Humidity test in PLA filament

[Gregg Bone]

I did a very subjective test of print quality vs humidity in PLA filament vs print quality.  The test had some objective beginnings, but the translation to quality are purely my opinion.  Scale is on 0-5 with 5 being WOW! and 0 being failure.

I used as my test a sample print (.stl included) that has two rectangular pillars that connect together at the top (kind of bridging, but I designed in a slope to help it), a couple of very fine details (hooks) and a slight overhang.  I made judgements on surface quality on flat surfaces, at right angle corners, at detail quality, and for stringiness between the pillars as a non extruding move was done, as well as a few notes on visual things during printing.  The PLA used was Natural (ie no color) 1.75 mm.

I did this test because I was very impressed with the print quality I got when I opened this new spool of PLA, but over time it just got worse.

Printed at 187° first layer, 185° >1 layer, bed with blue painters tape and at 50°. (This part would have worked at room temp bed on glass as it is so small, but heated base with blue tape works for big parts, so now all my PLA parts are done this way.)

1. PLA stored 1 week at 47% relative humidity.  Just out in the room.  There is a processing sink in this room.
• Flat surfaces - 3 Occasional void or protrusion on flat surfaces
• Corners - 2  Occasional globs of PLA extending past the corners, many voids, especially on corner where non-extruding moves occurred to
  
• Bridge/Overhang - 1  Couple of dangling loops, they look like they start at a small blob.
• Detail - 2  the hooks are almost closed in with fine strings
• Stringiness - 0  The gap between the pillars was closed on one side where the non extruding move, inside corner to inside corner, was performed.  
• Comments:  Occasional spurts and sputters and pops from the nozzle, what looks like puffs of smoke or steam.
2. PLA stored 1 week at 28% relative humidity.  A room with less water usage - no sinks.
• Flat surfaces - 3 Occasional void or protrusion on flat surfaces
  
• Corners - 2  
  
• Bridge/Overhang - 3 no dangling loops but not a clean surface at all
  
• Detail - 2
  
• Stringiness - 1  The gap between the pillars was not closed on side with the non extruding move, but lots of strings.  
  
• Comments:  No pops this time, but still some sputtering.  But not much improvement overall in print quality.  Biggest improvement in the connection area.
3. PLA stored 1 week at 11% humidity closed airtight container with desiccant.  (I'm not real comfortable on the humidity measurement - my container is opaque, so to read it I have to open the container, but then the room air goes rushing in and the value is climbing)
• Flat surfaces - 5 very clean, well defined extrusions, you can see the roundness of the filament on each layer.
• Corners - 4 an occasional small blob (much smaller than before)   
  
• Bridge/Overhang - 4  fairly clean
  
• Detail - 4
  
• Stringiness - 3  A fair amount of scraping with a blade cleaned up the strings.  I tried a flame wash but that left blobs at the corners.
  
• Comments:  Well that helps a lot.  I really think the humidity is lower inside the box when it is closed.  I have too good of a humidity meter (fast response) and I can see that it's climbing pretty fast when I got the 11% reading.
  
4. PLA stored 1 hour (I used the reel from the 47% test) at 30 um Vacuum.  This is what I was most curious about.  I put the wet reel in my chamber and pumped it down with a 2 stage vacuum pump to about 30 microns (a pretty good vacuum).
• Flat surfaces - 5 very clean, well defined extrusions, you can see the roundness of the filament on each layer.
• Corners - 4.5 Nice crisp 90° corners. I do see some slight density variations so not perfect so therefore not a 5.
  
• Bridge/Overhang - 4  about the same as test #3
  
• Detail - 4  the hooks are great, a little better than test #3 but not enough to rate it better
  
• Stringiness - 4  A flame wash cleaned this part up well.  I think I need to work on my retraction setting a little.
  
• Comments:  Nice.  Seems that vacuum drying works pretty well, even better than keeping the PLA in a closed box with desiccant.  No sputtering during extrusion - just clean round lines.

Gregg

[jimc]

thats great since i happen to have a small vacuum pump and chamber that rarely gets used. thanks for the testing!

[Tim]

Have you tried baking out the moisture in an oven?  I'm curious as to how that compares to a vacuum pump and chamber.  Working in the integrated circuit chip industry, I'm used to the idea of baking moisture out of chip packages.  However, chip packages, even the plastic ones, can take a lot higher temperature than PLA or ABS.  But it should be mostly a tradeoff between the temperature and the time spent baking.

[Gregg Bone]

Other people have had success with oven baking, so I did not re-test.  I own and work at an olive oil production farm and everything needs to be done below 40°C to stay extra virgin, so we have no ovens other than the Pizza oven and commercial kitchen oven we use to do food pairing tastings of our olive oils..  The Chef (who though she is employed by me has tight control in the kitchen) does not like me baking things in "her" ovens.  So I decided to try the vacuum chamber to see how well that works.  The people who used oven to dry filament used lower temperatures and longer times, but I don't remember what those were.

[Tim]

I hadn't thought of that angle.  I regularly bake artwork in the kitchen oven.  My wife doesn't complain, maybe because some of the artwork is for her. . .