Tires inner and outer temperature

Discussion in 'General Discussion' started by Pablo11, Aug 8, 2015.

  1. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    Pressures are consequence of inner carcass tire temps, that rise slower than surface temp of the tire.

    But you flat your tire when you rise too much the surface temp of the tire (you reach or exceed vapor temp of the compound).

    See davidporeilly reply just above.

    Cheers.
     
  2. Gonzo

    Gonzo Member

    Joined:
    Oct 5, 2010
    Messages:
    927
    Likes Received:
    292
    The temps you re seeing are surface-temps, which are completely controlled by slide / load / contact of the tire. Internal tyretemp is another kind of story.
     
  3. Gonzo

    Gonzo Member

    Joined:
    Oct 5, 2010
    Messages:
    927
    Likes Received:
    292
    Having exactly the same tiretemp on inside and outside is important IS A MYTH which exists since simracing exists.
    In the past there was a setup-tutorial which recommended to balance the temp, which is somewhat bull... .
     
  4. Ronnie

    Ronnie Registered

    Joined:
    Jun 15, 2012
    Messages:
    1,123
    Likes Received:
    296
    Yup. Someone spoke the truth finally. ;)
     
  5. Gonzo

    Gonzo Member

    Joined:
    Oct 5, 2010
    Messages:
    927
    Likes Received:
    292
    Plus a tyre has hundreds of points and hundreds of temperatures. Even tyre-engineers cant measure everything.

    You have surface all around the tire, surface on the side, within-material-temperatures, inside-surface-temperatures, inside-gas temperatures, rim-surface-temp, rim-material-temp and all this around the tire. Those termo-cams from the f1 in the tv show surface-temps, which are never balanced, thats it.

    Just grater your hands and create some temp, vary the load from one to the other side. Your hand will end up with hundreds of different temp spots ;).
     
  6. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    Hello Gonzo,

    Debunking myths is great.

    I very much appreciate it, but please could you replace the myth by some scientific facts, and help us the average sim racers?

    1. What is the right manner to observe tire temps in order to adjust camber and tire pressures?

    As you I have seen termo-cams on F1 replays, what I have seen seems logical, inner temps hotter than center and outside temps in straight due to camber, rise of temps across the width when braking and cornering. What I have never seen is center of tire much hotter than the rest of the tire.

    On the DW12 I have been unable to bring down center temp of rear tires, but may be someone can invalid my comment by explaining how he succeeded in doing so.

    2. If visual tools supplied by ISI for temps are wrong or error inducing, why are they still present in the sim?

    3. Is the new ISI tire model finished and working as expected?

    4. If not what aspects remain to be developed or fixed?

    Thanks for sharing your knowledge as an ISI tester with us, answering the four questions I asked would help the average guy enjoying this sim.

    Cheers and thanks for your help.
     
  7. Max Angelo

    Max Angelo Registered

    Joined:
    Oct 5, 2010
    Messages:
    4,958
    Likes Received:
    10
    A point of forced approximation, even in cars supplied with the CPM, is the current HUD window ... trusting on it, you have to believe that 1/3 of the tire tread has a constant temp, and that 1 mm from, eg, inner to middle section, has a sudden temp change of, say, 10C ... the HUD was done for rF1, which had a very simplified tire model compared to rF2, but in its current state "it is an insult" to the complexity of the termodynamic brush model of rF2.


    I hope that ISI soon or later "will translate" the TGM window to the HUD, because the TGM window was created for rF2 and it is the only tool to see the correct temp distribution along the thread.
     
  8. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    9,501
    Likes Received:
    4,517
    I think you need to avoid looking at this specific game and trying to work out how you're supposed to use it; if you search around you'll find a lot of car setup guides for real cars on real track days, and some guides specific to certain entry-level racing cars. The danger with a game is you can have telemetry that shows very specific, and very current, figures that just don't make sense in the real world. (there are more advanced gadgets these days, even for lower budgets, but a lot of the traditional guides that simracers have long based their own processes on are based on driving a number of laps and then measuring tyre temperatures in the pits. Very different to mid-corner or even mid-straight analysis)

    This thread is about inside vs outside; the centre temperature issue is something else entirely, I'm pretty sure ISI still say the contact patch isn't final so I wouldn't expect it to be 'completely' realistic yet. I can't comment on how much the new contact patch model has improved this aspect.

    You can see average surface temperatures in the game, probably fairly accurately reflecting what the real life surface temperatures would be. Do they relate to a lot of setup guides? Probably not. Telemetry has more figures available, but you still need to be aware they may not match the methods used by a real driver/team.

    Simpler games and sims had specific targets the programmers decided on, and code that used those 'magic numbers' to help determine your grip. rFactor1 tyres have an optimum camber value in the tyre file, which is an example of a magic number. rFactor2 models camber thrust from the physical design of the tyre, so you no longer have a figure you can plug in... you need to test specific scenarios to find out what works best.

    Never finished, and still acknowledged to be needing work. If you ever read that a game/sim has a finished and perfectly accurate tyre model be careful - if they believe it, they don't know enough about it. More likely they know it's not true but it sounds good for sales.

    The latest ISI tyres still have temporary parameters awaiting enhancements to the tyre model.
     
  9. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    Hello Lazza,

    Thanks for your reply, I respect a lot your knowledge, and I am a very grateful user of your tools.

    For my tests with the DW12 I used Motec and I focused on mid corner observations.

    If I play with RF2, it is because I am not interested in a sim where the only way to tune a setup is by pure trial and error, with no tools giving reasonable hints about the direction to go. This would be a reward to the people who can have the leisure to grind the greatest number of laps on every track with a given car.

    What I fear is that such a complicated tire model may end :

    - not being universal (same for all cars with different parameters), but different for every car.
    - so complicated that most of the cars available will end with "fantasy" tires.
    - be precise to the millimeter on some parameters but wrong by a mile on others.

    I hope that my fears will be proved wrong by the facts.

    Cheers Lazza.
     
  10. Gonzo

    Gonzo Member

    Joined:
    Oct 5, 2010
    Messages:
    927
    Likes Received:
    292
    1) It depends, even best paid f1 tire engineers have big ussues understanding tyres completely. Just remember the tyrewars and how difficult it was for them to understand those tyres. Increasing camber increases wear (on the inside).
    If you sprint 10 laps and wear is not an issue, inceasing camber might help. But it has to match the rest of the balance. Adding more frontcamber to an already oversteering car might increase oversteer and you slide more. With more sliding you got way more wear, etc.

    I had this in a leaguerace about 4 weeks ago. About 1 sec faster than the rest in the qual and a lot faster in the race-practices, too. With increasing trackgrip i decided to tweak my setup a little against the understeer. FAIL! After the start i was able to get a gap of 2 seconds and then i startet sliding in the fast corners which increased my tiretemp, wear, etc. now the three cars behind me were faster than me and catched up. I had a really hard fight with them and i needed a bit luck to stay in front. Then at the pitstop i decreased my front downforce and that gave my back the pace.
    This is rFactor2 ;)!

    2.) Those tools are not wrong. They just show the surface temp. In the Grand Prix they cant measure surfacetemp while driving (except with additional thermo-cameras). They ususally use infared thermometers immediately after the pitstop.

    3.) What does finished mean? If finished means that you get realistic racing with the leading tiremodell, then it is finished. If finished means that you can do way better racing than with rF1 or even with the initial release of rF2, than call it finished. If finished means you cannot improve it further more, than the answer is: no, its not finished. With current processors there should be some room left to improve, but steps are getting smaller.

    4.) I leave this up to the physics-pros. The tyre-modell is that complex, i am just a driver with advanced physics and rf2-sim knowledge. Assuming that too much details might be covered by my nda.

    Hope that helps and hope you enjoy rF2 as much as i do. ISIs guys are reading and collecting here carefully. So be sure that they know whats going on here and what are the right things to do in the future to have an even greater racing-experience. Sorry for spelling mistakes. Just rushed this down, got to go.
    Bye
     
  11. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    9,501
    Likes Received:
    4,517
    Bear in mind the more complicated the tyre model the more any 'facts' will be obscured. Already we're at a level where people can feel changes and improvements where there likely aren't any, and actually that's been happening for a long time (am I the only one who released an updated mod with good feedback, only to realise I'd packaged up the wrong files and everyone was still using the exact same thing as before? I doubt it, but it shows the power of suggestion).

    As for your fears, the first one is actually reduced by a better (more complicated) tyre model, the second is absolutely possible but a less accurate tyre model could be seen as always producing (comparatively) fantasy tyres. Ultimately it comes down to the data being used, the checks against real performance data, and the reliability of the model. ISI is trusted by people who ultimately make money from the use of their simulators which has to be more valuable than any gamer's evaluation (though they have their strong opinions regardless). Third party mods will vary, and could be harder to produce because of the tyre model complexity, but contrast that against a simpler tyre model (like rF1) that was easier to tune... but tune to what? Basically to what some guy sitting in his bedroom/loungeroom/basement decided was right, and a bunch of other games decided seemed about right.

    No simulation will ever be perfect, but they can and do get better. Unfortunately more accurate means more complex, which also means it's more difficult to tune a setup. There's a reason top racing series have teams employing literally tens or even hundreds of engineers, and even they don't always arrive at the same point.
     
  12. ErnieDaOage

    ErnieDaOage Registered

    Joined:
    Feb 26, 2013
    Messages:
    180
    Likes Received:
    0
    I think as long as ISI is providing different tyres along with different cars, modders can still use ISI tyres on their cars till there is more knowledge available how everything works. For example look at reiza and their team, those people gathered their knowledge of rf1 and his tiremodel over years before they launched GSC. For example i don't care if i race a 2015 GT mod with tires from the 2010 Corvette as long as the tires behave like real ones. I haven't done a lot of modding but i think its rather an easy job to change the size and width of a tire to fit to a specific mod than doing it all from scratch.
    The problem with simulations is still that you have to create a closed model in which everything works to its borders. But as soon as you cross that border its gets unrealistic. Like Spinelli wrote a lot of times his observations with the tiremodel and sims based on the gmotor engine with sliding and catching a car. If the model of rf1 is designed to allow for example to go 30% over the given limits in the closed model it will be correct till the 130%. If you go over it you will end up with not realistic behavior, but if you stay within its limits its pretty perfect. (stay on the road and on the curbs, stay within the limits of the car) That is also the point why a lot of race teams used rf1 and its engine for their own sims and driver trainings. the physics engine is pretty much universal but i wouldn't do a motorcycle mod on it and blame it unrealistic because its developed as a car sim. Same goes for me a bit with drifting and loose surface racing its a different kind of racing when you have a lot of controlled wheelspin. You have to make additions to the engine if you want to implement these things and i think thats also a point why ISI said that dirtracing will not come in the near future. Also they are working a long time now on the stock car content since it was announced because i think they are still working on the implementation of new features which come along with proper oval racing. Tim stated the stock cars will come along with a new build to get all features going depending on oval racing. And with the stock car tires you have at least 2 tires, indy and stocks, people can use for the oval car modding.
     
  13. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    @Lazza

    "As for your fears, the first one is actually reduced by a better (more complicated) tyre model"

    More complicated doesn't automatically mean better in simulations. You very often get better results with 10 parameters than with 1000.

    Cheers.
     
  14. Jamie Shorting

    Jamie Shorting Registered

    Joined:
    Sep 11, 2013
    Messages:
    2,628
    Likes Received:
    2
    Why don't you just go and play something else if the sky is falling?
     
  15. Jamie Shorting

    Jamie Shorting Registered

    Joined:
    Sep 11, 2013
    Messages:
    2,628
    Likes Received:
    2
    Is that a fundamental for todays tires, or a fundamental from racing sims in the past?
     
  16. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    Dear Jamie,

    Thanks once more for your contribution to this thread, that helps us a lot to understand the inner working of cars physics and the tire model.

    As you are full of suggestions for everyone, why don't you refrain to post when you have apparently no interesting contribution to bring to the topic debated, but a misplaced malevolent comment?

    Cheers and have fun racing.
     
  17. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,409
    Likes Received:
    919
    If those 1000 parameters are needed to take into account 1000 effects that happen in real life, then a good simulation must take them into the equation even if their "weight" is not detectable by the end user. If programmer skip 990 parameters because they are to hard to code, and the effort is not worth, then we have a problem, and the problem is that the software is not a simulation. Simulation fans are strange beasts, they want to know that those parameters are used, even if they matter only for the 0.001% of the time. This is the only way they take thing seriously, and they are right, if the programmer for first skip on the hard and tiny stuff, how can they be serious about that product?
     
  18. Gonzo

    Gonzo Member

    Joined:
    Oct 5, 2010
    Messages:
    927
    Likes Received:
    292
    This discussion is moving into the wrong direction.
    Please hold the topic and stay constructive.

    Keep in mind that testers and dev dont have much time for forum-readings. If you dont want them to get bored, keep short, keep constructive and they ll have more time to develop what we all love : rF2.
     
  19. David O'Reilly

    David O'Reilly Registered

    Joined:
    Jun 12, 2012
    Messages:
    1,015
    Likes Received:
    704
    I did some backwards testing of data.
    IE. I took a good set up and ran some laps.
    Decreased camber to a worse performance level and ran some laps.
    Put it back and lowered tyre pressures to minimum and ran some laps.

    What I found: (all are mid rubber readings)
    Decreasing camber increased the outside temp reading mid corner thus reducing the delta.
    With my fastest GT cars set the i/s and o/s temp delta was usually 25 deg C.
    Middle temp was somewhere around only 5 deg cooler than inside. So EG: 100-95-75 deg.
    Decreasing tyre pressure lowered tyre temp (opposite to my expectation in fact).
     
  20. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,246
    Likes Received:
    329
    Hello Davidporeilly,

    Did the car you tested use the new tire model?

    I ran some test with the new DW12 and my findings correlated yours, decreasing camber increased the outside temp. I reached a "normal" spread of temps in the front axle, I was unable to reach this on the rear axle.

    I found that even if tire temps looked stabilized after 5 laps, pressure wouldn't reach an equilibrium until 10 laps. As you I expected an under inflated tire to reach a higher pressure after 10 laps that a more inflated tire (due to carcass flex and deformation, as it is the case in real life).


    See my post in the new DW12 that I copy there for ease of access.

    Thanks for your contribution to this very interesting topic.

    "What I am saying is that as you know tire pressure Pi (i for instant) is base pressure Pb + Ph cumulative increment of pressure due to heat generated during driving.
    Resuming Pi = Pb + Ph.

    First eye opener for me was that even if I thought that after 5 laps tire temps were stabilized, Pi was still raising through 10 laps, I didn't have time to test on a longer stint. So if tires are not to burst they must reach a limit Pl (at least they do in real life).

    I thought a little more about the facts I have seen (tire pressures take at least 10 laps to stabilize), even if tire temps look stabilized after five laps. The heat from the tire must be exchanged between the tire (carcass and thread) and the air inside (Nitrogen for a racing tire), air is an excellent isolation (Nitrogen chosen to be even better), so it is to be expected that the gas inside the tire will heat at a much slower pace than the tire itself.

    So they are many question to answer:

    - in how many laps do you reach Pl depending on base pressure Pb?
    - does Pl vary depending on Pb?
    - should we wait to have reached Pl to observe tire temps and adjust suspension geometry as it seems logical?

    In real life, an under inflated tire runs hotter that a correctly inflated tire, and logically at a higher pressure (and consequently with a different geometry when subjected to loads).
    Even if competition tires are inflated with nitrogen which reduces the effects of heat build up compared to air inflated tires."
     

Share This Page