DD-Wheel with Hoverboard Motor

Hello Benu
Now I have a bigger problem
When mmos has ffb output, turning the motor will feel like a gear.

I know this problem occurs when the connection between the motor and the encoder is unreliable
Now I keep testing, the problem is still there.

I think Ive had that before. It was some kind of combination of the phase cables and the encoder cables being round the wrong way

I have also tried exchanging the connection of phase A and phase B of the encoder, exchanging the connection of the motor uvw, more combinations

Update on my solution utilizing the original hoverboard control board.

Through communication with developer of EMC, he has created a version which can output PWM signal in a format that match a tradition RC servo, and can be used to command the original hoverboard with hacked firmware. This simplifies connections, but surprisingly had little change in performance. Please contact EMC directly if you want to try it.

Also, to knock down the oscillation problem I added some mechanical friction directly on the motor hub. This eliminated the oscillations, even at high torque settings.

Next step:
-- Clean up wiring and pack in one housing.
-- 3d print mount for friction wheel

Swapping the wires of encoder A and B would have been my guess too.
Is this only happening with the new firmware I sent you or with the previous firmware too?

The new firmware has the same problem as the previous firmware

I am a bit confused now. To be sure that I understand the problem correctly I will summarize. Please correct me if my summary is wrong.

Every time MMOS board sends force feedback signals to the ESC you feel a cogging torque while turning the wheel. That is, when you turn the wheel, it does not move smoothly. It moves a small distance and then stops moving. To move it again, you have to push harder. It then moves to the next position and stops moving again. This is the same at every position of the wheel.

In an earlier post you said, your opto-isolated encoder setup was working as expected.
You were using the first firmware I sent you for this test?
During that test there was no problem with cogging torque?

Really confused

Your description is exactly the same as my question, with a strong sense of cogging. When I increase the mmos Kraft Maximum value, the cogging is also enhanced.
I counted about 15 cogging (seems to be the same as the number of pairs of motor poles).

I ran the Motor Profiler again to test the motor and it ran normally. Then re-upload the firmware you gave me. I turn the steering wheel to the angle limit or turn on the spring force(MMos/Desktop Effekt/Feder). I only feel the cogging torque when I turn the wheel. (I also removed the optocoupler intermediate connection)

Use the firmware you sent me the first time and the second time, which have been tested normally before (FFB output steering wheel is very smooth).
But now I have no solution

Ok, from your description I take, that firmware (both versions), ESC and encoder are delivering a smooth force feedback.
The problem only occurs when you
- turn on (MMOS/Desktop Effekt/Feder)
or
- hit the angle limit (is that the wheel rotation limit set within MMOS?)

If you do not hit rotation limit and do not turn on (MMOS/Desktop Effekt/Feder) everything is working smoothly.

Previously tested normal
The test is not normal now
mmos has no pwm output, smooth and no resistance when rotating.
mmos has pwm output and feels cogging torque when rotating. Any feedback force is expressed as cogging torque(Including steering wheel turn angle limit; opening spring force)。If you enter the game, it shows cogging torque。

I uploaded a video
1. Connect usb
2. Turn on the power...

Thanks for the video. I am not at my rig for the next few days, so I cannot test this with my setup.

I would say the cogging at and beyond the angle limit is normal behaviour.
When the wheel hits the angle limit, MMOS board simulates a mechanical stop. To do that it creates an opposing force, to prevent further turning of the wheel beyond the angle limit.
If you you use higher forces to overcome the opposing forces, you enter uncharted territory.
When you turn the wheel back within its angle limits, everything is back to normal operation.
Usually I use 900° (Lenkanschlag) while driving and let the racing sim to the rest. So far I had no problems with that.
To make it more difficult to go beyond the angle limits you can increase the opposing forces with (Lenkanschlag - Verstärkung / Kraft Maximum).

For the second problem (Desktop Effekt - Feder), I have never used those and don't know what they are supposed to do.
What are they used for?

grateful
Desktop Effekt-Feder keeps the steering wheel always in the center. When you turn the steering wheel (left or right), you will feel a reverse force, and the steering wheel will return to the center position when your hand leaves the steering wheel.

As you can see from the video, when the steering wheel is turned to the angle limit, the steering wheel will be stuck at the angle limit, even if your hand has left the steering wheel. Then continue to turn beyond the angle limit, the steering wheel will be stuck in the next position, even if the hand has been released.
The correct feedback should be that when the steering wheel is turned to the angle limit, there will be a continuous reverse force, and when the hand is released, the steering wheel will be reversely rotated to within the angle limit instead of being stuck in a certain position.

Normal cogging is slightly imperceptible. During the previous normal test, I didn't feel the cogging phenomenon at all, it can be said that the perfect force feedback!
I only encountered this problem after I rearranged the wiring (without changing the connection method), and now the steering wheel only emits this strong cogging force, and I can't recover it by trying various methods.

Check the voltage on all 3 phases and 0 volt (Minus) on the supply. They should all be at almost the same amplitude. I suspect that one of the output NFET have failed, and puts out a constant 0 Volt or 12 volt or whatever your supply is, alternatively, one of the phases are not connected to the motor winding.

I don’t have professional equipment. Can I use the Motor Profiler to judge whether the UVW output is normal?
Now I run the Motor Profiler and the motor rotates normally.

That was strange. I had problem with one of the encoder inputs needed more pullup, but if it works in the profiler it should be ok. Have you tried to operate it in torque mode from the profiler?

Sorry, I only saw the content on the picture. It can work normally with different speeds.

I found this setting and it works

@fzxj520
As far as I can tell, the motor profiler does not use the encoder. So it still might be an issue with the encoder.

There seems to be no cogging (within the angle limits) when Desktop Effekt Feder is zero or the wheel is not pushed beyond angle limits. In those cases no force feedback is applied to the wheel.
Only when force feedback is applied, Desktop Effekt Feder turned on or pushing beyond angle limit, the cogging starts.

Maybe there is interference from power supply or motor phase wires to the encoder wires.

I had this exact problem you are describing. I solved it by trying every possible combination of phase wiring. Make a little chart to make sure you tried all combinations, if you haven’t already.