I’m not too familiar with the design of the microdrivers–I just bought them and used them (via the SPI instead of the CAN interface). But based on my experience debugging it, I had the red LEDs come on during early testing whenever my power supply could not supply the required current when moving motors. It then enters a fault state with the red LED and noisy motors, which can be exited by pressing a physical reset button on the board. There are some other threads in this forum describing this behaviour, for example here: Motors stop working after a fast movement - #2 by fgrimminger
I’m hoping someone more familiar with the hardware/firmware of the micro driver can confirm/follow up!
Sorry, I realized that my suggestion was wrong–I see in your first video now that the red LED isn’t on there. So my suggestion of current limits is likely not the issue.
I haven’t encountered this type of behaviour myself, so I don’t have suggestions for solving this. The only other possibility I can think of are the motor phase wires–if you swap two of the motor wire connections, do you get the same behaviour?:
Sorry about that. I’m hoping that someone else could give additional suggestions.
we haven’t seen this problem here so I asked the designer of the board Jonathan Fiene @jfiene for advice.
Here is his reply:
That is indeed really noisy! I’m skeptical that the copper thickness would explain it, but it is possible. I’d suggest to measure the PVDD on the DRV driver ICs with a oscilloscope, as the thinner copper could introduce enough resistance to reduce that voltage below the optimal level.
The voltage level actually has a wide valid range (4.4 to 45V), but it does need to be reasonably stable. I’d suggest to watch that pin with a scope to see if it’s stable, and perhaps compare with a TI board.
Did you order multiple micro driver boards?
Did you test the other ones to see if this problem occurs with all of them?
I don’t know why the capacitor values are different.
But all the micro driver boards that we have here are working and were ordered based on the values in that BOM.
So that has to be fine.
Good luck with finding the issue.