Rotating Magnetic Sensor - Keeps Pausing Printer

T3P3Tony

@alankilian thanks, if you are happy to make the upgrade then please follow the following steps.

First upgrade to 3.3:

download the Duet2and3Firmware-3.3.zip file from https://github.com/Duet3D/RepRapFirmware/releases/tag/3.3
and use "system" -> "upload system files to upload it. it will prompt you to install the updates say yes.
Assuming that all goes as expected DWC should reload and you should see 3.3 versions reported for DWC and RRF in the General and Machine Specific tabs in DWC.
To get around a issues with future upgrades (we switched from /sys/ to /firmware for firmware binaries between 3.2 and 3.3) the best thing is to re- upload the Duet2and3Firmware-3.3.zip again at this point. you don't need to re-install the updates. this just ensures stuff is in the right place for future upgrades.

Second upgrade to 3.4b5

get the Duet2CombinedFirmware.bin and DuetWebControl-SD.zip files from here: https://github.com/Duet3D/RepRapFirmware/releases/tag/3.4.0beta5
Upload the firmware binary first and say yes to install the update. then upload the SWC zip.
check that the firmware and DWC are now reported as being 3.4b5

Finally upgrade to the 3.5b5+1

Get the Duet2CombinedFirmware.bin from the dropbox link posted earlier. There is no updated DWC.
uploaded it as in the previous step, say yes to the update
the Machine specific tab/M115 should report 3.4.0beta5+1 (2021-10-29)

alankilian

@t3p3tony

Board: Duet 2 Ethernet (2Ethernet)
Firmware: RepRapFirmware for Duet 2 WiFi/Ethernet 3.4.0beta5+1 (2021-10-28)

Started a print and almost immediately I got "To much movement"

Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 28.80mm/rev, allow 60% to 160%, check every 3.0mm, version 3, mag 133 agc 114, measured sensitivity 16.39mm/rev, min 100% max 100% over 12.0mm

Resumed and checked again:
Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 28.80mm/rev, allow 60% to 160%, check every 3.0mm, version 3, mag 132 agc 114, measured sensitivity 25.89mm/rev, min 94% max 102% over 78.4mm

Printing completed.
Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 28.80mm/rev, allow 60% to 160%, check every 3.0mm, version 3, mag 132 agc 114, measured sensitivity 25.82mm/rev, min 94% max 103% over 1697.9mm

alankilian

Second try: Too little movement:

Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 28.80mm/rev, allow 60% to 160%, check every 3.0mm, version 3, mag 130 agc 114, measured sensitivity 26.55mm/rev, min 2% max 105% over 118.3mm

Since my AGC is almost at the high limit, I'll do a little sanding to improve it and try again on Monday.

T3P3Tony

@alankilian thanks - it might be the agc, if not then please share the file that causes the too little movement.

alankilian

@t3p3tony

Here's the test file: CalciumCup.gcode

I'll adjust and try printing a few more times on Monday.

My sensor is just below the extruder and moves around a little bit with the bowden movement.
I should probably move it above the extruder.

I'll also try that on Monday after getting the AGC in a better range.

alankilian

@dc42

Unfortunately, I can't get the AGC under about 115 without the pointy-bits of the brass roller starting to scratch the surface of the sensor chip and not moving all the time.

You mentioned stronger magnets. Are they available?

Otherwise I may try sanding the top surface of the sensor chip a little.
I figure there's enough plastic above the sensor I can get the tiny amount of space I need to get a better AGC value.

T3P3Tony

@alankilian yes if you drop me a line (info@duet3d.com) i can look at arranging a replacement assembly.

wouldstain

@t3p3tony im actually still running 3.3 but I will upgrade today to 3.4b5 and 3.4b5+1 progressively.
I just wanted to confirm the file(s) I needed. Should I be using the combined version?

thank you!

T3P3Tony

@wouldstain the instructions i posted above for alankilian should work for you if you are using Duet 2. do Duet 3 substitute the appropriate firmware binary for your board.

T3P3Tony

I have done a quite a few tests now and over the short print snippets I have been using I have got consistently good results (wider window than 90%-105%) until PA gets greater than 0.4, with higher pressure advance i see drops down to 70% but still nothing like the 10%, 0% -n% that we were seeing before.

I have more tests to do with different retraction distances and print speeds but this is looking very promising. I would be interested in gcode files (preferably short!) that still get poor results with the new build of firmware (3.4b5+1) where all other factors are limited (i.e. good mag readings, short filament path between the extruder and the sensor).

wouldstain

@t3p3tony Updated firmware and DWC as requested. Printed up a batch of those drywall anchors today with no problem out of PETG.

As soon as I switched to PLA it started happening again:

Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 26.50mm/rev, allow 1% to 200%, check every 3.0mm, version 3, mag 130 agc 75, measured sensitivity 30.28mm/rev, min 0% max 121% over 97.7mm

m572 d0
Extruder pressure advance: 0.070

PLA_Test-xyzCalibration_cube.gcode

Ran again:

m591 d0
Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 26.50mm/rev, allow 1% to 200%, check every 3.0mm, version 3, mag 130 agc 75, measured sensitivity 40.48mm/rev, min -1% max 154% over 33.8mm

T3P3Tony

@wouldstain thanks for the report.

Other than the temperature is the only difference between your PETG prints and PLA prints the pressure advance?

You may have already posted it but please post your config.g so i can use those settings in my test.

dc42

@wouldstain is the only difference between the PLA and PETG prints is the hot end and bed temperatures? If so, then it's likely that your extruder is skipping steps or losing its grip on the filament, and the filament monitor is reporting correctly. PLA is harder for the extruder to push through the nozzle then PETG is.

You could test this by printing at a high temperature, to make it easier for the extruder to push the filament through. It may result in rubbish print quality, but if it improves the filament monitor readings then it will have identified the problem.

T3P3Tony

@wouldstain also what results were you getting with PETG on the drywall insert?

Also looking at the PLA gcode you share, you are printing it at 235C.. have you had success with that temp for PLA in the past?

wouldstain

@t3p3tony I have had good results with PLA at higher temps.

I have tried printing at lower temps. the only difference on test / calibration prints seems to be the strength of interlayer adhesion.

I can certainly dial it down if you think it would make a difference.

wouldstain

@dc42 I normally have higher retraction distance and speed settings to reduce stringing on PETG, so aside from decreasing both while using PLA, really no difference.

I've tried printing PLA with higher temps (accounting for either toolsteel or hardened steel nozzle) but it doesnt seem to matter.

@JohnOCFII has the following in his prusa config:

{if printer_notes=~/.PRINTER_VENDOR_PRUSA3D./}M900 K45{else}M572 D0 S0.06{endif} ; Filament-specific Linear Advance or Pressure Advance

but I've also tried slicing with this commented out and gotten the same results.

wouldstain

@t3p3tony

about to run test @250C. Will post console output (assuming it pauses)

config.g attached. modified slightly from previous to account for addition of garolite sheet on build plate.

WS_110321_config.g

T3P3Tony

@wouldstain said in Rotating Magnetic Sensor - Keeps Pausing Printer:

I can certainly dial it down if you think it would make a difference.

no i don't - that was just an question seeing it at 235.

wouldstain

@t3p3tony gotcha.
Just ran it at 250C. got this at less than 6%:

Duet3D rotating magnet filament monitor v3 on pin e0stop, enabled, sensitivity 26.50mm/rev, allow 1% to 200%, check every 3.0mm, version 3, mag 130 agc 75, measured sensitivity 26.29mm/rev, min 1% max 107% over 136.1mm

same cube.

T3P3Tony

@wouldstain I printed that on the test rig:

That was before I got your config so i need to change to your config.

Can you try printing one of the drywall anchors in PLA with the PLA settings to see if you get the same issue (trying to see if its related to the PA/retract settings which are different). Also a short print to recreate the issue would be good.