Pacejka scaling factors

[Silente]

Hi,

i have tried to look on the forum for an answer to my question, but didn't find anything useful.

I am using tire data from TTC, but need to correct maximum grip because, as reported in other posts, it seems to be to high. I am using then in a Pacejka tyre model.

I did thais grip correction using the peak grip coefficient for both longitudinal and lateral forces. But i noticed that doing so, also the shape of Force - Slip angle (or Force - SLip Ratio) curve is changing.

In this phase i am more interested to the longitudinal but i think it applies in the same way to both lateral and longitudinal forces.

My question is: does anybody know if and how to use also the other scaling factors (i am thinking to the shape factor in particular) to let the graphs to mantain a similar shape to the original one? Is there anything on how graph shape is changing because of poor road grip around on the web?

Just to be completely clear, the factors i am talking about are LMUX ora LAMBDAMUX, and LMUCX or LAMBDAMUCX.

Thanks

[CameronBeaton]

I understood up till this point....WHAT?!? I am guessing they are the variables for the normalised curves


Originally posted by Silente:

Just to be completely clear, the factors i am talking about are LMUX ora LAMBDAMUX, and LMUCX or LAMBDAMUCX.

Thanks

If you just want to scale down the graph why don't you use you own constant that multipies the whole equation. I know roughly how to use the scaling factors and for what you just said I don´t really think its worth knowing.

[Silente]

my question is connected also to how the shape of the curve is changing in real life when the road grip is changing.

Do we have to expect the tire to reach maximum force at different slip angles or slip ratios or does this peaks values don't change?

[BBolze]

I like to refer to the Adams website information on the various tire models. I've thought that a Pacejka '94 is enough depth of information for our cars (Do we really need dry steer information? I think that's more a production car concern for parking lots and such).

In that website is an overview of what each variable does specifically. This was very informative, but what it doesn't inform you of is how sensitive the data is with respect to these variables. When changing one thing, often many things change like you've just experienced, though the peak factor should not have really changed the shape much because as CameronBeaton said, it's just a coefficient applied to the whole function (usually 1/1000th of what you want peak to be or very close to that neighbourhood to get you started).

Most basic level instruction of Pacejka models use the "BCD" variables. These examples don't include the effect of weight, just a way to do a line function of slip angle vs grip at 1 particular weight.

The Pacejka '94 model i mentioned goes into more depth, just includes more variables to include the effect of weight. If longitudinal is what your mostly interested in your lucky as there are MANY fewer variables involved.

My first result when googling Adams Pacejka 94 was a PDF of the information i'm referring too; here's the URL

http://ti.mb.fh-osnabrueck.de/...06_Pacejka_89_94.pdf (http://ti.mb.fh-osnabrueck.de/adamshelp/mergedProjects/tire/Chapter_06_Pacejka_89_94.pdf)