Exsiccator for dry filament

[jumpmobile 0]

This is the continuation of my post on the water-content of filament.

(original post here: http://umforum.ultimaker.com/index.php?/topic/1711-tga-of-filament/ )

The stroy:

I experienced that when mounting a new filament on my printer the first few prints generally have nice surface quality with little blobbing and neatly aligned lines.

But after some days the quality lowers. Not dramatically but still visibly, the blobbing returns and retraction needs to be readjusted.

The explanation:

Everyone has water in his filament as you can see by extruding into the air for a bit. At some point the extruded filament will have a bend or a bubble.

As you can easily imagine bends and bubbles in your extruded filament will cause an imperfection on your print surface.

The quotes:

Marplex (in own words "leading supplier of quality plastics in Australasia") provides on it homepage a variety of interesting paper on handling ABS plastic.

http://www.marplex.com.au/technical/

This one is about water content and pellet drying.

http://www.marplex.com.au/technical/A%20guide%20to%20ABS%20Pellet%20Drying.pdf

For extrusion purposes they recommend a water content of no more than 0.02% (Edit: That value is wrong. should be 1.33%, see below). This is close to nothing, compared to the average air humidity of about 40-50%.

The conclusion:

I've been to the hardware store and got myself a plastic box, a hygrometer, foam-seal tape and some desiccant (calciumcarbonate http://www.uhu.com/en/products/moisture-absorbers.html )

I put everything togehter like this:

http://www.thingiverse.com/thing:62826

...just to see how low the humidity would go with the desiccant and voilá: after about half an hour the hygrometer showed 6% remaining air humidity.

This is much more than 0.02% 1.33% but maybe we can see an improvement to standard stored filament.

Next step is to compare quality of dry- and wet-stored filament.

If that doesn't give the desired results then i will try reducing the moisture even further to the recommended 0.02% 1.33%

Update (year later):

- The next day the hygrometer showed 0% humidity.

- Had problems with cracked filament (look for pictures in my galery). Haven't seen improvement to regular "humid" filament.

- Dried filament is so brittle that it easily cracks even inside the bowden tube leading to friction and underextrusion.

- Tightly coiled filament can lead to underextrusion even when not cracked.

Final Conclusion:

Only dry filament that was delivered on big spools like the ones used by UM (sorry for advertising).

Humid filament is fine for printing.

 

Greetings Fabian

 

[jonnybischof 60]

Very interesting research, keep it going!

Until now I never minded the humidity issue, but I'll have some kilograms of filament laying around soon, so I'll probably want to copy your solution. Thank you!

 

[gr5 2,176]

Cool! I want to get me that stuff...

However you are confusing humidity with % by weight.

Some people think 100% humidity means you have 100% water and 0% air. But no, 100% is when you can't add any more water (as a gas) to the air and adding any more water will be in droplet (liquid) form.

At 100% humitidty, 1 cubic meter of air at 30C holds 30 grams of water

And the air alone weighs about 1200g

So even if humidity is at 100%, this is only about 2.5% water by weight (that's normally how you do it -- by weight).

So if your goal is to keep water under .02% then the humidity only needs to get down to .8% to have the *air* .02% by weight. I'm not sure how this translates to qty of water in PLA though. But getting humidity down to 1% should help a lot!

 

[jumpmobile 0]

@Johnny:

Cool. If you do it please share your experience.

BTW: make sure to buy some rubberjoint-belt for windows. (i don't know the english expression for it). In german it is "Fensterdichtband".

These are cheap and will help you seal the box. I forgot them in the first post though (now edited).

And just to be fair: This wasn't my idea in the first place. I just copied this from other people (see the thingiverse-link in the first post).

@gr5:

Jep, i got that wrong. Thanks! :smile:

So let me recalculate that:

Figure 2 on the second page of that paper:

http://www.marplex.com.au/technical/A%20guide%20to%20ABS%20Pellet%20Drying.pdf

shows a correlation between relative humidity in the air and equilibrium moisture concentration (EMC) in the filament. You can see that 0.15% filament moisture (general purpose type) relates to a 10% humidity. So what we need is not an air humidity of 0.02% but only 1.33%. That sounds more realisitc....

 

[jumpmobile 0]

Well to give you an update on this:

I have tried this for a few days now. For the parts used see my gallery:

http://umforum.ultimaker.com/index.php?/gallery/album/307-exsiccator/

The setup is really simple.

Greets Jump

 

[jumpmobile 0]

Yesterday I experienced something really weird:

I had this delivery of filament: Nylon,Laywood, Bendlay and 2kg of white PLA from mexhibit (german vendor):

http://www.mexhibit3druck.de/#!product/prd1/996213934/fdm-filament-3d-druck%2C-natur-weiss%2C-3.0-mm%2C-2.3-kg

This PLA was delivered on two very small spools a 1kg each, that wouldn't fit onto my UM. So i took the material of one spool and put it on an empty 2kg-reel i had laying around. Put this in my newly build exsiccator and left it in there for about a week now.

And yesterday I came home to this:

http://umforum.ultimaker.com/index.php?/gallery/image/1862-pla-drying-crash/

read on here:

http://umforum.ultimaker.com/index.php?/gallery/image/1861-spool-of-dried-and-cracked-pla/

Sorry for the double post. I just didn't want to mix up these two things. The principle of drying filament seems to be justified for already dry Nylon/ABS but in wet filament there seems to be some changes (hydrolysis?) in the chemical consitution of the filament.

 

[jonnybischof 60]

Ouch :-/

Any idea on how to prevent this?

Maybe put new-bought filament very loosely on it's spool, and use a separate chamber with not too much desiccant inside? Then wait 1-2 days until you start using it.

I think I'm going to try something like that (my spools have been laying around for some time now...) when I'm done with my desiccator (exsiccator is german btw ). Might take some time because right now I'm preparing a bigger heated bed & direct drive upgrade on my UM.

 

[jumpmobile 0]

I don't think that i want to prevent this. This filament just snapped because i changed the spool-size. The other spool i had (which i didn't change) has been in there for about two (?) weeks and still looks good.

It is just a hint so you know not to change new bought wet filament onto bigger spools before putting in the dry-chamber.

P.S: "Exsiccator" in german is "Exsikkator". But "desiccator" sounds better since i am using 'dessicant'

 

[yzorg 9]

Yesterday I experienced something really weird:

I had this delivery of filament: Nylon,Laywood, Bendlay and 2kg of white PLA from mexhibit (german vendor):

http://www.mexhibit3druck.de/#!product/prd1/996213934/fdm-filament-3d-druck%2C-natur-weiss%2C-3.0-mm%2C-2.3-kg

This PLA was delivered on two very small spools a 1kg each, that wouldn't fit onto my UM. So i took the material of one spool and put it on an empty 2kg-reel i had laying around. Put this in my newly build exsiccator and left it in there for about a week now.

And yesterday I came home to this:

http://umforum.ultimaker.com/index.php?/gallery/image/1862-pla-drying-crash/

 

I made the exact same mistake... i now have quite some unusable filament lying around.

i hope i can recycle it with my filastruder in future.. but i need to build a shredder first.

 

[jumpmobile 0]

Yep.

I got myself a Filastruder for that purpose. Will arrive at the end of january.... sooo excited.

 

[gr5 2,176]

So what we need is not an air humidity of 0.02% but only 1.33%. That sounds more realisitc....

 

I was reading a paper about PLA this morning and found this graph.

.02% is 200ppm which on the graph looks like about 15/5 or 3% humidity. The graph is on page 561 of this document:

http://www.jimluntllc.com/pdfs/polylactic_acid_technology.pdf

So it looks like 3% humidity is your goal.