[MakerBot] Short Melt Zone on Extruder: No Problem

[Brent Crosby]

(Note: This is a wrap-up of these two subjects:
http://groups.google.com/group/makerbot/browse_thread/thread/eb1a5c66b7aa00d2
http://groups.google.com/group/makerbot/browse_thread/thread/6b92930480adb08b
They should all be one thread--I really do not understand why they got
split, and are now both are buried, even though there are new posts. I
re-posted under a different subject, including a new photo to try to
"reset" google groups.)

We got this extruder finished and it is printing better than anything
we had tried with the stock extruder.

Here are pics:
Thermistor placement: http://yfrog.com/06n58j
Thermistor thermal coupling: http://yfrog.com/6w8b2j
Entire extruder in operation: http://yfrog.com/5sa0nj
Super-short melt zone: http://yfrog.com/55jmrj
Additionally a bunch of build-in-progress pics are available in my
main yfrog account: http://bit.ly/d2yi9P

What have I learned from this? Here is the summary (keep in mind this
was all done with ABS, natural and black, and it is still early):

1) The melt zone can be very short ( http://yfrog.com/20r3qj and
http://yfrog.com/55jmrj ). The heated section in this design is really
small. The entire heated brass element is only 15mm or 0.6 inches
long, and of that only 10mm or 0.4 inches is in contact with the ABS.
There is NO REASON to have a long brass tube heated and clogging.

2) Having the heated zone very small gives the default heating drive
circuit and nichrome element plenty of heating authority. The light
blinks happily as it is printing, where in the original design the
heater seemed to be full on all the time. It comes to temperature
quickly and controls nicely. There is NO REASON to have a huge metal
washer thermally coupled to the heated element.

3) Pressing the low COE (ref http://www.engineeringtoolbox.com/linear-expansion-coefficients-d_95.html
) brass OVER the higher COE plastic (Torlon tested, I think there
would be similar success with PTFE and PEEK) seals and does not leak
over temperature. Hotter = tighter if the low COE material is on the
outside. The original design puts the PTFE with its high COE on the
outside, and the hotter it gets the looser it gets (and not in a good
way like on spring break), plus the PTFE gets soft at high temps.

4) To lower the active thermal mass, I used the Torlon "cap" (
http://yfrog.com/0anruj ) to cover the heated brass element, and then
also used it to thermally insulate the steel washer from the brass. I
expected the steel washer to be fairly cool, but it was not nearly as
cool as I would have liked. This means that Torlon is more thermally
conductive (poorer insulator) than I had expected. I will need to re-
think insulating of the heated element.

5) As shown in this photo ( http://yfrog.com/55jmrj ), the melt zone
is very short. Certainly within 1/2 inch of where the brass is pressed
onto the Torlon. This means that the whole assembly should be able to
be reduced in size greatly. This prototype was made to fit in a stock
MakerBot, so the top Torlon is long. I think a much shorter length can
work, and it would reduce the "compression spring" bit between the
plug and the extruder pinch wheel. Shortening the length should make
the responsiveness between the motor commands and actual extrusion
more coupled.

6) I did not make a dedicated place for the thermistor. In the next
version I will extend the extruder so there is a hole to put the
thermistor into.

7) Not shown on the drawings, but I made several counter-bores on the
back side of the nozzle to keep the 0.45mm bore as short as possible.
The idea is to reduce the friction of pushing the plastic through a
long skinny hole. Trying to keep that "to the 4th" term small. (ref:
http://hydraraptor.blogspot.com/2009/03/rheology.html )

8) I think the small flat area around the extrusion hole is too small.
It looks like skeinforge uses this area to smooth the top layers and
mine is so small it leaves grooves. (one of Nophead's designs
http://hydraraptor.blogspot.com/2009/11/no-compromise-extruder.html
has a very large flat area)

9) I think that an important design element is that the design "steps
out" from the guide hole in the Torlon to the heated brass area. This
essentially takes nophead's taper and expands it to 180 degrees. In
the next design I may try a slight taper, but I am not sure it is
needed at all.

Coincidently, Richard is working on a very promising design:
http://www.makergear.com/blogs/frontpage/1636862-peek-ptfe-hybrid-ins...
which looks very like it might work well.

Intentionally, Eberhard is working on the "Complextruder":
http://pleasantsoftware.com/developer/3d/2010/05/15/complextruder/
which also looks very promising.

My plans for the next revision:

1) Reduce complexity
2) Try out lower-cost materials PEEK or maybe even PTFE rather than
the super expensive Torlon
3) Reduce the size (this will require new or modified "dinos")
4) Accommodate the thermistor (which will likely increase the size,
meh)
5) Re-think the insulation around the heated element

The end goal is to have a design that is simple to manufacture and
assemble, has very few components and those are as low-cost as
possible, works every time, and never, ever sticks. Now that I can
actually create objects in MakerBot #1024 (aka MiseryBot), it may be a
while till I get back to trying to improve on this design.

As usual, comments are welcome.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

[Brent Crosby]

Here is a video of the extruder in action:

http://www.youtube.com/watch?v=MnMNGLq8euE

The biggest failing so far is the lack of a way to securely mount the
thermsitor. I have some 1mm thermistors on order from DigiKey, so the
next version will fix that.