Sim Racing servo ffb systems : OSW & Bodnar

Discussion in 'General Discussion' started by Adrianstealth, Jun 1, 2015.

  1. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    What do those values relate to?
     
  2. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    I don't get caught up in all this Nm talk although it's a guide to a servos capability

    think my servo is capable of 21Nm but purposely restricted to 16Nm by bodnar's firmware
    ( perhaps for safety or maybe to run the unit well within its limits, perhaps this is why it doesn't clip )

    at the settings I run at it can be very powerful yet my settings are no where near max

    if I turn it up there is a stupid amount of power, wrists/fingers will ache the next day lol
     
  3. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    The bodnar uses a kollmorgen AKM52G motor. The stall (i.e. zero speed) torque it is capable of sustaining indefinitely (at 40 degrees Celsius room temp) is just over 8Nm. The peak torque (i.e. the maximum short duration torque output deemed safe by the manufacture) at stall is somewhere around 21Nm, which is usually a rating that is only meant to occur for a few seconds (as it builds up heat and fast).

    Leo may have found that 16Nm was the best compromise to use as the peak torque of the motor for sim racing use and hence why he has software or hardware capped it to that. Hypothetically for example motor performance and consistency only start to decline after 1 minute of 16Nm continuous torque output (a scenario you would not expect to encounter in the sim on track anyway) whereas a short stints of demand of 21Nm from the motor over a few laps may start to decline the motors performance significantly. And so there is a much lower chance (perhaps even non-existent chance in practice) of the motor dropping in performance with 16Nm as the artificial limit when using it for sim racing.
     
    Last edited by a moderator: Jun 5, 2015
  4. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    yes that's what I said
     
  5. Joe

    Joe Registered

    Joined:
    Nov 14, 2013
    Messages:
    372
    Likes Received:
    21
    The setting shall be car and track dependent, as well as your driving dependent. Full true simulation, as you would like to have (since you paid that much $$ for), you shall calibrate your wheel to 1:1 as real car, or close that.
    (you probably will never able to get true 1:1). For OSW white paper they did just like that for a recorded lap with iRacing’s Lotus 79 at Mosport. See attached chart.
    Now you know why you need to set the wheel at 30Nm not 16Nm.

    "if I turn it up there is a stupid amount of power, wrists/fingers will ache the next day"
    No you will not. You need to turn down the in-game FFB scale. So your FFB distribution is close to 1:1 torque curve. Otherwise, you spend high % portion of time with high torque.
    View attachment 16907
     
  6. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    I experienced that with Adrians bodnar with only set 30-40% overall strength (making it's peak torque around only 5-6.5Nm which is only a little more than the peak torque output possible from my T500). I don't think the sensation of extreme brutality i received from it actually stems from the torque output alone but from the combination of torque output and most importantly the significantly shorter transient response time of steering position change caused by torque output. Part of that due to the DD nature of the system (when comparing it to other ffb wheels) but when comparing it to real cars, it stems from the motor having a very significantly lower rotational inertia than a real cars steering system (which the ffb wheel is meant to try and simulate) rotational inertia (which is vastly greater on a real car).

    So even if you managed to mimic the torque output 1:1 of the real car on your ffb wheel, the significantly lower inertia of the ffb wheel results in a significantly faster and larger steering wheel change in direction and position (which is unrealistic ofc). For example, if the ffb wheel had a 22x smaller inertia than the real/virtual car it's trying to simulate, then for the same torque outputs on/from the virtual steering system/wheel, reproduced perfectly 1:1 on the ffb wheel will also generate an velocity and distance change that is 22x higher. The result of which is a very snappy feeling steering wheel than perhaps it should/would feel like in reality. You feel that you must apply a death grip that greatly exceeds what you should need to apply in real life to control the steering wheel.

    There are some rudimentary ways to try and minimise/reduce this problem (with artificial spring and damper settings) but non are perfect and can be much better but not without a complete redesign. ;)
     
    Last edited by a moderator: Jun 5, 2015
  7. Joe

    Joe Registered

    Joined:
    Nov 14, 2013
    Messages:
    372
    Likes Received:
    21
    Well, feel on a wheel itself seems really personal preference. Even real racing car drivers change their wheels all the time. We cannot invoke this into context of 1:1 meaner. Ideally, sure one shall use a real race wheel if one goes to 1:1
    Then what is real race wheel?

    A race driver made every thing his own, just fit him perfectly in order to win! Even a pad of brake pedal:
     
  8. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Ah, you misunderstand me due to my poor wording. Where i said "the virtual steering system/wheel" i should have just said "the virtual steering system" which involves the steering wheel, steering column/shaft, steering rack and the most significant of them all the two front steering tyres. The rotational inertia of a steering wheel represents only around 5% of the total steering system rotational inertia of a real/virtual car. So even if you used a real car steering wheel on the ffb wheel, the problem i described would remain.

    In summary, using ffb wheels (even the best ones) to simulate a real cars (or in this case the virtual cars) steering system rotational inertia is like trying to simulate/predict/calculate the resultant change in motion of a bowling when a given force is applied (or more accurately speaking when a given amount of momentum is transferred to the bowling ball) by using the mass/inertia properties of a baseball. The simulation/prediction just won't be correct. If you transfer the same amount of momentum to a baseball as to a bowling ball, the baseball will be given some 20x greater change velocity than the bowling ball if the baseball was 1/20th the mass of the bowling ball.
     
    Last edited by a moderator: Jun 5, 2015
  9. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    I have it set much differently now dripper

    88% force ( in the bodnar profiler ) , unbelievably it seems each extra % makes the system behave just subtly differently
    ( unlike other wheels I've used ), obviously the im-sim settings are much lower

    it feels good through the settings of course, I think Barry had his set at 40% (is that right?)
    at higher levels it's a tad more aggressive (in-sim force settings down for some compensation )
    & subtle detail is slightly amplified ( maybe good for me as I have x7 vibration units on my rig now )

    also "heat fade" your referring too, on these kind of systems the servo will not reduce power or degrade if its in high powered hold situation for a period of time .......it will start to heat up, then cut out to protect it self ( no damage to system ) the performance will not degrade at all in the very slightest to this point ( obviously nothing once cut out though lol)
    so there is no heat fade ( performance degradation ) on these systems

    if some Mellon head is quoted these claims on Google then they are not running one ( a servo motor) in a system that's a part of a full industrial grade system

    p.s I doubt anyone will ever have a safety system protect cutout happen for sim racing application
     
    Last edited by a moderator: Jun 5, 2015
  10. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    ps about strength Nm's

    please ignore these claims ( & charts ) if you can, they have use but I think they are giving you a false sense of logic, go & try a servo system & ramp the settings up ....good luck

    Google is sometimes not your friend
     
  11. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    pps dripper did your fingers ache the next day after using my wheel ?
    ( as that's how it use to be set , it's set a tad stronger now)

    strange as an ex-formula Renault driver visited ( check out some of the latest modifications etc on my rig), he said it was almost unbearably powerful

    I just think his muscle stregth has reduced and he's tuned into the fairly weak
    (in comparrison) forces of the t500
    (the t500 can give some force but then clip city)
    ( to add the t500 is a brill wheel for the price )
     
    Last edited by a moderator: Jun 5, 2015
  12. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    I'm speculating here that most sims are the problem, rather than trying to use the wheel at 20, 25, 30 etc. Nm. My thinking is that because if a sim is not "boosting" forces, and all the FFB is programed to run at 1:1 with physics, then setting your wheel's dynamic range of FFB from (e.g.) 0-30 Nm should not be introducing any insane type of finger-killing sensations in most normal racing situations. Afterall, these are the same type of torque figures recorded in real-life, therefore the sim's FFB output would be to blame rather than the fact you have your wheel set at a max Nm which covers most in-game cars, most of the time, on a 1:1 scale.
     
  13. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    Dennis@OSW builds OSWs with all 3 motors (Lenze, big MiGe, small MiGe) for people. The following is what he said about the lower resistance ratings:

    source --> http://isiforums.net/f/showpost.php?p=355650


    Another quote...
    source --> http://www.bsimracing.com/simxperience-accuforce-pro-available/#comment-1769586586

    [​IMG]



    Getting frustrating trying to figure all this out. I can't decide whether to use an over-amped (apparently they over-amp safely and reliably with just a bit of minor cooling added) small MiGe due to the lower inertia - which is still quite a bit higher than the Lenze - or the big MiGe :(


    P.S. If I remember correctly from working at an F2000 school for a year, the steering had almost no resistance when the car was lifted in the air. Therefore almost all the steering resistance/dampening should come from the car being on the ground. That would mean that the resistance/damping should come from the game's/motor's FFB output rather than the natural wheel's motor...Therefore, if a wheel's low natural resistance/damping/inertia is making the FFB TOO reactive/quick/sharp etc., then I would presume that means the game's FFB output is causing that behaviour rather than the actual raw motor/wheel/unit itself. No?
     
    Last edited by a moderator: Jun 6, 2015
  14. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Well, if you did had some ffb clipping before, increasing the bodnar's profiler ffb strength and lowering the in game ffb multiplier would certainly reduce/remove it and feel better.


    Oh ok, makes sense, considering the premium on of the motor they would most likely use thicker wires for the winding wires to keep energy losses from heat production to a minimum (thereby avoiding the heat fade torque lose issue present on many mainstream ffb wheels, which was pretty awful on my CSR-elite).

    Just saw the akm52g has a thermal limit of 155 degrees celcius around the windings. So as long as the average use torques does not cause the internal temperature of the windings to exceed 155 degrees celcius then it's all fine.

    How hot does your motor feel to you? It would be interesting to see how hot it feels if you set it to 100% overall effects strength (i.e. 16Nm) and used the optimal ffb multiplier for a while of continuous driving.
     
    Last edited by a moderator: Jun 6, 2015
  15. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    I'm looking at kollmorgens own data mate. non random google searches, lol.

    I think so but most likely from the constant death grip required to tame it. But i don't believe that is due to the torque (as i explained it was almost the same as a t500) but rather the reactiveness of the motor due to it's significantly larger power output.

    Very interesting (thanks for sharing) but i'm not surprised. The issue is not caused by your ffb wheels inability to output the same/similar amounts of torque as would be applied to the virtual cars steering system from the interaction between front tyres and road. But that it's a result of much larger accelerations on your ffb wheel than it would have happened to the virtual/real cars steering wheel. And that over acceleration is seen by the extremely snappy nature of your DD wheel.

    This is not a jab at you wheel though. The problem is not limited to your wheel but to all ffb wheels but just more so the more powerful the ffb motor gets. There are some rudimentary means to try and minimise this problem (by adding some artificial damping in the bodnar profiler for example) but it's inherently unfixable. However you do get stronger and more refined feel for the car with more powerful motors still so it's still a more decent compromise to go with more power.
     
    Last edited by a moderator: Jun 6, 2015
  16. Jokeri

    Jokeri Registered

    Joined:
    May 17, 2014
    Messages:
    204
    Likes Received:
    32
    Just stop over analyzing and go with small mige. You wont be disapointed.
     
  17. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Bingo! it helps to mimic the effect of having a higher rotational inertia ffb wheel whilst still producing 1:1 (or in the direction of 1:1) torque outputs.


    Does the stall torque feel strong enough but when it comes to torque output in motion, it feels too little? If so, that maybe due to the fact that torque output of stepper motors falls dramatically as rotation speed increases. So whilst letting go there seems to be enough torque to accelerate and keep the steering wheel spinning by itself quickly….if your hands are on the wheel however (increasing the mass/inertia that the force has to work against) the same steering wheel result/effect does not happen.

    Does that sound about right?
     
  18. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    But isn't it the sim's fault rather than the DD wheel's if it's acting too reactively and/or with too much acceleration? I mean, the wheels are capable of less torque and slower accelerations than their max (obviously) therefore it's the game that's telling/making the wheel behave the way it is. No??....

    I know, I know :). All this inertia talk has just gotten me all worried now.

    First I thought the small MiGe was better because of the lower inertia compared to the big MiGe. Now I'm hearing that - until a certain point - more inertia is actual better for simracing.

    I guess the smaller MiGe is probably the best bet because it falls in the middle inertia wise...
     
  19. Christian Rosén

    Christian Rosén Registered

    Joined:
    Feb 7, 2012
    Messages:
    137
    Likes Received:
    1
    Hmm, do I understand this right; for the moment high inertia is better since the FFB output doesn't calculate the turning friction from different tires against asphalt in different speeds yet? Meaning the force I feel that build up in the wheel when I go faster is only aero? If this is true I would think the steering wheel should be as transparent (low inertia) as possible since probably the FFB/output from sim will be upgraded with this effect too? Or is it not possible perhaps (?) so its better to buy a high inertia engine for the OSW? Lol, Im confused. :confused:
     
  20. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    response times on these servo motors are so dam quick at those sort of levels I wouldn't worry about how it varies vs intertia,
    seems that the very very subtle differences in feel ( for those that can detect it ) are subject to a personal preference

    +remember each gram of difference in your attachments ( hub/wheel/paddles) will also effect intertia in the same way

    I've contributed along with others & an OSW is being built & posted to the Iracing devs to help with any further dev of their ffb
    ( ISI's ffb is so darn good I don't think they need one lol)
     
    Last edited by a moderator: Jun 6, 2015

Share This Page