Laser filament monitor
-
If you want to calibrate volumetric extrusion, then I agree, in principle you need to measure filament diameter too. But standard practice is to calibrate linear extrusion. If the filament is of high quality, then the diameter will be tightly controlled so it will also give you sufficiently accurate volumetric extrusion too. Nevertheless, different filaments swell by different amounts after extrusion, so even accurate volumetric extrusion doesn't guarantee that you won't need to adjust the extrusion factor.
-
Sure measuring die-swell would be hard. The only idea that might be feasible for true auto calibration of the extruder would be to extrude onto a glass section at the side of the bed, from a precise z-height, and measure using a CCD scanner sensor, or something similar the width of filament extruded then change the steps/mm and extrude again. In theory it would account for all necessary factors.
-
IMO, even with a perfectly calibrated extruder there may still be a need to alter the extrusion multiplier to compensate for say models which have a lot of short slow moves vs models which have predominantly long fast moves. Another factor which is related to that is temperature and by that I mean the temperature throughout the filament from outside edge to the core. The filament itself is a very poor thermal conductor so when printing at high(ish) speed, there is less time in the melt chamber so the inner core may be significantly cooler than the outer surface. So even measuring die swell at one fixed speed/temperature combination might not be a practical solution.
-
The schematic for Prusa's laser sensor has been published in his github. The part number on the print is PAT9125SEL
https://github.com/prusa3d/PRUSA_Laser_filament_sensor/blob/master/rev.02/laser-sensor.pdf
Slightly off topic but he has also published the power panic circuit. From the picture, I think it may need a slot routed in the pcb between L and N to meet UL creep distance.
-
Thanks, that confirms that the sensor we chose is the same one.
-
The schematic for Prusa's laser sensor has been published in his github. The part number on the print is PAT9125SEL
https://github.com/prusa3d/PRUSA_Laser_filament_sensor/blob/master/rev.02/laser-sensor.pdf
Slightly off topic but he has also published the power panic circuit. From the picture, I think it may need a slot routed in the pcb between L and N to meet UL creep distance.
Can confirm slot! Just received MK3.
-
Happy to hear that the laser sensor is on the way ;).
Would be happy to test it.
Had problems lately with filamentgrinding.
That would be a thing of the past then. -
For a while I've been toying with the idea of a filament thickness detector.
I often do 20 hour prints on 2.5kg spools, so the thickness of filament can vary a bit within a single print.
Supplier states 1.75mm +/- 0.05mm but I have found it to be closer to 0.08mmThe idea I've had so far is to use a digital tire gauge caliper, it's inexpensive, fairly accurate, and has a built-in serial output.
Reading the output with a small microcontroller (ESP8266) I can then adjust the flowrate.
For actual filament movement; A 16 bit resolution rotary encoder with a extruder gear on it.
I know these solutions are more mechanical than the current suggestions, but maybe they can inspire some new thoughts.
-
Does the section of your filament vary that much?
You need to measure the diameter in both directions, as the filament can be oval, and the section remaining the same…
-
@fma:
Does the section of your filament vary that much?
You need to measure the diameter in both directions, as the filament can be oval, and the section remaining the same…
Over a whole 2.5kg spool, it can vary quite a lot.
Yes, checking for OVALness will be very difficult, I have not thought of a solution to it yet
-
There are some very expensive laser-based diameter and roundness testing devices but they're $1500… Its whether there is a way to do it for a lot less, rotating the laser filament monitor that Duet3D are playing with around the filament?
-
…The idea I've had so far is to use a digital tire gauge caliper, it's inexpensive, fairly accurate, and has a built-in serial output.
http://www.ebay.co.uk/itm/NEW-Digital-Tyre-Tire-Tread-Brakes-Shoe-Pad-Wear-Depth-Gauge-Caliper-0-25mm-/321762696307
Reading the output with a small microcontroller (ESP8266) I can then adjust the flowrate.Nice idea, but would you press the gauge against the filament using a spring or something? I guess when a whole spool of filament passed this plastic toy, there would be a huge abraded hole in it.
-
I've just had notification that the prototype PCBs for the laser filament sensor are on their way to me from HK! I only ordered them last Sunday.
-
That's good news David
-
Nice to hear.
-
…The idea I've had so far is to use a digital tire gauge caliper, it's inexpensive, fairly accurate, and has a built-in serial output.
http://www.ebay.co.uk/itm/NEW-Digital-Tyre-Tire-Tread-Brakes-Shoe-Pad-Wear-Depth-Gauge-Caliper-0-25mm-/321762696307
Reading the output with a small microcontroller (ESP8266) I can then adjust the flowrate.Nice idea, but would you press the gauge against the filament using a spring or something? I guess when a whole spool of filament passed this plastic toy, there would be a huge abraded hole in it.
Idea was to have a small pulley attached to the end of the guage, (similar to pulleys found in extruders) so there will be no friction.
-
I've just had notification that the prototype PCBs for the laser filament sensor are on their way to me from HK! I only ordered them last Sunday.
Great news!
Is there still space on the Beta List? -
Sorry to jump in but what is the advantage of this compared to a simple micro switch?
-
Sorry to jump in but what is the advantage of this compared to a simple micro switch?
These monitors track the amount of filament passing through the system. You can use that to tell if the extruder drive is slipping, not just if there is filament present.
David has raised the possibility of auto-calibration of e-steps using the system, which would be very neat if it works.
-
Ah I see, very clever!