Problems with Phoenix 6, SwerveGenerator (with and without pathplanner)

We’ve been struggling with tuner x Swervegenerator and migrating to phoenix 6.
We took the tunerconstants.java and dropped it into
Phoenix6-Examples/java/SwerveWithPathPlanner/src/main/java/frc/robot/generated
(all the required tuner x tests passed for calibration and rotation and so forth). We can interact over tuner x and drive individual motors too.

Code builds and deploys. Oddly, however, the right controller makes only the 2 left swerves rotate, while the left controller makes them rotate very fast (as if this was the drive motor maybe?). Never see the drive motors actually do anything either. Hard to understand how this is happening, given the reports on CD of how easy it was for folks to bring this up. We must be doing something dumb.

We also are getting these errors:
ERROR -10021 Device firmware could not be retrieved. Check that the device is running v6 firmware, the device ID is correct, the specified CAN bus is correct, and the device is powered. pigeon 2 1 (rio)
ERROR -10021 Device firmware could not be retrieved. Check that the device is running v6 firmware, the device ID is correct, the specified CAN bus is correct, and the device is powered. talon fx 5 (rio)

But everything on the can bus is phoenix 6, recently updated (days). We also don’t have anything on can 5 (if that’s what that means).

If I hadn’t seen this exact same robot working before updating to phoenix 6 (pro), I would suspect hardware, but that seems unlikely.

We also tried the autogenerated code and are finding that the steering kinda looks like it is working, but nothing happens when we drive.

I would first suggest posting a screenshot of what you see on tunerX when connected to the robot.

Good suggestion, forgot to get this for completeness… However the matter has been resolved.

We ran tuner x project builder on a fresh robot and make sure to copy the tunerconstants.java file into the resulting repo, then built it and it did reasonable things.

I did a diff of the above repo against the one in the last line of my original post, above, and found that the tunerconstants were the only difference.

My conclusion is that we didnt copy the generated tunerconstants file into this repo the prior day?

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.