After my previous failings at using using solder to cap an extruder barrels open end I have now turned to brazing. The solder method may have worked but it turns out I don’t have any high temperature solder and it was just normal solder that I have been using. I obtained some silver alloy brazing rod off a friend and after shaping the extruder barrel, set about capping one end.
Brazing a flat plate onto the end of the extruder barrel.
The metal I used to cap the end is just a section of pipe that has been cut open and hammed flat. I found it quite hard to get the pipe up to temperature with just the propane burner. A MAPP gas torch would be more suitable for this job. To solve the problem I did the brazing on a cinder block which helped insulate the tube and allowed it to get hot enough for the silver brazing rod to flow.
Once sealed, I cut away the excess metal and cleaned it up with a bench belt sander. After drilling a 2mm hole in the end I wrapped it in polyimide tape I fitted it with fresh nichrome wire and added the thermistor.
The extruder assembly half complete.
Its a little hard to see, but in the image above the left hand side is the brazed end. THe middle section is where the two sections meet. Just enough pressure is applied to stop any plastic seeping out. The auger drill bit only extends in as far as where the two pipes meet. The aim is that after use, everything cools down but the feed pipe will still contain none melted granules, and so can be removed if needed.
After adding roofing insulation and reattaching in to its holder on the drill press it was ready to go.
The extruder finished and ready for testing.
A short time later it was warmed up to some temperature around the 180 degree C mark and began to extrude a beautiful filament of HDPE. However I was still having to do the work of turning the auger drill bit manually. So now the next point to focus on is gearing down the DC motor to drive it. Then some time after that make an automated spindle to collect the filament after its extruded.
Once the mechanical side of things is made reliable enough I can then start to experiment with different additives to make my own custom ‘reprap HDPE’ filament.
Capolight Electronics Projects. 
Great work. Good to seem somebody making progress on this.
Wow, nice work! The filament looks pretty good already. Maybe it can become even more consistent if you put feedback on the motor? (if you want constant speed instead of (more ore less) constant force) or perhaps a more geared down motor? Maybe a higher or lower speed also does the trick.
It may require changes in the thickness and smoothness of the die, though, which may be harder to experiment with because it would require re-assembly.
What is the diameter of the filament?
Just to loosen your constraints a bit, it might be worth it to aim for any diameter smaller than or equal to 3mm that you can produce consistently and then adapt your RepRap extruder to that. This would give you some more design freedom. A smaller diameter extruder feed is probably a bit easier. Though differences in diameter would (proportionally) start to matter more.
Just thinking out loud… 🙂
What is the thinking behind using a smaller DC motor instead of the AC one inside the drill press?
Yes the speed will set the pressure which will affect the die swell and hence the filament diameter. It needs to be very consistent, so I think measuring it with a camera and controlling the motor with a feedback loop will be needed in the end. Or it could be measured as it enters the RepRap machine.
Smaller filament does need less force to extrude but you might get problems with the softer plastics like HDPE with it kinking on its way into the extruder as the strength drops with the fourth power of diameter but the force drops with the square.
Also smaller filament will need more force to when you are making it I think.
I also wonder whether it would be easier to control the diameter using some kind of pinch wheel extruder that would be mounted beyond the die and pull the filament once it has re-frozen. Pulling faster than the extrude rate will make the filament thinner and pulling less fast will allow the filament to be thicker. Feedback would still be required using a webcam as nophead suggests.
Thanks for the feedback, they are all very interesting ideas.
The level of die swell is indeed huge. It appears to be about twice that of the 2mm die diameter. I will measure the filament once I get the DC motor sorted. I’m crossing my fingers and hoping that due to the size of the melt chamber (around 5.6cm^3) any variations in the DC motor RPM will be smoothed out leading to a uniform filament diameter. If not, then I will start looking into the more complex feedback options that have been suggested.
@Giles: As far as I can tell the speed of the AC motor already attached to the drill press is fixed according to the AC frequency (50Hz) and the number of its poles. So even if I reduced the AC voltage with a variac the RPM would still remain the same but with reduced torque. It would also be too difficult to reduce its speed through gearing alone due to its size and position. Hence I’m trying to use the small DC motor which can be varied by changing the supplied power. This job would be made much simpler if I already had a mendel!
I believe its possible to control the speed of an induction motor as can be found in your drill but until you mentioned it I didn’t even know that the speed of an AC motor would be tied to its supply frequency, so chances are I am wrong. Seems a shame not to utilize it as it looks pretty beefy, but I would imagine that a DC is as you say much easier to control.
Ah I have just noticed on the page I linked that you would need a VFD to control it, I bet they are expensive!
I have had a rethink of the problem and maybe I will try and use the AC motor after all. I Just did some quick testing and even under a very heavy load the motor still maintains a near constant RPM as expected. So although it might be difficult to gear down the pay off would be a very constant RPM,and hopefully a very consistent diameter filament.
So what first appeared to be a drawback of the drill press may turn out to be a great benefit.
Thanks again for the input from all.
What is the amount of real power you need? You are gearing down a lot, but why not go for a motor that has way more torque in the very low RPM’s? You already have such a motor, as it is an ordinary stepper motor. You probably still need to gear it down, but not with such large values. You need a good stepper driver on it that can do current limiting and larger is better if you want more torque from the motor.
The design above is a little out dated. I’m currently in the process of building an extruder like the one linked below.
http://objects.reprap.org/wiki/Web4Deb_extruder
This uses a different auger drill bit with a reduced pitch and so will work with a slightly higher RPM. It also uses a windscreen wiper motor which has more than enough Torque and is already geared down by a large amount.
I think a good method would be to move the filament horizontally away from the nozzle
(the first vew cm may be vertical) and then guide it directly through an upper and lower light switch. Based on that the (quite near?) spool could be driven with a stepper motor.
Almost no tension and almost no tension variation should be possible.
Airbubbles could effect the diameter. Did you encounter any in your experiments?
btw what are the specs of the auger drill bit you’ve used in your succesfull extrusions?