25.95mm/rev, min -33% max 162% over 9615.6mm

T3P3Tony

@nuroo with that filament path - that might be the range of what the filament monitor is actually seeing as there will be loads of elasticity in the feed tube. short retractions probably hardly move the monitor. Have you tried with it close to the extruder like @themelle

themelle

@T3P3Tony The measured maximum value may be quite accurate, but 0 or negative numbers as minimum values do not really make any sense (at least to me), do they?
As @nuroo and me are having comparable issues (min values way to low, frame and/or pol errors in sensor communication) with different sensor placements I assume we are either missing something fundamental setup-wise, or there is an issue with the sensor itself, the way it is wired up, or how robust the firmwares' communications portions are. I do have high hopes to resolve those issues with you and @dc42 within an acceptable time frame, and in turn may contribute to make the Duet rotating magnet sensor more reliable for every user out there.

samlogan87

Hmm maybe I shouldn’t have just ordered one

themelle

@samlogan87 Of course you should - the more of it in the wild, the better to resolve any random issues. It's a quite well designed product, and I am confident that it will perform as expected, once we have a larger user base and therefore more feedback as to what works and what (and more importantly: why) not.

samlogan87

Yeah I do agree, it’s more just if changes need to be made to the firmware on the sensor I am a bit screwed. I am really keen to get one so I guess we will see what happens when it arrives

Regards
Sam

nuroo

@T3P3Tony im not sure I want to move the sensor above extruder. Certainly not ideal for a corexy setup. Already moving the mass of extruder, stepper motor, hotend etc around..... I didnt realize this sensor needed to be close to the extruder when I bought it. Is that requirement listed on product page?
Can we not account for different setups in firmware with options for a setup like this.

samlogan87

@nuroo I will be in the same boat. Mine will have to go at the end of 650mm of tube. I am running a flex3drive and mounting the sensor on the carriage with the drive tube in the way will be impossible

Regards
Sam

dc42

The filament monitor must be right next to the extruder drive input or output. Any other arrangement allows the filament to flex too much in the tubing between the two, in particular when the direction of motion reverses.

nuroo

Well ok. I need to make an extension and try again. I wish this requirement was noted to customer on product page or wiki. I'll report back.

nuroo

New Setup

=== Filament sensors ===
Extruder 0: pos 188.09, ok, measured sens 24.89mm/rev min 95% max 103% over 1015.3mm, errs: frame 0 parity 0 ovrun 0 pol 1 ovdue 0

M591 D0
Duet3D rotating magnet filament monitor v2 on input 3, disabled, sensitivity 24.80mm/rev, allow 70% to 130%, check every 3.0mm, current pos 44.6, measured sensitivity 24.90mm/rev, min 95% max 103% over 1021.3mm

much better numbers, but will I get false positives when I enable sensor?
M591 D0 P3 C3 S1 R100:110 L24.90 E3.0 ;

nuroo

2 successful prints, no false positives with new setup.

M591 D0 P3 C3 S1 R90:110 L24.96 E3.0 ;

now to try a real filament outage

themelle

@nuroo Congratulations on getting it to work as intended!

Now I'm left alone again with my "frame errors"/"sensor has stopped working" problem.
May try to send it back to Filastruder as I am out of ideas now

T3P3Tony

@nuroo said in 25.95mm/rev, min -33% max 162% over 9615.6mm:

Well ok. I need to make an extension and try again. I wish this requirement was noted to customer on product page or wiki. I'll report back.

I am glad you got it working and thanks for the feedback, i have updated the installation section of the documentation to note the requirement: https://duet3d.dozuki.com/Wiki/Duet3dFilamentMonitor_RotatingMagnetVersion#Section_Installation

@themelle i think asking filastruder to swap it will be a good test to see if this is a board that passed testing but is marginal in your setup.

themelle

@t3p3tony Thanks for your honest feedback.
May I direct Filastruder support to your post confirming a warranty swap?

T3P3Tony

@themelle yes

themelle

@t3p3tony Thank you!

nuroo

Well I thought I was good to go with correct setup.
2 small 1 hour long prints no false positives. 11hour print - false trigger to little movement, about 5 hours in. Had to disable sensor so I didn't have to sit there having to keep resuming print.

nuroo

Are there any other troubleshooting measures I could try?

themelle

@nuroo I got my sensor swapped as it seemed to be faulty. Since then I have been getting much better values - but I am running 70-130% allowed movement to compensate for spike values.
Maybe you also should loosen the grip a little bit

nuroo

@themelle good to hear they swapped yours out. Figured I'd give the sensor another look.

I love that you can gather data while not actually having the printer stop.
Im current running:
M591 D0 P3 C3 S0 R85:115 L25.12 E3.0 (which worked on shorter prints)

S0 - disable filament monitoring.

So with my current longer print I did:
M591 D0
Duet3D rotating magnet filament monitor v2 on input 3, disabled, sensitivity 25.12mm/rev, allow 85% to 115%, check every 3.0mm, current pos 210.9, measured sensitivity 24.33mm/rev, min 1% max 110% over 6134.2mm

The high side range seems ok. Max I got is 110% but room for error still with 115%.
That min 1% is strange?? It would have surely triggered a pause had filament monitoring been enabled. But I have moved the sensor location as suggested. It reminds me of the min -33% I got starting this thread.

How do I over margin min 1% or min -33% so that these results are within good range?
For min 1% do i enter 0%? ie (M591 D0 P3 C3 S1 R0:115)??
For min -33% do i enter -40%? (M591 D0 P3 C3 S1 R-40:115)??
Can your algorithm handle such low values?
Do these low values still point to an issue with my setup?