Pacejka probs

Hi, I've started using pacejka's magic formulae in my sim, but one of the *MANY* of my problems is that my pacejka lateral force equation returns a fairly large (about 100N) force when the slip angle is zero, so when my car is standing still it starts to rotate! I've checked the ceofficients with Racer's pacejka player, and I think they are right, but I have noticed Racer's cars do not turn on the spot as mine does! So its obviously a fault at my end somewhere, but i've compared my calculation of lateral force with racer's and it appears no different. Any ideas grateful! Then there's the angular acceleration of the wheels......aaAAAaargh! That can be a different post

>Hi, I've started using pacejka's magic formulae in my sim, but one of the >*MANY* of my problems >is that my pacejka lateral force equation returns a fairly large (about >100N) force when the slip >angle is zero, so when my car is standing still it starts to rotate! > That's odd. Mine actually switches between+/-90 degrees when it's standing still, making some really big forces, but they switch so quickly and the forces usually are fighting each other pretty effectively, so the car doesn't move... I'd check first to make sure the forces are going the right way (easy mistake). Obviously if you've been driving it around and it runs fine already, then that's probably not the problem.. If you can, run it in super slow motion (every integration step) and have the forces displayed in another window or in the 3-D view so you can watch what's happening. Better yet, draw the force vectors coming off the wheels. I've found and fixed a LOT of problems that way :-) You might need to get rid of the vertical offset in that data. Unfortunately, I don't know which Pacejka constant(s) do it. >Then there's the angular acceleration of the wheels......aaAAAaargh! That >can be a different post > Hehe.. It can be MANY posts... Todd Wasson --- Performance Simulations Drag Racing and Top Speed Prediction Software http://PerformanceSimulations.Com My little car sim screenshots: http://performancesimulations.com/scnshot4.htm

"Alex Smith" <[email hidden]> wrote in message news:a8g42p$rlj$[email hidden]...

Quoted text (7 lines)

> Hi, I've started using pacejka's magic formulae in my sim, but one of the > *MANY* of my problems > is that my pacejka lateral force equation returns a fairly large (about > 100N) force when the slip > angle is zero, so when my car is standing still it starts to rotate! > > I've checked the ceofficients with Racer's pacejka player, and I think

they

Quoted text (13 lines)

> are right, but > I have noticed Racer's cars do not turn on the spot as mine does! So its > obviously a fault at > my end somewhere, but i've compared my calculation of lateral force with > racer's and it appears > no different. > > Any ideas grateful! > > Then there's the angular acceleration of the wheels......aaAAAaargh! That > can be a different post > >

Hi Alex, Ah this a question I asked quite a long time ago, but I still dont see an answer. What Ruud (i think) and I have down is simply ignore this part of the pacekja equation. From memory I think you just remove the +Sh from the equation. What Sh gives you (actually it may be Sv?) is the force that is generated in the tyre from ply steer and conicity, I am pretty sure that this should be dependent on velocity. If you find a satisfactory answer to this I would be very interested in hearing it! Hope this helps a little All the best, Ash http://www.siroccoracing.com

Hi Ash! This is all a problem with the low speed behaviour where the slip angle and ratio aren't well defined. If you use the relaxation length approach towards calculating slip angle and ratio, a 'slip angle' will develop automatically through dynamics that will offset the conicity effect when the car is standing still (as it would all other lateral forces like standing on the hill). -Gregor On Thu, 4 Apr 2002, Ashley McConnell wrote:

Quoted text (20 lines)

> > Hi Alex, > > Ah this a question I asked quite a long time ago, but I still dont see an > answer. What Ruud (i think) and I have down is simply ignore this part of > the pacekja equation. From memory I think you just remove the +Sh from the > equation. > > What Sh gives you (actually it may be Sv?) is the force that is generated in > the tyre from ply steer and conicity, I am pretty sure that this should be > dependent on velocity. If you find a satisfactory answer to this I would > be very interested in hearing it! > > Hope this helps a little > All the best, > Ash > http://www.siroccoracing.com > > >

Aaaaaahaaaa, **penny dropping** Ash http://www.siroccoracing.com "Gregor Veble" <[email hidden]> wrote in message news:[email hidden]...

Quoted text (17 lines)

> Hi Ash! > > This is all a problem with the low speed behaviour where the slip angle > and ratio aren't well defined. > > If you use the relaxation length approach towards calculating slip angle > and ratio, a 'slip angle' will develop automatically through dynamics that > will offset the conicity effect when the car is standing still (as it > would all other lateral forces like standing on the hill). > > -Gregor > > On Thu, 4 Apr 2002, Ashley McConnell wrote: > > > > Hi Alex, > > > > Ah this a question I asked quite a long time ago, but I still dont see

an > > answer. What Ruud (i think) and I have down is simply ignore this partof > > the pacekja equation. From memory I think you just remove the +Sh fromthe > > equation. > > > > What Sh gives you (actually it may be Sv?) is the force that isgenerated in > > the tyre from ply steer and conicity, I am pretty sure that this shouldbe > > dependent on velocity. If you find a satisfactory answer to this Iwould

Quoted text (10 lines)

> > be very interested in hearing it! > > > > Hope this helps a little > > All the best, > > Ash > > http://www.siroccoracing.com > > > > > > >

Quoted text (9 lines)

>> This is all a problem with the low speed behaviour where the slip angle >> and ratio aren't well defined. >> >> If you use the relaxation length approach towards calculating slip angle >> and ratio, a 'slip angle' will develop automatically through dynamics that >> will offset the conicity effect when the car is standing still (as it >> would all other lateral forces like standing on the hill). >> >> -Gregor

Also useful to handle camber thrust when stationary, no doubt... :0) Todd Wasson --- Performance Simulations Drag Racing and Top Speed Prediction Software http://PerformanceSimulations.Com My little car sim screenshots: http://performancesimulations.com/scnshot4.htm

"Alex Smith" <[email hidden]> wrote in message news:a8g42p$rlj$[email hidden]... <snip> > I have noticed Racer's cars do not turn on the spot as mine does! So its<snip> Actually, many of them do. It's very very slow, but noticeable... park your car, go watch some TV, and come back to find you've rotated 30 degrees or so. Doesn't affect all cars.

What is the relaxation length approach? Is it in Genta or Milliken? Gregor Veble <[email hidden]> wrote in message news:[email hidden]...

Quoted text (10 lines)

> Hi Ash! > > This is all a problem with the low speed behaviour where the slip angle > and ratio aren't well defined. > > If you use the relaxation length approach towards calculating slip angle > and ratio, a 'slip angle' will develop automatically through dynamics that > will offset the conicity effect when the car is standing still (as it > would all other lateral forces like standing on the hill). >

Neither, its in a SAE paper."Tire Modelling for Low-Speed and High-Speed Calculations" All the best, Ash http://www.siroccoracing.com "Alex Smith" <[email hidden]> wrote in message news:a8k6p4$61k$[email hidden]...

Quoted text (12 lines)

> What is the relaxation length approach? > Is it in Genta or Milliken? > > Gregor Veble <[email hidden]> wrote in message > news:[email hidden]... > > Hi Ash! > > > > This is all a problem with the low speed behaviour where the slip angle > > and ratio aren't well defined. > > > > If you use the relaxation length approach towards calculating slip angle > > and ratio, a 'slip angle' will develop automatically through dynamics

that > > will offset the conicity effect when the car is standing still (as it > > would all other lateral forces like standing on the hill). > > > > >

On 04 Apr 2002 16:45:11 GMT, [email hidden] (J. Todd Wasson) wrote:

Quoted text (11 lines)

>>> This is all a problem with the low speed behaviour where the slip angle >>> and ratio aren't well defined. >>> >>> If you use the relaxation length approach towards calculating slip angle >>> and ratio, a 'slip angle' will develop automatically through dynamics that >>> will offset the conicity effect when the car is standing still (as it >>> would all other lateral forces like standing on the hill). >>> >>> -Gregor > > Also useful to handle camber thrust when stationary, no doubt... :0)

Yikes! Hope that is true; my inclusion of camber meant nothing but trouble with cars slamming to the side. ;-) As for Racer, it indeed has the jittery 'lateral forces bang back & forth all the time' effect at low speed (which is why the conicity forces don't really hurt). The latest beta (which does have a lot of huge camber problems) has a G-force indicator; the longitudinal version is steady enough but the lateral G-forces just go crazy when you're standing still. Hopefully indeed, Barnard's SAE950311 will clear this up (relaxation lengths). I'll wait for that till I get Pacejka's book on Tyre Physics; you never know what nice tire models are in there (or a more elaborate description of the Magic Formula, such us in what units you must input camber). Ruud van Gaal Free car sim: http://www.racer.nl/ Pencil art : http://www.marketgraph.nl/gallery/

On Thu, 4 Apr 2002 18:30:10 +0100, "Jim Seamus" <[email hidden]> wrote:

Quoted text (10 lines)

> >"Alex Smith" <[email hidden]> wrote in message >news:a8g42p$rlj$[email hidden]... ><snip> >> I have noticed Racer's cars do not turn on the spot as mine does! So its ><snip> > >Actually, many of them do. It's very very slow, but noticeable... park your >car, go watch some TV, and come back to find you've rotated 30 degrees or >so. Doesn't affect all cars.

Hehe, ok. Try downgrading the sim frequency (increase debug.ini's timing.integration_time) and the problem gets worse. Never EVER park your car in Racer and go out for lunch! ;-) Ruud van Gaal Free car sim: http://www.racer.nl/ Pencil art : http://www.marketgraph.nl/gallery/

On Fri, 5 Apr 2002 12:53:56 +0000 (UTC), "Alex Smith" <[email hidden]> wrote: >What is the relaxation length approach? Instead of directly calculating slipAngle, you make it a variable and update that. Makes the variable vary more smoothly based on the turning speed of the wheel. >Is it in Genta or Milliken? No, as Ash said, it's in an SAE paper, 950311. It's also mentioned, though very quickly, in: http://www.lib.ncsu.edu/etd/public/etd-373202299842941/etd.pdf (a thesis from Erik Lowndes). Ruud van Gaal Free car sim: http://www.racer.nl/ Pencil art : http://www.marketgraph.nl/gallery/

On Fri, 5 Apr 2002, Ruud van Gaal wrote: > > Never EVER park your car in Racer and go out for lunch! ;-) > Or in any Italian city for that matter (can easily cost you the radio). ;) -Gregor

On Fri, 5 Apr 2002, Ruud van Gaal wrote: > On 04 Apr 2002 16:45:11 GMT, [email hidden] (J. Todd Wasson) wrote: > > Yikes! Hope that is true; my inclusion of camber meant nothing but > trouble with cars slamming to the side. ;-) > Helps with camber thrust as well, naturally :). -Gregor

Thanks for the help guys! Although I haven't really progressed with it yet, the discussion is most appreciated! And its also nice to know I'm not alone with finding it difficult, no - near impossible :) Alex

On Fri, 5 Apr 2002 17:52:11 +0200, Gregor Veble <[email hidden]> wrote: >On Fri, 5 Apr 2002, Ruud van Gaal wrote: >> >> Never EVER park your car in Racer and go out for lunch! ;-) >> > >Or in any Italian city for that matter (can easily cost you the radio). Add to that Haarlem in Holland, where a couple of my CD's were stolen about 2 to 3 weeks ago. :) Funny how no-one in Grand Theft Auto 3 ever steals YOUR car. ;-) Ruud van Gaal Free car sim: http://www.racer.nl/ Pencil art : http://www.marketgraph.nl/gallery/