USF2000 Damper issue

Discussion in 'Bug Reports' started by benborp, Nov 13, 2017.

  1. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    I recently discovered that the USF2000 Championship car has some very extreme damper values. The slow damper settings reach as high as 33800 N/m/s. The spread of values is such that there are barely any useful damper settings that correspond with the spring ratings available.

    I have not had any success in researching the real world values of the dampers used in this series.

    Is this a feature of the car or an error?
     
    Lgel likes this.
  2. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,267
    Likes Received:
    365
    I don't know how you read those value (encrypted physics).

    By observing motec logs of suspension histograms and travel speeds, I find plausible velocities, that don't indicate such a gross over damping.

    Cheers.
     
  3. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    3,035
    Likes Received:
    1,654
    Isn't there a line in the player.JSON that lets you display values rather than "steps" for the dampers? Unfortunately what you won't have is the motion ratio for the springs, so calculations can be off if you assume they're 1:1.
     
  4. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,267
    Likes Received:
    365
    Yes you are right, the parameter is Damper Units that you can change to true to display units instead of clicks.
    I had it changed to true, last update to 1109 or one in the middle changed it to false.
    See nothing strange my rear slow damper setup reads 4100 N/m.

    Cheers.
     
  5. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    Which, because the base value is 1400 N/m/s adjustable in increments of 2700 N/m/s(!), is the only useable setting. All the other values of slow rebound are pretty much useless. The default damper values, where the slow rebound is 11100 N/m/s, do result in histograms and velocity traces that reveal gross overdamping. This is also true for the vast majority of the dampers' range of adjustment.
     
  6. Lukas Lichten

    Lukas Lichten Registered

    Joined:
    Oct 14, 2017
    Messages:
    96
    Likes Received:
    242
    Through my experience with developing carstat I know some things about the hdv file, were a lot of things are defined for the car, including the suspension settings.

    Now this is the rear left damper setting for the skip barber (avaible in the ModDev section of every rf2 install):
    Code:
    DamperMult=1.0                 // take into account suspension motion if damper is not attached to spindle (affects physics but not garage display)
    SlowBumpRange=(9250, 100, 1)
    SlowBumpSetting=0
    FastBumpRange=(2200, 100, 1)
    FastBumpSetting=0
    SlowReboundRange=(15000, 250, 1)
    SlowReboundSetting=0
    FastReboundRange=(5250, 250, 1)
    FastReboundSetting=0
    
    Now how ranges are defined in rf2 is that the first value is the lowest, the secound the step size and the third is the number of settings. With this you have SlowBump of 9250 and only one setting for it, but if it were 5 then the max would be 9650, which seams reasonable.
    The Setting is only which value is the default (here rf2 counts from 0, which is displayed ingame as 1).

    Here is the rear left damper settings for the usf 2000:
    Code:
    DamperMult=3.61
    SlowBumpRange=(3000, 2700, 11)
    SlowBumpSetting=3
    FastBumpRange=(2250, 36, 13)
    FastBumpSetting=5
    SlowReboundRange=(1400, 2700, 13)
    SlowReboundSetting=5
    FastReboundRange=(5900, 150, 13)
    FastReboundSetting=5
    
    Just look at the SlowBumpRange and SlowReboundRange. The step size is so high that the value gets nearly doubled from the first to the secound step (with the Rebound it is more then doubled).
    This is just broken.
    The fast settings look completely normal, and are how it should look like.

    Now this does not mean the lowest value is the best. Looking at the skipy you see that the slow bump and rebound should be higher then the there fast settings (with the skipy 3 to 4 times more). But the usf 2k is not an skipy, and I lack experience with other cars (and irl dampers in general), to give any guide were to go here.

    The only thing that I can say is that a damper that can multiply it's strength by 10 (24 times on the rebound) looks very very odd.
     
    SRGP, TeroD and Lgel like this.
  7. Lgel

    Lgel Registered

    Joined:
    Jun 24, 2014
    Messages:
    1,267
    Likes Received:
    365
    You are right, when I drove the car I felt the car was over damped with default setup, and reduced values, I didn't explore other values.
    Cheers.
     
  8. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    I'll do some work to see how the DamperMult affects performance. It could have a profound effect on these figures, but either way it makes one or other of fast and slow damper settings even more nonsensical.
     
  9. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    A DamperMult value greater than 1 would seem to increase the effective strength of damping. In the case of the USF2000 that means that the extreme slow damper values are not mitigated and it looks even more like an error.
     
  10. Lukas Lichten

    Lukas Lichten Registered

    Joined:
    Oct 14, 2017
    Messages:
    96
    Likes Received:
    242
    I think the multiplier is not the problem. It is just there to simulate the influence of mounting angles while maintaining a realistic strength. The issue of course is that this makes the max dampening even more severe, even though it is still "only" 10 to 24 times from the lowest to the highest, the N/m/s per step will still increase from the already way to high value, making increasing the dampening a change as subtle as sledge hammer
     
  11. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    I agree. I could envisage ways that the multiplier could be used that would mean a wide spread of damper values would be effective, but minimal testing would seem to show that is not the case.
     
  12. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    This doesn't seem to be coming from the claimed real manufacturers data. Jumping from 1400 to 4100 doesn't seem very logical IMO.

    The problem is that only when big errors like this are used, we can see that probably devs do not have as much data as they claim to have.

    It is the same as when Radical was released claiming to be the most accurate car and you can see that LH weight distribution is completely unbalanced, uncomplying the given recommendations of the owner manual.
     
  13. benborp

    benborp Registered

    Joined:
    Jul 20, 2017
    Messages:
    110
    Likes Received:
    89
    This looks very similar to a data entry issue to me. Roughly tenfold error to realistic values. Extra digit or misplaced decimal point.
     
  14. lordpantsington

    lordpantsington Registered

    Joined:
    Oct 5, 2010
    Messages:
    849
    Likes Received:
    79
    I have obtained some realworld data, granted it was for the shimmed (instead of DSSV, pre 2010) dampers, but a high value setting of 33800 N/m/s for low speed bump seems within reason for this car. If anything the low value end of the setting scale might be a bit too low. As a general rule it is better to find data, then draw conclusions. FWIW the suggested fitting position is 6 steps from fully closed. If those steps are linear this is somewhere in the range of 15-20 KN/m/s for rear low speed compression.
     
    Last edited: Nov 18, 2017
    benborp and Slamfunk3 like this.

Share This Page