Soft PLA troubles

[chuckmcgee 7]

Hello all,

 

I have searched for anything current on this topic and can't seem to find anything.  I have an Ultimaker 3 Extended (relatively new printer).  I have Soft PLA from Matterhackers that I just bought and have been having trouble getting it to print.  After extensive troubleshooting with an engineer at Matterhackers, I can't believe the issue is me, but you never know...so I am bringing it to this group in hopes of finding an answer.

 

The problem is that I can't get the filament to feed reliably...it under extrudes (or stops extruding altogether) on every attempt.  Here is what I have tried:

• Bypassing the automatic load feature of the feeder and manually feeding the filament through the Bowden tube to the hot end
• Printing at slow speeds (tried 15mm/s and 20mm/s...nothing faster or slower)
• Set hot end at 240 to make sure it wasn't a chore to get the filament through the hot end
• Retraction off or as short as 0.50mm

 

Unfortunately, none of these things have made a difference.  I read older posts online about oiling the Bowden tube, but I didn't do that and don't think friction in the tube is the issue.  The filament feeds freely (diameter of 2.40 as measured with calipers).

 

In my last attempt (manually fed filament to the hot end, speed of 20mm/s, retraction at 0.50mm, and hot end temp of 240), filament stopped coming out even at the jump when it dispensed into the blob on the corner.  When I stopped the print and removed the filament (manually) from the printer, this is what the filament looked like inside the feeder.

 

Is it me or should I just return this filament and give up on the UM3 for printing anything flexible...(and I purposefully chose Soft PLA with a value of 98A as a less flexible option with some rigidity). 

[chuckmcgee 7]

My Lord...I may be the largest noob ever.  I hope I am wrong.  I have a print going now that seems to be goign well.  My big mistake was to create a custom material in Cura 3.1.0.  That was not overriding the settings in the default profile in Cura.  So while I thought I was setting the values above - alas - they were not the values actually being printed.

 

more to follow...going at 20mm/s speed, 235 hot end temp, retraction 0.5mm, and using the TPU 95A material type (with modifications).

[chuckmcgee 7]

I am wrong...under extruding again.  Stopped feeding altogether...but I did get almost all of the brim done.  Removing the filament shows the same issue...a corkscrew effect on the filament inside the feeder.

[SandervG 1,521]

Hi @chuckmcgee, thank you for sharing your findings with us so thoroughly. Looked like indeed you did some research.

I think dropping a few drops of sewing machine oil down the bowden tube could help, and if nothing else perhaps it could eliminate the bowden tube as a cause and helps

us closer to the actual problem through elimination. But, there are also other things we can try first. I'm wondering, have you changed the feeder tension on your Ultimaker, and if so, what tension is it set now? If not, perhaps you could try with a tighter feeder. 

 

And I think I would disable retractions all together by the way.

[rebekah_harper 60]

Darn it Sander, I was going to suggest that. lol

 

I'd try a lower temp, It gives a wide range from 210 to 235.  I haven't printed with soft PLA before but I found blockages are very prominent at higher temps when PLA is concerned so I am kinda leaning towards this idea. 

[SandervG 1,521]

2 minutes ago, rebekah_harper said:

Darn it Sander, I was going to suggest that. lol

 

1

 

haha you snooze you loose  

[rebekah_harper 60]

6 minutes ago, SandervG said:

 

haha you snooze you loose  

you better have enjoyed those Jaffa cakes lol 

[chuckmcgee 7]

9 hours ago, SandervG said:

Hi @chuckmcgee, thank you for sharing your findings with us so thoroughly. Looked like indeed you did some research.

I think dropping a few drops of sewing machine oil down the bowden tube could help, and if nothing else perhaps it could eliminate the bowden tube as a cause and helps

us closer to the actual problem through elimination. But, there are also other things we can try first. I'm wondering, have you changed the feeder tension on your Ultimaker, and if so, what tension is it set now? If not, perhaps you could try with a tighter feeder. 

 

And I think I would disable retractions all together by the way.

 

Thank you for the reply, Sander and Rebekah!  I will try the oil, but would like to do so as a last resort.  I don't know why, but it seems so radical and unnecessary since the printer shows TPU 95A as a default material type.

 

As for the feeder tension, I am assuming that is the meter on the side of the feeder which is controlled by the hex screw on the top of the feeder assembly where the Bowden tube exits up to the hot end.  If so, that is currently at the midway point.  Seems ample tension as I am having to squeeze pretty hard on the trigger to manually push the filament through the tube to the hot-end.

 

Finally, as for temps - happy to attempt to lower, but wondering why being excessively high wouldn't eliminate the hot end as the bottleneck.  Since the issue is under extrusion caused by inconsistent feeding, just figured using a temp at the upper end of the range might make for a messier print with lots of stringing.  At the moment, I would welcome that issue

 

I will await your reply for what to try next.  Sounds like it may be worthwhile to still attempt some of the things above...just looking for confirmation based on my responses.

 

Thanks again for the reply!!

[SandervG 1,521]

9 hours ago, rebekah_harper said:

you better have enjoyed those Jaffa cakes lol 

 

Hah, I certainly did, they were delicious! But as according to the original agreement, they all went to R&D! 

 

@chuckmcgee, I'm not really sure how flexible soft PLA is compared to TPU95A, but just because it is flexible does not mean it is more or less the same (it can tho, but not necessarily). 

Regarding the feeder, yes that is what I mean. The middle is considered 'default' tension. You may want to increase the tension a bit, so the teeth have better traction on the filament while pushing it up. 

 

Hotter is not always better. Depending how fast the filament is being extruded through your nozzle, (speed movement & layer thickness) the material can stay in your nozzle for too long and even burn or the heat will crawl up through the filament making it softer higher up 'in the chain', where it will expand due to the pressure it receives from the feeder, trying to push more down in the bowden tube. When it has expanded, it may become significantly more difficult to push further down, into the nozzle, and you can get under extrusion.

 

A good indicator of whether or not it expanded, is when you pull the filament back out through the bowden tube. Good luck, and curious to hear about your findings!

[yellowshark 153]

Firstly with a Bowden tube printer, using flexible filament is always likely to be problematic but that will vary depending on the actual filament being used. Direct drive printers are much better but people have successfully printed flexible filament with the Ultimaker printers. I have successfully fed flexible filament (Colorfabb probably) through my bowden tube printer but I could not achieve a decent surface finish so gave up.

 

One thing does worry me, you state your filament diameter as being 2.40. This is very thin; the Ultimaker and similar printers are designed to use the standard 2.85 filaments. I am not sure as to what the outcome of this diameter would be as I have never tried it.

[SandervG 1,521]

20 minutes ago, yellowshark said:

you state your filament diameter as being 2.40

Good point, completely overlooked that. That is indeed thin. And Cura assumes by default it is 2.85. Did you adjust this value?

[chuckmcgee 7]

Thanks.  The 2.40 is coincidental and I have a replacement roll being sent from MatterHackers.  As for changing within Cura, I did...not on the material manager but in the profile settings.  

 

I will I’ll report back after I use a roll with 2.85 and with slightly tighetening the stepper tension. 

[geert_2 557]

What you could also do to diagnose the problem, is to remove the bowden tube at the feeder end, and manually insert and move filament. Manually dial nozzle temp up to various settings, and extrude some material. Then you might be able to feel where it gets stuck and why?

 

[chuckmcgee 7]

4 minutes ago, geert_2 said:

What you could also do to diagnose the problem, is to remove the bowden tube at the feeder end, and manually insert and move filament. Manually dial nozzle temp up to various settings, and extrude some material. Then you might be able to feel where it gets stuck and why?

 

Thanks Geert.  I had a similar thought and wondered if I could do the whole print, unattended, while bypassing the Bowden tube.  Didn't seem like a smart idea, but I like your suggestion.

[gr5 2,172]

DON'T FEAR THE OIL!   lol.  It works so well.  Just do it.  It will not mess up your printer.  It will not affect your print.  It works very well.  Just do it. 

[rebekah_harper 60]

2 hours ago, chuckmcgee said:

Thanks.  The 2.40 is coincidental and I have a replacement roll being sent from MatterHackers.  As for changing within Cura, I did...not on the material manager but in the profile settings.  

 

I will I’ll report back after I use a roll with 2.85 and with slightly tighetening the stepper tension. 

hiya,

 

there is a roll with 2.40mm filament?  never seen that before so it's a first for me.  I am hoping that it is physically that diameter.  If it measures at 2.85 then that is a problem.  the g-code is going to assume that it needs to feed more/faster to provide the same amount of filament.  but at the same time if it is 2.40mm then there may be too much space at the hot end and it buckles slightly causing substantial resistance at the hotend. 

 

hopefully you get it solved

[chuckmcgee 7]

On 12/21/2017 at 10:11 AM, gr5 said:

DON'T FEAR THE OIL!   lol.  It works so well.  Just do it.  It will not mess up your printer.  It will not affect your print.  It works very well.  Just do it. 

 

Will do gr5.  Bought some last night and will try it out today.  Any suggestions on how exactly to apply it?

 

On 12/21/2017 at 11:09 AM, rebekah_harper said:

hiya,

 

there is a roll with 2.40mm filament?  never seen that before so it's a first for me.  I am hoping that it is physically that diameter.  If it measures at 2.85 then that is a problem.  the g-code is going to assume that it needs to feed more/faster to provide the same amount of filament.  but at the same time if it is 2.40mm then there may be too much space at the hot end and it buckles slightly causing substantial resistance at the hotend. 

 

hopefully you get it solved

 

It's the physical diameter.  I believe it was meant to be 2.85mm.  The support folks at MatterHackers are sending me a replacement roll.  I am going to try a different roll that measures 2.85mm to see if that helps...between a better experience with the feeder and the use of the oil.  We shall see.  Trying today.

[cjs 34]

@chuckmcgee Please don't use oil! (Have a look at: " Friend's don't let friends use oil... " for reasons to not use oil) 

Oil shouldn't be necessary to print flexible filament. Your filament (92A according to matterhackers) is more flexible compared to Ultimakers (95A) TPU filament but not by much. 

If you wan't to print flexible filament, a well constraint filament path is needed. Otherwise the filament will buckle up. The filament path of the UM stock feeder isn't well constraint and can cause issues. There may be some printable add on by the community to help print flexible filament.

With the Bondtech DDG installed on my UM3 I can print UM TPU 95A at 40-50mm/s with no problems without using oil or anything else to help. Youtube video of Bondtech QR printing flexible filament

Disclaimer: I'm one of the designers of the Bondtech DDG

[chuckmcgee 7]

Just now, cjs said:

@chuckmcgee Please don't use oil! (Have a look at: " Friend's don't let friends use oil... " for reasons to not use oil) 

Oil shouldn't be necessary to print flexible filament. Your filament (92A according to matterhackers) is more flexible compared to Ultimakers (95A) TPU filament but not by much. 

If you wan't to print flexible filament, a well constraint filament path is needed. Otherwise the filament will buckle up. The filament path of the UM stock feeder isn't well constraint and can cause issues. There may be some printable add on by the community to help print flexible filament.

With the Bondtech DDG installed on my UM3 I can print UM TPU 95A at 40-50mm/s with no problems without using oil or anything else to help. Youtube video of Bondtech QR printing flexible filament

Disclaimer: I'm one of the designers of the Bondtech DDG

 

Thanks @cjs.  I have mostly been testing with the red filament which measures at 2.40mm in diameter (a mistake).  I am going to try all of the ideas (minus the oil for now) with filament measuring 2.85mm.

 

As a point of clarification, the MatterHackers Soft PLA I am using is 98A...harder than TPU95A.  I chose this because I wanted it to be a little sturdier than the 95 or less.

[chuckmcgee 7]

Also...let me distinguish a difference between what that article says about oil and what was suggested earlier on by @SandervG (and likely as @gr5 meant).  I believe the article is discussing the use of oil as a way to season the filament and I believe SandervG is suggesting a few drops in the Bowden tube to help with flow through the tube.

 

Not sure if it fundamentally is different from the premise of needing oil vs not needing it, but I see no harm in the Bowden tube as a way to help.

[cjs 34]

24 minutes ago, chuckmcgee said:

Also...let me distinguish a difference between what that article says about oil and what was suggested earlier on by @SandervG (and likely as @gr5 meant).  I believe the article is discussing the use of oil as a way to season the filament and I believe SandervG is suggesting a few drops in the Bowden tube to help with flow through the tube.

 

Not sure if it fundamentally is different from the premise of needing oil vs not needing it, but I see no harm in the Bowden tube as a way to help.

Yes you are correct. The article is discussing the use of oil in a slightly different scenario, but if you use a few drops in the tube it will end up in the nozzle at some point. 

I really don't think the tubing is the problem. From my experience flexible filament is easier to push through the tubing compared to normal PLA especially at 2.85mm diameter. Best way to test this is to remove the tubing from the printer and pushing a piece of filament through the tube by hand.

 

Other reasons I'm against using oil is the fact that it can affect your build plate adhesion as well as inter layer adhesion for example.

Edited December 23, 2017 by cjs
Added reasons against using oil

[gr5 2,172]

Pushing by hand is an okay comparison but make sure you put tension at both ends - in other words fight the filament at one end and push it at the other because if there is no resistance in the print head then you shouldn't need any oil.  But add a little back pressure from the print head and now the oil makes a difference.

 

One drop every meter is plenty.

 

yes the oil of course gets into the head eventually but it doesn't seem to affect the print in any way.  It's quite benign.

 

[AreDigg 8]

14 hours ago, gr5 said:

yes the oil of course gets into the head eventually but it doesn't seem to affect the print in any way.  It's quite benign.

 

 

I have printed a lot with ninjaflex on the ext2+, and oil makes wonders. But one has to be careful, the oil must not seep into the feeder, I realized that after a while some of the oil dripped into the feeder wheel, and thus negating the effect from having a slick Bowden. Oil in the nozzle seem also to make a more uneven temp exchange. The small plastic dust created from the knurling wheel tend also to love the oil. I now got a more slippery Bowden tube from 3dsolex and it seem like it helps with most semi flex plastic, but I still add a little oil for ninjaflex. I use a bit isopropanol occasionally to remove oil residue in the nozzles, and also clean the knurling wheel in the feeder. 

[fergazz 60]

I'm quite curious to try Capricorn bowden tube, the dark blue one, seems to be very slippery... JAMN approved it.

[SandervG 1,521]

On 23/12/2017 at 3:33 PM, chuckmcgee said:

and I believe SandervG is suggesting a few drops in the Bowden tube to help with flow through the tube.

 

Yep, that is what I meant. With only 2 or 3 drops, or the 1 drop per meter as @gr5 suggested it should not affect the build as far as I know. If you put too much in it, I can see how it may undermine your bed adhesion (if the first extruded filament is all oily), but I did not see it influence layer adhesion. I'm not saying oil is the holy grail, but it is an easier thing to try than making modifications to the feeder or bowden tube. The constrained filament path argument is by the way not related to why you would want to use oil. Oil is there to reduce friction in the bowden tube. 

Anyway, usually there are different roads that lead to Rome @chuckmcgee , did you have any success in the meantime?

 

@fergazz , I don't know that one. Do you know what it is made of? And what is the inner diameter and outer diameter?