Magnetic filament sensor false errors.

[claustro]

Hi guys,
I am having a lot of random false alarm with my filament sensor. It worked like a charm in the past.
The typical behavior is that the min % gradually, very fast, goes to 0% and stays there forever event the filament extrusion remains regular.
The error appears in almost every kind of print and print materials
During printing the led of the filament sensor is almost green but in specific head movment it become first off and then red.
Here a video that show it

https://youtu.be/3U9-kPt0wGs

my machine:
Board: Duet 2 WiFi (2WiFi)
Firmware: RepRapFirmware for Duet 2 WiFi/Ethernet 3.4.0rc2 (2022-02-22)
Duet WiFi Server Version: 1.26

15/5/2022, 10:46:01	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 30.31mm/rev, min 0% max 123% over 51.3mm
15/5/2022, 10:46:00	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 30.31mm/rev, min 0% max 123% over 51.3mm
15/5/2022, 10:45:58	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 30.31mm/rev, min 0% max 123% over 51.3mm
15/5/2022, 10:45:57	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 30.03mm/rev, min 0% max 122% over 48.3mm
15/5/2022, 10:45:55	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 27.68mm/rev, min 38% max 112% over 44.5mm
15/5/2022, 10:45:53	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 27.68mm/rev, min 38% max 112% over 44.5mm
15/5/2022, 10:45:52	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 132 agc 77, measured sensitivity 26.11mm/rev, min 53% max 106% over 40.5mm
15/5/2022, 10:45:44	M591 D0
Duet3D rotating magnet filament monitor v3 on pin e0stop, disabled, sensitivity 25.00mm/rev, allow 50% to 150%, check printing moves every 3.0mm, version 3, mag 131 agc 77, measured sensitivity 26.35mm/rev, min 54% max 107% over 34.5mm

this is my filament line config.

M591 D0 P3 C"e0_stop" S1 R50:150 L25 E3.0

this is my hardware setup

The only thing I changed lately in my setup is the extruder switching from a E3d V6 to a Revo six

[jay_s_uk]

@claustro it may be an issue with the housing.
I suggest printing the replacement one in this thread https://forum.duet3d.com/topic/27780/redesigned-the-mfm-housing?_=1652607235984

[Lo-Fi]

What you're describing certainly sounds like a mechanical problem. From the photo, it doesn't look like you're running PTFE in your sensor housing (?), which may a factor. It serves to stabilise the filament path, rather than relying on the idler to provide enough pressure to overcome any force the filament may be applying as it's entry or exit angle changes. It wouldn't surprise me if you found the filament hole in the housing has worn slightly and that's why you're seeing a problem now.

Some further notes in the thread Jay suggests above. I've had zero issues with my sensor since moving to my updated housings and had some good feedback

Keep us posted!

[claustro]

@lo-fi Thank you all for your help.
I printed your mod and another original one using a SLA printer I'll remove the sensor and I'll check it.
I never had an issue in not using the PTFE tube because the sensor is mechanically fixed to the extruder, anyway I'll can add it .
thank you

[Lo-Fi]

@claustro a pleasure!

Only need little short runs of PTFE, it's just there to provide a nice low friction, non wearing entry and exit and to keep the filament straight as it's passing the hobbed section. Particularly important on setups where the angle changes lots as it looks to be doing with your rig. Without that support, it's not impossible for spring in bending filament to overcome the idler spring and move away from the roller. The holes for the filament in the original design are fairly tight and to some extent serve that function, but will wear over time without the PTFE. May be why you're only seeing the problem down the line.

Interested to hear how you get on, anyway.

[Phaedrux]

@claustro said in Magnetic filament sensor false errors.:

Firmware: RepRapFirmware for Duet 2 WiFi/Ethernet 3.4.0rc2 (2022-02-22)

I suggest updating your firmware to either 3.4 final release, or the recent 3.4.1 rc1

[claustro]

Thank you all for helping me.
I solved opening the case of the magnetic sensor and cleaning it. A lot of debris of plastic was around the ball bearings.
I made 2 hours of printing without issue.
Thank you