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
    Be careful with the IBT-4. It has no built-in overheat protection like the IBT-2. I have used the IBT-4 in another project, but i'm going to change to IBT-2. I fried one IBT-4, maybe because the cooler wasn't properly fastened.
    • Old Old x 1
  2. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    I tried the wheel with the STM32 board, same settings. Basically there is no difference in the feeling. A board does what a board has to do. Be it blue or green. :)
    For now, I won't reassemble it for belt drive. I will wait for a proper 24V PSU and see if I can get some more juice out of it.
    I have a hacked 12V wiper motor with the gearing removed, I will assemble the belt driven version for that motor. I'm curious how it behaves.
    The goal is cheapness, no matter the cost. ;)
  3. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    I think the higher voltage will not make any difference. Remember that the motor mostly are working in "stall" mode, which doesn't require so much voltage to reach maximum current. That's why i realized (posted in a earlier post) that a 100 volt motor
    should be better suited because of more turns of winding = more low rpm torque with less current. (Please correct me if I'm wrong) The higher voltage is only required to reach higher rpms when the the motor starts producing back EMF. Actually, I guess that your current motor will work with a 5 volt supply. Have you checked the resistance in the motor? That will tell you how much voltage you need.... Big thanks to you for testing the STM32 board. Actually, i have plan to try the "ESP_WHEEL_0.3" just to have the ability to invert the input, activate internal pollups and eventually make a simple linearity function to fix the deadzone.
    I remember there were someone on this forum that had started work with it. If you read this, you are welcome with comments and feedback...
  4. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    My latest observations regarding the controller tells me this time it's not a thermal shutdown but a current limiting one. On paper, the IBT-2 is advertised as being capable of 43A, but every time it went anywhere close to 30A it shut down immediately. I checked the heatsink and it felt only warm. I suppose I could raise the voltage to keep the amps below 30A for the same torque.
    I measured the motor resistance, it says 0.37Ohm.

    By the way, I jumped back to my G27 for a quick lap to test the difference. Today's lesson: Don't compare a G27 to anything else that is heavier. I already forgot how slowly it turns and how numb the force feels.
    • Like Like x 1
  5. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    So that means that it is already a better wheel than G27? That sounds promising. About voltage/current, that's what I'm trying to explain to you; The torque is proportional to the current. You don't need any more voltage than the resistance require. So if you have 0.37 ohm, you can make a current of 12/0.37 = 32.4A at stall. So rising the voltage wont change the fact that the IBT-2 shutdown at 30A. I guess the only difference you will see with 24v is that you must go from max 70% force to 35%, and that will actually lower the resolution steps for the PWM output. Maybe a IBT-4 is the solution, it works until it die... Be aware that it is limited to max 15 V because of the voltage boost (charge pump) function I think.
    Last edited: Dec 3, 2017
  6. 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
    Change the pwm frequency to 20khz and the squealing noise will go.

    I have drilled holes in my motor end cover to let the heat escape

    I think it may be possible to run 2 x ibt2 if you are exceeding the current limit.
    Last edited: Dec 3, 2017
  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
    i have just ordered up a stm32 board - are there any advantages to getting the indexing to work ? is the software interface any better ?
  8. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    It's already at 20kHz. At 4 or 8kHz it's more like a high pitched sine wave, but at 20kHz it sounds exactly like a pig running for their life. :)
    At higher torque the sound changes to something like between a ballon being scratched and a stubborn plastic screw without lubrication. :)

    How is it possible to sync 2 controllers without blowing up the whole thing?
  9. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    The wiring for my encoder is exatly the same. I plugged in the same wires to the STM board and it worked. In my early tests I could make the index work without the logic inverter gate, so it's more forgiving.
    The Leonardo had some strange issue with wheel centering, it was 10 degrees or so off from center, but I compensated by turning the wheel to the same error the opposite way and center that way. After that, it saved the correct value. The STM did not have this issue. However, the STM didn't save the force feedback values when saved to eeprom, I had to change them, save, change them back and save again. Strange.
    I couldn't get the analog axes for pedals calibrate properly on either board. Probably the STM needs a 3V reference instead of 5V. The load cell input with HX711 seems worthless on the Leonardo as I don't see how calibrating the range works. If anyone has a golden trick for the pedals it would mean a lot.
    Both have their ups and downs but at the end of the day, both do their job.
  10. 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
    mine does not ring at the highest frequency - must be something to do with the thickeness and winding design in the motor - fitting a capacitor across the motor may help

    try wiring the ibt-2 so they are both linked together on the same pins
  11. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Well, still only received the motor, but when I was working with a another project with IBT-4 cards, I took the opportunity to try out the strength of the motor and how the the IBT-4 handle the current. Had no power supply with high current, so I took one from an old computer that i don't use any more. According to the label on it, it should deliver 18A on the 12VDC output, but it went over 20A at 40% pwm.
    With 50% pmw output, it already felt much more stronger that the G27. I used a wrench and it was not easy to hold still... But the IBT-4 got hot almost immediately, so there is no chance it will last with the standard cooling. >60% pwm, the supply cutted off. My multi-meter cannot measure current over 20A, so I have no idea of what current it reached.

    Another notice is that it needed 4% pwm to overcome it's own inertia and start to rotate. With the wrench fitted, it needed 5% to start rotate, although with a "cogging" rotation. With a wheel mounted on the axle, the "deadzone" shouldn't be more that 6-7% i think.
    Edit: Changed to 20 khz pwm, and it actually surprised that it even got stronger, and it start rotating at 1% with the wrench.
    Conclusion so far is that for strength, the current supply may be enough, but still worries about how the cogging will impact when no or low force are applied from the car sim.
    My theory earlier is now confirmed also that the voltage don't need to be high at all to reach the desired current because the motor is working in stall mode. This is a 36 volt motor, and it consumes over 20 A at 5 Volts when you hold it.


    IB4 test.png
    • Like Like x 2
    Last edited: Dec 9, 2017
  12. 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
    what rpm is your motor rated at ? is there a chart of torque vs rpm ?

    the cogging is due to the spacing of the magnets i think

    my wheel does not seem to have cogging issues - i am using a 350w motor ibt2 and a 12v 77 amp power supply
  13. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    It's the 1000W36V version. I have study dc-motors a lot before and even after I ordered this, and Independent of speed, the torque is proportional to the current. So this motor will give 3.2 Nm at 35.6A no matter what the speed is. So the better choice had been to buy a motor with even higher voltage (=more turns of winding) to get the same torque but with lower current, because we don't need the high rpms.

    If you look at the 48V version, you have the same torque but with lower current. More turns of winding means that the motor generates more back EMF and therefore needs higher voltage to get higher rpms. But for our use, it's a better choice, because we don't need 3700 rpm.

    Have you tried to connect two H-bridge in parallel? It should work in theory I think, but if there are some difference in tolerance (capacitance etc) between the cards maybe there will be some sync error.

    The cogging is due to the spacing of the windings. (There are only 4 magnets in this motor)
    But when i connected the motor to the bridge, already with no force, it felt little better with more continuous resistance due to the shorted
    winding through the h-bridge.

    [​IMG]
    • Like Like x 1
  14. danove_b

    danove_b Active Member

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

    There are also a 60v version where you get 3.2 Nm with "only" 21.5A
    [​IMG]

    Or what about 2.74 Nm with only 7A! I'm sure 24VDC is enough to reach rated torque and the speed for our needs.


    https://www.aliexpress.com/item/JS-...-f234-4979-8e57-1ecbaa6a733e&rmStoreLevelAB=0

    [​IMG]

    Now when I know what to search for, there are several better choices, but the price increase also...
    https://www.aliexpress.com/item/JS-...-f234-4979-8e57-1ecbaa6a733e&rmStoreLevelAB=0
    • Like Like x 1
    Last edited: Dec 10, 2017
  15. Mbakos

    Mbakos Member

    Joined:
    Aug 30, 2017
    Messages:
    51
    Occupation:
    Datacenter Technician
    Location:
    Hungary
    Balance:
    545Coins
    Ratings:
    +45 / 2 / -0
    It's an interesting idea to choose motor with a higher nominal voltage. If you will have some hands on results, that would be super useful.
    Meanwhile I built the wheel into my rig and had a proper test run. Drifting is awesome with this rotational speed, but it still feels weak overall, not more than 1-1.5 Nm of torque by feel.
    Also found out the squealing noise is coming from the server power supply I'm using. There might be a chance my protection shutdowns at higher cuurent is produced by the PSU rather than the H-bridge.
    I will try it with a car battery sometime.
    I failed to set up the pedals and the G27 shifter with the STM32 board, the output was too jittery and I was unable to calibrate it at all. Tried some tricks from random forums but no luck.
    Therefore I used an arduino Leonardo and uploaded a firmware for standalone G27 pedals and shifter. That works great.
    Here are some pics of the PSU and controller box and the wheel installed. :)

    Attached Files:

    • Like Like x 2
  16. 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
    An experiment for you @danove_b . Place a scale next to the wrench and you can directly measure the force.

    If you want to use motor drivers in parallel and you are using the STM32F board you will need something called a " fanout buffer". This is a chip that will enable duplicate signals to run in parallel (1 in - 4 out) as the STM board does not have enough output current to drive multiple motor drivers. Not sure why this is but I've had to do this in the past when I made a FFB wheel with multiple Monster Moto drivers.
    • Informative Informative x 1
  17. 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
    2.74nm at 7 amps ! how big is the motor ? does it have a lot of weight ? will it take more amps than rated ?

    what would 7amps 110 volts be equivalent to in amps at 12 volts ?

    something that could be done if someone wanted more torque is to use 2 motors that are inline - however if you research the OSW you need to turn it down to around 30% to make it feel realistic
  18. xiatian

    xiatian Member

    Joined:
    Sep 5, 2017
    Messages:
    78
    Location:
    Xi'an, China
    Balance:
    262Coins
    Ratings:
    +28 / 0 / -0
    My Motion Simulator:
    2DOF
    My FANATEC steering wheel feedback is a bit small, I am very interested in your project, I hope it can be successful.
    • Like Like x 1
  19. danove_b

    danove_b Active Member

    Joined:
    Nov 11, 2017
    Messages:
    209
    Location:
    Sweden
    Balance:
    1,792Coins
    Ratings:
    +79 / 3 / -0
    Because the motor is hardly rotating in "our" application and therefore doesn't produce any back EMF the current will be equal to I=U/R, and again; Torque is proportional to the current no matter what the voltage is.
    • 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
    Received the rotary encoder today, and thought that I should do a quick test, but than I discovered that the Pro micro board doesn't have D11 connected! Damn! So the test was limited to force only in one direction. Also my current 18A pc supply trips almost directly when I try to test the steering spring stop over 30%. At that level it feels similar to the G27:s spring stop. Leonardo ordered already, but aiming at this right now:
    https://feelvr.game/


    [​IMG]