Magnetic Filament Monitor Sensitivity Increasing

Eduma

I just installed a magnetic filament monitor and as my print runs the sensitivity continues to increase.

Duet3D magnetic filament monitor on input 3, disabled, sensitivity 25.00mm/rev, allow 90% to 110%, check every 3.0mm, version 3, mag 130 agc 84, measured sensitivity 25.05mm/rev, min 96% max 103% over 108.9mm

As the print continues the sensitivity continues to grow
Duet3D magnetic filament monitor on input 3, disabled, sensitivity 25.00mm/rev, allow 90% to 110%, check every 3.0mm, version 3, mag 131 agc 84, measured sensitivity 38.10mm/rev, min -2% max 180% over 510.5mm

Duet3D magnetic filament monitor on input 3, disabled, sensitivity 25.00mm/rev, allow 90% to 110%, check every 3.0mm, version 3, mag 132 agc 84, measured sensitivity 39.29mm/rev, min -2% max 186% over 636.1mm

Eduma

Should mention I'm using a duet Wifi with firmware 2.05.1 and filament monitor is V1.7a

Eduma

I ran some additional prints this morning and I continue to see measured sensitivities increasing throughout the print with unusable percentages. Based on other posts my mag and agc values seem to be within range, although I haven't found any specifics on how these are used.

Has anyone run into this and found a soltuion?

T3P3Tony

@Eduma something looks to be slipping. is the housing getting hot? the AGC being the same means the magnet is being read well so i suspect the filament slipping past the hobb. Potentially an idler tension issue or the filament path is bent and forcing the filament in an arc away from the hobb.

fractalengineer

@T3P3Tony my diy pulse generating sensor is varying quite a bit too; I'm not sure if it could be due to the retractions causing confusion or a lag between expected/measured filament progression...

For me it's mostly just fluctuating a lot. I have to set a pretty large measurement length and tolerance for it not to trip

Eduma

@T3P3Tony I'm not feeling temperature difference. Once my current print is complete I'll pull the sensor apart and ensure everything is correct.

Am I correct that there is not way to see filament monitor readings without running a print?

@fractalengineer I saw your post yesterday and the design certainly caught my eye. If the filament monitor doesn't pan out for me I was thinking about doing something similar with an E3D or bondtech hob. Do you think your seeing slip on your printed wheel? Or do you think print variation on your encoder disk could cause distance between pulses to inconsistent?

fractalengineer

@Eduma Excellent point; yeah I made that encoder wheel to be very sensitive; there are 28 slots per revolution. I can imagine the slightest irregularities in those may have a larger impact at this scale.

I'll test a less sensitive wheel at 16 or even 8 slots per rev. Thanks for the insight

I am not noticing any slip though; the grip is pretty good especially considering there is little to no resistance thanks to the ball bearings.

The only issues I might see with that friction wheel/ptfe tube design is 1-ptfe wear over time 2-wheel slip if greasy filament (and even then; there's no resistance to the shaft rotation anyway)

I've already been thinking of a pro version using a hob gear and spring loaded idler for that matter; might be a more reliable long term approach.

Eduma

@T3P3Tony sorry for the delay in getting back to you but i was able to confirm that this issue is filament slip. If my guide to at the entry of the filament monitor is at too much of angle I see slip. If I maintain a straight feed preliminary testing looks like i see accurate readings.