Filament Drying Time Calculator
Scale a published drying time to your dryer and your filament — temperature enters exponentially where time enters linearly.
Time against dryer temperature
How dry is dry enough
The approach to equilibrium is exponential, so the last tenth costs more than the first nine.
Ten degrees short doesn't mean ten per cent longer
Temperature enters exponentially where time enters linearly. Diffusion follows Arrhenius, so a dryer running cool doesn't want a bit longer — it wants a different order of time. At a typical 40 kJ/mol activation energy, 65 °C dries 2.45× faster than 45. Across the plausible 30–50 kJ/mol range that factor is 1.96 to 3.06, so the exact multiplier is uncertain and the direction isn't. This tool reports the band rather than pretending to a single figure.
And time scales as the square of the radius. 2.85 mm filament takes 2.65× as long as 1.75 — the same 2.65 that appears in how much plastic it carries per millimetre, and for the same reason: both are d² quantities. It also means the spool geometry barely matters. What has to happen is water crossing the strand radius, not escaping the coil.
The glass transition produces an inversion worth knowing. The temperature you may dry at is capped by Tg — above it the spool softens and the coils fuse. PLA has the lowest Tg here, so it must be dried coolest and therefore slowest, despite absorbing the least water of anything on the list. Nylon absorbs several times more and tolerates 70 °C, so it diffuses fastest. The material that needs drying most isn't the one whose molecules move slowest.
What this does and doesn't do. The scaling is calculable from the diffusion equation; the absolute time isn't, because how much water a material holds and how dry is dry enough vary too much between grades. So take a published time as the anchor and use this to correct it — for a thicker filament, or for a dryer that won't reach the stated temperature. That's the question people actually have.
- A dryer is not an oven, and humidity is half the job. Diffusion carries water to the strand surface; something has to carry it away. In a sealed box saturated with what's already come out, nothing does — which is why a vented or desiccant dryer beats a hotter sealed one.
- The last tenth takes as long as the first nine. 90% dry is 0.335 in dimensionless time; 99% is 0.733. That's why "leave it overnight" is sound — the marginal hours are cheap when you're asleep.
- Weigh the spool if you want to know rather than guess. Nylon gains enough water to register on kitchen scales, and weighing before and after tells you both that drying worked and when it stopped making progress.
- Never dry above Tg. It's the one hard limit here, and it ruins the filament rather than merely deforming it — the coils weld and it comes off the spool oval or not at all.
How to use
- Start from a published time and temperature for your material.
- Measure what your dryer actually reaches rather than trusting the dial.
- Never set a temperature above the material glass transition.
- Weigh the spool before and after if you want to know rather than guess.
Frequently asked questions
How long should I dry filament for?
Start from the published figure for your material — roughly 5 hours at 45 C for PLA, 6 at 65 for PETG, 12 at 70 for nylon — then scale it for what your dryer actually does. The absolute time depends on how much water the material holds and how dry is dry enough, which vary by grade; how it SCALES with temperature and diameter is calculable.
Does a lower drying temperature just take proportionally longer?
No, and this is the thing most worth knowing. Diffusion follows an Arrhenius relation, so temperature enters exponentially where time enters linearly. At a typical activation energy, 65 C dries about 2.45 times faster than 45 — so a dryer running ten degrees under the published figure wants roughly half as long again, not ten per cent more.
Why does 2.85 mm filament take longer to dry?
Because moisture leaves by radial diffusion and the time scales with the SQUARE of the radius. A 2.85 mm strand takes 2.65 times as long as a 1.75 mm one — which is the same 2.65 that appears in how much plastic it carries per millimetre, and for the same reason: both are functions of diameter squared.
Does the size of the spool affect drying time?
Much less than people expect. What has to happen is water crossing the strand radius, not escaping the coil, so a full spool and a half-empty one dry at similar rates per unit of filament. What the spool does affect is airflow — a tightly wound coil in a sealed box has nowhere to send the moisture once it reaches the surface.
What temperature should I dry filament at?
As hot as the glass transition allows, which is the real constraint. Above Tg the spool softens and the coils weld to each other, which ruins the filament rather than merely deforming it. That is why the published temperatures differ so much between materials — 45 C for PLA against 80 for ABS is a statement about Tg, not about how much water there is to remove.
Why does PLA take so long to dry when it absorbs so little?
Because its glass transition is the lowest of the common filaments at around 60 C, so it must be dried at 45 or below — and at that temperature the water molecules simply move slowly. It is an inversion worth knowing: the material that absorbs least takes longest, because the temperature it tolerates is what actually sets the pace.
Which filament needs drying most urgently?
Nylon, by a wide margin — it absorbs several times what anything else does, and a spool left out overnight is measurably heavier. It is also among the quicker to dry, because its glass transition rises as it dries and it tolerates 70 C. The material that needs drying most is not the one whose molecules move slowest.
Can I dry filament in a kitchen oven?
With great care about temperature, and with the door cracked. Domestic ovens overshoot their set point substantially and cycle by tens of degrees, which is fine for a casserole and fatal to a PLA spool sitting fifteen degrees under its glass transition. If you do it, verify with a separate thermometer and keep well under the published temperature.
Is a sealed filament dryer better than a vented one?
Usually worse, which surprises people. Diffusion carries water to the strand surface and something has to carry it away from there — in a sealed box saturated with what has already come out, nothing does. A vent or a desiccant sachet outperforms another ten degrees, and cracking the lid of a sealed dryer for the last hour helps more than extending the time.
How dry is dry enough?
Ninety per cent of the removable moisture is the usual target and it is a reasonable one, because the approach to equilibrium is exponential — going from 90 to 99 per cent more than doubles the time for a further nine per cent of the water. That is why leaving it overnight is sound advice: the marginal hours are cheap while you are asleep.
How can I tell if my filament is wet?
Printing symptoms first — popping and crackling from the nozzle, stringing that was not there before, a rough or foamed surface, and layers that separate more easily than they should. Weighing is the only quantitative test available: nylon in particular gains enough water to register on kitchen scales, so weighing before and after tells you both that drying worked and when it stopped making progress.
Does drying filament restore its strength?
Largely, yes, if the damage is moisture rather than hydrolysis. Water in the melt causes bubbles and poor layer bonding, and removing it removes those. What drying cannot undo is chemical degradation — nylon and polycarbonate left wet and then printed hot can hydrolyse, which shortens the polymer chains permanently and no amount of drying afterwards recovers it.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.