Moisture is the most misdiagnosed problem in 3D printing. It gets blamed on retraction settings, on the printer, on the supplier — and the actual cause is that the material absorbed water from the air, either before you received it or after.
Key facts
- Moisture reduces part strength, not just surface quality. Water turning to steam in the melt zone disrupts layer bonding, and interlayer strength is usually where a part fails.
- PA (nylon), TPU and PETG absorb moisture fastest. PLA is the most forgiving common filament but is not immune.
- Popping or crackling during extrusion is close to conclusive. That sound is water flashing to steam inside the hotend.
- A sealed bag is not indefinite protection. Bags leak, desiccant saturates, and material can arrive already wet if it was packed humid.
- Drying is a process step for engineering grades, not a remedial action. For qualified parts it should be recorded like any other parameter.
How to tell
Wet filament announces itself in several ways, and the audible one is the most reliable.
| Sign | How diagnostic it is | What is happening |
|---|---|---|
| Popping, crackling or hissing at the nozzle | Near conclusive | Absorbed water flashing to steam in the melt zone |
| Visible steam or wisps at the nozzle | Strong | Same mechanism, more advanced |
| Rough, hairy or pitted surface | Strong when combined with sound | Steam disrupting the extrudate as it leaves the nozzle |
| Stringing that resists retraction tuning | Moderate | Moisture reduces melt viscosity and promotes oozing |
| Weak parts splitting along layers | Moderate | Steam voids preventing proper interlayer fusion |
| Inconsistent extrusion width | Weak on its own | Could equally be a worn nozzle or a mechanical issue |
| Filament snapping on the spool | Material dependent | Some polymers embrittle with prolonged moisture exposure |
Why it matters more than it looks
The visible symptoms are cosmetic, which is why moisture is often treated as a finish problem. The mechanical consequence is more serious and much less visible.
When water in the filament vaporises inside the hotend, it creates voids and disturbs the flow at exactly the moment the new layer should be fusing to the one beneath it. The result is reduced interlayer adhesion — and since printed parts are anisotropic, with the interlayer plane already the weakest direction, moisture attacks the property that was least able to spare it.
For a decorative print this is irrelevant. For a functional part it means a component that passed visual inspection and will fail earlier than designed, in service, along the layer lines. This is why industrial buyers treat drying as a recorded process parameter rather than a habit.
Which materials, how fast
| Material | Moisture sensitivity | Practical implication |
|---|---|---|
| PA / nylon | Very high | Can absorb enough to affect printing within hours of exposure. Dry box during printing is effectively mandatory |
| TPU | High | Prints badly when wet and is already the most demanding material to feed |
| PETG | High | The most common source of "why is it stringing" complaints in workshops |
| PC / PCTG | High | Dry before every run; wet PC also loses clarity |
| Filled grades (CF / GF) | Follows base polymer | A PA-CF is as hygroscopic as the PA it is built on |
| ABS / ASA | Moderate | Noticeable but slower to develop |
| PLA | Lowest of the common filaments | Still absorbs over months; the "PLA never needs drying" claim is overstated |
Drying guidance
General starting points. The manufacturer's datasheet for the specific grade takes precedence — particularly on temperature, since drying too hot can soften or fuse the spool.
| Material | Typical drying temperature | Typical duration | Note |
|---|---|---|---|
| PLA | 40–50 °C | 4–6 h | Stay well below the glass transition or the spool may deform |
| PETG | 60–65 °C | 4–6 h | Very responsive to drying |
| ABS / ASA | 60–70 °C | 4–6 h | — |
| TPU | 50–60 °C | 4–8 h | Lower temperature, longer time |
| PA / nylon | 70–80 °C | 8–12 h | Longest cycle; keep in a dry box while printing |
| PC / PCTG | 70–80 °C | 6–10 h | — |
Storage that works at production scale
Drying repeatedly is a symptom of a storage problem. At production volume the aim is that material never gets wet in the first place.
- Keep spools sealed until use, with desiccant, in the original vacuum packaging where possible.
- Treat desiccant as consumable. Saturated silica gel is doing nothing; use indicating desiccant so the state is visible, and regenerate or replace it.
- Print from a dry box for PA, TPU and long PETG jobs. A multi-day print in an open workshop will absorb moisture as it runs.
- Buy to consumption rate. Large purchases that sit open for months are a false economy for hygroscopic materials — a point that particularly affects institutional buyers spending year-end budget.
- Check packaging integrity at goods-in. A failed vacuum seal in transit means the material arrived wet, which is a supplier conversation rather than a workshop one.
Commonly mistaken for other problems
| Blamed on | Actually moisture when… |
|---|---|
| Retraction settings | Stringing persists across a wide retraction range and the filament has been open a while |
| Nozzle temperature | A temperature tower shows poor surface at every temperature |
| A worn nozzle | A new nozzle changes nothing and there is audible popping |
| Bad supplier batch | The same spool prints cleanly after drying — it was storage, not manufacture |
| The printer | A different printer produces the same defect with the same spool |
FAQ
How do I know if my filament is wet?
The most reliable indicator is sound: popping, crackling or hissing at the nozzle during extrusion is water flashing to steam and is close to conclusive. Supporting signs include visible steam, rough or hairy surfaces, and stringing that does not respond to retraction tuning. The fastest confirmation is to dry a sample from the same spool and reprint the identical file.
Does wet filament make parts weaker?
Yes. Water vaporising in the melt zone disrupts fusion between layers, reducing interlayer adhesion — which is already the weakest direction in a printed part. The visible surface defects are cosmetic, but the strength loss affects the plane most likely to become the failure surface, so a part can look acceptable and fail early in service.
What temperature should I dry filament at?
As general starting points: PLA at 40–50 °C for 4–6 hours, PETG at 60–65 °C for 4–6 hours, ABS and ASA at 60–70 °C, TPU at 50–60 °C for up to 8 hours, and nylon at 70–80 °C for 8–12 hours. Always check the manufacturer's datasheet for the specific grade, since drying above the glass transition temperature can deform the spool.
Can I dry filament in a kitchen oven?
It is risky. Most domestic ovens lack accurate control at these low temperatures and cycle well above the setpoint, which can soften or fuse a spool and ruin it. A purpose-made filament dryer or a food dehydrator with reliable low-temperature control is safer. Drying with the spool still sealed in its bag achieves nothing, since the moisture has no route out.
Does PLA need drying?
Less often than other filaments, but the claim that PLA never needs drying is overstated. It absorbs moisture more slowly than PETG, nylon or TPU, yet a spool left open in a humid workshop for months will show the same symptoms. If PLA is printing worse than it used to for no other apparent reason, drying is worth trying before more complex diagnoses.
How should filament be stored in a production environment?
Keep spools sealed with indicating desiccant until use, print hygroscopic materials from a dry box, and buy to consumption rate rather than stockpiling material that will sit open. Treat desiccant as a consumable that needs regenerating, and check packaging integrity at goods-in — a vacuum seal that failed in transit means the material arrived wet.