I’ve done some work on filament sensors. The electronics and software part is working at least, but the mechanics is not yet great. Better solutions welcome : )
Packaging the assembly for a direct drive/direct mount extruder is going to be challenging but for a Bowden it’s not too far from being able to implement.
@dstevens_lv I guess the rotary encoder could be a replacement for the radial bearing pushing against the filament in a direct drive extruder.
For the encoders I’ve bought 111688538683 of flE-Bay to play with. How about soft silicone tube pushed over the encoder shaft with an adjustable spring tension 608 bearing idler like a wade’s extruder. For direct drive, a Bowden tube secured to the sensor mechanism and to the extruder should work. I would love to have a woking system like this for my replicape.
the problem with measuring at the spool is that as the spool is used, it rotates more quickly for the same ammount of filament.
how about this? http://hackaday.com/2015/06/16/prevent-failed-prints-with-a-filament-speed-sensor/
also, i’ve seen one tiny sensor wich i couldn’t find anymore. It was to samll (maybe 3mm dia) rubber rollers, and one had a magnet glued. Put that inside a little box wich hold them appart at the correct distance, trow in a hall sensor, and you have a jam sensor. i’ll try to find it, since it was the simplest and most compact setup i’ve ever seen.
Is there anything in the future for encoders on the XY steppers? Like maybe a daughter board which if a skipped step has occured, then pass it to redeem to correct it?
@tizzledawg BeagleBone has support for rotary encoders in hardware (counting steps) which I think might be necessary in order to not spend too much resources on the encoders. Or one could go for stepper drivers with support in hardware.