Conicity, ply-steer and others effects in Pacejka tire model

[ThreeColours]

Hi, folks. I've been reading you for a long time.

I'm studying about Pacejka tire model and I don't understand why we don't correct the effect of the conicity and ply steer in the model. For instance I've seen some projects about that and they didn't correct these effects in the model. Why ? These effects are due to the manufacturing process, they are different for each individual tire, and we can know it easily.

Cheers

[BillCobb]

Many users do correct for ply steer, ply R.A.T. and conicity in their tire models. Their efforts depend on many needs. You incorrectly state that 'we' don't address this. We sure do.

FYI, conicity and plyrate are specified in O.E.M. tire specification to address various performance attributes in production vehicle engineering. In race tires, these characteristics are 'managed'.

Search for S.A.E. papers written on this subject and read them.

[ThreeColours]

Thanks for your reply, BillCobb.

When I say "we" I don't speak about you or others forum users. I speak about some team members projects that I could see about suspensions design. Sorry but my english sometimes (always, perhaps) is not enough.

In Pacejka model (at least in pure side slip) is extremely simple to correct the vertical and horizontal shifts: For zero camber and fix vertical load I've calculated Sh and Sv and I've deducted this amount for the following curves. For this reason I don't understand why some team members didn't. I'm newbie I thought that, perhaps, I was forgetting some important aspect. Of course, I don't want that.

Sorry, What do you want to say with "O.E.M" ?

Cheers

[JT A.]

O.E.M. stands for Original Equipment Manufacturer, meaning the companies that produce cars like Honda, Toyota, GM, etc.

I am also trying to decide if I should correct these effects out of my tire model or just leave them in.

On one hand I've heard the argument "conicity & ply steer do exist in real life, so why would you remove them from your model? It will only make your simulations less realistic"

On the other hand, I think leaving the effects in the model can do some weird things. For example I've just recently started creating a combined Fy and Fx tire model to use for a lap time simulator, and the tire data says our tires produce ~500 pounds of forward thrust at 0 slip ratio. I'm not sure what the name for that effect is, but I think it has something to do with the slip ratio sweeps not being entirely steady state, and there is some lag in the response of the tire. When I start working on the lap simulator this may give me some problems, but I'll have to figure out how to deal with it when I get there.

[ThreeColours]

Originally posted by JT A.:
On one hand I've heard the argument "conicity & ply steer do exist in real life, so why would you remove them from your model? It will only make your simulations less realistic"

I don't think so. I think it is better to correct this effects because conicity and ply-steer are different for each individual tire. You know what is the conicity and ply steer of the tested tire but you don't know what is the conicity and ply steer of your tires.


Originally posted by JT A.:
On the other hand, I think leaving the effects in the model can do some weird things. For example I've just recently started creating a combined Fy and Fx tire model to use for a lap time simulator, and the tire data says our tires produce ~500 pounds of forward thrust at 0 slip ratio. I'm not sure what the name for that effect is, but I think it has something to do with the slip ratio sweeps not being entirely steady state, and there is some lag in the response of the tire.

I think we should correct these effects. My problem is do well. For instance, now, I have a negative camber thrust. I have Fy=0 when slip angle is -0.01º at 2º camber. At 2º camber I expected some positive lateral force to zero slip angle but I have a negative lateral force http://fsae.com/groupee_common/emoticons/icon_confused.gif.

Cheers

[exFSAE]

Ply-steer is not a random number for a batch of tires. It is a design feature, as BC mentions. You will find that it is a repeatable quantity. At the very least, if you were to take a batch of tires and look at these values on the aggregate they are non-zero.

Depending on the application you may or may not want to eliminate things like ply-steer. For that matter, you can learn quite a bit with a Pac tire model stripped down to its most simple form with arbitrary coefficients. Personally though, I don't see any good reason to strip these effects out and rigidly say that Fy = 0 at SA = 0.

JTA - You're probably looking at FX vs. slip ratio (SR) rather than vs. longitudinal slip (SL). Slightly different meaning between the two. You will find that in the latter case, Fx = 0 at SL = 0, and is probably what you want in a tire model.

[Edward M. Kasprzak]

JTA - You're probably looking at FX vs. slip ratio (SR) rather than vs. longitudinal slip (SL). Slightly different meaning between the two. You will find that in the latter case, Fx = 0 at SL = 0, and is probably what you want in a tire model.

Thumbs up!

[ThreeColours]

Originally posted by exFSAE:
Ply-steer is not a random number for a batch of tires. It is a design feature, as BC mentions. You will find that it is a repeatable quantity. At the very least, if you were to take a batch of tires and look at these values on the aggregate they are non-zero.
Thank you very much, exFSAE. You have a point there. Ply-steer effects are due to tire design therefore we have a narrow distribution of values. I love ply-steer.


Originally posted by exFSAE:
Depending on the application you may or may not want to eliminate things like ply-steer. For that matter, you can learn quite a bit with a Pac tire model stripped down to its most simple form with arbitrary coefficients. Personally though, I don't see any good reason to strip these effects out and rigidly say that Fy = 0 at SA = 0.
But Can I separate the two effects? For instance, Can I say ? "Sh (vertical shift) is due conicity and Sh (horizontal shift) is due ply-steer." I want to eliminate conicity. I haven't seen anything like that in my books but I think I've seen in... http://fsae.com/groupee_common/emoticons/icon_confused.gif ... I don't remember, damn !!! http://fsae.com/groupee_common/emoticons/icon_mad.gif.

Honestly I don't think that is possible. If I'm not wrong it's like finding A and B knowing that their sum A+B is worth 10.

Cheers http://fsae.com/groupee_common/emoticons/icon_wink.gif

[BillCobb]

Not with the data you have here. You need a high resolution test with a tire run in both forward and reverse directions in order to separate the two effects. I don't mean left and right turn directions.

[ThreeColours]

Originally posted by BillCobb:
Not with the data you have here. You need a high resolution test with a tire run in both forward and reverse directions in order to separate the two effects. I don't mean left and right turn directions.
Yeees, that's so true. Thank you very much, BillCobb

[exFSAE]

Just my opinion, but my gut feeling is not to worry about separating out the conicity. It's going to be a small number - I'd bet the majority of zero slip behavior is ply-steer and camber thrust.

In any event, as BC says - given the available data you won't be able to tare it off anyway.

[ThreeColours]

Originally posted by exFSAE:
Just my opinion, but my gut feeling is not to worry about separating out the conicity. It's going to be a small number - I'd bet the majority of zero slip behavior is ply-steer and camber thrust.

In any event, as BC says - given the available data you won't be able to tare it off anyway.
Yes, I agree with you. Thanks. I won't get gain on it. Cheers