Emco PC turn 55 upgrade..

I commented out the BASE_PERIOD that I had added, changed P values to 1000, and turned the velocities back up to 25mm/s and accel down to 500mm/s^2. I still get a following error at the same spot. I'm using fusion 360 so between the facing op and the profile it retracts to z5 and then when it begins the profile op it advances in the -z direction towards the lead in for the profile. It's at this reversal that I get the following error. I have backlash compensation on, and the section in the manual just says backlash compensation is turned on as a possible cause of following errors.

Added the .ngc file.

I commented out the Z backlash and I'm still getting the error. I don't know if that actually turned off backlash on the z axis though.

Based on comments in my .ini file, try increasing STEPGEN_MAXACCEL

I already have STEPGEN_MAXACCEL set to double MAX_ACCELERATION.

Make it bigger, say 300%.

I did push these up a few times to get rid of following errors on my config but I am not using backlash compensation.

I already have STEPGEN_MAXACCEL set to double MAX_ACCELERATION.

If you have backlash compensation, the STEPGEN_MAXACCEL value should
be about 2.5x the MAX_ACCELERATION value

This is because with backlash compensation, the actual requested acceleration
can be 2x MAX_ACCELERATION and the STEPGEN_MAXACCEL needs about 1.25x
headroom above this

I just tried it at 500 and 1500 with the backlash value back in, and I got the same errors. I should maybe mention that I've got 2 errors, but I've assumed that the first one is just because the machine is stopped.

also change

setp pid.x.maxerror .0005
to
setp pid.x.maxerror .01

and the same for Z

(this is a pncconf bug = too low value of pid maxerror for mm machines)

Maybe this does have something to do with my G-code. It's trying to go from 2000rpm to 0 to 2000 again instantaneously. If I delete the M5 from the code it seems happy.

Did you change the pid.(N).maxerror values?

All bets are off if these are still .0005 mm

Its possible but unlikely that there is a trajectory planner bug. more likely faulty steggen tuning