Annoying silk-like ooze

It's a bit hard to describe, but take a look at the attached image. I get these silk-like filaments oozing between the outer shells of all parts on a print bed. In all my novice tuning in Slic3r, I have been unable to get better results then this.

Surly better results are feasible though, right? These fibres are not as easy to clean off as one may expect.

I'm printing with PLA on a RepRapPro Huxley with 0.3mm nozzle and a 0.18mm layer height.
In Slic3r, I have retraction turned on to pull back the filament 3mm at 120mm/s and lift the Z axis 0.6mm. I also have cooling turned on which is controlling a large fan affixed to the printer.

Any info on a fix is appreciated, thanks.

Edited 1 time(s). Last edit at 11/28/2012 12:38AM by Physics_Dude.

... try with rising your extruding temperature ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
VDX
try with rising your extruding temperature

I've been using 218C with PLA. Any idea on how far I should increment up?

Edited 2 time(s). Last edit at 11/28/2012 10:09AM by Physics_Dude.

... what's the measured temp on the extruder tip? - This can be toatally different from your settings!

I have some 'virgin' PLA pellets that are specified between 190C to 240C for best use in injection moulding systems ... don't know, how it behaves when printing with same material as filament

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Measured nozzle temp? I don’t have the tools to precisely measure that, but the minimum temp before it jams when printing is around 180C-190C.

... seems to be off some ten degrees ... so try with 230C or even 240C - PLA won't burn, but will get more fluid with this temp-rising ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

If anything I think you are printing too hot.
Also your retract speed setting is very high, it's probably being capped by the firmware, but you might want to put a mark on the gear and verify it's actually retracting and returning.

3 mm of retraction is far too much. Try using 1 mm. 30 to 45 mm/s retract speed is more normal as well.

I tested various new settings using an ooze test block from Thingiverse (in PLA), here are the results. (A score of 10 out of 10 is most preferred)

1. 235C, 3mm retraction at 120mm/s - Score 3/10
2. 225C, 3mm retraction at 120mm/s - Score 5/10
3. 225C, 5mm retraction at 120mm/s - Score 5.5/10
4. 218C, 3mm retraction at 120mm/s - Score 6/10 (seen in original post)
5. 218C, 5mm retraction at 120mm/s - Score 7/10
6. 218C, 10mm retraction at 120mm/s - Score 7.3/10
7. 218C, 3mm retraction at 120mm/s, sans Z axis retraction - Score 8/10*
8. 218C, 1mm retraction at 35mm/s - Score 1/10
9. 205C, 5mm retraction at 120mm/s - Score 8.3/10
10. 218C, 3mm retraction at 25mm/s - Score 5/10
11. 210C, 5mm retraction at 120mm/s - Score 8.5/10
12. 210C, 3mm retraction at 120mm/s - Score 8.7/10**
13. 210C, 2mm retraction at 120mm/s - Score 8.6/10**

*tough ooze drips adhered to walls and X/Y axis jamming is likely. Minimal fibres though
**still not perfect, slightly thinner fibres still seen.

Note: All tests showed full length fibres between shells; most of the better tests had fibres thinner and less common then others.

I know 120mm/s exceeds the cap for that speed, but whatever that cap is, it seems to show the best results.

Also, my data seems to be slightly confusing and inconsistent, perhaps more tests are in order.

You have a bowden setup (extruder separated form the hotend and pushing filament through a PTFE tube), don't you ?

Stringing is a well know drawback with those setups, retraction helps but is not a perfect solution (by the time your retraction movement have propagated to the other side of the tube, the compression of the filament will have already extruded some). It takes a mix of various corrections to minimise that problem, and I'm not sure it can be totally eliminated (depends on the filament).


Possible action 1 : It is recommended to keep only the shortest possible length of tube between the extruder and the hotend (but keep in mind that the hotend have to access all the bed areas without the tube pulling it back).


Possible action 2 : I don't know if slic3r has something similar, but skeinforge offer the "Dwindle" module which can help mitigate this. It allows to slow down (or reduce the plastic flow) gradually before an non-extruding move, thus reducing the filament pressure.
I haven't tried to configure it yet, but there have been a topic with contradicting opinions about how to do it : [forums.reprap.org]


Possible action 3 : Another very important point to reduce that stringing is to use the lowest possible extrusion temperature.
You can find it by manually pushing the filament in the tube (or manually turning the extruder gear).
Do that starting at 160°C to get a feeling of how hard or easy it is to extrude. Increase the temperature gradually by 5 or 10 °C steps, and repeat the pushing. There's a temperature at which it will feel easy enough to manually extrude, which depend on the filament.
You might want to add a few °C to that to compensate for the cooling effect the flowing filament has on the hotend.


Possible action 4 : Also get sure that your non-printing movements are done at the maximum possible speed : faster travels, less time for plastic to flow out of the nozzle.

---

Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.

Ah yes, the RepRapPro Huxley has a bowden type extruder. In that case, 3 mm of retraction is on the low side. Everything that DeuxVis suggested is good.

In response to DeuxVis,
Yes, that would make sense. Given the slight wiggle room inside the PTFE tube, there would be a delay at the hot end.

I much prefer Slic3r over Skeinforge, but it does not include its own related Dwindle feature.
Perhaps, if the Hardware allows it (Melzi V2.0), then a faster retraction speed cap may be in order. Currently it seems like 45mm/s is the cap. I'm not sure if it's Pronterface, the Melzi's firmware, or both, but does anyone have any advice on whether or not it is possible to change this, and if so, how?

You would need to change the speed cap in firmware. You may not be able to get much more than 45 mm/s without the motor skipping steps, but it's worth a try.

Beside those "hair" all over, those are very nice prints.

Try to use a stiff brush to clean most of the hair and then using a blow torch go over all the remaining hair "VERY QUICKLY" it should take care of it.

I imagine you could also try a very high grit sand paper (800 and up) high enough not to scar the plastic and a quick rub all over would take care of the hair without affecting the surface finish.

Don't give up, your having nice results. If you need, take a few days off 3d printing... I almost went nuts trying to fix mine at one point lol.