UPDATE 01/31/18 - I keep getting comments on this so I assume people are still interested. However, it's a really poor design, and I'd suggest one of my other two designs:
which is a very precise direct displacement extruder like this one or
which is a pneumatic extruder so it works a little different
An update as I've had a chance to start using it now:
My strategy has been to calculate steps/mm just like regular filament (change the diameter of the filament to be the ID of the syringe).
You'll need to send "M302" to allow for cold extrusion, and the correct "M92 EXXXX" value to calibrate the extruder for this to work. I just add it to my start g-code in slic3r but you can just send it prior to printing.
My calculated steps/mm values (a second pair of eyes would help):
steps/rev microsteps / gear-reduction belt-pulley-teeth) / mm-per-tooth
(belt-pulley-teeth and mm-per-tooth are most likely different)
20016/1220/2.5 = 2133 steps/mm
In application, I've measured ~1600 gives 1mm of travel
Then, a filament diameter of ~14.5mm has yielded good results.
I'd love any input or corrections!
The idea is super simple, but I had a lot of issues with the latest universal paste extruder (the biggest of which was there was no easy way to mount it on an i3!)
Thanks to alexspeller for that fix (http://www.thingiverse.com/thing:217044) and for inspiring me to learn how to import/modify stl files with openscad. I knew a fair bit of openscad, but this is its most useful feature IMO.
I've attached all the files to build a universal paste extruder that can mount on an i3. Also, in the zip file, I have all of the original files, the openscad files that modify them, and the resulting files.
A breakdown of my changes:
- Body - added the i3 mount holes. I also want to add a little reinforcement and add a fillet to help feed the belt through the first time, but haven't had a chance (you'll know when you first assemble it.
- Idler - the roller bearing was off by about 2mm. I had to do the openscad modifications in two steps because it wouldnt execute all of them at once, no matter how I repaired the stl. I also added a guide that angles the belt away from the syringe since my belt was interfering with itself and the teeth would catch on each other.
- the motor gear's base was too large and it was scraping on the main body's block. I didn't want to open up the spot for the motor more, so I made the motor gear base smaller. I also made the ID for the stepper motor axis larger, you may need to tweak this to print right for you.
- Double gear - I also made the ID bigger for the m4 bolts
- Third Gear - I didn't like the way the bolt secured to the gear so I flipped it around and took out the set screw (on the gear itself). If you get 60mm hex head cap screws (mcmaster part number 91287A060) they will fit perfect and be the right length. any m4 nuts will do.
- Syringe top - good as it is
- Belt pulley - the original design said use a t5 10 tooth metal gear for the belt or print your own. I included both possibilities, T5 (10 tooth) and T2.5 (20 tooth) that both work well, fit, and wont change the calibration parameters either way.
That's it, put a bunch of work into it so I hope it helps someone. It was easy to make, the main body is the only large piece and with these tweaks assembly should be a breeze.