Baking Water out of PLA

[gestalt73]

While I've been baking my ABS, Nylon, and PET spools for a while now, I thought I'd try the same thing with PLA to see what happens.
Link takes you to the full size image.

On the Left: you can see the shimmer of all the little air bubbles in the calibration square print, how the part turned out. Uneven layers, with the airbubbles clearly seen.

On the Right: you can see the results. No air bubbles, and great dimensional accuracy and perimeters.

I suspect this could be an issue with any roll of PLA, it's just easier to see because there's no pigment in this natural PLA.
I'm sold, and I'll be baking all my PLA filaments before running prints.

4-8 hours at 150-175 Fahrenheit
(the limit isn't the glass transition temperature of the filament, but of the spool. Go too high and the spool melts)

http://i.imgur.com/SNpiyQh.jpg
[img]http://i.imgur.com/SNpiyQhl.jpg[/img]

[joe]

That is good to know. Thanks for going through the trouble of documenting it. I'll give it a go.

[joe]

How often would you do this?

[gestalt73]

You'll definitely want to do this before using any new roll of filament.

After that it's just a factor of your environment and how long you tend to leave spools out in the open air.

I keep all my filament in sealed containers with desiccant packs. Spools are brought out for a print then placed back into a sealed container after the prints are done.
If it's a client print and I haven't dried the spool in a while, I'll throw it in the oven for 4 hours to make sure it's still dry, then run a calibration print to fine tune the extrusion multiplier for that roll.

As long as you're limiting exposure to outside air you may only need to do it once.

[Qdeathstar]

Cool info,

When i read this, i have to be honest that when you mentioned that the temp limit wasn't the limit of the PLA but of the spool, my first thought was about how many spools were lost to melt and how many stoves where damaged during the testing phase ^_^

[Jimustanguitar]

I've been sitting on the parts to make a heated vacuum chamber for my filament dryer. As I understand it, the trick is to get the filament up to temperature first, because the vacuum isn't a good conductor of heat afterwards (which is why a Thermos works so well)... Maybe I'll get back to that project soon. I've got some neglected PLA that could use a refresh.

[Qdeathstar]

If you do build, i'd be interested in a guide. Sounds like a fun project.

[nebbian]

I've noticed that since building a chamber for my printer, and using a flying extruder (which makes the filament spend some time in the heated area before going into the bowden cable), that bubbles are reduced a lot in the print. It's gotten to the point where if I use a roll that's well saturated with water, the first few layers show bubbles, but the rest don't. Which makes sense if you think that the first 15 cm of filament has gone straight from the roll into the bowden cable, and hasn't had the time to bake out at the top of the chamber.

At least that's my theory, it seems to be borne out in practice but I haven't really done any proper testing. Note that the filament only spends at most about a minute in the hot part before making its way into the extruder, and the chamber isn't very hot (40 degrees or so, measured near the bottom). Maybe it's hotter near the top of the chamber?

[gestalt73]

Ya know, honestly... only two. The first one was from when I was trying to find the right temp in my oven (ended up at 175f), and the second was from trying to find the rignt temp in the convection toaster oven I bought for this purpose (ended up at 150f)

A good way to tell if you need to or not is to run a calibration print before and after you dry the filament.
For whatever reason it looks like the Inland filament I'm getting from Microcenter benefits from it, which is interesting because the filament is vacuum sealed.
I'll run a calibration print before a client print job, and before drying, the filaments seem to want an extrusion multiplier of between 0.95 and 0.975. After they've spent a couple of hours in the oven, they're very close to 1.00

Cool info,

When i read this, i have to be honest that when you mentioned that the temp limit wasn't the limit of the PLA but of the spool, my first thought was about how many spools were lost to melt and how many stoves where damaged during the testing phase ^_^

[gestalt73]

Hey Jim, iirc Cope413 down in SoCal was talking about running a vacuum chamber, didn't hear much from him after the first flurry of posts.
http://forum.seemecnc.com/viewtopic.php ... 70&p=59513

I think he was even talking about offering it as a service, through MatterHackers I'd assume.

That would in theory be the best and fastest way to get the water out. Is there anyone out there that's successfully using a vacuum chamber?

I've been sitting on the parts to make a heated vacuum chamber for my filament dryer. As I understand it, the trick is to get the filament up to temperature first, because the vacuum isn't a good conductor of heat afterwards (which is why a Thermos works so well)... Maybe I'll get back to that project soon. I've got some neglected PLA that could use a refresh.

[rurwin]

Vacuum works by reducing the boiling point of the water. You should get the same or better effect as heating it. (175F is not boiling point) There will be other effects though; you won't get the annealing effect of heat and boiling the water too rapidly may crack the PLA. Probably nothing that would make a difference though.

Our oven's first graduation is 80C (175F), I don't trust it to be telling the truth that low, and I'd have to book time time on it. Four hours of plastic smells would not be looked on favourably.

Once I'm up to speed and put a couple of projects behind me, I'm thinking of building a desicator -- much lower heat but using dry air. Now I'll have to make sure it's big enough to fit a spool of filament.

[RocketMagnet]

In my experience with ABS you don't need to bake it if you store it "correctly". Simply placing the spools in air tight containers (a homebrew desiccator) with a layer of indicating silica beads in the base works 100%.. absolutely no heat required though a warm room helps. Never ever had any wet filament issues when I store my filament this way .. even if it's been left out long term some time in the Silica containers dries it out. If I had obviously wet performing filament by being lazy and leaving it on the printer for a couple of weeks (spitting etc) and I needed it "now" then careful heating can be used to speed up the process.. but in practice I've never had to resort to heating ABS ever.

I do keep my PLA in the same containers but never print with it anymore so not sure but I never had any water related issues when I did. Nylon is a different case and will benefit from oven drying IMO.. PLA and definitely ABS don't need the heat.

I recall the discussions around Vacuum drying of Nylon filaments.. I suspected at the time it would be more difficult than it sounds due to heat transfer issues (It's a Vacuum ) and the potential to damage the filament (lots of holes blown in your filament) when you apply the Vacuum, when you start to consider different materials and quality fluctuations it's very complex imo. As said above why bother with Vacuum drying ABS/PLA where is the benefit as a simple homebrew desiccator works, in fact you may make actually the filament perform worse?