View attachment 10924 All cells are as they appear except the green ones which have the code shown at bottom - same in all green cells. Graph is selected to show source cells. No idea of load, how do I see that? I have tried all sorts of values, including those you suggested. It all works as expected on flat surface, but as soon as I hit the 24deg banking it is as if it immediately goes back up to 1. Max grip. I have your TS, I will try to catch you again. EDIT formula is =1/((($A$4*50000)+C$1)^C$2)*((($A4*50000)+C$1)^C$2) (image was resized too small)
On further reflection I think this is going beyond my rather limited knowledge of tyre behaviour. In a static sense the pressure applied to the rubber is equal to the internal tyre pressure (I was thinking overall tyre load before, which makes no sense when the TGM is defining the rubber rather than the tyre). A tyre at 100kPa (100000 N/m^2) will pretty much by definition equate to a contact patch of 1m^2 if it is supporting 100000N of weight (obviously in practice you'd need a large tyre to do this, but anyway...). Halve the weight and the contact patch area halves, so the pressure applied at the rubber-road point remains constant. Variations over and above the tyre's internal pressure therefore come down to what the model is doing, and that's something my entry level physics knowledge can't deal with. This appears to be very different to the rF1 (and rF2 AI) TBC definitions, where the tyre load -> grip multiplier is explitly defined (I think). Hopefully someone with more knowledge in this area can shed some light on what might determine grip at varying loads. Certainly at least one of machine's test values above should give very little grip at any normal loads, so either something in the model can't cope with what's happening under heavy (tyre) loads or there's something else determining behaviour which we've overlooked.
Someome hasn't boobooed and set camber as deg off vertical instead of degs off perpendicular to body. That would make our camber 24+whatever we are currently using in a 24 banked corner. That could explain massive grip increase. Thinking out loud.
There is not the possibility of a visual representation, something like a virtual test bed for wheel alignment, the geometry in motion without texture but only the frame, drive, axes and such ? Seems very cumbersome anyway only with numbers to deal and just to feel the result but not being able to really see the result in testmode ?
It would be helpful if there was the parameters LoadSensLat and LoadSensLong like we had in rFactor1 that showed the grip to load curve. Or if we new which of the new parameters in rFactor2 showed us the grip to load curve. Speed1, are you a modder ? It doesn't sound like you are when I see your posts. This is the Car Modding forum.
Maybe I interpret the implication of your question wrongly, but does it matter whether he is a modder or not?
I think maybe he's saying speed1 is asking questions about what is available, which gives the impression he hasn't done any modding and therefore doesn't know what's there. How important that actually is, is another matter Joel.Brown, that's really the question I think. I mistakenly read the RPSP line as being equivalent to the old TBC lines because I was thinking (just for that line) in terms of the tyre as a whole, instead of the rubber. So I'm not sure what's equivalent or if there is one - same as there's no longer separate lat and long grip, because the physics model is (sort of) determining those in realtime. (yeah, that's a pretty gross simplification, but anyway...)
The last tires I built work way better on banked tracks. Still learning why tires act this way depending on how they are built. Part of the problem is it takes 4 hours to build a new tire to test with any big changes.
No just impressed how much patience people have while i have great respect for the doing but when i read some posts here, some people seem to have no idea of the what they are doing. Is it lack of information or could i just understand the stupidy as soon as i'm a modder. Seems a bit painfull and awkward to me. Sorry but it actually looks like amateurs without a teaching program in some uncertainty. Exactly but intrested, just don't know where to arrive atm. Thx.
I would like to share some information on this. RubberPressureSensitivityPower=(-1.8,5000,750000,1) Essentially, the 2nd value changes the curvature of the load sensitivity, and the 1st value alters the overall severity. Of course as it's a power function, they are somewhat dependent on each other, of you try for example, RubberPressureSensitivityPower=(-1.8,5000,750000,1) the tyres will lose so much grip so early that they will be like driving on ice, so you have to change them in combination to change the 'shape' of the curve, but if you just want to change the severity, lowering the first number reduces grip more quickly with load. From ISI and hope that helps others some. Still working with them to understand more. The values are examples not what I am using right now. You have to experiment with them to get the results you want but it does work. Thanks ISI!
Yeah, it's effectively a y = 1/x curve, so increasing the 'offset' will move you further past the initial steep slope (so if you normalize it, you flatten the initial slope). Again, if you put together graphs based on the PDF formula (as machine has done) you can see how the two values interact. But still, unlike rF1, you can't treat the input load linearly here. When you defined the curve in the TBC you could directly work out your grip percentage based on the input (suspension) load to that tyre; in rF2 you're defining the rubber response, which will be affected by the size of the contact patch plus any variation in the pressures within that contact patch (something it sounds like ISI still wants to do more work on, so who knows exactly what's going on there). The tyre design/construction will affect the size of the contact patch with a given load, which in turn determines the average pressure on the rubber, and that (as an input to the RPSP curve) will determine the grip multiplier. So the answer does lie in the tyre design, assuming nothing else is broken. And I'm assuming my view of a static tyre is simpler than what the model does with a rotating tyre. Hopefully if you are able to make some discoveries with the tyre construction you are able to test behaviour on heavily banked ovals and work out if there's an underlying issue there or not.
Just to add something that came up in discussions with machine, a question we don't have the answer to (as far as I know) is how much load is on the rubber with a given tyre load. If the tyres were big balloons the contact patch would increase pretty much linearly with load, and the contact pressure of any particular part of the rubber would remain constant. (so if your tyre/balloon pressure was 100kPa, and the load was 5000N, your contact patch size would be 5000/100000 = 0.05 m^2 with an average contact pressure of 100kPa. Same load but double the balloon pressure would halve the contact patch size and double the contact pressure. Original pressure but double the load would double the contact patch size and the contact pressure would still be 100kPa. So in simplistic terms your internal pressure determines the contact pressure.) Tyres, on the other hand, have a shape and rigidity that determines the size, shape, and uniformity (pressure wise) of the contact patch. So you'd expect the contact patch size to increase with load, but only to a certain degree. So, if your right front tyre is going to reach 4500lbs of load, that would be... what... 4-5x resting load? What I'm thinking is, depending on your tyre pressure and the way the load is put into the rubber at the contact patch, you could be reaching the 'nominal max' part of the RPSP line. Presumably it'll then just keep that final multiplier, so your grip will increase linearly with load from that point. So maybe increasing the nominal max will help with heavy loads. Been a long time since I played with the ttool... does it give contact patch size for a given load? I remember it doing deflection...
Construction plays a big part. I increased the density of the rubber in the construction which keeps it from deforming too much and creating the huge contact patch like you discussed. That was the suggestion from ISI for Stockcar tires. I have not messed with it in a few weeks as I needed a break. But plan on messing around some more soon. This has been a great discussion and believe we are making progress on understanding the new tire model better.
To keep a 1600kg car in a radius of 75m at 80m/s, I did a rough calculation, got 7700lbs on each tyre - thats at 90 deg to vertical. Not sure how to apportion the part acting at 90deg to 24 deg track. Probably less, as there is an upward force component also.
There's a little more to that math. Example: 'calc lateral weight transfer Wf = (Kf * (TotalWeight * (Ay) * (NRA) - (W - TotalWeight) * (Y))) / ((sFTrackWidth / 12) * (Kf + Kr - TotalWeight * (Ay) * (Y) - W * (NRA))) + (TotalWeight * (Ay) * (b) * (sFRC / 12)) / ((sWheelBase / 12) * (sFTrackWidth / 12)) Wr = (Kr * (TotalWeight * (Ay) * (NRA) - (W - TotalWeight) * (Y))) / ((sRTrackWidth / 12) * (Kf + Kr - TotalWeight * (Ay) * (Y) - W * (NRA))) + (TotalWeight * (Ay) * (a) * (sRRC / 12)) / ((sWheelBase / 12) * (sRTrackWidth / 12)) 'calc banking weight transfer W1 = Cx1 * (W) * (b / (sWheelBase / 12)) W2 = Cx2 * (W) * (b / (sWheelBase / 12)) W3 = Cx3 * (W) * (a / (sWheelBase / 12)) W4 = Cx4 * (W) * (a / (sWheelBase / 12)) This doesn't take into effect aero and doesn't cover several other formulas that need to be performed before you get to this point.
