How do I generate a slip ratio vs friction coeff graph in ptool?

Discussion in 'Car Modding' started by PTRACER, May 23, 2022.

  1. PTRACER

    PTRACER Registered

    Joined:
    Sep 26, 2021
    Messages:
    6
    Likes Received:
    2
    When I use the tyre tool to generate a slip curve, it seems that even at a slip angle of 60 degrees, the tyre is generating 93-95% of its friction compared to at peak slip angle.

    All tyre modelling theories / graphs I have seen online and in other racing sims have a drop-off down to 70% on a dry surface and even lower for a wet surface.

    I am presuming that is because I need a slip ratio graph rather than just a slip curve?

    And if so can I use pTool to generate such a graph?
    [​IMG]
     
    doddynco and djbob like this.
  2. Jokeri

    Jokeri Registered

    Joined:
    May 17, 2014
    Messages:
    204
    Likes Received:
    32
    Use the tyre spreadsheets.
     
  3. PTRACER

    PTRACER Registered

    Joined:
    Sep 26, 2021
    Messages:
    6
    Likes Received:
    2
    I am already using the spreadsheets provided by Studio 397. However as I said, the slip curves for all tyres, even the vintage ones, don't go below a normalised friction coeff of 0.95, even at extreme slip angles. At 8.25 degrees, the BT20 tyre produces a peak friction coeff of 1.35, at 60 degrees it is 1.30.

    I'm looking to generate something more like the green and blue graphs above, which I presume requires the slip % to be involved in the output.
     
    Last edited: May 26, 2022
  4. redapg

    redapg Registered

    Joined:
    Jan 16, 2012
    Messages:
    3,959
    Likes Received:
    2,809
    @PTRACER Maybe the Spreadsheet and ttool can only generate the "usual" Graphs, because the Pacejka Calculation Model gets used?
    Some Time ago, i have written a Script, following his Calculations, with that you can generate Slip Curves and that also doesn't allows too big Differences in the Curve.
     
  5. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,346
    Likes Received:
    6,572
    The ttool testing uses the full tyre model, so simplified calculations shouldn't be an issue.

    I haven't personally played with tyres enough to give solid examples (I've run through the guide to produce a table from ttool, and that's it), but I would be wary of expecting the normal slip curves you see around the place. Here's Niels with an interesting look at isolated vs combined tables:
     
  6. Robin Pansar

    Robin Pansar Member

    Joined:
    May 20, 2018
    Messages:
    139
    Likes Received:
    107
    As most people have mentioned here, you can't really adjust the slip curves (much) without adjusting things on the tyre construction that affects how the slip curves behave. It's very advanced stuff. Change things to help X and you may also affect A, B, C, Z Y, etc - so you basically need to know what you are doing unless you want to spend hundreds of hours doing trial and error.
     
  7. Simulation_Player

    Simulation_Player Registered

    Joined:
    Mar 4, 2022
    Messages:
    707
    Likes Received:
    449
    I see , so in rfactor 2 you are quite literally making a tyre , so every changes you make affects other parameters.
    Wouldn't it be better to separate all these parameters so that changes made to X don't affect A,B,C etc ?
    is this change possible in rf2 ?
     
  8. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,346
    Likes Received:
    6,572
    Then it wouldn't be a physical model, so you couldn't generate the multitude of internal lookup tables used in realtime. At that point you're back to an rF1 style tyre with a handful of slip curves.

    Conceptually what you say is possible, where you might say "use this construction, but then make X bigger" and to a small extent this can be done through the tyre's realtime parameters, but the more you allow that the more you move towards fantasy tyres (and handling) than simulation.
     
  9. Simulation_Player

    Simulation_Player Registered

    Joined:
    Mar 4, 2022
    Messages:
    707
    Likes Received:
    449
    even if it this makes fantasy approach to tyre creation, for my point of view it seems to give more freedom to devs to match virtual tyres to real data. I think i heard about a hybrid tyre model, this way we get best of both worlds.
     
  10. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,346
    Likes Received:
    6,572
    Where did you hear that?
     
  11. Simulation_Player

    Simulation_Player Registered

    Joined:
    Mar 4, 2022
    Messages:
    707
    Likes Received:
    449
    hybrid might be a bad word , i think it is knows as semi-emprical and semi-physical based tyre modeling. ofc idk if that improves on rf2.
     
  12. PTRACER

    PTRACER Registered

    Joined:
    Sep 26, 2021
    Messages:
    6
    Likes Received:
    2
    Hey thanks for that, I found your site a while ago and found it useful.

    Could you tell me exactly how you got this data? It is exactly the sort of thing I am looking for - curves that drop off from 1.000 down to about 0.700, as opposed to curves that go from 1.000 to 0.960
     
  13. redapg

    redapg Registered

    Joined:
    Jan 16, 2012
    Messages:
    3,959
    Likes Received:
    2,809
    Do you mean the Data for the Formulas?
     
  14. PTRACER

    PTRACER Registered

    Joined:
    Sep 26, 2021
    Messages:
    6
    Likes Received:
    2
    Yes, the raw data for the actual curves. i.e.:
    http://meetme.bplaced.net/rF2_onlineTools/rF2GGeninfo/Dateien/TBC/JPG_th/Eve68MS10.jpg

    I am testing these tyres in ptool and getting slip curves like this. This is from the Eve MS10 rear tyre at a load of 1400N :

    [​IMG]
    How is it possible for the tyre to be providing basically the same level of friction from 3 degrees all the way to 20 degrees?
     
  15. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Good question. I have seen many curves like this. I must say first of all, I don't know. Secondary, I think it could be just basic curve which doesn't include secondary effects that happen when tire is sliding completely for extended time, and also at different speeds - higher speeds sliding is harder. I think this only shows mechanical part, which involves static vs sliding contact patch portions dynamics, basically what is seen there is diminishing of static friction area till it goes away, and further on it is all kinetic friction by fully sliding tire. Obviously it will change depending on how much difference is there for static friction vs sliding friction. Just guessing there, I think what we don't see there is further drop of curve that would happen due: heat accumulation, different velocities of sliding, shredding off of a rubber and perhaps few more factors that makes a difference in event of loosing grip. Also for aero cars there would be loss of normal force due loosing downforce with yaw angle, but it is next thing after tires.

    Speaking of real life curves, IIRIC I have read in some iRacing blog, that the curve much further past peak is usually not even tested. Maybe because IRL they just suppose it isn't worth it because tire is not meant to be used in that range to begin with. Surely for simulation it is very important range, as it can make so much difference on the way car must be driven and level of concentration. In my personal opinion, it is often the case that static friction is set bit too low, and eventually gets compensated with higher sliding friction, simply because laptimes comes from net friction - sliding friction+static friction. Old bias ply tires should be expected to have less peaky tire, is this tire good in rF2 ? For that matter I simply do some mad observing of old footage, and then do proper testing in game. And right now top motorsport is going hard with simulation, so I'd suppose they should pursue to acquire such critical data. Maybe it is very secret, maybe it can't be properly tested in testing rigs controlled environment.

    In case that is complete slip curve with everything involved, then it is just weird to me.
     
    Last edited: May 28, 2022
    doddynco likes this.
  16. redapg

    redapg Registered

    Joined:
    Jan 16, 2012
    Messages:
    3,959
    Likes Received:
    2,809
    That Data is coming from the tbc Files of the different Cars.
    I have to say that i'm no Engineer :), so i can not explain in Deep the Calculations that the Pacejka ’94 Longitudinal Formula uses.
    And i also can not say why they look like they look or how these Curves get handled by the Games Code.

    On the Tools Site i have linked a Site that gives some Info to the Formula, where you can see that the shown Curves look like that and another Site with some Information from ISI (the original Developer of rFactor), that can maybe help you.
     

Share This Page