Experimenting with my DIY Y-switch dual extrusion setup.

Experimenting with my DIY Y-switch dual extrusion setup.

I think i’ve nailed down the G code retraction sequence for the PLA I’m testing with - nice pointed retraction ends without any “whiskers”.

The key (for me) is a F1000 retraction of 10mm, then a 2mm reinsert at F250, then the final retract past the Y split at F7000.

Reinsertion is F7000 to the 10mm before the melt zone, then F250 for the final 10mm into the melt zone.

I’ve been printing with these settings for a few hours, and have yet to jam.

DIY is the way to go!

Still some tuning needed on the retraction - got a whisker jam about 10 minutes after that post!missing/deleted image from Google+

Awesome stuff! I wanna get a 2 in one setup going myself. Love all this progress lately with it. Hope to see more posts from ya about it!

Really interested in this, please post the finished gcode and steps for setup. :slight_smile:

Awesome. Keep it going!

Maybe try the E3D ram purge technique?

@Ryan_Carlyle that’s a bit different because now it requires a “dump bucket” for the filament that comes out during the ram. A possibility might be to make the prime tower have a hole in the middle of it, and the Extruder can spit into that. But now we’re into atypical software implementations into the various slicers.

@Ryan_Carlyle - the whisker jam wasn’t due to a jam of the whisker into the melt zone, but the whisker extended past the Y after the whiskered filament was retracted, preventing the second filament from reaching past the Y and into the extruder.

I don’t think that ‘ram purge’ will help in this case. Just need to work on speed, timing, and movement to prevent the whiskers, or just pulling WAY far past the Y on retraction.

I’m considering making a glass window version of the 2-in-1-out adapter so that I can observe the filament ends in-situ.

2-colored cat seems to be very good at causing jams - had to clear 3 so far.

Very good test article, as it has color change on every layer, and a lot of retractions for each color.

(0.8mm nozzle at 0.25mm layer height)
missing/deleted image from Google+

I know that the jamming is caused from whiskers but would adding a little filament oiler to each filament path help a little? Just spitballing

@Jason_McMullan look at how Makerware (preferably 2.4) does dualstrusion purge walls, it builds a bucket for purging as part of the wiper wall. Admittedly not a straightforward thing to use with non-Sailfish machines.

My thinking on the ram purge is that it could extrude out the semi-molten zone that draws down to a string as you retract. Maybe. I don’t know.

Promising! I wonder if you could use the cooling tower as the dump vessel for the ram purge. For repeatability, a purge and ripe would be nice.

with my DIY Y-switch Enjoy the weekend​:smiley_cat::cat2::cat::smiley_cat::cat2::cat:

@Ryan_Carlyle it’s a thought, but a bigger change than what I’m banging around with right now.

I’m playing around with doing a slow retraction to “draw the tip” (like around F100 for 4mm) then retract very quickly to attempt to prevent whiskering.

Running tests now…

Oh, and Cura’s wipe and prime tower does a toolchange, even if there is no dual extrusion on a layer. Annoying.

@Griffin_Paquette so far I don’t have any excessive path friction - I can test cold (no hotend attached) the filament switch for as long as I like.

That said, fabricating the Y guide in PETG should make for even less friction.

One more thought - for a robust system, one would need filament presence detectors at the Y input sides

Then filament changeover could be completely automated by the firmware.

Are you using geared steppers?

@Ross_Hendrickson no, not geared - just very overpowered NEMA 17s with a simple hobbled bolt.

The friction of the feed path isn’t much higher than my regular Bowden setup.

Cool, just watched the Prometheus review by Tom and was noting the similarities to your design. Very cool nonetheless. If you can reliably get it working then that will be really neat.