I agree that humidity has a role since something changed during the night.
But still, when the extruded PLA filament is short enough (not enough weight pulling) it tangles right after extrusion, everytime and always in the same direction. Is that due to the molecular structure alone ??

I have read several papers/sites stating that the long polymer chains in thermoplastics "remembers" the cirkular motion from the feed screw. It is therefore recommended to use a breaker plate.

Google pitfallsinmoulding+extrusion and the second hit is an interesting read.

Will try to trace down the other papers soon.

I just finished my extruder today, and I only have PLA pellets avialible at the moment.
The extruder is based on both filastruder and the lymans extruder.
It has both a breaker plate and a cone shaped nozzle. It is a bit over engineered, but it might be rewarded...
Hopefully I will get good results during the next few days.

Do anyone know if watercooling at the end of nozzle will do harm to the filament? I have the equipment to implement this, but I am not sure if the filament will absorb moisture. We do have a climate chamber at work, so baking/drying of the filament is not a problem.

I suspected that because the nozzle's back is flat, previous polymers (ABS) never get completely flushed out (I could see lines and tiny bits of white ABS in the clear PLA) so I removed the nozzle and drilled its back in a funnel shape. I put it back on and set the temperature to 180°C so that any remaining ABS that didn't melt at 160°C would now melt and be evacuated.
I also carved out wood from under the fan so that more air would come to cool PLA.
I turned the motor on and a little while after extrusion started I saw some weird tangling patterns that could indicate that indeed tangling is due to molecular structure alone.
I lowered temperature to 150°C and this time for some reason it extruded fine (straight filament and motor didn't slow down too much) so I continued with this temperature.
Result:
extrusion speed: 80cm/min
filament diameter: 1,325 ±0,011mm
(I didn't pre-dry it)

You can't slow motors with a potentiometer.

I also doubt that the marks on the inside of the nozzle are to blame, especially since it didnt change shape overnight, but your results did. The PLA was able to absorb humidity overnight, though.

If you wanted to get rid of the machining marks, you could ream
the nozzle.

I think that I have found the source of tangling, waves, bumps and other issues including diameter variations (to some extent). It is due to the fact that the hole of the nozzle is twisted (drill's shape) instead of being smooth (I don't know the appropriate terminology). I have the same problem on my Ultimaker nozzle but not on my Replicator nozzle.
Thinking about it as writing, if it's not due to the nozzle it might be the movement of the auger that is transmitted to the melted pellets, in this case it could be solved by increasing the extrusion speed.

CornGolem wrote:

How does one pre-dry PLA ?

Depending on the specific grade, most manufacturers recommend drying crystallized PLA at 150ºF-190ºF (65ºC-87ºC) using dehumidified air with a dewpoint of -40ºF.

http://www.plasticstoday.com/mpw/articl … ations-pla

Ok but how do I do that, in other words, what equipment do I need exactly ?

How does one pre-dry PLA ?

Depending on the specific grade, most manufacturers recommend drying crystallized PLA at 150ºF-190ºF (65ºC-87ºC) using dehumidified air with a dewpoint of -40ºF.

http://www.plasticstoday.com/mpw/articl … ations-pla

It seems to me that if the pressure was higher it would extrude PLA at a larger and more constant diameter. To do this we need to bring more pellets but we can't control the motor's speed on the filastruder. We can however change the motor, what about the faster one that was mentioned in an update ?

It'll be in the store that launches at 8pm today.

I'm not sure it will solve PLA, but it definitely will create a higher pressure. You can overvolt the motors (old and new) if you like by 20%, but make sure the motor casing doesn't get above 35C when overvolting. You may need additional cooling when overvolting. Heat is a motor brush's worst enemy.

How does one pre-dry PLA ?

I ran it again today and the extruded filament was very different, slower, thicker and wavy (significant diameter change every few cms).
extrusion speed: 49,25cm/min
filament diameter: 1,421 ±0,075mm

I also tried at 150°C but it was too cool for the polymer and the motor was slowing down a lot. The diameter jumped to 2mm, the wavy pattern became more frequent and the plastic became more opaque.
I then tried 155°C but it was not better than 160°C.

It seems to me that if the pressure was higher it would extrude PLA at a larger and more constant diameter. To do this we need to bring more pellets but we can't control the motor's speed on the filastruder. We can however change the motor, what about the faster one that was mentioned in an update ?

I have read here that PLA is more challenging than ABS, and I imagine that the increased stiffness of 3.0 vs 1.75 may make coiling on the floor (or anywhere) more difficult.  I'm running at 160 per Tim.  I can, of course, 1) switch to 1.75, 2) switch to ABS, or 3) build a spooler.  I'm trying to get ready to show off the Filastruder next Wednesday, and am beginning to think that I may have to do that without a fully successful setup.  In the long run, the winder seems like the best approach, but I have other projects that are taking up a lot of my time. 

Any thoughts about how to make progress with 3.0 PLA?

Sorry Mate, I am exceptionally busy right now!

I have my PhD viva exam coming next week and working a full time PostDoc as well :S,

In two weeks hopefully I will have a broad smile and a lot more time to post my results!

I'm reluctant to give specific advice without testing it at least twice, although if anyone would like experimental recommendations I'm happy to oblige,

Just don't want anyone ruining there shiny new filastruders oin my back,

It's my job to break and fix these things

Watch this space,

All the Best,

Cheers,

Rob

Can you compare/contrast come results with quantitative data? What materials? What filament variation? Pictures? Deets, man!

Tim

Just too point something out to people,

The winder can take up room, but it is the way to go!

Also the space the winder takes up is much less than the space required to let the filament lay onto the floor!

So actually the winder takes up less room and gives a better result!

Hats off to Ian for working so hard on the winder project!

Cheers,

Rob