Can't get DEV branch to run


I can’t get the dev branch to work at all. Not in open loop or closed loop, the encoder is reading fine. As soon as I go back to the master branch it all works fine. Is there anything else that needs to be called in the dev? I’m using stm32f405 and platformio via VS Code.

Hey Matt,
I was not aware that there is such an issue. The branch has been updated yesterday and is the testing phase. I’d avoid it for the next few days.

If you wish to use the dev branch please go back to the commit earlier than yesterday.

We’ll keep you in the loop.

Hi Antun,

Okay thanks, will leave it for now. Just having trouble running and tunning low resistant 150kv 0.039ohm motor. Sometimes it will be perfectly fine with velocity LP-Filter set from 0.5-1.0 and after a restart, it needs to 3-5. The velocity feedback is quite noisy even when running in an open-loop with steady vq. I have downsample set to 1.0 anything greater than 1.0 seems to be worse. The motor does have high cogging so this does not help. But I did have it running really well from 3r/s to 75r/s and with exact same tuning paremeters after a reset it won’t run at all and very unstable and to the point motor makes a high pitch noise and stalls. Note the motor is the Odrive 150kv motor and runs well on the Odrive.

I have also come across some issues with start that caused major issues and couldn’t get the motor to run at all. I have done a temp fix and am not familiar with Github and pushing them to the dev branch to get added. Basically if the phase resistance is set and motor does the encoder index search it will not use the sensor align voltage and just uses calcultated voltage limit which is way to high for open loop angle. I have added some logic on the while loop in the absoluteZeroSearch function to detect the zero search is active and not to use the voltage limit and to use a new variable I have added. Index search voltage. The reason for this is the index search voltages needs to be alot lower than the sensor align voltage. If the sensor align voltage is the same as needed for the Index search (0.5) I get bad data for electrical angle etc and pp search fails. I use 0.5 just for index search and 3 for voltage align I get good data and pp ok.

All this is probably fine for high resistant gimbal motors, but with low resistant motors it’s causing issues. I have had really good results a JKLB42 BLDC motor and runs very well and experiences none of these issues. What are your thoughts?

Any updates on the low resistance motor tuning? I’m trying to initiate a 0.053 ohm BLDC with hall´s. I guess it is because of cogging, that it can’t read movement and therefore fail init. Will you share your code?