Bitsfrombytes extrusion barrel experiences

I have been trying to get my extruder working. Here are my experiences

1st, I blew a transistor in the PWM, God knows why, I was operating at 12V.

2nd, luckily there is three channels and only two reserved so I jsut used the spare channel. I found the barrel could not be heated up enough to melt HDPE. Element coil resistaance was almost exactly 8 ohms. So I conclude the barrel cools too quickly. Maybe we should heat insulate it somehow, any ideas? Anyway, I just ran the PWM module off a variable voltage supply. At 17V I was able to melt HDPE. My thermistor said that was about 145 degrees.

3rd Genrally the HDPE starts oozing at a lower temperature, but the extruder can't put enough pressure on the filament to squeeze the HDPE out the nozzle. When I get to about 150 degrees the HDPE is extrudes with ease, but it is very nearly smoking at that stage. Is that smoke dangerous? It seems such a fine line between whether the HDPE is burning and the temperature required for it to be extruded easily.


Anyway I am quite looking forward to getting this extruder up and running. My axis work and so I am just about ready to reprap!


Tom

Hi Thomas,

HDPE does not smoke at 150C. I extrude it at 240C with no sign of any smoke or discolouration. It melts at about 130C, the lowest temperature I can make it extrude reliably at is about 160C. To get it to bond to the previous layer on large objects that have cooled requires about 240C but you can get away with less for small objects.

I use 12V across an 8R heater which is only 18W. My PWM operates at about 70%. You are using about 15V (Darlingtons loose about 2V at 2A) which gives 28W so if you only get 150C the BfB heater must be a lot less efficient. Most people use fiber glass loft insulation around the heater if they need insulation.

---

[www.hydraraptor.blogspot.com]

1- Use heatsinks on the pwm transistors - they get really hot

2- Maybe something with your temp-sensor is wrong, i had no problems heating up the BfB barrel up to 230 C

3- I'm not sure, but if this is PTFE-smoke then it is dangerous! PTFE is stable up to 250-300 C --> maybe (re-)calibrate your sensor manually --> [reprap.org]

And last but not least - the pressure problem
I tried the following things - but it didn't work:

- Sharpening the screw
- Add a second washer to press against the screw
- Use other springs to get more pressure on the screw
- Make a new nozzle with bigger diameter

Next problem - the extruder leaks where PTFE attaches the nozzle because PTFE deforms above 200 C
Maybe this could be solved using PEEK but it is quite expensive....

Tobi,
You can solve the leaks with PTFE plumbers tape in the thread and a pipe clip. The picture belows shows HDPE being extruded at 240C through a 0.5mm nozzle at 8mm/s.



Edited 2 time(s). Last edit at 08/24/2008 05:11AM by nophead.

---

[www.hydraraptor.blogspot.com]

I know but it isn't that easy, the BfB Extruder design is different, the heater barrel just touches the PTFE, but the parts aren't screwed together so the HDPE always finds a way through the half-molten PTFE

Hmm, I can't see how that is going to work, even if the PTFE does not soften it's a lot of pressure to expect an interference fit to withstand without a seal.

---

[www.hydraraptor.blogspot.com]

Actually I had spare thread exposed on my bfb heater barrel where the PTFE met the barrel. I ground in a countersink with my rotary tool (no thread). It took ages but I could then screw the PTFE on even without threading, so the seal is pretty good.

I think you guys are right in that my sensor is not calibrated properly. I'll double check my headers in the Ardiuno firmware. I have no equipment to check my temperatures though, but I am confident that whatever temperatures my reprap software states I can manually recalibrate mentally. So I just needed an idea of whether yours smokes and stinks during a successful PTFE extrude session (clearly not).

Cheers for the help everyone.

Thomas Larkworthy Wrote:
-------------------------------------------------------
>
> I think you guys are right in that my sensor is
> not calibrated properly.
>

That was my guess when you talked about the HDPE "nearly smoking" at 150C earlier. It shouldn't be doing anything of the kind at 150C.

OK well temperature sensor is way off. I crossvalidated against water. The water imediatly sizzles when the barrel thinks its at 50 degrees. I havn't gone any lower. So that is a side issue to be sorted later.

I have dissambled my extruder toolhead. I am finding the HDPE wont feed into the barrel nicely even when the PTFE is removed. The barrel hole is wide enough when cool, but when hot the surface of the HDPE filament melts and it exudes sideways relative to the direction it should be feeding. I beleive that when the PTFE is there, this side extrusion stuff cools later on and wrecks everything, but apart from that I think it absorbs much of the pressure from the feeder, so you lose alot of pressure. I beleive that to be the real problem. My barrel is hot enough but the feeder energy is not well converted into extrusion out the nozzle. (I now get ooing out the side of the PTFE as mentioned above).

My attempt at a solution so far is to countersink the brass heater barrel hollow rod, so there no longer a 90 degree corner to snag the desending HDPE filament. I also bored a little with a 3.5 mm drill bit into the barrel so the counter sink closes from about 5mm hole to a 3.5 mm hole which later closes down to the desired 3mm hole. Unfortunatly during all this drilling my heater filament snapped off twice. Leaving me with almost nothing to make an electrical connection with. Its just about doable but I think I will do that in a couple of days time. If I have another heating filament accident I will have to build a brand new heater barrel. I don't really want to rebuild a part that is less than ideal though anyway.

Tom (I'll get back to everyone about this when I experiment a little bit more)

I have had the same trouble with HDPE and side extrusion I added a M5 washer above the PTFE and this then puts more pressure to closed the leak, the new heater barrel is longer which allows the PTFE to form over the end of it as Thomas stated and also would allow for a clip if required.

The heater barrel has a 3.2

---

Ian
[www.bitsfrombytes.com]

I have carried out some basic limited heat & feed tests - got leaks and the PTFE was visibly swelling.
I have made up a replacement with machinable ceramic but I havn't run it yet. I gotta build my repstrap bot next and experiment with my roller nut feeder.

Richard,

your post is a while ago but, did you manage to get successful extrusion in the end?
I recently finished the BfB v2.0.1 RR and, for quite some time, struggling on extrusion as well.

regards,
Johan

[1m93.blogspot.com]

Edited 1 time(s). Last edit at 03/24/2009 04:50PM by Johan van Oostrum.

I'm also having problems with the BfB extruder and ABS.
I drilled the PTFE and heater barrel to 3.5mm which fed my ABS much smoother: however, the screw still has trouble driving the filament. I can extrude slowly using extra hand force.

You might try lowering the tempature if you are using 240c which has been mentioned a lot here. In my case it works best at 232c (haven't built anything yet, just extruding a thread).


Nophead has an explanation here about the plug that can form at higher temperatures.

[hydraraptor.blogspot.com]

After drilling my BfB barrel to 3.5 mm I was able to examine the effect of the plug formation.

Disassembling the extruder after a 240c extrusion shows that the melted section flowed all the way up the barrel right into the PTFE - about 4mm into the PTFE the plug stopped. This meant pushing is against friction along the whole length of the barrel.

Lowering the temp *should* cause the melting to occur lower down.
I'd suggest loosening the drive screw so you can use hand force to extrude: then try in 10c increments from about 150c up. Using hand force alone you can often feel where the best point is, and then refine using the motor. Note that hotter is not always better.

One possible problem would be while the motor is paused, the heat will cause the filament to melt further up, so that when you restart the force required has gone up. With my extruder experiments, feeding continuously does feel easier than the starting off.

Nophead's suggestion that the hot section should be as short as possible certainly makes sense with my experiments.

Hi!

you've had the same problem we did... almost simultaneously... the bits from bytes extruder works very well! on paper... when you use it on the real world, it has some flaws... But they can be corrected!

About the extrusion strength: the current design only works for about 5 min... after that, doesn't allow to push the wire with enough strength... the way we found to correct that is to take the screw out, and with a hand saw make the threads deeper, around 1mm deeper. after that, get a triangular sander and sharpen the edge of the threads, so that they can carve well on the plastic wire. last, but by far not least, get a new ball bearing... you will simply use it to press the screw up, instead of the silly washer... that will only send you screw to hell... with the bearing, you manage to press the wire against the screw, and manage to carve it well, and no more problems...

About the heat, only with thermal insolation we manage to roughly get 200 degrees in the barrel (measured with an independent calibrated device)... So, time to upgrade again... longer barrel, with 2 independent heating wires, heated by pwm 1 and 3 (just hotwire the pwm3 signal to pwm1 in the arduino/sanguino). after that, 260 is easy to get, without insulation!

about the reflux, that is a little trickyer to solve... insulation or not, you need to put something very tight with the wire, so that the melted plastic doesn't come out. even so, the extrusion speed must be slow... one way which worked not for long was to make a hole in the PTFE barrel, and screw it to the heating barrel! but the contact was not perfect,a cavity was present between the screw and the barrel, which induced some plastic to stuck in there, and obstruct the exit of the wire... still working on that detail... I hope that tomorro it will be solved... the only true problem, is that when we have everything right, and do everything you're supposed to, it doesn't move when you press the final print....