The .tgm, and +ttool

Discussion in 'Car Modding' started by Tim Bennett, Jan 17, 2012.

  1. Michael Borda

    Michael Borda Car Team

    Joined:
    Oct 21, 2011
    Messages:
    187
    Likes Received:
    143
    Adjust wear using

    [Realtime]
    TemporaryAbrasion=1e-10 // m^3/J volume of rubber sheared per Joule energy (should be affected by temperature or possibly WLF in future)

    Keep in mind wear is a bit limited at the moment.
    None of the items under [Realtime] require regeneration of the entire lookup table. A bit more information will be provided with regard to these when the TGM doc gets an update.


    Almost certainly you need to use a DT multiplier of less than 1.0


    Information about tyre design can be found here;
    http://www.nhtsa.gov/staticfiles/safercar/pdf/PneumaticTire_HS-810-561.pdf
    You probably won't want to read all of it, although it's mainly focused at road tyres, it offers a great insight.

    For a quick reference I remember Rosberg once saying, F1 tyres are constructed of "Nylon and Polyester" woven together, while the belt is almost certainly made primarily of kevlar. Either way, common material choices are outlined in the above text.
     
  2. Bill Malicoat

    Bill Malicoat Registered

    Joined:
    Oct 6, 2010
    Messages:
    39
    Likes Received:
    0
    For anyone that is confused on cross section of the nodes, a picture is worth a 1000 words. Here is one of the M2009 tires, I did a quick plot of the nodes in CAD.

    nodes.jpg
     
  3. ArnoldWong

    ArnoldWong Registered

    Joined:
    Oct 5, 2010
    Messages:
    9
    Likes Received:
    0
    Does anyone know the unit of the "Young's modulus" in the tgm files?
     
  4. Niels_at_home

    Niels_at_home Registered

    Joined:
    Dec 30, 2010
    Messages:
    22
    Likes Received:
    1
    Usually it is very consistent with SI units, which would make it Newton / m^2 although in the past rFactor used kilopascal for pressure which is 1000 N/m^2 ..

    I put my 2 cents on N/m^2
     
  5. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    I think we need to some solid methods to convert load index, rolling resistances, shore rating of rubbers, speed rating etc. to values that are used in rF2 tires.

    No point for everyone to invent a wheel so to speak.

    For example load index of 82 is 475kg, which is suitable for car that has corner weight of no more than 237.5kg, but currently I don't know how to translate this to tire file, but I have not looked very much into it.

    Then there is speed rating, for V rated tires that is 240kph, now that needs to be converted to radians/sec^2, to get that we need to find out how many revolutions tire turns for 240kph and for that we need tire's circumference.

    185/60/14 tire for example has radius of 0.288 meters, so it will be 2 x 0.288 x 3.14159 = 1.809 meters

    As we know that speed is 240kph that is 66.67 meters per second (kph * 3.6 = m/s), we can now get that in one second at 240kph tire turns 66.67 / 1.809 = 36.85 rounds per second.

    We still need to convert that to radians, which is perhaps possible, but I have no idea how, so I convert to degrees first, as one revolution is 360 degrees, then 36.85 * 360 = 13266 degrees per second.

    To convert that to radians, there is formula of degrees * Pi / 180 so that gives us 13266 x 3,14159 / 180 = 231.53 radians per second.

    RotationSquared= requires unit to be (radians/sec)^2 so I would imagine that is 231.53^2 = 53608.54

    For me it looks plausible, but as I am mathematically challenged these days, there might be some very large error in this.

    What I would like to do is to make everything opened to this level, but I would need to understand at least some things or know few formulas to be able to do so.

    Tire size is made with nodes, that I have figured out too.

    Here first one is perhaps something similar to DryLatLong in TBC, but maybe value is not comparable as my understanding is that this is static ie. when tire is not rolling (or was tire static even when rolling?), maybe this is then scaled by lookuptable and other factors to make final level, but is this unit something I can get from test reports, for example Toyo semi slicks I remember seeing 1.45, some gravel rallying rubbers I have seen 1.3, direct translation from my language is friction multiplier, but I guess that is Mu, I don't know that much, so feel free to correct anytime:
    StaticBaseCoefficient=2.250

    This, for me looks like to be when tire is sliding, but I don't understand, rubber is same rubber sliding or not, why it is different?
    SlidingBaseCoefficient=1.44

    For street tires a lot less anyway, maybe max 1.1 like with TBC?

    Shore reading then, range for street summer tires is A 60-80, our winter tires can go below A 50, racing tires might be quite low too, but haven't really seen info from those, most I know comes from winter tire reviews as they tend to list tires by shore rating too, it affects also to tire wear, but adding of silica helps to get softer (lower) shore rating while still having less tire wear. This affects grip and also to wear, but lets say shore A 70, how to convert that to appropriate gmotor units?

    Rolling resistances then, haven't even yet looked into this at rF2, but I managed to find realworld data and tried to convert it to TBC file in rF1, but I still don't know how to do it for sure, my attempts are however bottom of this post, you can start reading from Edit2: if interested, rest of stuff on that tread is quite meaningless for rF2 and maybe in rF2 that is calculated differently, but data is there for yours to grab:
    http://www.nogripracing.com/forum/showpost.php?p=1478003&postcount=20

    Anyway longer TGM quide might explain some of those when it gets ready, but to convert real world units to gmotor units is still something we must find out how to do it, that is at least my guess.

    There is now some seeds for discussion, I hope that all answers will be found eventually so I can add it to my translation book, whatever that will then be called, rFactor for dummies like myself? :p
     
  6. Bill Malicoat

    Bill Malicoat Registered

    Joined:
    Oct 6, 2010
    Messages:
    39
    Likes Received:
    0
    This is because Friction = Force pushing / normal force (force pushing down on object)

    If you look at any Friction chart between two materials there will be a static and moving friction for each. Once and object starts to slide it takes less force to keep it sliding.
     
  7. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Ah that it is, just did not popped to my mind, thanks :)
     
  8. krivjur

    krivjur Registered

    Joined:
    Oct 5, 2010
    Messages:
    64
    Likes Received:
    7
    When you think it through radians from start, you get pretty simple formula:

    Angular velocity [rad/s] = Full circle in radians * rounds per second
    Angular velocity [rad/s] = Full circle in radians * Vehicle speed / circumference
    Angular velocity [rad/s] = 2 * Pi * Vehicle speed[m/s] / (2 * Pi * radius[m])
    Angular velocity [rad/s] = Vehicle speed[m/s] / radius[m]

    RotationSquared = (Angular velocity [rad/s])^2

    I checked Megane's front tire using the tire specs from .tbc comments and RotationSquared from .tgm. Should give you roughly 70 m/s = 252 km/h for veh speed. Matches close to the car specs I've seen.
     
    Last edited by a moderator: Jan 24, 2012
    1 person likes this.
  9. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Thank you krivjur, that will make nice and clean formula to Excel :)
     
  10. Domi

    Domi Registered

    Joined:
    Oct 11, 2011
    Messages:
    747
    Likes Received:
    44
    How can we adjust the grip level from the tires now? And heating? It's a mess now with 2 files, and so little info...
     
  11. Devin

    Devin Member Staff Member

    Joined:
    Dec 30, 2011
    Messages:
    235
    Likes Received:
    189
    Also my F1 tires are deformed too much in corners, even on straight. How do I change it?
     
  12. fanlebowski

    fanlebowski Registered

    Joined:
    Oct 19, 2010
    Messages:
    132
    Likes Received:
    1
    temporarybristlespring in first, IMO

    this is apparently in N/m per points
     
  13. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    This is done in TGM file, there is section called REALTIME, where you can find two grip parameters, sliding and static. You can change those and no need to recalculate tires, everything that is in realtime section can be altered without recomputing tires.

    Engineering toolbox says that rubber has 0.8 coeff when sliding, but then again tires are rubber, metal and air, so that does not necessarily apply. Chronus (at Nogrip board) says that street tires should have Mu 1.25 or bit higher, I can only guess that it is something that should be in static grip, or then that realworld value does not apply anymore and we need to figure out what does. Rubber has static coeff too, but I don't know if that either applies at all to tires and static parameter of TGM.

    There is lot to learn, I'm looking mostly how to get realworld information translated to rF2 units, like that speed sensitivity above which is currently only thing I know of, load index is still bit of mystery.

    I don't think we should pay too much attention to how tires deform at the moment as that feature is WIP confirmed by ISI.
     
  14. fanlebowski

    fanlebowski Registered

    Joined:
    Oct 19, 2010
    Messages:
    132
    Likes Received:
    1
    we can't translate coefgrip in rfactor1 directly. hum, sure we can but we'll have to "cheat" after with the loadsens lines to have the final grip excepted.

    about deflection, sure this will apparently change in future,but anyway, you'll have to know the deflections you want ;)

    in my point of view, search realdata as rf2 ask is a waste of time. find datas about slip curves,angle, ratio, deflections is already a nightmare. who will give you the accurate description of the tire ? :( at this moment, i still don't know how we'll able to do something with this new model.

    i don't see other way that ISI will offer a real modder ttool with enough outputs and examples to give an idea of the direction to take at the beginning.
     
  15. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Now I don't know which is true as I get two kind of information :p

    I think it is not idea to find data in units rF2 needs, but to learn how to translate behaviour in real world to rF2.

    For example that speed rating of tire is something that can be converted to rF2 units, many other parameters can be made so too, some needs perhaps bit more work, but for sure they are not made up or mystery units in rF2, so there have to be realworld equivalence and in realworld that depends from some other things, certain load index of tyre means it has certain stiffness etc. and knowing which parameters and in what kind of range means that there is lot bigger chance to have more realistic tires than when everyone tries to figure out same thing without being able to translate realworld to simworld by data but only by behaviour, which then results lot of mess which have been seen in rF1.

    That is why I think it is important aspect and certainly not wasted effort.
     
  16. fanlebowski

    fanlebowski Registered

    Joined:
    Oct 19, 2010
    Messages:
    132
    Likes Received:
    1
    well, we didn't talk about the same thing ;)

    actually, i was talking about find carcass information as rf2 ask. and others parts, but especially this one.

    sure all datas you can have and unterstand is important to make a physics, in all sims.

    but my point is on a game who is supposed to be a modding platform, IMO, we'll have to enter kind of datas we can find in realworld and the software will convert it as he want, as he need, and not the other way.

    this can be done by a tool, ttool or other, but at this moment it doesn't exist.
     
  17. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    This I agree completely, it would make great deal of improvements if sofwtare would even suggest parameters based on realworld tire, when one would input what kind of tire, load index, speed rating, size, it would put some base level of values, then one could tweak all as is now.

    But I think with this ttool it is perfectly possible for us as community to work together and make several sizes of different kind of tires that we as community consider to be realistic and share that on rfactorcentral or similar, it would make big difference already.
     
  18. Domi

    Domi Registered

    Joined:
    Oct 11, 2011
    Messages:
    747
    Likes Received:
    44
    I have tried to run some tests, but it keeps always on "Percent done: 0.005". Is there someone with the same problem? Any ideas?
     
  19. Tim Bennett

    Tim Bennett Registered

    Joined:
    Nov 4, 2011
    Messages:
    63
    Likes Received:
    0
    Lower your DT Multiplier Domi.
     
  20. Domi

    Domi Registered

    Joined:
    Oct 11, 2011
    Messages:
    747
    Likes Received:
    44
    Thanks Tim!
     

Share This Page