E3D Tool Changer Stall Detection

fcwilt

Hi,

I was looking at some sample code for the tool changer system from @dc42.

The tool coupler stepper was configured to use stall detection.

On my 6HC I setup the same settings as used in the sample code but with logging enabled there was no message appearing in the log showing the stall detection worked.

These are the relevant bits from the config.g file:

M906 C500 I30        ; set coupler stepper current

M913 C50             ; reduce coupler stepper current

M574 C1 S3           ; endstop using stall detection for coupler at low end of its range

M915 C S6 F0 H200 R0 ; stall detection settings for coupler

These are the relevant bits from homec.g file:

M913 C60	    ; update coupler stepper current
G1 H2 C-1000 F10000 ; move towards low end to the end of travel

I enabled logging with M929 and changed the R0 in the M915 to R1.

The log file was created but only contained the header:

2022-05-09 18:56:30 [info] Event logging started at level debug
2022-05-09 18:56:30 [info] Running: Duet 3 MB6HC v0.6 or 1.0: 3.3 (2021-06-15 21:45:46)

Any ideas what might be wrong? Or does stall detection even work with such a small stepper?

Thanks.

Frederick

engikeneer

@fcwilt I believe that the coupler motor gives quite inconsistent results, especially if it gets warm.
Mine just drops the current a bit and drives against the endstop so it keeps stalling then uses a G92 to set the position. No stall detection involved
Machine is 2nd hand and came like that so not sure if that's the default

dc42

@fcwilt it's as @engikeneer says. The high resistance of the small stepper makes it impossible to get consistent stall detection in my experience.

fcwilt

@dc42 said in E3D Tool Changer Stall Detection:

@fcwilt it's as engikeneer says. The high resistance of the small stepper makes it impossible to get consistent stall detection in my experience.

Thanks for the feedback.

Frederick

fcwilt

@engikeneer said in E3D Tool Changer Stall Detection:

@fcwilt I believe that the coupler motor gives quite inconsistent results, especially if it gets warm.
Mine just drops the current a bit and drives against the endstop so it keeps stalling then uses a G92 to set the position. No stall detection involved
Machine is 2nd hand and came like that so not sure if that's the default

That's what I was doing but then someone gave me a link to some dc42 sample files which had code for using stall detection. I guess the sample code is just out of date.

Frederick