Guide: Optimal FFB settings for rFactor 2 - The key to being in the "Zone" :D

Discussion in 'General Discussion' started by DrR1pper, Mar 26, 2014.

  1. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    I agree, bodnar wrote a peice about this very issue and a possible solution is to factor in user force inputs at the wheel (which would include any friction of the wheel) that are fed back into the sim.

    The current method is certainly not bad though, so long as you have a really good wheel with a very high refresh rate as well, i think the issue can be avoided all together.

    Edit: on second thoughts i'm not so sure bodnars solution is fixing a real problem in the first place. I mean he says the ideal ffb wheel measures what force you put into the wheel and this is sent back to the sim but is it really needed? I mean, in reallife and in the sim, you react against the forces at the (ffb) steering wheel by applying your own force and depending on the net force outcome, the wheel finds it's next new position (or not if the forces are equal). The net force is decided at the ffb wheel (between the motors and the user) and a new steering position arrises as a result which is sent back to the sim and sets a new front wheel position. This is no different as a physical steering wheel setting a new front wheel position through the mechanical linkages, no?
     
    Last edited by a moderator: May 17, 2014
  2. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    And there is no way to eliminate mechanical backlash in G'series just by the software.
     
  3. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    well, there is no way to elimiate any mechanical (ffb) play by software because it is a physical/mechanical tolerance issue. Belt wheels reduce this signicantly and I think the bodnar wheel (being direct drive) has absolutely zero mechanical (ffb) play whatsoever.
     
  4. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    Yes Leo's and servo drive wheel's in general are another level, while the belts doing a pretty good job the G'series could be more precise without the play, than a belt drive because it is similar to direct drive.
     
  5. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    But just to clarify, the servo motor in servo wheels is not a benefit for sim racing (in fact it's not used at all). It is used like any other normal motor, only servo motors are likely to be of higher quality (i.e. greater and more precise force production) and longevity in general (due to using brushless motors....though a servo motor does not automatically imply it's brushless....just that robotic servos are always brushless).

    I wonder if the motor itself can cause play? e.g a motor with a small number of poles.
     
  6. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    I don't have much knowledge about it but I don't think so, delay perhaps but no play in rotatory direction.
     
  7. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    I dont think so either. The force depends on the electromagnetic field which must be continuous. Play is mechanical not electrical.

    enviado mediante tapatalk
     
  8. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    Yes, in relation to mechanical game Yes, with respect to the magnetic field No, I think was what I meant.
     
  9. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Sorry, i should have said ffb play.
     
  10. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    I had understood you I just was not understood as it seems but no problem. :)

    I'm quite happy with the result of the FFB here with the T500RS, it is not optimal but at this point I'm not sure, is it the technology/hardware or software, I think both but I can not increase more no matter what I change. The standard currently achieved seems is the limit atm.
     
  11. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    I would reserve that judgement until you get to try yourself a bodnar. :p
     
  12. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    Probably but the T500RS (in this context) is enough atm.
     
  13. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Ah, yes ok i understand and agree. But i wouldn't mind getting my hands on the accurforce wheel when it arrives. :p

    The more dynamic torque range the better still.
     
  14. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    That's true and I'm sure I'll change something in the future but when and what I will see. :)
     
  15. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    hiya

    I'll try and make this quick

    I think you'll find that a servo has to be operated as a servo ( & not a standard motor ) even in sim racing
    a servo is a closed system & will not function unless the control board is receiving high frequency data from the encoder
    (located inside rear of the servo unit), Leo bodnar would have coded within the firmware which mode & how the system copes with "error correction" -in sim racing terms what the system does when the wheel isn't in the position the servo wants it to be in.

    the encoder constantly monitors the precise speed acceleration / de-acceleration & feed back this information to the control board

    the ffb information is one way traffic coming from the PC to the control board of the ffb device (the Leo interface) it's what happens between this interface & the servo which is what makes a servo based system very different to a standard motor based system

    a standard motor can be rigged out with an encoder and have positional tracking/error reporting/power correction etc in the same way but the resolution is so low due to the fact a standard motor isn't capable of correcting/reacting like a servo is

    if an application requires a servo companies have to pay 4 to 5 times the cost for a servo ( servo vs motor )


    Leo's servo system isn't the only one available for sim racing , there are a few others available also (I think one is being sold in France)

    this gives an idea of precision of servos -a demo from ABB

    http://youtu.be/SOESSCXGhFo
     
  16. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    But that's the thing, sim racing ffb does not work by targetting a wheel rotational position. Servo motors are specifically designed for the perpose of rotational positional accuracy, sim racing ffb only (cares about) outputing a target force/torque percentage for a motor to produce and this is achievable by any electric motor.

    I therefore surmise that the specific function of a servo motor (it's positional accuracy) goes completely unutilised by racing sims.

    Very impressive video indeed but all i see are very quick responding, high torque output motors coupled with high quality servo control units that make for very accurate positional control.

    Don't forget that the error sensing & correcting controller that turns a regular motor into a servo motor can make a significant portion of the cost, especially when it comes to these manufacturing grade robotic arms. The significant price difference does not just come from the higher quality motors used in servo motors. I surmise that leo used a servo motor not for it's servo function but simple because servo motors use higher quality motors. By higher quality i mean in terms of the max torque output, torque output linearity and consistency (due to more poles), torque output responsiveness (i.e. quicker transient response) and pretty much all industrial servos employ only brushless motors (a significant price jump over brushed motors...not just for longevitiy but for higher speed and torque potential as well....it's a complete win-win with brushless motors).

    If a servo motor is not receiving positional data but instead force data (as with all racing sims), it's not being used as a servo but just as a motor.

    I'm very sure this is the case but if i'm wrong i would be more than happy to stand corrected.
     
    Last edited by a moderator: May 18, 2014
  17. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    hiya dripper

    there will be a mode (for the servo to operate in) & a set of conditions
    (tailored to give a convincing effect for sim ffb) in the firmware made by Leo which facilitates the way the servo system operates eg the 2 way traffic (of data) between the encoder and the control board (within the Leo interface box)

    this is within the firmware

    I've noticed a difference in the last firmware update, there is more "slope" or falloff of forces, less aggressive detail which is better, unlike a motor based system where the unit is "dumb" and just receives power at different levels
    regardless of wheel position / latency / motor performance (perhaps effected by heat & wear )
    a servo based system is "live" and a 2 way system
    -it would be pretty pointless to try and downgrade it to work as a standard motor would

    apart from the obvious advantages of a servo system, the biggest thing I initially noticed was fine detail forces even when the wheel was under power ( banking on an oval or mid corner etc) this is helped by the firmware/control board-encoder
    ensuring that detail is delivered above the already present more dominant force
    (that would simply wash away the smaller forces in a standard motor based system)

    I have a few engineering friends that could explain this to you in a lot more technical terms than I have
    (I switch off as soon as technical phases and sentences start to sound boring lol
    & I really don't work in a field that requires me to know this stuff, a basic understanding is good enough for me)

    if your passing near silverstone book your self in with Leo his guys will show you the parts that go into their system
    ( control board / signal amplifier etc) & explain to you how any servo based system is different other than just a posh "motor"
    -this is not a sales ploy, I took one of my engineering friends with me on my first visit to Leo's place, thanks to them (the engineering company I know)
    I already knew about how servo systems differ
    (forgot half of it now though lol)

    some cheap/smaller servos can have brushes usually installers of industrious applications keep away from these (or should)

    a servo based sim steering system is a tad overkill for sim racing though as it's capable of a lot more, for sim racing the servo is in a state of semi retirement lol
    only limited by the software that's moving it, only an option for those that want the best feedback possible and don't ever want to buy another ffb system again

    yes that YouTube vid is very impressive hey, the penny drops as to what it is they are demonstrating if watched all the way through
     
    Last edited by a moderator: May 18, 2014
  18. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    No idea how these special products are developed but torque control should produce very accurate outputs and servo systems offer this sort of control, and as far as I am informed rf2 provides the torque of the steering column.
     
  19. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    A servo motor can accept force input only as well ofc which would make it function as just a motor. It makes zero sense to me to use the servo specific function for race sims (nor can i imagine how it's applied) because it has no part to play in reaching a force output target. Servo's do no have force sensors to check that force output is correct and you wouldn't use positional data (what the servo function deals with) to ensure that a force output is correct in sim racing. You can use it to roughly check if the force output is linear or not (like i've done with wheelcheck.exe program for my t500) but only if there is no counter force applied onto the wheel by a user and it's only accurate with unit step force inputs (unlike the constantly changing forces we experience in a racing sim). I just can't see how positional data (and also the speed and accelerative data you attain by integrating the positional data over time) is going to be useful when you factor in the human component of applying counter-active force to the motor.

    Not at all if servo motors are the only options out there that offer a motor with the performance characteristics that leo bodnar was after for his wheel. Of course the fact that it is a "servo motor" certainly doesn't hurt being marketed and highlighted. But nowhere do i read that the servo specific function being used for sim racing from my searched online. The only thing i can find is one of the bullet point for the components saying "High precision and strength servo motor" which doesn't expressly imply that the servo function is used at all.

    A firmware update by leo does not automatically imply it has anything to do with the servo function. I remember my csr-elite wheel getting firmware updates that noticeably changed the way force output occurred which is why many of use rolled back to older firmwares which produced both higher torque and were more aggressive with torque application/build up (i.e. gave a more raw and less damped ffb sensation). What you describe by your firmware update is exactly what i experienced with the firmware changes when i had the csr-elite.

    I would love to hear their explanation of how the servo function can be used in race sim ffb.

    Would love to if i get the chance to meet him. ;)

    I'm sorry Adrian but I still don't see what it is that you're seeing differently. It's a servo, designed to reach a positional target without overshooting (of great importance for robotics and manufacturing/engineering applications).

    edit: speed1, servo motors do not perform error sensing and correction on the force output. It only modulate the force output based on the intended positional path that is desired and the feedback of how close to it's next position it is so that it can plan accordingly so as not to overshoot a positional target. It does so by reducing the torque output each time it's positional readings get closer to the target. But this has nothing to do with hitting specific torque output goals and in fact it can't do this even if someone wanted it to.
     
  20. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    no probs, happy for you to continue on your train of thought etc, I can assure you they work very differently though

    a servo works very differently to a motor, do nip into silverstone (Leo's place) or if you know an engineer working in the field of robotics ( has dealing with drive systems and motors and servos ) he'll maybe convince you where I have failed to do so
     

Share This Page