Ok, I can confirm that this (in the input channel gcode):
; Set dummy filament diameter and then load filament settings
M200 D1.75
M703
; report filament diameter?
M200
Does get me into volumetric extrusion mode, and that the M703 does result in a change in the filament diameter. Somewhat bogus way to have to do things, but this does solve my problem for me.
My best guess is that something is going wrong with the +5V and/or +3.3V supply rails when you apply VIN power.
Does it do the same thing if you disconnect everything except VIN and USB from the Duet?
Does the A4403 chip (small square chip close to the VIN terminals) look OK?
By disconnecting everything, the same thing happens.
I looked with a magnifying glass and everything seems normal
Do two test prints. Once with G29 S1 active, and again with G29 S1 commented out. You should be able to tell if it's working as intended.
I don't think going back a firmware version to avoid a warning message is such a great idea. You also give up any bug fixes and improvements. But to each their own.
All the warning is saying is that the Z0 point wasn't established by a probe before the heightmap is loaded. As long as the Z0 point of the endstop is the same now as it way when the heightmap was created, you should have no issues. The reason the probe is preferred is that it's getting a current reading from the actual bed surface each time. No if your endstop is accurate and the build quality good (as it would appear yours is) then the endstop method is fine.
I'm printing from SD card. This GCode does have lots of small segments, but only on the perimeters and they are printed at 6-12 mm/s.
The problem areas seem to be the infill or support, which are merely rectilinear with small zig zags.
But actually, now that I think about it, the dense support structures from PrusaSlicer draw loops that conform to the curve of the model. This could be it, since I was printing the supports lightning fast with PA enabled.
The only thing that exists in the slicer related to starting/ending a print are calls to macros
print_begin.g
print_end.g
; ----- print_begin.g -----
;
; must occur after...
; - all axes have been homed
M106 P2 S255 ; lights on
M106 P6 S255 ; lights on
M106 P7 S255 ; lights on
G90 ; absolute moves
G1 Z75 F1200 ; position for cleaning
G1 X0 Y-145 F6000 ; position for cleaning
T0 ; select tool 0 so extruder commands below will work
M703 ; configure selected filament (sets temp set points)
M291 R"Waiting on heaters to reach set points..." T0
M116 ; wait for temps to reach set points
M291 R"Priming extruder..." T0
G92 E0 ; reset the extruder logical position
M83 ; extruder relative mode
G1 E10 F120 ; prime the extruder
G92 E0 ; reset the extruder logical position
M400 ; wait for extruding to finish
M291 R"Clean Nozzle and Bed" P"Click OK when ready to print" S3
M98 P"configprobe.g" ; configure Z probe
M98 P"probeZ.g" ; set Z=0 datum
G29 S1 ; load height map
M291 R"Starting to print..." T1
If you are looking for a quiet (fan-less) power supply that stays cool, has proper safety certifications and is from a know brand, I recommend the Meanwell UHP series, they come in different voltage and power specifications, e.g. https://www.amazon.com/gp/product/B07GPDRV7V
It appears to be working after going back through and remaking the configuration in the configurator, maybe something didn't like transferring over from 2.0 firmware with the updates but it's working now! thank you for the help!
FWIW we have prototypes of a CAN-connected closed-loop stepper motor controller board for Duet 3. I think that or a variant of it will be the best solution for your extruder.
@Veti that link defeats the purpose of forum help, Try to be more specific in your help, instead of just links to massive amounts of info
He pointed to the answer. M558 probe type 0 lets you probe without a probe. If you can't be bothered to read the suggested existing material why should we type it out again in a forum post?
The point of a forum is to discuss the existing material. If it's misunderstood, you can get clarity, but we can't reinvent the wheel over and over.