Filament storage hygrometers: reading and checking measurements

What the display inside the box is actually measuring

The hygrometer reports the relative humidity of the air around your spool, as a percentage, along with the temperature. It does not measure how much water the filament has absorbed. Drop a spool that is already soaked into a dry box and the display will eventually show dry air while the plastic stays wet. Prusa's documentation is explicit here: the dry box slows moisture absorption using silica gel packets, but it does not dry filament. A spool that is already moist needs a dryer first, which we cover in our guide to drying filament.

Relative humidity also depends on temperature. Warm the same air and the percentage drops even though the amount of water in it has not changed. That is why the display always pairs the two numbers. If your box sits next to an enclosed printer that heats up during a job, the reading will drift over the day without any real change in the box's water vapor. Reading a hygrometer means reading a temperature and humidity pair, at one moment, in one specific volume of air. Keeping that in mind will save you from overreacting to a number that ticks up or down.

How to read the accuracy the manufacturer publishes

To interpret a spec sheet, look for three things: the measurement range, the accuracy, and the display resolution. According to Prusa's documentation, the TPM40 hygrometer that ships with the USS Drybox covers 10 to 99 percent RH and minus 50 to plus 70 degrees Celsius, with an accuracy of plus or minus 3 percent RH and plus or minus 1 degree Celsius. It runs on two LR44 button cells with a stated battery life of around two years.

As an illustration of the nominal band, 20 percent with ±3 points gives 17–23 percent. This assumes applicable specification conditions and does not certify actual box humidity. Resolution and accuracy remain distinct; two displays may disagree without that comparison alone establishing a fault. TPM40 is specified between 10 and 99 percent RH, so Prusa's ideal ASA target below 10 percent lies below that range. For another sensor model, check its own limits instead of applying the TPM40 specification.

Where to put the sensor and when to read it

Placement changes the number. A sensor pressed against a wall, tucked under the lid, or sitting right next to the silica gel packet sees local air that may not represent what surrounds the spool. In the USS Drybox, the hygrometer has a position designed into the container. If you are building your own box, put the sensor at mid-height, clear of the walls, away from the desiccant, and not touching the filament.

Timing matters just as much. Prusa's documentation notes the humidity value can take up to one hour to drop and stabilize after the box is closed. Reading it two minutes after you shut the lid tells you nothing useful. Prusa also recommends visually inspecting the seal every time, avoiding unnecessary opening, and making sure the small side covers used to rotate the spool are properly closed afterward, since frequent or improper opening raises internal humidity. A good habit is to take the reading at the same time each day, say first thing in the morning before anything is opened, and jot down both humidity and temperature. A series of readings under matching conditions is worth far more than one isolated glance. For setting up the rest of your workspace, see our page on essential accessories.

What the NIST calibration setup illustrates

Calibrating a hygrometer means comparing it to a reference whose uncertainty is known and much smaller than the device's own. NIST, the U.S. national metrology institute, describes the process: the instrument goes into a test chamber where air with an accurately known moisture content, produced by a humidity generator, flows through at constant and uniform temperature and pressure. The expanded uncertainty NIST states for relative humidity in that setup is 0.5 percent of the value. That is a different universe from the plus or minus 3 percent of a dry box display.

This matters so you do not fool yourself. Setting two hygrometers side by side and declaring the one that matches the other to be correct is not calibration. Both could be off in the same direction. Likewise, a kitchen-table method does not create a reference of known uncertainty: without a controlled chamber and a reference instrument, you cannot claim a specific offset. NIST's capability is mentioned here as a benchmark for what traceable measurement demands, not as a practical path for a home shop and not as an endorsement of any inexpensive hygrometer. For filament storage, the realistic ambition is consistency between sensors and over time, not absolute truth.

Compare trends without claiming calibration

For an informal comparison, place sensors in the same air volume without contact with filament or desiccant and record temperature too. Follow each device's placement and settling instructions. Prusa's possible one-hour settling period concerns the USS Drybox here; it is not a universal period for every container or sensor. Repeating observations under comparable conditions can show trends without creating a reference of known humidity.

A persistent difference between sensors is a reason to investigate, not proof that either is outside its specification. Local conditions, settling and uncertainties matter. Even matching readings can share an error. Do not correct one device from another without a suitable reference. Record the model and conditions with each reading; Prusa says to replace the TPM40 batteries if the display stops working. The filament dryer guide separates this air measurement from any treatment the material may require.

Applying the error band to storage thresholds

Thresholds only mean something when you read them through the sensor's accuracy. According to the table in Prusa's USS Drybox documentation, PLA should stay below 30 percent RH with short-term tolerance up to 40 percent, ABS below 30 percent and ideally below 15, ASA below 30 and ideally below 10, while PETG, TPU, PVA, PC, and BVOH should stay below 20 percent, with TPU ideally under 15. Prusa says to replace or regenerate the silica gel when the RH exceeds the maximum recommended value for the material.

With a sensor rated at plus or minus 3 percent, a display of 19 percent for PETG does not guarantee that actual humidity is below 20 percent. A margin below a threshold can support a decision, but does not certify humidity or replace the specification conditions. TPM40 is specified from 10 percent upward; a lower reading cannot quantify humidity within that published range. An instrument without declared accuracy does not support this nominal-band calculation. For more on moisture-sensitive grades, see PETG and nylon.

Sources and limits

• Hygrometers | NIST Lock
• Prusa USS Drybox | Prusa Knowledge Base Prusa Prusa Prusa

We organize manufacturer and community documentation; we have not measured these products ourselves. The manual for your exact model takes precedence over general guidance.