Run the numbers on the air and the mystery goes away
Take a 14 by 10 by 6 inch interior, 40% full of contents, at 70 degrees and 70% humidity. The free air is 0.29 cubic feet and weighs about 9.9 grams. At that condition air carries 76.6 grains of moisture per pound of dry air, so those 9.9 grams of air are holding 0.11 grams of water — two drops.
Now open the lid. Even if every molecule of that air is replaced with room air at 45%, you have lost 0.04 grams. A four ounce charge of medium holds 113 grams. It refills the air in minutes, and the hygrometer needle that dips when you open the lid and recovers over the next hour is showing you exactly that.
So the openings are not the problem. At four air changes an hour, two lid openings a day are 1% of the total loss and the seal is the other 99%, because a seal leaks continuously whether you are there or not. One opening is worth about nine minutes of leakage.
Grains per pound, not percent
The driving force for moisture leaving a box is not the difference in relative humidity, it is the difference in absolute moisture content, measured in grains of water per pound of dry air. Relative humidity is relative to temperature, so two rooms at the same RH and different temperatures hold different amounts of water.
| Room condition | Grains per lb | Against 70°F / 70% inside | Days per 4 oz charge |
|---|---|---|---|
| 70 °F at 70% — inside the case | 76.6 | — | — |
| 72 °F at 50% | 58.3 | 18.3 driving loss | 45.0 |
| 70 °F at 45% | 48.9 | 27.7 driving loss | 29.7 |
| 68 °F at 30% | 30.3 | 46.3 driving loss | 17.8 |
| 78 °F at 60% | 86.1 | 9.5 the other way | no loss at all |
The bottom row is the summer case: in a warm humid room the box gains moisture rather than losing it, and the medium sits idle for weeks. The row above it is a heated house in February, where the refill interval falls from 45 days to 18. A case that behaves perfectly for six months and then drops ten points every winter is not faulty; it is being asked a question two and a half times harder.
Measure your seal instead of guessing at it
The leakage figure is the one input that decides the answer, and nobody knows theirs. The good news is that it is directly measurable with a kitchen scale that reads to a tenth of a gram.
Charge the medium fully, weigh it, put it in, close the box and leave it alone for a week. Weigh the medium again. Divide the grams lost by seven to get grams a day, then divide by the one-air-change figure the calculator reports — 0.039 grams at the default conditions — and divide again by 24. That is your air changes per hour, measured rather than assumed. Do it in winter, when the grains difference is largest and the weight change is easiest to read.
The wood is a slow, invisible reservoir
Wood in equilibrium with air at 70% relative humidity carries roughly 13% of its own weight as water, and that figure moves by around a tenth of a percent of the wood weight for each percentage point of RH. Three pounds of lining is therefore holding about 180 grams of water in total. The default here discounts the theoretical 1.4 grams per percentage point down to 1.05, because much of the wood responds far too slowly to matter day to day — and even that is ten times what the whole air volume holds.
This is why a well-made wooden case drifts slowly and a plastic one with the same seal does not, and it is why a new case needs a long seasoning period before it holds a reading: until the lining reaches equilibrium it is absorbing everything the medium puts out. It is also the mechanism behind the wood movement calculator, which handles the dimensional consequences of the same moisture exchange.
Related
For a whole room rather than a box, the humidifier sizing calculator does the same grains arithmetic at house scale, and the wine cellar humidity calculator adds the cooling coil that dries a cellar out. A single pair of conditions converts in the dew point calculator, and the growing-room version of the same physics is the VPD calculator. If the case has gone the other way and is too damp, the dehumidifier sizing calculator is the wrong scale but the right direction.
Questions people ask
How often should I refill a humidor?
The calculator gives the interval from your own numbers rather than a rule of thumb, because the honest answer varies by more than ten to one between a tight box in a humid room and a loose one in a heated house in winter. For the default case here — a 14 by 10 by 6 interior at 70% inside, a 45% room, four air changes an hour and a 4 ounce charge — it works out at 29.7 days, losing 3.8 grams a day. Take the room down to 30% humidity and the interval falls to 18.7 days; put the box in a 78 degree room at 60% and it loses nothing at all.
Does opening the lid ruin the humidity?
No, and the arithmetic is what proves it. The free air in a desktop case holds about 0.11 grams of water, two drops. Swapping 60% of it costs 0.04 grams against the 113 grams in a four ounce charge, and the medium replaces it within minutes. At four air changes an hour, one lid opening is worth nine minutes of leakage, and two openings a day come to 1% of the total loss. What you see on the hygrometer is a real dip and a real recovery, but if the humidity is genuinely falling, look at the seal rather than at your habits.
How do I work out my seal leakage in air changes per hour?
Weigh it. Charge the humidification medium, note its weight, close the box for a week without opening it, then weigh the medium again. Grams lost divided by seven gives grams a day. Divide that by the grams-per-air-change figure the calculator reports for your conditions — 0.039 grams at the defaults here — and divide again by 24. That is your air changes per hour. Do the test in winter, when the difference between inside and room moisture is largest and the weight change is easiest to read.
Why does the same box behave differently in summer and winter?
Because loss is driven by the difference in absolute moisture, not relative humidity. A 70 degree room at 45% holds 48.9 grains per pound; the same room at 30% holds 32.5. Against 76.6 grains inside, that is a driving difference of 27.7 in one case and 44.1 in the other, so the winter loss rate is about 60% faster and the refill interval falls from 29.7 days to 18.7. In a warm humid summer room the difference can reverse entirely and the box gains moisture.
Does the wood lining matter or is it decoration?
It matters more than the air does. Three pounds of wood in equilibrium at 70% humidity holds around 180 grams of water and exchanges roughly a gram for each percentage point of relative humidity — around ten times what the whole air volume holds. That is what makes the humidity in a wooden case drift slowly instead of stepping, and it is also why a new case takes weeks to settle: until the lining reaches equilibrium it absorbs whatever the medium puts out.