Friday, December 19, 2008


Paste extruder motor test

I wanted to find out which motor lends itself best to a paste extruder design. I came up with some rough prototypes and pitted Fab@Home's non-captive stepper from Haydon (left) against our old favourite, the Solarbotic GM3 servo (right).

Both designs worked great, however, the non-captive stepper design was radically simpler with significantly less fiddle and faff (hats off to Evan from F@H for going with this principle). Unfortunately, this particular non-captive stepper (Haydon, 28000 series) is worth $100+, and with cost of the RepRap an ever-pressing issue, we are investigating cheaper non-captive alternatives. From the design point of view non-captive is the answer. Could we make the GM3 non-captive?

Adrian demonstrated a great hack to convert a captive stepper into a non-captive by boring a clearance hole into the shaft and leaving an internal thread on the last few millimeters:

Investigations continue...

Aside, if we can hack a non-captive onto the GM3 and somehow prevent the filament from rotating, that would make for a *damn* simple thermoplast extruder. Sprung pinch wheels below perhaps?
This is somewhat beyond the scope of this post, I know, but what *is* the difference between a captive and non-captive stepper motor?
A captive stepper motor has a fixed shaft attached that rotates. A non captive stepper motor has a hole through with rotating threads, and a threaded rod that is put through the middle, if I understood it right. So a non-captive stepper motor is ideal for threaded drives, I believe Forrest is using them for his 3D printer Tommelise for example.


Normal captive

I'm going to take the final drive out of a GM3, make a silicone mould for it (see, and cast a replacement with embedded M3 nuts...
Oh no I'm not...

Just taken it apart. The final drive includes a co-axial idler shaft on which one of the other pinions rotates, making running a thread down the middle virtually impossible...
"Aside, if we can hack a non-captive onto the GM3 and somehow prevent the filament from rotating, that would make for a *damn* simple thermoplast extruder."

How about:

A) Two counter-rotating, coaxial gears: one with an M3, one an adjustable friction clamp.


B) A single M3 nut, plus a scalpel blade mounted to cut a longitudinal groove
Adrian, et al,

For the non-captive mechanisms, what prevents the threaded shaft from rotating with the rotor?

I'd guess there'd be some sort of spline groove cut into the threaded shaft, with a stationary (mating) spline/key resisting rotation. However, I don't see any such in the photos. How do they prevent shaft rotation?

-- Larry
Tests show that the seal friction in the syringe is more than enough to prevent shaft rotation (nice bonus!) - and the bigger the syringe diameter, the better this gets.

Steve, like scalpel idea!
I have an solution for this problem. I made a solid model and put an animation on Youtube:

If the animation is not clear I can make a drawing. Simply put, the motor and the captive nut are separated by a tube that transmits the torque. Let me know what your think.
Syringe dispensers very cool. great for solder paste dispensing and possibly pick and place.
Geo, that's great! One day all mechanical explanations will be that slick ;-)

I like the solution... obviously not as compact as a direct non-captive shaft, but still it does the trick.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]