Took me weeks of “mystery” failures to figure this one out, so sharing in case it helps anyone.
Symptoms I was chasing: random stringing, zits/blobs, brittle prints, occasional popping sounds. Looked like a hardware problem — I even swapped nozzles.
Turned out my filament had absorbed moisture sitting out in a ~65% RH room.
Quick checklist I now run before touching any hardware:
Listen for popping/clicking during extrusion — the classic wet filament sign
Check surface quality — rough/stringy on an otherwise fine print profile
Swap in a fresh or dried spool — if the issue vanishes, it was the filament
Keep the same settings when you test, so you’re isolating one variable
These days I keep my spools sealed with desiccant and try to stay under ~20% RH — the “mystery” failures basically disappeared.
Anyone else chase a hardware ghost that turned out to be moisture? What were your symptoms?
It’s kind of sad here at times. Some almost rebuild their entire printer chasing wet filament. Wasted time and money and it still doesn’t solve their problems.
As an aside, you can test filament by putting it in a ziplock bag with a hygrometer and sealing it. After about 30 minutes at room temperature you can read the filament “humidity”.
For my printer/setup, 25% on PLA is my cutoff and 20% on PETG. At both of those there are minor water effects but it’s printable. I shoot for below 10% though.
Lots of info here on dry storage too. Storing in high humidity isn’t good for filament. (Note, this said too many dryings are bad for filament but as long as temperature isn’t excessive there may not be significant or maybe any damage. - ty @Bratling) Water diffuses across the walls of plastic bags without special coatings. New filament, even vacuum-bagged, can have too high of water content to print properly.
But for those suffering water issues, they are pretty easily dealt with. Good on you for sorting it! Many don’t.
Thanks! The ziplock + hygrometer test is a great trick — I’ll definitely use that instead of guessing. Your cutoffs (25% PLA / 20% PETG) are useful benchmarks too; I’ve been hovering around ~20% with my sealed box, so good to know that’s in a workable range. And yeah, the “rebuild the printer for wet filament” stories are wild — that’s exactly what I almost did before realizing it was moisture.
About the bag/box test though, hygrometers are a little slow to respond so you need to wait a bit to get a good reading.
Also, using desiccant to dry filament is pretty inefficient. It works but not on any reasonable time scale. Heat is really needed to get the water to let go of the filament.
Also, when using desiccant, it’s important to keep it more dry than however dry you want your filament to be. If you don’t, desiccant can even add water to filament. And you can see how wet it is the same way you can test filament.
What was I supposed to notice? Many don’t understand the mechanics of moisture content. For someone bootstrapping themselves into fixing water issues, he seems to have gotten things pretty well.
Now I see you deleted that. I’m even more confused now.
Really appreciate the detailed write-up — all three points land. The hygrometer lag explains some readings I chased early on that made no sense, and “desiccant drier than your target” is exactly the kind of detail you only learn the hard way. The heat point is the one I’m still chewing on — my sealed box keeps things dry once they’re dry, but I can see the argument for a proper dryer on anything that’s already absorbed moisture. Good stuff, thanks for sharing.
On heat, it’s the same reason clothes dryers heat. It helps put energy into the water which “evaporates” it. What it’s doing is getting the water molecules wiggling so hard that the attraction they have to the filament isn’t enough to keep it stuck down so it flies off and can be vented away.
Desiccant with filament at room temperature still dries to an extent because even at room temperature there is heat to make some water let go but not enough to get it all. Also the process is a bit slow without added heat.
So heat both speeds up drying as well as dries to lower moisture content than you can get without heat.
In storage in a box or bag, the purpose of desiccant is to scavenge water that leaks in. Nothing needs to be heated to help water let go because you are grabbing it out of the air instead of trying to get the water to let go of the filament.
As an aside, since you seem to be a water person, if you want your filament even drier than 20%, you can purge your filament dryer with dry air. It sweeps out the moisture and helps more water come off the filament faster. It also allows hitting lower moisture contents in your filament.
With a dry air purge in my filament dryer, when the filament cools, it will pull the humidity in a container below 10% RH. The filament alone.
Full disclosure - this is my model (there are two - Arid Air 1 & 2). 2 is smaller than 1 and uses a plastic bead column housing (Nalgene water bottle). 1 uses a glass column that has more optimal geometry but is a more involved print. 2 is better overall IMO.
That mechanism explanation makes it click — heat just makes the water let go, and desiccant’s job in storage is only to catch what leaks in. That reframes my whole setup: seal keeps it dry, desiccant catches the leaks.
And the Arid Air is a neat idea — using dry air to sweep instead of relying on heat alone. Might print it once I finally get my head around this moisture game. Appreciate the deep dive!
Tell me more about this. Why is it a problem? I could see if you are drying at too high a temperature and are going over the glass transition point, each time will make a spool of filament more likely to stick to itself and jam. But why would cycling through wet/dry a few times be a problem?
@77919919 I don’t mean to be rude, but your comment history is very chatGPT-like. Are you a real human?
Just to be clear, it’s okay if you are real but use ChatGPT! I have met people who use it to write their replies, for perfectly valid reasons. But ChatGPT has a very distinctive writing style, so just in case I want to ask if you are or are not a bot.
I don’t actually know that it is. I’ve just seen it repeated by others. The too high of humidity prt damaging filament is very real. But I don’t know for a fact repeated dryings also do damage.
I did a search and do find mention of removing plasticizers and modifiers/colorants with repeated dryings but as much saying the changes are small or nonexistant.
I’ll modify the original post. I’d have to say it looks to not be a big issue as long as temperatures aren’t higher than recommended.
Haha, fair question — English isn’t my first language, so I write replies carefully (and yes, sometimes with help from a translator), which probably comes across a bit stiff and “structured.” Definitely a human, just a careful one. Appreciate you asking rather than assuming!
The “change one variable” step is the key. I’d add a simple control: run the same short model with a known-good dry spool, keep the printer and profile fixed, then change only the suspect spool. Storage-box RH describes the environment, not necessarily the moisture inside the filament. For drying settings, I’d follow the material maker’s current guidance because PLA, PETG, nylon, and spool materials differ. A log of material, storage time, RH, dryer setting, and symptoms would make the checklist easier to reproduce.
Following up on this: In the CNC Kitchen review of the Sunlu i10 filament cabinet, he mentions the risk of degrading the filament through repeated heat-drying cycles. I don’t know if he has data on that, or is just repeating the common belief. But, it’s something.
Good additions. The “change one variable” control with a known-good spool is the part I actually use most when I’m not sure if it’s the filament or the printer. And yeah, the log idea would make this way more reproducible.
That’s the idea of a “blast” oven. Bambu is using a similar approach with their AMS by venting the air. I don’t know exactly how the performance compares with a blast oven, but its better than trying to dry filament in a sauna.
Interesting point. I haven’t tried the blast oven approach, but the sealed storage + desiccant route works well enough for me that I haven’t needed to push it. Curious how the blast oven compares long-term.
