I MUST build a new extruder advice please.

This has been a lot of fun and a great experience but my extruder is a big failure prone pain. I know the original makerbot one is as well.

1st, I had no idea what I was doing when I bought my Geared Extruder printed parts made of PLA, this is really dumb, when the motor heats up the whole thing is very pliable and bends.

2nd, there are so many circumstances that seem to get the whole filament pipeline backed up, of course everything continues as if it were fine and things get worse. Has anyone figured out a way to sense this and shut things down?

lastly, and most damaging, is the pressure build up has finally pushed the hollowed out bolt out of the ptfe insulator, stripping the threads. I'd like to avoid using such a soft material if possible next time.

Any suggestions on the next etruder I should build (besides the experimental soft grip stuff I was talking about) Do they all have the same hot end / insulator issues? If I'm going to have to take thing thing apart and clean it regularly I'd at least like to have some parts that make it easier to do and last a while.

thanks in advance.

#1 - set your motor current more carefully next time. I've been running nothing but PLA extruders for years and have not melted one yet. A warm motor is fine, but if it feels hot, turn down the current.

As for hot ends, I'm looking for a better one as well, although both my PTFE hot ends are still working fine. One did fail two years ago when I had a thermocouple come loose, causing the hot end to overheat and melt it's way out of the PTFE. Good temperature control helps a lot, which also means that good solid wiring helps a lot too.

Adrian's new hot end looks promising, I haven't tried it yet, but it looks a bit stronger: [reprap.org]

You'll notice he's building them out of PLA as well.

I haven't seen anyone work out a way to tell when the extruder is failing yet, although I do use non-welded thermocouples with some code to check to see if they've come loose; it shuts down the reprap if there are any problems with the temperature control. I've printed 3 Mendels in a row now with my extruder without any failures though; they can be quire reliable.

Wade

I still think my glass nozzles, by eliminating the PTFE, will eliminate that kind of problem.

Yes, the extruder base and structure must be made of higher temp materials than PLA. I have not operated on yet, but I think the ABS based ones are more reliable. Wood or aluminum might make something stronger still, but can no longer be made by a reprap or repstrap.

Also not having used an extruder yet, I think that the wade geared extruder is the best current design. If you want to add one of my glass nozzles to it, we either need to make a two piece clamp/adapter to hold the flanged nozzle in the opening made for the PTFE. The other is to thread that opening (Maye have to fill with expoy first if the thread is too small for the opening already there) and screw in the compression fitting holder to secure it in place. I think the hole is small enough to tap directly, so the compression fitting is the fastest approach, but I think the flanged clamp assembly will be the most long term reliable.

There are many other variations on the PTFE, PTFE & PEET, stainless steel, and other hot ends for the extruder. Some fix the problem you just had by adding a pair of long bolts parallel with the filament, a steel plate across the end of the nozzle with another thermal break washer, so that the tip sticks through, put the pressure is born by the strap and bolts, not the brass threading into PTFE joint.

As for whole filament pipeline, it seems that running the filament through a bowden cable (or even just plastic tubing) rigidly mounted on the filament reel holder and passing though a micro-switch mounted on top of the extruder to feed directly into the drive gear(s). The micro-switch controls a motor to rotate the filament reel to keep the loop big enough that the plastic tubing does not close the switch. This relieves the extruder from having to pull the filament so hard that it can turn the reel (especially bad for 5-10 kilo reels!) and keeps a fixed length loop that is less like to tangle in the structure of the reprap.


Hope this help.

Mike

thanks guys, I thought of a partial/obvious solution as well... I bought my hollowed out bolt but I think I'll work on making them myself. If the fit of the filament in the bolt was tighter it wouldn't have as much room to build up the back pressure around the sides. I'm sure I've seen that explained and illustrated somewhere before.

Mr. Rocket Scientist... I'd be happy to help test some of those fancy looking glass nozzles you have. If you want to send me some to try and adapt and use send my an email at anthony.aragues dot gmail dot com

Quick update, I think I came up with a very simple / cheap solution to sensing a common failure to take action...

on my new extruder I'm putting a thermistor as high on the brass tube as I can, then will trigger a series of actions of the temp of the top thermistor is with a defined range of the extrusion temp. This way for a few $ I can at least detect if the cooling is not sufficient and the meltzone is getting too high up the brass tube.

feedback appreciated.

Anthong Redbeard Wrote:
-------------------------------------------------------
> Quick update, I think I came up with a very simple
> / cheap solution to sensing a common failure to
> take action...
>
> on my new extruder I'm putting a thermistor as
> high on the brass tube as I can, then will trigger
> a series of actions of the temp of the top
> thermistor is with a defined range of the
> extrusion temp. This way for a few $ I can at
> least detect if the cooling is not sufficient and
> the meltzone is getting too high up the brass
> tube.
>
> feedback appreciated.

Not that it isn't a good idea, but is the time period that the sensor is either polled, or it's response time sufficient to give you a meaningful result? In other words if the melt zone moves up the barrel faster than the sensor can react, or than the micro checks to see if it's within tolerance, the only thing you are doing is detecting the failure after it's too late.
I don't know the answer to that question, but I'd find out before making the system more complex.

If it polls at least as often as it polls the extrusion temperature I think it will be fine.

I was assuming that looking for a temperature that is a defined range of degrees cooler than the extrusion temperature I could detect and act on a failure before it happens.

If I can get it worked out the added complexity would be an extra thermistor with leads on it, so I think the cost/benefit ratio checks out.

Perhaps a homemade rotary encoder below the idler wheel of the extruder? if the filament jams in the hot end it wouldn't spin the encoder, and if it doesn't match the movement of the stepper it could trigger a shutdown.

not really a simple solution, though.

I have also thought about a pressure sensor connected very close to the tip. When extruding, the pressure should go up, when you stop or reverse the gears, it should go done. If the opening gets jammed, the pressure will become very high. If the melt zone gets too big and a plug forms, the pressure will drop while the extruder stepper is running.

Mike