DD-Wheel with Hoverboard Motor

Hi manu79,

you can find a basic connection diagram in post #304 on this page. From MMOS board use PWM 0-50-100 (PE9). If you need further guidance feel free to ask.

If you will send me some information about your (motor)-setup I can send you a custom made firmware.
What I need is:
- motor pole pairs
- motor phase resistance
- motor phase inductance
- motor nominal current
- supply voltage to the esc
- encoder ppr (and gearing if applicable)

-Benu

Sorry @Benu . I tried to reporter from smartphone and it was not a good idea
I said:
I saw the pattern, but I miss knowing a few things. Like the exact value of the resistance, the diode bridge, the capacitor. Also, what is the one integrated on the resistance in the photo?
In short, a detailed scheme like Etienne's for his osw project with leonard would be fantastic !!
I had thought of sending the 12v engine.
If you think it is better to power it at 24v or 48v tell me.
I have a classic 27-pole howerboard motor.
The engine has hall sensors, but I think they are 90ppr. Too small. So I will install a 1000 ppr encoder. Directly on the axis.
I am going to lock the engine and attach the steering wheel crown to the central axle. Less weight equal to a better FF.
Honestly other information at the moment I do not know.
As I said, I had already started thinking about this project before meeting you. And I bought a cheap "jyqd v7.03i" controller.
Of course I will buy a "B-G431B"
Thank you very much!!!!!

pcb is too small! After some practice, finally welding is completed

Simple heat dissipation

Hello there
I think the voltage should be selected according to the motor and driver board you can get
B-G431B supports 12V and 24V

New here. Passing the lockdown time by taking up sim racing and decided I need a force feedback wheel.

Already have two hoverboard motors, hoverboard control board, arduino leonardo, 24v powersupply left over from my previous remote control golf cart project.

What is the status of the project with regards to utilizing the hoverboard control board? figured not having an encoder input would disqualify that board for this application, but it seems like there might be a workaround. Would be happy to experiment, even if you blow up the board, replacements are cheap. Kind of confused if Danove_b got it working.

I also have a 48V Odrive which should work, but it sounds like there is concern over the PWM rate. Has anyone got an Odrive to work?

Any info are best starting point for me would help.

Thanks in advance.

For anyone interested I've posted the STL files I used to build my wheel in the first post of this thread
https://www.xsimulator.net/community/threads/3d-printed-hoverboard-motor-wheel.14454/

@fzxj250 I think you may needmore heat sinking than that, this is what I did

You can get away with a lot less but in my book over kill is better than under kill
Dont look too hard at the wiring, thats all going to change when I make a housing for it

The hoverboard control board might work. But as far as I know @danove_b switched to the B-G431B-ESC1 board.

The odrive board might work too, but I have not tried it and do not own one. By the time I started this, the odrive was not that easy to get in Europe.

The current working options are:
- B-G431B-ESC1
- NUCLEO-F303RE + IHM08M1

The quickest way to a hoverboard DD-Wheel base would be to get one of the above combos.

I've got a working version with hoverboard control board, but it has no pwm input for the torque value. You need to set the torque value by serial communication. I tried to make a software pwm input just for experimental purpose, but it got glitches, so nothing you want. There are also some minimal torque ripple, but you hardly notice it when driving. The ESC board is very cheap, so I wouldn't put any time tinkering with hoverboard control board. I did it just because I'm a technical nerd, and just wanted to make a working solution where I understood every line of code...
I have shared the code for it earlier, but just let me know if you want me to share it again.

Yes please share code again. I probably saw it, but on information overload, so not sure where I saw it.

I would like to make the hoverboard controller work if possible. I have been collecting used hoverboards for various projects, some more successful then others.

There is a version of hoverboard firmware hack that provides torque control via PWM. It is RC style PWM, so might need to adjust. Can't add the link yet, need three posts.

Thanks

Here is the link to the Hoverboard firmware hack I have used on past projects.

https://github.com/EmanuelFeru/hoverboard-firmware-hack-FOC

Just got a pair of b-g431b-esc1’s delivered this week and got one soldered up and connected to my hoverboard motor. @danove_b and @Benu helped me out getting it all to this point. Thanks guys! I’m not quite there, but close!

I’m using a MMOS unit and flashed the esc with Benu’s FW. When I connect everything and run it, the motor starts humming. If I turn on a simple spring effect the motor gets very notchy, like I just connected the windings directly to a power source, worse the further from center I go.
What could this be? Got the windings mixed up maybe? I have no idea which one is which..

I've spent weeks, trying to build a FFB-wheel using a hoverboard motor, and now I found this thread!

I took a cheap 4S ESC, ripped out the atmega 8 bit MCU and soldered wires from the FET drivers to an arduino nano to which i also attached the hall sensors from the hoverboard motor. Took me a while to write the new ESC code, but after a while I had a sensored ESC (reinvented the wheel here, but a fun learning process)...

To control it, I use a Leonardo board with a binary i found online (RFR, AI-Wave) and an AMT encoder. It works well, especially after running wheelcheck & lut-generator and using that .lut in assetto corsa.
The downside, I used a binary and have no chance to alter the code for the leonardo board. The calibration procedure is quite scary when it starts to spin uncontrollably.

My next idea is to get rid of the leonardo and simply write a feeder/receptor program for vJoy that runs on my computer and sends torque-commands to the nano via serial. This is where I'm stuck at the moment... I've managed to aquire FFB packets, but the SDK documentation is not so great for the C# wrapper, so don't really know what to do with the packets.

@danove_b How do you forward the data from vJoy to your controller?

Well, during my time doing this I realized that many sim/game providers have their own way sending the signal. I actually haven't been that successful solving it, but with some reversed engineering I have managed to make it work with my only use sim rFactor2, and also LFS. The AMS2/PCARS2 seems to use the same effect (periodic) as rFactor2, but instead there is a problem with the steering input. rFactor1 works (using Constant effect) but not with "full" effects. It might be that the version of vJoy that I use (2.1.9) still have som errors, I don't know. I recently found this:
https://github.com/njz3/vJoyIOFeederWithFFB
Someone seems to have released a 2.2.0 version of the vJoy driver, but doesn't seem compatible with my current code, the ffb works, but the inputs doesn't. My advice is that you take a look at the vJoyIOFeederWithFFB code, it seems that is a ready to use code of what you are planning to do. At least you can look how they solved it.

Now I do! After some shuffling around I ended up with a working wiring! (It is tough being color blind )

@Granberget Looks like you and I are in about the same place on this project. I started working on a vJoy feeder program last night and was able to get inputs from the wheel to vJoy pretty easily. Was even able to drive a few laps in iRacing.

Next step is to get FFB from vJoy to send to my motor controller. (like you, through arduino), Unfortunately, iRacing didn't recognize vJoy as FFB wheel, so I am stuck. Any ideas on how to set up vJoy to appear as FFB wheel? Did it "just work" with assetto corsa, I tried various versions of vJoy with no luck. Any help is appreciated.

Got my unit up and running today!
Big improvement from the old one.

I haven't really tried to see if any game recognizes the vJoy device ass FFB yet. I might be able to pick the cherries from the github link that danove_b just shared.

However, as i said - it's quite scary to have the wheel spin uncontrollably when "calibrating" and so i decided to make the wheel buttons/shifters wireless (learned the hard way). So if anyone is interested, i've successfully managed to write a feeder in kotlin (like java 2.0) that gets the input from a wemos D1 mini and send it to vJoy.