Core XY high speed closed loop tuning

[Tech_Sam03]

I wanted to get in touch again. I've put the closed loop project on hold for a while and now I want to get back to it. My main problem is the big error when accelerating and decelerating.

[dc42]

@Tech_Sam03 what amount of error are you are getting when accelerating and decelerating, and is it during printing moves, travel moves, or both?

[Tech_Sam03]

@dc42

About */-~1.5 Steps of error in the closed Loop Tuning Plugin. I have set my error boundary to E2:3. I don't get any messages while printing, but I do get them when traveling.

[dc42]

@Tech_Sam03 thank you. Is this with the Pid A term optimised for travel moves or for printing moves? What acceleration are you using for travel moves?

[Tech_Sam03]

@dc42
I use a travel acceleration of 40k mm/s. I tuned it between travel and printing Speed.

[Tech_Sam03]

@dc42

I did a quick, rough tuning to see if I was making any mistakes.

Step 1 Speed and Accels

Speed too low

Speed Ok

Step 2 A therm

A too low (A30.000)

A therm is ok(A40.000)

A too heigh odd spikes in P therm(A50.000)

Step 3 V therm

V too low(V300)

V ok(V400)

V too high(V500)

Step 4 D therm

D ok(D0.1)

D better but to loud(D0.2)

Step 5 I therm

P20 A40.000 V400 D0.1 I0 Current tune no steady error. Do i even need a I therm because i dont really have a steady-state error?

I500

I1000

I1500

System specs:

-Version 3.6.0-RC3
-LDO 42STH48-2504AC i pressed in a longer shaft in the rotor to fit an encoder to the back
-48V
-2500A for the motors
-AMT10E3 encoder 2000PPR

[dc42]

@Tech_Sam03 before you added the D term you had a lot of ringing in the P term. You would also have seen this in the Current Error had you plotted it. That's an indication that the P term is too large. A small amount of ringing can be countered by adding a small D term, but as you discovered, larger D terms make the motor noisier.

[Tech_Sam03]

@dc42

This is the plot befor adding the D therm

Should i use 30-50% off the current P therm and test again?

[dc42]

@Tech_Sam03 you may not need to reduce the P term by as much as that. Try 50 to 70% of the current value.

[Tech_Sam03]

@dc42

I tried a P therm of 15. But i get those high peaks (Error/PID) in the accelereation segments of the move.

[dc42]

@Tech_Sam03 that plot suggests that the A term was too low or the maximum motor current too low.