Drying is the most over-prescribed remedy in desktop printing. It is offered for warping, for layer shifts, for bed adhesion and for prints that failed at three in the morning, and in most of those cases it is irrelevant. It is also, for a specific and well-defined set of symptoms, an almost complete cure — so the skill is knowing which list you are on before you commit six hours to a box.
The materials that actually need it
The stored drying requirement is not a matter of opinion, and it sorts filaments into three groups.
Always, before any serious print. Nylons of every kind, polycarbonate, PET, PVA and BVOH. These absorb water quickly enough that a spool left in open air overnight is measurably worse the next morning. PA6 is the extreme case on this site and it is not close.
Yes, and people skip it anyway. PETG, ABS, ASA, TPU, and every fibre-filled grade — including carbon-filled PLA, where the PLA barely cares but the fibre does. This is the group where drying converts a frustrating spool into an ordinary one.
Only when it has been badly stored. PLA and its cosmetic variants, and polypropylene, which is barely hygroscopic at all. A PLA spool that has been open on a shelf for a year is worth drying; one from a sealed bag is not.
The temperature ceiling that ruins spools
Every published schedule is a compromise between drying faster and softening the plastic. Drive the box too hot and the filament on the reel fuses into a solid block, or the reel itself deforms and the filament will not unwind.
The rule is that the drying temperature must stay clearly below the material's glass transition, and the stored schedules already respect it:
- PLA and its blends: 45 °C. Note that PLA's deflection range starts at 52 °C, so a dryer set to 60 °C — a common default — is inside the range where PLA softens. This is exactly how people weld a spool shut.
- PETG, ABS, ASA and TPU: 55–65 °C depending on the material.
- Nylons: 80 °C.
- Polycarbonate: 90 °C.
- ULTEM and PEEK: 120 °C, which no consumer dryer reaches.
If your dryer has one setting, it is a PETG dryer. Do not put PLA in it and walk away.
How long, and why the published hours vary
Water leaves a spool from the outside in, so the time depends on how deep the water got, not just on the material. The stored schedules give 4 to 6 hours for the easy materials and 8 to 12 for the nylons, and those figures assume a spool that is damp rather than saturated.
A spool that has been open in a humid room for months needs longer — often double. The way to find out is not to guess, which brings us to the only part of this page that really matters.
The check that proves it worked
Weigh the spool before and after, on a kitchen scale.
Absorbed water is mass. A badly-stored nylon spool will shed several grams in a dryer and a saturated one noticeably more; a PETG spool sheds less; a spool that was already dry sheds nothing. Write the figure on the reel with the date.
This single habit replaces every argument about drying times, because it converts a belief into a measurement. If the spool stopped losing weight after four hours, it was dry after four hours. If it is still losing weight at eight, keep going. If it lost nothing at all, the problem was never moisture and you have saved yourself from chasing it further.
The secondary check is the print itself: run a short retraction test before and after. If the popping and the fine hairs are gone, you had a wet spool. If they are unchanged, you have a retraction problem wearing moisture's clothes.
Equipment, honestly ranked
- A purpose-built filament dryer. Correct temperatures, a timer, and on better models a vent that lets the humid air out — which matters, because a sealed box just recirculates the water it evaporated. This is the sensible purchase if you print PETG or nylon regularly.
- A food dehydrator. The classic budget answer and a good one. Check the actual temperature with a thermometer rather than trusting the dial; many run well above their markings, which is fine for PETG and fatal for PLA.
- A kitchen oven. Workable but risky: domestic ovens overshoot their setpoint substantially on the way up, and the overshoot is where spools fuse. Only with an oven thermometer, only for materials with headroom, and never with the fan off.
- Printing from a heated dry box. Not drying at all, but for PA6 and PA6-CF it is the only thing that keeps a long print dry from start to finish. A spool dried and then left in room air for a twelve-hour print is wet again by the end.
Silica gel in a sealed box is storage, not drying — it maintains a dry spool and will not rescue a wet one in any useful time. That distinction is worth keeping straight, and storing filament so it stays printable covers the maintenance side.
What drying will not fix
Be clear about the boundary, because the time is not free.
Drying fixes popping and hissing at the nozzle, fine stringing that survives retraction tuning, a hazy or rough surface on a material that should be glossy, unexplained under-extrusion, and nylon that snaps on the reel. Those all trace directly to water flashing to steam in the melt zone or to the polymer chains being cut by it.
Drying does not fix warping, corner lift, layer shifts, bed adhesion, dimensional error, ringing, or a nozzle that has partially clogged with something solid. If you are dealing with one of those, the diagnosis is elsewhere — what moisture does to filament explains why the boundary falls exactly there, and it is a boundary set by chemistry rather than by convention.