My PLA went runny for a short while

Hi,

I have just returned to printing PLA (after a few weeks spent printing ABS) and I did all the expected things like cold-pull on the ABS and hot-flush with the PLA. I produced a few prints in PLA successfully after this.

On trying another print I noticed that PLA was oozing very rapidly out of the extruder during the pre-heat stage of start_PLA.gcode. There was a significant build-up of material around the nozzle so I abandoned the print and wiped the nozzle clean.

Examining the heated bed there was a large blob of very dense PLA where the nozzle is parked for the pre-heat and the skirt was not printed at all.

I thought that something might be wrong with the temperature sensing e.g. thermistor reading low but that is not the case. The thermistor is tracking ambient reliably - which it would not if it were reading low.

I have just printed another object successfully so it appears to have been a material problem. The question is, what would cause PLA to come out runnier than normal at correct temperature?

This is Reprapsource White PLA and I have had no trouble with it up to now.

Regards,
Neil Darlow

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.

Unlike PLA, ABS does not take on a "rubbery" consistency as it cools down which makes getting the stuff out of the hot end using the same "cold pull" method as carried out for PLA is not completely successful. When moving back to PLA after printing with ABS, I have found that I need to flush all of the ABS out of the system first. I set my hot end to around 230C and then manually wind PLA through the extruder until all traces of ABS have disappeared - not easy to tell if you are using the same or similar coloured filament!

This does not necessarily explain what you have described of course but the technique seems to work.

Alan

Edited 2 time(s). Last edit at 11/29/2014 10:53AM by Alzibiff.

Hi Alan,

The cold-pull almost went perfectly!

I tried an earlier pull at 160C as the Build Manual advised but the ABS stretched and I wasn't sure it was all cleared. For a subsequent pull I dropped the temperature to 130C and the plug came out intact although I think the recommended 120mm/min speed is a bit slow because the plug was squashed by the hobbed bolt and wouldn't come out of the Wade's block (I had to remove the idler).

Examining the removed plug, it followed the conical profile of the nozzle at the end and was only missing the extruded part. The PLA flush at 220C will have pushed that out of the nozzle, I'm sure.

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.

I used to get this problem when running Fabberdashery PLA at 220C. It seemed unstable at 220C as most of the time it extruded as a filament but occasionally during warm-up or the first layer it would pour out like syrup. I couldn't explain it but when I dropped the temp to 185C after adding the silicone tape mod the problem went a away.

---

[www.hydraraptor.blogspot.com]

Hi,

I believe I have determined the cause of this and I can mark it as solved. It actually took a change of filament supplier to exaggerate the problem and cause me to investigate.

Having just switched to rigid.ink I was suffering the problem of plastic welling-up around the hotend during warm-up and the print subsequently failing.

I did a cold-pull on the filament and disassembled the Wade's block idler assembly to check the state of the hobbed bolt (a slight diversion but it was in pristine condition with no signs of grinding). Examining the plug from the cold-pull I could see the filament had taken on a brownish colour which led me to believe that the filament was being subjected to heating for an excessive amount of time.

It occured to me that the pre-heat sequence that we are accustomed to using might be the cause of the issue. Typically we:

1. Set the extruder temperature
2. Set the bed temperature and wait for it to be achieved
3. Set the extruder temperature again and wait for it to be achieved

There is a problem here if the extruder achieves target temperature before the bed does. The filament in the hotend is subject to additional heating until the bed reaches its target temperature plus the 60 second residency time set in the firmware.

So, what I have done is to remove the initial extruder temperature setting and although this lengthens the warm-up sequence it minimises the amount of time the filament is being heated while static.

I have also reduced the residency time in the firmware from 60 seconds to 30 but that is still subject to evaluation and I am not committed to keeping it at this lower value.

Regards,
Neil Darlow

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.

My start sequence for PLA starts the bed heater off at a target temperature which is 10C less than what I want it for my first layer. Once this temperature has been achieved, the bed is commanded to heat up to its final target temperature and only then is my extruder set to heat up.

Very little time is wasted as the bed is heating while the nozzle is travelling down towards the print bed and I have found that on my machine, it takes the nozzle about the same amount of time to reach its extrusion temperature as the bed takes to go its final 10C.

Start code (edited) below noting that M104 sets a temperature for the extruder and then moves on to the next command whilst M109 sets the extruder temperature and waits until this is achieved until progessing to the next GCode line. ....

G28 ; home
G1 X0.0 Y98.0 F9000
G1 Z0.05
M190 S[BED FINAL TARGET MINUS 10C]
M104 S[PLA EXTRUSION TEMPERATURE]
M190 S[BED TARGET TEMPERATURE FOR PRINT]
M109 S[PLA EXTRUSION TEMPERATURE]
G92 E0
G1 X50.0 E5 F200

Alan

Edited 2 time(s). Last edit at 10/20/2015 06:35AM by Alzibiff.

On an un-modified Mendel90 the bed reaches PLA temperature faster than the hot end. That is why the start sequence is the way it is. For ABS and for metal beds I don't start heating the nozzle until the bed reaches the target.

The residency time is to allow the hot end length to stabilize and the plastic to melt inside the melt chamber. 1 minute is from experience with Hydraraptor, so may be longer than necessary on Mendel90. If it is too short the first layer height gets lower as the layer progresses,

---

[www.hydraraptor.blogspot.com]

Hi,

Actually, I think some of my problems are exacerbated by the rigid.ink PLA filament.

It seems that this filament is not tolerant of excessively long heating which results in it becoming very runny and leaking out of the nozzle. This effect not only happens with higher temperatures but with (what I would call) normal ones when they are applied for a significant amount of time.

I have found that this material needs to be printed at 180C and a minimum of 40mm/s (it is ok to print the first layer at 20mm/s). Slowing print speeds down and imposing long layer times cause the stuff to cook in the hotend or actually brown when printing slowly. It seems to have a very narrow flow vs. temperature envelope.

On the plus side I would imagine that you can print at insane speeds (with sensible temperatures) if your printer is capable and it does seem to handle overhangs very well.

Regards,
Neil Darlow

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.