its safe to say that its lacking something to simulate the reinforcments in a tyre

[Gunn]

all radial tyres have threads running through the sidewall (usually kevlar afaik) which dont strech much at all ... at least not half as much as they do in lfsWhat has that got to do with the quote you just posted? Taller tyres are more prone to flex, so what's your point actually?

[Tweaker]

judge the visual feedback freely and contructively suggest with pics and vids how it can improve, but please, dont waste everyones (i guess) time, by typing physics essays based on how 7 polys (2+3+2) bend over and look.
Yeah, it is only visual I think. Because the physics feel fine and well... if you watchin the live suspension view, the tires aren't flexing on that view :really: So I am inclined to think this is only a visual effect (most likely is) and has nothing to do with the physics.

I think it can improve by flexing with a few needed traits of the tire.

So if an 'economy' tire or Road_Normal tire were to flex, it would have a pretty heavy amount. Not a whole lot, but noticeable. Also dependant on the height of the sidewall/tire too. Road_Supers would have to flex with a 'sharp' edge on it kinda... and it wouldn't flex that much because of its typical low profile and better construction. Racing slicks would definitely still hold their shape and structure, but flex aswell... but I think they would stay 'solid' a whole lot more than that of road tires... regardless of the height of the sidewall persay.

Other than that, I think it would be something to consider. Sidewall height, and structure/strength of the sidewall depending on the tire type you choose. Oh and, minus the stretching ;)

[danowat]

This was my question in another thread, is the visual representation on the tyre model, true to the physics model going on behind the scenes?.

I would appreciate 2 points being answered, by the devs if possible.

1) Are the sidewalls at all reinforced

2) Is the visual represention of the tyre model true to the physics model

In any case, the pics that shotglass posted clearly show that the sidewall isn't deforming realisitcally (at least not visually)

Dan,

[Honey]

What has that got to do with the quote you just posted? Taller tyres are more prone to flex, so what's your point actually?
he meant that while the fact that sidewall flex is very true, the sidewall stretching is very false.

in real life stretching is very limited and a sim model could neglect the "tire stretching" and still being very very accurate.

the other thing that is wrong imho is the "tire folding" effect wich makes teh central part of the tire to fold innerwise and this is not clear if it's due to because of the wrong "tire stretching" alone or is also due to the excessive amount of modeling of sidewall load effect.

to answer to danowat: to me is very clear that the sidewall stretching affects phisics, because if you put very low pressure the car behaves like if you were lowering the roll bars value.

another thing i would say is that comparing with previous patch P, this problem seems to be already there but the stretching effect is much much less, the car that seemed most affected in patch P is the xfr.

i made the conlcusion, also reading between the lines of the autosimsport interview, that scawen knew he had to tune longitudinal and lateral grip and (imho) he tuned the laterla grip by increasing the sidewall stretch...but of course this is only a speculation of mine...

PS i always want to remark that i think that lfs is always the best sim with the best phisics and the latest patch was overall a huge improvement, but this do not mean we have to shut up and not trying to give the much as feeedback as possible to help devs make lfs everyday more close to perfection :thumb:

[KiDCoDEa]

tyre deflection was tweaked at bf1 patch time, (...) the contact patch does move around more (requested). In future this value will probably be tweaked. Also dif tyre types are gonna be considered.
I said that 2 threads ago. I dunno what else people want to debate.

slip curves did not change due to this request. this was a graphical feedback request that involved tyre deflection code, coz the anim of flex is based on the physics. no one in the team, bases their physics suggestions on how 7 polys look. clearly thats not the case in this forum. and thats why this is the public forum. where the public can write whatever they want.

[Shotglass]

What has that got to do with the quote you just posted?

because the quote was about theflexing being flawed

Taller tyres are more prone to flex, so what's your point actually?

flexing yes ... streching no

Yeah, it is only visual I think. Because the physics feel fine and well... if you watchin the live suspension view, the tires aren't flexing on that view :really:

the tyres in that view are solid blocks and dont strech at all