Glad it's working now.

@alankilian
It quickly flashes green when printing so I assumed it was working... It always says no data until after about 30 or 60 seconds (enough Filament has gone through the sensor). It's currently pauses before it has a chance to get through that much filament. And when it pauses it doesn't show the sensor status on the console. Just shows an error for not seeing enough movement and says paused.
I will check the wiring as it makes sense to think one wire may have been slightly pulled loose.

Thanks! I will look at it tomorrow and follow up. (Printing now)

@ziggymanpopo I think what you're asking is already in the M591 command, using the S parameter:
https://duet3d.dozuki.com/Wiki/Gcode#Section_M591_Configure_filament_sensing

Sn 0 = disable filament monitoring (default), 1 = enable filament monitoring when printing from SD card. Supported for all filament sensor types in firmwares 1.21.1 and in 2.0 and later. In firmware 1.21 this parameter is not supported for sensor types 1 and 2. Filament monitors accumulate calibration data (where applicable) even when filament monitoring is disabled

So just set S0 and it will still measure (so you can calibrate) but not stop printing.

If you want to do this per print, maybe put the line in your start code?

@Veti said in Strange error with magnetic filament sensor:

@the_dragonlord

i think you need to calibrate the sensor again.

ok, thanks

@Hiroaki said in Filament laser sensor shows too short movement in a minute.:

Maybe because of the filament is dark blue ?
Maybe because the sensor has a dirt on it ?

Yes, it's possible.

Have you read through this page?

https://duet3d.dozuki.com/Wiki/Duet3dFilamentMonitor_LaserVersion

Hi there
Yes, i already did the PID tunning, i have an E3D with a sock, but the weird thing is that the malfunction happens only while autoleveling, if there is no temp error after autoleveling the printing goes perfect, the temperature is very stable.
referring to the filament monitoring, i tried like you said and looks like works 🙂

We have a few sets of housingless MFMs at Filastruder, drop an email at support@filastruder.com.

@Phaedrux thank you so much!

@scottypres said in M591 Simple Filament Sensor Trouble:

E0

I think the E0 should be removed.

https://duet3d.dozuki.com/Wiki/Gcode?revisionid=HEAD#Section_M591_RepRapFirmware_1_21_to_2_x

If you can feed the filament past a toothed wheel before it reaches the filament monitor, that may put enough texture on it for the laser sensor to see it better.

@T3P3Tony
Thank you, that worked. I believe the issue was that no magnet was detected. Mounted the board in it's proper location (next to the magnet) and it reported v3. With it just plugged and hanging in it reported v1. This is a Duet3 no SBC on v3.1.1. Thanks again.

I have it on the work list to add a new macro file for a pause caused by a filament problem.

I think it should, but AFAIK that combination has never been tested.

@makemerush No problem. The orange part is that limit switch which is was specifically chosen because of it's activation distance as well as the tension it provides. Essentially this is killing two birds with one stone. The switch acts as an idler assembly to provide tension against the hob and at the same time also acts as a switch to statically detect filament presence.

The switch can be purchased here:
https://amzn.to/2SkgX9U

The switch is attached to the switch assembly with a single screw so it can pivot. The hole on the top side of the switch assemble is used so you can insert a screw to increase or decrease tension and activation distance to tweak your setup. The screw just pushes the switch tighter against the hob.

Screenshot 2020-10-01 at 10.09.59 PM.png

The second thing that may not be obvious. Is the mount. I got great results once I used a rigid mount. You don't want this flopping around and only held on by the tube. I hope that helps.

BTW, with the new firmware beta released the L parameter can be used in the config.g to normalize the results between 0-100%.

@claustro in that picture you do not have PTFE Tu e between the extruder and the filament monitor. You need that to constrain the filament path otherwise retractions especially will not be registered properly

The reactions need to push the filament back past the sensor consistently.

@ziggymanpopo I find the laser monitor works well with indirect sensing. I adapted a housing developed by others (https://www.thingiverse.com/thing:3717158) ; this approach pretty filament agnostic, and with the calibration parameter in the 3.2b1 M591command gives excellent results. On my 1.01 duet wifi it works best on the lcd_conn with a shielded cable.

if ever there was a solution looking for a problem...................

Which firmware version?

In RRF3 C has different value - more like M591 D0 P3 C"e0stop"