I have a 2-axis force feedback joystick build that I have up and running with Odrive pro’s and RMD-X8 PRO 1:9 motors. Its working great, but whats key to it performing at peak is minimising cogging wherever possible.
I have done the anticogging calibration with a default set of suggested parameters from the docs (same settings recommended for the 8 pole NEMA 34), however I want to do some experimentation on settings. The problem is disassembling my stick each time to run the process is a bit of a headache!
I was hoping someone had some guidance on setting selection to minimise the trial and error I might need to go through?
Hi! You could do a cogging test by moving the motor in velocity control and measuring velocity ripple. I made a script to sweep across anticogging parameters (worth running overnight – takes a while!) and then another one to plot the resulting cogmaps. It also captures velocity data – you could add a heuristic such as RMS ripple to find the best anticogging solution. Though I think there’s probably better heuristics – e.g. giving the motor a torque impulse to spin up a bit, then letting it spin down at zero torque and measuring the RMS velocity ripple as it coasts to a stop. But that might not work so well depending on your system’s friction
Here’s my messy scripts, definitely worth modifying for your needs.
Great idea thanks! Appreciate the existing code as its well outside my capability to mock that up from scratch. Ill have a play around with this and see what I can come up with
You can search for relevant posts and discussions on the ODrive Community Forums to receive advice on setting picking and reduce trial and error. Users discuss a range of topics in the forum, offering their solutions and experiences.
