The new extruder as described in a previous post was connected to a new x-carriage using L-shaped aluminium strips and m3 bolts. The x-carriage is the same as the x-carriage made for the makerbot MK4 extruder. I decided to keep the old carriage with MK4 extruder untouched to be able to switch back an forth between extruders easily by just replacing the whole x-carriage. Extruders not being the most reliable part of the 3D printers...
I used a new nozzle and PTFE insulator from mendel-parts.com. Though this nozzle and insulator are meant to be used as in standard mendel fashion (glued at the insulator), I decided to make the construction a bit stronger. The filament is pushed down with quite a bit of force into the nozzle and when the insulator is getting warm it may fail. To prevent the downward force to be taken up by the insulator most modern extruders have some kind of support connecting the nozzle to the rest of the extruder above the insulator. I used a thinned M6 nut on the nozzle and stip of alumium attached by 2 long m3 bolts to protect the insulator.
The aluminium strip is not ideal as it conducts heat too well. The contact area between the nut and the aluminium is kept as small as possible and there is a bit of kapton tape in between; the nozzle does not touch the aluminium. A new design for the mendel extruder which can be found in the wiki uses a piece of PEEK, but I did not have that available. Below the aluminium strip a short part of the nozzle is left to put the nichrome wire around and to attach a thermistor. All is electrically isolated with kapton tape. As I did not have proper heat insulation around the hot end of the extruder like in the MK4 extruder and the metal strip is leaking some heat, I see it takes a bit longer for the extruder to warm up properly and the temperature seems to vary more than it used to. This does not seem to hurt the operation and the extruder is working fine. A hose clamp around the insulator is an additional bit of protection, not sure if that is really needed, but it probably won't hurt.
The stepper motor is controlled by a TCA3727 integrated circuit via a separate arduino board. The TCA3727 is quite old and better alternatives can be found, but is works fine. A quick-and-dirty program was written to make the motor turn at a constant speed using full steps. The circuit allows for quarter and eighth step but I did not think that would be necessary; that was a wrong, because while the stepper motor turns at the right speed, it vibrates horribly. Especially since the x-carriage is pulled down by gravity only (no ball bearings under the horizontal metal rods), one full step lifts the carriage an invisible fraction everytime before it falls back. When PLA is extruded in the air and left to flow down, it is visible that the extrusion in not constant but it is a bit thicker for each step. Some more software should fix this later.
The extrusion motor control is not integrated with the rest of the electronics or the rest of the software. I manually start extrusion and the printer software is still sending signals to the MK4 gearmotor, though that is not connected anymore. This works surprisingly well. Of course there are ugly plastic lines from head movements where the extrusion should be off but isn't here an there, but depending on the object to be printed these could be tolerated. Of course this will be fixed later, having precise control over the extruder was one of the reasons for building a stepper controlled extruder in the first place.
So far I used the MK4 extruder only with ABS plastic. The pulley which is used to get a grip on the plastic filament has "teeth" which are quite wide. The ABS was clearly stretched and clear marks of the pulley are visible on the filament. Since PLA plastic is much harder (when cold) than ABS. I did not even try it, though some people did get it to work. The new extruder is using many more, but much smaller "teeth" to get grip on the filament. PLA has the advantage of warping much less allowing larger objects to be made. The largest object made so far with ABS is the big gear on the new extruder. It too is warped a bit, but it is just "flat enough" to work.
Summarizing things that can be improved still on the extruder:
- extruder connected with 2 aluminium strips on each side, a better design should allow that to be just one part
- motor control from 3D printing software. I currently use replicatorG but I have not found how that can be integrated with a stepper controller yet, maybe it is time to switch over to the FiveD reprap software which has inbuild extruder stepper motor control.
- Wiring should be fixed.
- 8th stepping to be implemented.
The extruder works though and printed without issues for more than 2.5 hours in a row on a single object.
No comments:
Post a Comment