Hi, I am thinking to build a simple hotend clogging sensor. My intention is to use it on delta printer but could be used for any other printer design. I would monitor flow of the string of PLA (or any other material) on the extruder. I was thinking to use optocoupler and measure if the string is going through or if it is clogged. In case that clogge whould be detected for some amount of time than printing would be stopped. Then the user would have possibility to carry on with the printing when problem is solved and the print would be saved. It would also require some modification of the repetier firmware but I think it should not be so hard.

What do you think about this idea? Or do you have some experience with similar stuff already? Would be somebody interested to work together on it?

Thanks for your replyes.

As it happens, at Duet3D we've been working on this for a little while and we have a few prototypes, based on measuring the rotation of a magnet that is turned by the filament. The latest release of RepRapFirmware supports this type of filament sensor, using gcode M591 to set the parameters (see [reprap.org]). We won't have it in production for several more weeks.

This idea isn't new; there is also the Tunell filament sensor, which is based on a similar principle but uses a mechanical rotary encoder instead.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I understand that a lot of people have problems with extruders failing to feed filament. I went though about a year of that before I finally understood the problem and how to solve it. I have found 5 main factors that stop filament from feeding.

1) proper extruder pinch roller adjustment- too light and the extruder chews divots into the filament and it stops feeding
2) high push force from the extruder- geared extruders multiply motor torque and will push the filament even when things get "less than ideal" in the hot end
3) installing a decent hot-end- one that has a proper heat break and a cooling fan
4) proper handling of filament spools to prevent knots- don't let the end of the filament loose, and don't use a spool holder that lays the spool in its side
5) a spool holder design that prevents filament from springing over the flanges of the spool

They are others, such as malfunctioning controllers, broken filament, bad g-code, wiring problems, etc., but these are relatively rare and all will usually create many other problems before or even as they cause filament to stop moving.

Most people who would install such a detector are doing so because their printer has stopped feeding filament many times. But if they already know they have filament feeding problems, how is the detector going to help? How many times does your printer have to stop feeding filament before you decide to fix the problem instead of just getting notified that there is a problem... again?

Edited 1 time(s). Last edit at 08/17/2017 09:47AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Thanks for your replyes. I agree with you that it is always better to correct problems which lead to the filament clogging rather then using the sensors. Anyway I think that the sensor can help a lot during the time when you try to deal with the filament clogging. In case that it is not an expensive part of the printr I think the user can benefit from it.

I would like to make a sensor based on the mechanical rotary part with the optocoupler.

You mean something like this?
http://www.3ders.org/articles/20170202-38-filament-roller-uses-sensor-prevent-3d-printer-filament-stoppages.html
or
http://hackaday.com/2015/06/16/prevent-failed-prints-with-a-filament-speed-sensor/
or
https://www.youtube.com/watch?v=IeAWIFibcoc

Yes exactely. The hackday link you provided is very close to my idea. As in the discussion there, it is also mentioned the same princip as in the wheel mouse. That is exactely what I was thinking to build. I would like to make it as simple as possible.

What about to build it with following optical sensors and use two of them to also get the direction of filament:

Lm393 Beam Photoelectric Sensors


Then it could be build for less then 5$.

Edited 1 time(s). Last edit at 08/17/2017 05:47PM by boylucky.

As far as I understand the Hackaday-Link, it is something simple like this:
Rotary Encoder
with an arduino nano giving alarm. If you program the nano to set an output pin and connect it to an alarm port of your printer-controller (e.g.: out of filament) it's easy.

if you are familiar with the printer-firmware, you could even directly connect the encoder to your printer-controller and include the stop code there. (maybe some firmware already has this included?)

The smaller you make the wheel, the faster the pulses are generated, the faster you get to know if the filament is stuck, or ran out.
If you want even faster detection, you could use a small and a big gear to increase rotation-speed of the encoder.
This encoders already give you direction information.

So I do not see any need for new design - everything already here :-)
(My last clogging was over a year ago - better solve the problems instead of getting to know when it ocures - but just my opinnion)

Edited 1 time(s). Last edit at 08/18/2017 03:38AM by Dancer.

Quote
Dancer
As far as I understand the Hackaday-Link, it is something simple like this:
Rotary Encoder
with an arduino nano giving alarm. If you program the nano to set an output pin and connect it to an alarm port of your printer-controller (e.g.: out of filament) it's easy.

if you are familiar with the printer-firmware, you could even directly connect the encoder to your printer-controller and include the stop code there. (maybe some firmware already has this included?)

The smaller you make the wheel, the faster the pulses are generated, the faster you get to know if the filament is stuck, or ran out.
If you want even faster detection, you could use a small and a big gear to increase rotation-speed of the encoder.
This encoders already give you direction information.

So I do not see any need for new design - everything already here :-)
(My last clogging was over a year ago - better solve the problems instead of getting to know when it ocures - but just my opinnion)

As a hacking project that's OK, but it's a poor choice for serious use. Those mechanical rotary encoders have contacts that are typically rated to 30000 cycles. It's not clear from the datasheets whether a cycle is a contact closing/opening cycle or a complete revolution of the spindle; but even if we assume it's a complete revolution, 30000 revs will be reached after about 1km of filament. That's just three 1kg spools of 1.75mm PLA.

For our filament monitor, we chose a magnetic encoder that doesn't wear out. It returns the angle of the shaft to the nearest 1.4 degrees, so no need for gearing.

Edited 3 time(s). Last edit at 08/18/2017 06:27AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].