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
    Adrian your misrepresenting what I'm saying by ignoring the constraints I've placed on when I say the performance is comparable.

    Above a certain speed, the industrial servo wins hands down, of that there is no doubt. If however the speed requirement is not higher than what steppers configured well can achieve, performance should be similar. Industrial servos use AC asynchronous/synchronous motors not only for their much higher speed range (e.g. Up to 6000RPM vs steppers usual max around 100-400RPM) but for other reasons too that as you rightly said at the end go completely unused/wasted in a sim racing application.
     
  2. metalnwood

    metalnwood Registered

    Joined:
    Dec 31, 2011
    Messages:
    164
    Likes Received:
    4
    Berney has tried to be cost effective in choosing a stepper but as I said pages back that doesnt mean in this scenario it is going to be worse. As pointed out there are constraints and in a business doing this sort of thing you will not always pick a servo. Even choosing a particular motor has a lot of thought go in to it so that you dont spend more money than you need to. An extra few $$ spent adds up to a lot over many units.

    I don't think Berney could not pull off as good a wheel while using a stepper. If anything he underestimated what people wanted and it wasnt until people tried the OSW etc that they decided perhaps a bit more is better. If anything, perhaps he has picked the wrong stepper to use.

    Think of it like this (but you already know) I would use a servo in my CNC machines because I want the speed when I am doing rapids. For just cutting it hardly matters as they can perform equally most of the time but when working over a 5Mx3M area the rapids become important.

    OK, then take an extrusion based 3d printer. servo or stepper? Easy, stepper.. but the servo is the better motor.. In what way? In this application it cannot perform any better as it has zero opportunity to use any advantages it might have as a motor. The stepper is half the price. Clearly the stepper is the winner in this scenario and saying the servo is a better motor in know way would make my 3d printer work any better.
     
  3. cookie

    cookie Registered

    Joined:
    Mar 7, 2012
    Messages:
    144
    Likes Received:
    1
    Hey, DrR1pper I see Adrian making the same comments in other forums about the A/F wheel without actually try one himself. I have a OSW (waiting on the IONI) and a A/f wheel there's no way I could make comparison to the Bodnar just in theory and what I have read on the net. FFB is a personal thing and for me I think the A/F is going to be a better wheel, mainly because I use Simx FFB only and not game FFB which the OSW is not capable of.
     
  4. Adrianstealth

    Adrianstealth Registered

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

    overall are you happy with your AF ? I bet your nicely getting tuned into it now
    ( & use to the settings etc )

    ---------------

    metainwood

    I think simX pulled off a pretty good wheel ( I rate it as a good upper mid end / lower high end which demotes others like frex, 7000 ) , it's just that maybe simX shouldn't have been so aggressive in their claims during their persistent marekting campaign (that seems never ending) ( that at one point seemed to take over the Iracing forums ) ,

    then someone who received in first batch took his AF to pieces & stated what it actually was hardware wise
    ( not a servo ....surprise surprise lol ) which contradicted simx's previous claims ......cue the confusion on what a servo is ( which seemed to be frantically fuelled by simX )
    ... then the forums get filled with all the google stuff that refers to the hobbiest type servo as found in radio controlled cars etc

    during which time the OSW was getting some momentum (again mainly on Iracing forums) so actual comparisons are bound to follow,
    I thought the AF came off really well by the way I though it was going to get panned ( think its a decent system )

    now we are into the occasional AF being put up for sale on the forums as some want to switch to mainly OSW ( some the Bodnar) , people that post factual down to earth information or disagree with the ongoing claims of simX get deemed as simX "haters" & generally leads to people arguing & falling out lol

    that's how I've seen things + my summary of it

    I'm sure the AF is success for simX though, I'm happy it is, they make some good stuff

    soon it will be VR wars, if the ISR presenter dosnt get targeted again ( no offence DG if your reading )
     
    Last edited by a moderator: Jun 15, 2015
  5. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    cookie your right I've not tried one for myself, there's other ffb wheels ( mainly the fanatic range ) that I've not also tried but have a pretty good idea of where they are placed in the ffb charts due to the type of hardware & various points from users ( good & bad )

    would I think a fanatic csw could outperform a servo system ....I also think not
     
  6. traind

    traind Registered

    Joined:
    Jun 30, 2013
    Messages:
    129
    Likes Received:
    26

    Now that is crazy... someone on the internet keeping an open mind, trying to learn and doing research that they are sharing openly. Be careful unless someone kicks you off the internet for good!

    Thanks for your posts.
     
  7. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Adrian, i realised i could have been a little more specific/clear regarding the stepper vs AC synchronous servo motor performance.

    A servo motor used in closed loop (which is what makes it a servomechanism) will/can generally perform better than an open loop stepper in the same performance envelope achievable by whichever motor has the smallest working envelope (i.e the stepper which has a much lower peak RPM before torque falls off drastically).

    However, in that same achievable envelope, if one converts the stepper into a closed loop servo the performance characteristics are comparable if not the same to a closed loop servo motor with the same working envelope. It also tends to be much cheaper.

    The third way to compare them is to not use the servo motor as a servomechanism (i.e. just use it's motor to output torque instead of position control) and to not use the stepper in either an open or closed loop (i.e. to also just use it's motor to output torque instead of position control). Provided both motors are compared in the same same achievable working envelopes, performance characteristics (i.e. independently of available torque range) should again be similar if not the same.
     
    Last edited by a moderator: Jun 16, 2015
  8. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    I remember you shared this video before. I don't deny it's very impressive but i fail to see it's relevancy here. What point are you trying to make/prove? That servo motors can have awesome position control and accuracy? Sure. But that's also only half the equation. The other have of what causes a servo motor to be so accurate is it's driver. But regardless of that, since sim racing does not operate on position control, the prowess of industrial servo motors positional accuracy and precision is irrelevant. Why? Because sim racing simply tells a motor to output X amount of torque. Nothing more, nothing less. Position control is whole other kettle of fish.

    Now explain this. What/where would the difference in performance lie between two motors where just producing X amount of torque is required? The only i can think of is the difference in peak torque and for how large a range in speed/RPM the motor can sustain that peak torque (i.e. how long it can sustain the acceleration rate) before rapid fall-off. From the data we have, it's looking quite plausible as to why they perform characteristically different when the stepper should be capable of performing similarly with enough power.


    As great and true as that is…I don't see the relevancy here. Neither motors have to be used in a region of performance that causes impaired performance due to things such as heat fade (which has to be pretty hot anyway to affect performance). And as we know of the AF, it doesn't heat up. Interestingly, it likely would if it were used in it's stepper motor fashion but since it's not, power is only delivered when a specific amount of torque demand is sent to the wheel. When steppers operate in their open loop stepper function, they build up heat more easily because there is always a maximum amount of power delivered to the motor (which is a required for steppers to work in open loop position control but it has its benefits, just non that are relevant to sim racing like the servomechanism of servo motors).



    If you used a servo in it's open loop format, i would agree very likely to be a "bash city". However, use a similar sized stepper motor and in a closed loop fashion (i.e. converted to a servo) and provided the speed of the operation does not lie outside the steppers performance envelope, there is no reason it can't perform as well as the precision example shown in the video. Independent of the torque-speed region operated in, you're conflating the precision you are seeing in the video coming from the motor instead of it's servo driver/controller.

    The difference however is that the RPM/speed at which the position control precision would start to faulter and hit a can is much lower/sooner for the closed loop stepper because of it's some inherent limitations that are caused by it being designed to operate formally in open loop position control. But within the same speed range and in non position control, it should be able to perform very comparably.


    But that statement makes no sense because race sim ffb does not operate on position control and so there is absolutely no requirement for the precise movement control as what was being demonstrated/highlighted in that video. You've tried to use this as your argument for why servo's in sim racing is better before but fail to realise it's a trait that is completely wasted on sim racing. Completely.

    You've accused me and/or others of being sucked into the marketing hype of the AFPro and yet (pardon me for saying but) you seem to be doing exactly what you accuse others of doing with your own ffb wheel. It would be fine if your arguments/points against the stepper were applicable to sim racing.

    What you do get out of an industrial servo however generally is higher peak torques (though as you know you should know you can get equally sized steppers) and very very large speed/RPM envelope. But, as already explained, you don't really take advantage of that insanely large RPM range in sim racing. It looks like the AFPro is maybe not being powered enough for what we need for sim racing use. But with a little more voltage, no reason that working RPM envelope should not expand to produce comparable performance in the working speed/RPM range required for sim racing.

    My closing thoughts: it seems you so much want to believe in the story that the servo is better in every application/scenario but the reality is that the science does not agree. More so, from how you speak about all of this whilst even admitting/claiming it's an area you're not completely/very knowledgable on or pay much attention to when it comes to the technical facts surrounding motors, you seem awfully over certain/sure in your assertions.

    Another thing of note i should maybe have mentioned earlier. Stepper motor operation when used just to output X amount of torque is pretty much identical as that used by AC synchronous motors (i.e. in the bodnar and OSW). If you research how they are both driven…it's actually basically the same but with some physical differences that cause the stepper to have a much shorter constant peak torque speed/rpm envelope. But provided both motors are operated within that region, the stepper should be able to perform comparably. In fact, put both in closed loop control (i.e. servo mode) and within the same envelope, performance characteristic should also be very similar/comparable.

    At present, there is enough evidence/data to suggest the stepper is being held back giving an unfair comparison, regardless of whether it was intentionally made to perform like so or not by Berney because he believes it's more realistic.

    Interestingly, it would make sense that it's sort of more realistic because from the graph i made of Paul's data, it seems that (although there is some weird ffb clipping going on from 40% torque and up) torque demand signal is producing a constant velocity rather than an acceleration which may very well be more realistic. The reason being, correct simulation of the steering wheel momentum involves simulation of constant velocity but constant accelerations produce constantly changing velocities. Just an interesting observation into possible why Berney has set a speed limit (if that's truly the reason for it).
     
    Last edited by a moderator: Jun 16, 2015
  9. Jokeri

    Jokeri Registered

    Joined:
    May 17, 2014
    Messages:
    204
    Likes Received:
    32
    DrR1pper, you write way too much.
    People, if you can afford small mige OSW you cant go wrong with it.
    99% will be happy, 1% are those who brag with motor specs although gain is marginal.
     
  10. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    True. Sorry about that. :rolleyes:
     
  11. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    dripper -very good, if what your making of this science that you've mentioned you should get yourself a stepper based system, no problems, each to their own , I'm happy for you as long as your happy too, all good

    ps your posts are to long -but I read a bit, honest
     
  12. Kristoff Rand

    Kristoff Rand Registered

    Joined:
    Dec 23, 2011
    Messages:
    1,055
    Likes Received:
    10
    Me and my DFGT are feeeeeling very inadequate. :(
     
  13. Katsushika Matsumoto

    Katsushika Matsumoto Registered

    Joined:
    Oct 5, 2010
    Messages:
    24
    Likes Received:
    1
    Cool video thanks :)
     
  14. Katsushika Matsumoto

    Katsushika Matsumoto Registered

    Joined:
    Oct 5, 2010
    Messages:
    24
    Likes Received:
    1
    Yes did that. I'll send you my iRacing name in a PM ;)
    I thought the same about the headroom, but apparently most cars need even more power :D

    I'm cool for now with the small MiGe, the Argon provides enough flexibility to upgrade if I want to.
    Did you see the 23NM Bodnar upgrade? :D

     
  15. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    Ha! 23 bondar , I like to try it too,
    Let us in's how you get on with te Mige. ( see my post on iracing earlier about these servo motors vs more expensive ones etc )

    Thanks for the pm
     
  16. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Thanks Paul. The general response curve shape characteristic remains the same however, two things of note.

    [​IMG]

    Firstly, the initial deadzone is much smaller. Did you change some setting on your wheel? But it's strange how much less rotation occurs. Any idea why?
     
    Last edited by a moderator: Jun 20, 2015
  17. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
  18. traind

    traind Registered

    Joined:
    Jun 30, 2013
    Messages:
    129
    Likes Received:
    26
    Is Berny talking about that in the owners forum? Can you share a little more information with us? I am getting close to purchase so this information would be a somewhat important consideration. Thanks :)
     
  19. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Well, if we consider the log an anomaly but the general pattern correct, you can just visually scale it up to size and see how reducing the friction, dampening and inertia settings has reduced that deadzone region.


    I really think it's just the speed limit of the stepper since we see a maximum velocity reached. A very interesting test to confirm and compare this would be for a bodnar or OSW user to run the step2 test with the lock set to unlimited degrees of rotation (or to whatever highest) and run the time period per run at something like 1000ms. We should see a speed limit occur, though most likely outs 300ms time interval.
     
  20. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36

Share This Page