Two pieces of advice circulate, and they are both wrong.
"Dry everything before every print" wastes hours on materials that were never wet enough to matter. "Drying is overblown, I have printed for years without a dryer" is true right up until someone prints nylon.
The materials data on this site settles most of it. Twenty-seven of the thirty-five filaments here carry a drying requirement. The eight that do not are PLA, PLA+, silk PLA, matte PLA, high-speed PLA, glow-in-the-dark PLA, metal-filled PLA and polypropylene — six of the seven PLA grades, one filled PLA, and one outlier.
If you print PLA and nothing else, the sceptics are describing your experience accurately. The moment you buy anything else, they are not.
Three different problems wearing one name
"Wet filament" covers three failure modes with different mechanisms, different consequences and different urgency. Conflating them is why the advice is so contradictory.
Chemical damage. In polyesters and polyamides, water in the melt attacks the polymer chains directly. The chains get shorter, the material gets weaker, and no amount of subsequent drying puts them back — the part is permanently compromised, and it does not necessarily look wrong. This is the serious one.
Steam. Absorbed water flashes to vapour at the nozzle, blowing bubbles in the extrusion. You hear it popping and crackling, the surface goes rough and hairy, and the layers bond badly because half the interface is a void. Ugly, obvious, and completely reversible by drying.
Additive misbehaviour. Some filaments carry a filler that holds water even when the base polymer does not, and the filler causes its own trouble.
Where it is chemistry, and therefore not optional
The materials where moisture does structural harm, with the schedules stored here:
| Material | Dry at | For |
|---|---|---|
| Nylon 6 and the other polyamides | 80 °C | 8–12 h |
| Polycarbonate | 90 °C | 6–12 h |
| Unmodified PET | 70 °C | 6–8 h |
| PPA | 100 °C | 8–16 h |
| ULTEM and PEEK | 120 °C | 8–24 h |
Nylon 6 is the most moisture-sensitive material catalogued here, by a wide margin. The timescale on which it goes bad is not weeks or months; leave the reel on the bench in a damp workshop and by morning it will print visibly worse than it did the previous evening. That is the number that surprises people.
Polycarbonate prints cloudy and weak when wet, and the cloudiness is the visible symptom of a strength loss you cannot see. For PET the note in the data is blunt — non-negotiable — because hydrolysis in the melt is the normal failure mode rather than an edge case.
For these materials the correct mental model is not "drying improves the print". It is "printing wet destroys the part".
Where it is steam, and therefore dramatic and fixable
PETG is the material most people meet this on. It absorbs quickly, and the popping-and-stringing complaint that dominates every PETG thread ever written turns out, more often than not, to be a moisture problem rather than a retraction one. The schedule stored here is a few hours at 65 °C. TPU is the same story: wet TPU bubbles and prints furry, which is very common and very fixable.
PVA deserves its own line because it is a special case of the same mechanism taken to an extreme. It is a polymer whose entire purpose is to dissolve in water, so its appetite for atmospheric humidity is not a side effect — it is the same property doing its job at the wrong moment. Leave the reel out of its bag and you have days, not months, before it is scrap. Buying PVA without somewhere sealed to keep it is buying a consumable with a clock already running.
If your symptom is popping and crackling at the nozzle or filament snapping in the feed path, this is the category you are in, and the fix is a few hours away.
Where it is the filler, not the polymer
Wood-filled PLA is the instructive case. Base PLA does not carry a drying requirement here; wood-filled PLA does — because wood flour is hygroscopic even when the PLA around it is not, and damp spools clog. A partial clog on a long unattended print is an expensive way to learn that.
LW-PLA has a different mechanism again. Moisture interferes with the chemical foaming reaction, so the expansion ratio becomes inconsistent and the part comes out neither the density nor the dimensions you calibrated for. Drying here is not about surface quality at all; it is about repeatability.
Silk PLA is the mildest case on the list: wet silk loses its sheen before it loses any strength. Cosmetic, and worth knowing if you bought it for the finish.
The one where drying really is optional
Polypropylene is barely hygroscopic, and the data says so plainly — one of the very few materials here where you can skip it with a clear conscience. PP will punish you in half a dozen other ways, but not this one.
Drying is not storage, and storage is the actual fix
Here is the part that undermines a lot of drying effort.
A dried spool starts re-absorbing the moment it leaves the dryer. For the materials at the top of this page, the re-absorption is fast enough that a spool dried on Sunday and printed on Wednesday is meaningfully wet again. For the polyamides the recommendation stored here is not "dry the spool" at all — it is to keep the reel inside a heated enclosure while it feeds, which is a different piece of equipment and a different habit.
So the sequence that actually pays is: dry once, then keep it dry. A sealed box with fresh desiccant is a permanent fix; a dryer used reactively is a repeated tax. Storing filament so it stays printable is the durable half of this, and what moisture does to filament explains the mechanism if you want the chemistry.
What your dehydrator cannot do
A food dehydrator tops out well below what several of the materials above require. Anything asking for 100 °C or more needs a proper high-temperature dryer or an oven you trust, and "close enough" is not a strategy — below the target the water simply does not leave the polymer fast enough to matter over a realistic cycle.
And do not overshoot at the other end. PLA softens between 52 °C and 60 °C, which is why its stored drying temperature is 45 °C: dry a PLA spool at nylon temperatures and you will fuse the coil to itself. How long to dry filament and what temperature to dry nylon have the per-material figures; drying filament properly covers the equipment.
What it costs, which is almost nothing
A standalone dryer averaging 50 W over a full twelve-hour nylon cycle uses 0.6 kWh — about $0.11 at the United States average rate.
Set that against a scrapped nylon print and there is no argument to have. What drying costs has the fuller comparison, but the shape of it is that the cheapest intervention available is also the one people skip, for no better reason than that it takes twelve hours of not printing.