As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
The homemade hygrometer that produces a defensible number is a wet and dry bulb psychrometer: two matched thermometers, one with a wet wick over its bulb, air moving steadily across both, and a short calculation that turns the temperature difference into relative humidity. The hair tension version and the pine cone version demonstrate the principle without measuring it, and a bare sensor module wired to a microcontroller gives you a reading with no calibration behind it. All three are worth building for different reasons.
The four devices people mean by a homemade hygrometer
The phrase covers four distinct projects, and most disappointment comes from building one while expecting the behaviour of another.
- The psychrometer. Two thermometers, a cotton wick, distilled water and airflow. It is the only version on this list whose output traces back to physical constants rather than to a guess, and it is the method national weather services used for a century.
- The hair tension hygrometer. A human or horse hair under light tension, lengthening as it takes up moisture, moving a pointer through a lever or a pulley. Beautiful, historically important, and a demonstration rather than a measurement.
- The sensor module build. A packaged capacitive sensor on a small board, read by a microcontroller, displayed on a screen or pushed to a network. The sensing element was made and characterised by somebody else, bought without documentation, and the build is the enclosure and the software.
- The condensation or dew point improvisation. Chilling a polished surface until dew appears and reading the temperature at which it does. The principle is the gold standard for humidity metrology, and the kitchen version is limited by how precisely you can spot the onset of dew.
Wet bulb and dry bulb and the sum that connects them
Water evaporating from the wick takes latent heat with it and cools the wet bulb below air temperature. The drier the air, the faster evaporation proceeds and the larger the gap. That gap, called the wet bulb depression, is the measurement. Converting it into relative humidity needs two things: the saturation vapour pressure at each of the two temperatures, and the psychrometric constant that relates the depression to the drop in vapour pressure.
Worked through with a dry bulb at 22 degrees Celsius and a wet bulb at 16 degrees Celsius, at normal atmospheric pressure, the steps are short.
| Step | Working | Result |
|---|---|---|
| Saturation vapour pressure at the dry bulb | From the standard curve at 22 degrees Celsius | About 26.4 hectopascals |
| Saturation vapour pressure at the wet bulb | From the same curve at 16 degrees Celsius | About 18.1 hectopascals |
| Correction for the depression | 6 kelvin times 0.000666 per kelvin times 1013 hectopascals | About 4.0 hectopascals |
| Actual vapour pressure in the air | 18.1 minus 4.0 | About 14.1 hectopascals |
| Relative humidity | 14.1 divided by 26.4 | About 53 percent |
Printed psychrometric tables and sliding charts do exactly this and hand you the last line. The reason to see the arithmetic is the sensitivity it exposes. One degree of error in the wet bulb moves the correction term by about 0.67 hectopascals, which moves the answer by roughly two and a half percentage points at these temperatures. Two thermometers that disagree by a degree when both are dry will therefore produce a humidity result several points out before evaporation has even started, which makes the matching of the pair the most important part of the build.
Building a psychrometer that is worth trusting
Start by proving the pair. Hold both thermometers dry in the same still air for twenty minutes and read them. Any difference is an offset to record and apply, and if the difference wanders there is no point continuing with those instruments. Then the mechanical requirements, each of which exists to remove a specific error.
- A thin, clean wick. Cotton shoelace or muslin, a close fit over the bulb, long enough to wick from a reservoir. A thick or stiff wick evaporates slower and reads high.
- Distilled water only. Minerals from tap water build up in the fibres and reduce evaporation over weeks, so the instrument reads progressively damper than the air.
- Air movement across both bulbs. The traditional answer is a sling, whirled on a cord; the convenient answer is a small fan drawing air past both. Several metres per second is the target, and consistency matters more than the exact figure.
- Shade and shielding. Both bulbs must see the same radiant environment. Sunlight or a nearby radiator on one bulb and not the other corrupts the depression directly.
- Patience. Read when the wet bulb stops falling, not when the first number appears. A slung psychrometer takes a minute or two of whirling with reads in between until the value stabilises.
Done properly this is a real instrument, and it has a property no electronic hygrometer has: nothing in it drifts. The calibration lives in the physics and in the two thermometers, which is why a psychrometer is still a sensible way to check an electronic hygrometer rather than the other way around.
Where the hair and pine cone versions stop
A hair lengthens by a percent or two across the full range from dry to saturated. To make that visible you need mechanical gain of a hundred or more, through a lever, a drum or a pulley, and every bit of gain multiplies friction and backlash along with the signal. The result moves in the right direction, lags by hours, reads differently going up than coming down, and changes its behaviour as the hair ages. A pine cone or a twisted seed head does the same thing with less control.
These are excellent teaching objects precisely because the failure modes are visible. Hysteresis, drift, nonlinearity and temperature sensitivity are abstract words until you have watched a hair hygrometer take half a day to catch up with a room. Build one to understand what a sensor specification is promising. Do not use one to decide whether a cellar needs a dehumidifier.
Wiring a sensor module and what it does not solve
A packaged capacitive sensor on a breakout board, read over a two wire bus, is a weekend project with a satisfying result and one honest limitation: the measurement quality was decided by whoever made the sensor, and you have bought it without documentation. The parts you still have to solve are the parts that distinguish a project from an instrument.
Self heating comes first. A regulator, a radio and a bright display inside one small case raise the air temperature at the sensor, and because relative humidity falls as temperature rises, the reading comes out low for reasons that have nothing to do with the sensor. Finished products spend real design effort on this. The Govee H5074 mini humidity sensor states plus or minus three percent relative humidity and plus or minus 0.54 degrees Fahrenheit with a reading every two seconds, in a deliberately tiny enclosure with twenty days of onboard history and a comma separated export, and the small case is part of how it keeps the sensing element close to room air.
Then placement and multiplicity. One sensor answers for one spot, and damp problems are differential: a cold corner, a wall behind furniture, a crawl space. Building six nodes means building six of everything, whereas the GoveeLife mini sensor three pack with gateway states a gateway pairing up to ten sensors with a two year battery life per node and the same plus or minus three percent and plus or minus 0.3 degrees Celsius figures, which is the shape of the problem a homemade build has to reproduce from scratch. Adding a channel is harder still: the Temtop S1 Plus air quality monitor states temperature, humidity, particulate and an air quality index on a sixty day battery, and the particulate channel needs an optical chamber and a fan that no module on a breadboard supplies. None of these listings publish a long term drift figure or a response time for the humidity channel. Not stated in the listing bullets, which is the one place where a documented psychrometer genuinely beats all of them.
When building one is the right call
Build a psychrometer when you need a reference you can trust without sending anything away, when you want to settle an argument between two electronic units, or when you are teaching where a humidity number comes from. Build a hair hygrometer to see the physics misbehave. Build a sensor module project when the point is the logging, the network or the enclosure rather than the measurement, and accept the sensor on trust.
Buy instead when you need several points at once, when the readings have to be recorded while nobody is watching, or when the number has to be defensible to somebody else. That is not a comment on homemade instruments, it is a comment on scale and documentation. If the decision is between a build and a purchase, the specifications worth comparing are the tolerance, the response time and the drift, which is exactly the set examined in the Jedew hygrometer notes and the Allsun thermometer and hygrometer notes.
Frequently Asked Questions
How accurate can a homemade psychrometer be?
It is limited almost entirely by the two thermometers. With instruments resolving a tenth of a degree that agree with each other when dry, and proper airflow over the wick, the result is competitive with a consumer electronic hygrometer. With kitchen thermometers reading to the nearest degree, each degree of error in the depression moves the humidity answer by several percentage points.
Why does a psychrometer need airflow over the wet bulb?
Evaporation cools the wet bulb until the heat arriving from the air balances the heat carried away by evaporation. Still air next to the wick becomes saturated and evaporation slows, so the bulb never reaches the true wet bulb temperature. Moving air continuously replaces that layer, which is why instruments are either slung through the air or fitted with a fan.
Does a hair hygrometer work at all?
It responds to humidity, which is not the same as measuring it. A hair lengthens by a small percentage across the whole humidity range, the response is nonlinear, it drifts with temperature and age, and the mechanical amplification needed to make the movement visible also amplifies friction. It is a genuine demonstration of the principle and not an instrument.
Can I just connect a humidity sensor module to a microcontroller?
You can, and you will get a number within an hour. What you will not get is a calibration, a validated placement, protection from self heating by the board next to it, or any idea of the long term drift. Those are the parts that take the work, and they are the parts a finished instrument has already paid for.
What water should the wick be soaked in?
Distilled or deionised water, because dissolved minerals left behind as the water evaporates stiffen the wick and lower its evaporation rate, which biases the wet bulb reading upward and the humidity result with it. Replace the wick when it goes hard rather than rinsing it.