1. Do not share user accounts! Any account that is shared by another person will be blocked and closed. This means: we will close not only the account that is shared, but also the main account of the user who uses another person's account. We have the ability to detect account sharing, so please do not try to cheat the system. This action will take place on 04/18/2023. Read all forum rules.
    Dismiss Notice
  2. For downloading SimTools plugins you need a Download Package. Get it with virtual coins that you receive for forum activity or Buy Download Package - We have a zero Spam tolerance so read our forum rules first.

    Buy Now a Download Plan!
  3. Do not try to cheat our system and do not post an unnecessary amount of useless posts only to earn credits here. We have a zero spam tolerance policy and this will cause a ban of your user account. Otherwise we wish you a pleasant stay here! Read the forum rules
  4. We have a few rules which you need to read and accept before posting anything here! Following these rules will keep the forum clean and your stay pleasant. Do not follow these rules can lead to permanent exclusion from this website: Read the forum rules.
    Are you a company? Read our company rules

Cousin of OSW (Open Sim Wheel)

Discussion in 'DIY Motion Simulator Projects' started by Gadget999, Sep 23, 2017.

  1. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    It seems like we should change to the STM32 based solution instead. The Leonardo based solution have its weakness, and it seems that it's not gonna be some updates...
    http://www.racingfr.com/forum/index.php?showtopic=49848&st=1740
    (Use google translate...)
    I already have some Leonardo card, so I will test with that first. But if the rest of the equipment (driver, dc-motor) works good, I will order a STM32 card.
    I tried to register on the forum, but they didn't allow any new members. Although, I don't think he will give away the source code anyway...
  2. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF
    i have a stm32 board somewhere i will see if it is compatible

    is there a link to downloads and instructions ?
  3. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
  4. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    • Like Like x 1
    • Informative Informative x 1
  5. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Waiting for some update of your build... While waiting for my parts, I have continuing study characteristics of the brushed dc-motor, and have realized that I probably went for the wrong voltage on it. I was going in the right direction in my theories, but didn't go that deep before i ordered. Because the motor gonna work in a "stall" mode almost all the time, it doesn't need so much voltage to reach the rated current (=rated torque). The 60v 1000w version of the same motor which have a rated current of 21,5 A have the same torque output as the 36v 1000w version have at 35,6 A. So in the theory, I will only get 1,8 nM with my 24VDC 20A power supply. If I have ordered the 60v version, I could have reached 3,0 nM. My first goal wasn't to get more torque that the G27, but with with 20A I will not even reach the G27 level. The same type om motor with a nominal voltage of 100v should be perfect, at least in theory... Still haven't find that motor...

    https://www.aliexpress.com/item/MY1...tor-Electric-Bicycle-Motor-E/32817219626.html
  6. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    Looking at the motor you linked, the specifications seem to be a bit overrated. I can't imagine that could be anywhere near 1000W even if you raised the voltage to extreme levels. I'd say 500W at best, but I could be wrong. Chinese manufacturers tend to be generous with numbers when it comes to power output.

    See this link, it gives some explanation of how it works in practice.
    https://www.motiondynamics.com.au/xyd-6d-450w-24v-2600-rpm-with-chain-sprocket-en.html

    By the way, I have this XYB-6D motor rated 350W@24V. At first I will try it with a 12V 83A HP server PSU as that is the only one I have.
    I will also use a power analyzer I have for RC hobby. That should tell the stall current and some useful info.
    https://www.ebay.co.uk/p/G-t-power-RC-Power-Analyzer-Watt-Meter-130amps-I8o7/1081855724

    Finally here are some pictures of where I currently am at the build. :)

    Attached Files:

    • Useful Useful x 1
  7. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF

    go for a cheap server power supply - i just got 2 x 77amp supply for £10 !
  8. Alexey

    Alexey Well-Known Member

    Joined:
    Sep 23, 2014
    Messages:
    452
    Occupation:
    Electronics Technician
    Location:
    Adelaide, Australia
    Balance:
    8,060Coins
    Ratings:
    +620 / 2 / -0
    My Motion Simulator:
    3DOF, DC motor, Arduino
    The MY1020 was probably not the best choice for a DD wheel as it has quite a low torque at stall.
    The figures show the 36V/48V version only having a max of around 3 -4 NM whereas my 450W motor will have around 8.5 NM at stall.
  9. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    The MY1020 have at least the same stall torque as the 1016. The problem is the amount of current you need to achieve it. At stall, the torque is proportional with the current. So if I get 3.2 NM at 35 A, I will get 6.4 NM at 70 A. The problem is that most of that will be heat, a 1.7 kW radiator that will melt the winding very soon. I haven't analyzed
    There is no problem to get that torque of the 1020 also, that's not the problem. At stall the torque is proportional to the current, so if I get 3.2 NM at 35 A, I get 6.4 Nm at 70 A. I have a 1.7 kW radiator that will melt the windings in the motor
    The 1020 is good as 1016 when it comes to stall torque, that's not the problem I think. At stall, the torque is proportional to the current. So if i get 3.2 Nm at 35 A, I get 6.4 Nm at 70 A. The problem is that almost all that power will be heat at stall. A 1.7 kW radiator that will melt the windings... My idea of buying a 1000w engine, was that at least in the theory, it should handle more power losses. (36v *35A = 1260w * 0,22 = 277w/24v = 11,5A). So my motor should handle a average current of 11.5 A without broke down. And because current is proportional to torque I can get an average torque of 1 Nm out of the motor.
  10. Alexey

    Alexey Well-Known Member

    Joined:
    Sep 23, 2014
    Messages:
    452
    Occupation:
    Electronics Technician
    Location:
    Adelaide, Australia
    Balance:
    8,060Coins
    Ratings:
    +620 / 2 / -0
    My Motion Simulator:
    3DOF, DC motor, Arduino
    Ah, I see that the graph doesn't show the RPM vs current at 0 rpm.
    I could suggest adding active cooling to the motor by drilling holes in the ends of the case and adding a fan to pull air through the core.

    Unfortunately you are not going to escape having a large amount of current on these types of DC motors.
    If you want higher voltage then that brings in much more pricey motor controllers at which point you may as well purchase a simcube... Or experiment with stepper motors...
  11. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Looks great! Did you order the motor from that site?
  12. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
  13. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    My motor arrived today, and I am little doubtful about the strong cogging when I turn the motor with no current applied. (Haven't got any supply worth the name yet...)
    I assume that you have that in yours also, but the question is if it is noticeable even in use in the closed loop? It's jumps like 18° between every "teeth", so it's
    hard to believe it's gonna be any precise steering with it....
  14. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Didn't see this until now, but still interested of how you feel it Mbakos...
  15. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF
    You dont feel it when it is powered up - i guess the feel is different when electricity is flowing through it

    i have some motors from an electric scooter and these do seem to have a lot of cogging, i decided not to use these

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

    on a side note i have some military spec motors that are quite small - they are mounted to a gearbox and are rated at 28v and 7000 rpm !

    you can not feel any cogging at all

    the gearing must be quite something but because the motors are so small there is not much resistance

    these low inertia motors may make a nice ffb wheel, i may try one out when this project is finished
    Last edited: Dec 1, 2017
  16. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    I finally managed to build the wheel to be able to test it. It was quite a challenge as I had to figure out all the mistakes I made during the wiring and configuration. :) The power supply I was using is 12V 83A.
    The cogging worried me when I turned the motor by the shaft, but after I put a 350mm wheel on it, the effect diminished so much I could hardly feel it. When I let go of the wheel it settles into one of the "valleys of the magnetic force" but it's not much of an issue because in game there are larger forces. This cogging effect was kind of the same powered and unpowered.
    There is a certain minimum force the motor needs to get moving, whivh I figured out the hard way. Worrying about high current and torque, first I started with 25% main gain and 50% on all the effect sliders. The steering stop was set at 20% which felt quite strong, but not srtonger than what a G27 is capable of.
    When I started LFS, I did not feel any force at all, the wheel felt like it was unplugged from the power, but the steering stop from the wheelconfig software was there, with the expected force.
    After an hour of juggling with all the possible explanations, I found out I had to raise the main gain to 50% and all the effects to 100% to give large enough ingame PWM input to the wheel to even move the motor at all.
    This time I faced another challenge, that being overheating. Earlier I read a forum post about the IBT-2 controller not having the most efficient cooling, I followed a recommendation to unsolder and move all the terminals and the condensator to the other side of the PCB and add a CPU heatsink on top of the mosfets black case. I left the original heatsink on the bottom where it originally was and made a sandwich from the IBT-2 and the two heatsinks. The result was not what I was expecting. The smaller heatsink was too hot to touch and the big one on the mosfets' top was cold. When enough force applied in game, in a turn for example, it took no more than a couple of seconds for the IBT-2 to shut down in overheat protection. Unplugging and replugging the power helped it restarting but it heated up so quickly again. This time the force feeling in the wheel was still probably half of what I'm used to in the G27. The precision and the speed was there, but lack of force.
    The current reading on the power supply was a maximum of 15A for 1 second when it shut down. Then I added a 80mm fan to cool the controller, this time I could drive around normally for a minute, not more before it overheated again. Then it could keep up with 15A but failed at 20A. Not much improvement. The maximum main gain in wheelconfig I could achieve was 68% and the force was still weaker than the G27.
    I had to revert the components on the PCB to their original location and place the bigger heatsink to the underside in place of the original heatsink, and move the smaller heatsink to the top. This configuration is not tested yet, but even with the lower currents the motor also got worryingly hot. I could still touch it and keep my hand on it, but it was on the margin. I wouldn't want to heat it up more, the magnets could lose their power if heated up too much. Either way, it's the maximum this motor is capable of as a direct drive.
    The next is to add a 5:1 ratio with a HTD-5M 15mm belt and see the torque that way. With the cooling upgraded it could be stable.

    Attached Files:

  17. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    First, thank you for your honestly and describing report.
    Would be great for us all if you could do a wheelcheck test before you go to the 5:1 solution. What interest me most is the deadzone, that you describe as minimum force i think. (It's about 18% with G27, which is the G27 second biggest weakness with that wheel i think, the rattle sound is the biggest...)
    Remember to set at least 500° rotation in the wheelconfig.
    http://forum.projectcarsgame.com/sh...eel-FFB-Values-Google-Sheet-amp-FCM-Universal

    I'm still not convinced to continue with this project. Everything i have ordered except the motor could be reused if i go with the stepper motor solution from Craig. As soon as i get the rotary encoder, i will connect it temporarily with a belt just to see how it works before I put to much time fitting everything together.
  18. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF
    Is your motor 24v 450w ?

    The ibt2 is rated at 43 amps iirc

    The first time i powered up my wheel it got hot however some adjustment of the ffb strength in lfs and i was able to get the wheel to have more fidelity but no loss of maximum force

    I also have a ibt4 driver which i have not tried yet. It has a higher amperage
    Last edited: Dec 3, 2017
  19. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    Yes, my motor is rated for 24V @22A max. That means it could take twice as much on 12V. However, I'm afraid of giving it high current just by hearing is squeal like a stuck pig. :D
    Today I made a tidy little board for electronics. The breadboard version was a bit chaotic. Also the new cooling needs to be tested on the IBT-2. I did half a lap of Nordschleife in Assetto, 70% main gain. Average amps in a corner were 10-15A, I have seen a 24A peak once on a curb. After 5 minutes of cruising, the IBT-2 was room temperature, the wheel motor itself was not even warm. It must accumulate over time. I don't know if because of the wiring or the heatsink, but it seems more efficient so far. I might go higher in gain and see what happens. The cogging can be felt slightly under load, in corners. Just like a worn steering rack in a really old car. :D
    I ran wheelcheck at 70% gain, here is the .csv if you're interested.

    Attached Files:

    • Like Like x 1
  20. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Deadzone similar to G27, but it actually looks good i think. It is supposed to do 4 tests, and than the google sheet makes "normalized" (Average?) output, but I just put the same values i all four.
    upload_2017-12-2_22-41-54.png


    My G27:
    upload_2017-12-2_22-44-0.png