Hygrometer Accuracy: How Accurate Are They Really?

Hygrometer accuracy is the specification people compare and the one they are least able to check. Two units sitting side by side on a shelf will often disagree by five points, and there is nothing on either display to say which is closer to the truth.

This guide covers what an accuracy figure actually claims, why real-world error is larger, how much precision different jobs genuinely need, what makes one hygrometer more accurate than another, and how to measure the error of the unit you already own.

How Accurate Are Hygrometers?

It depends almost entirely on which end of the market you are looking at, and the spread is enormous.

The top and bottom of that table are separated by more than an order of magnitude, and both are sold as hygrometers. That gap is the whole reason the accuracy question is worth asking at all.

What a ±3% Accuracy Claim Actually Means

An accuracy figure is a statement about a batch of sensors, measured at a specific temperature, over a specific humidity range, when new. Every one of those qualifiers is doing work.

It is usually a typical figure, not a worst case. Sensirion’s SHT4x datasheet lists ±1.8 %RH as the typical accuracy of the SHT40 and SHT41, and refers the maximum figure to a chart — because the error is not flat across the range. Most sensors are at their best in the middle of the scale and worse near the dry and wet extremes, which is exactly where humidors and dry winter houses live.

It applies at one temperature. Specifications are typically given around 25 °C. Take the same sensor into a 5 °C crawl space or a 35 °C attic and the error grows.

It applies when new. Sensors drift. The SHT4x datasheet gives long-term drift as typically less than 0.2 %RH per year, which is excellent — but that is a premium sensor under normal conditions. Cheap sensors in a dusty, smoky or chemically loaded environment drift considerably faster, and analog mechanisms drift as their hair or coil element ages and stretches.

And it is a claim about the sensor, not the product. A well-specified sensing element sealed in a case with poor airflow, mounted next to a warm display driver, reports the conditions inside its own housing. This is the single most common reason a hygrometer with an impressive specification reads badly in a room.

Response Time Is Part of Accuracy

A hygrometer that is eventually right but takes four hours to get there is wrong for most of the time you are looking at it.

The SHT4x datasheet gives a response time of τ63% = 4 seconds — four seconds to cover 63% of a step change, in moving air. An analog dial takes hours to do the same thing, and a digital unit in a sealed case with no ventilation slots takes far longer than its sensor does.

That matters wherever the humidity moves. Opening a humidor or a guitar case swaps the conditioned air for room air; a fast sensor shows the dip and the recovery, and a slow one averages the event into invisibility. It also means every reading taken shortly after moving an instrument between rooms is meaningless, regardless of how accurate the unit is.

How Much Accuracy Do You Actually Need?

Less than the marketing implies, for most jobs — and more than most people have, for a few.

The pattern is clear: accuracy matters in proportion to how narrow the acceptable band is and how close it sits to the ends of the scale. Watching a house stay between 30% and 50% is forgiving. Holding a humidor between 65% and 72% is not.

What Makes One Hygrometer More Accurate Than Another

Setting aside the sensor itself, four things separate a hygrometer that reads well from one that does not.

• Whether it can be calibrated. The most important trait, and the one least often printed on the packaging. A ±5% unit you can test and correct outperforms a ±2% unit you cannot.
• Airflow to the sensor. Vents or a mesh window over the sensing element. A sealed decorative case is a microclimate.
• A temperature reading alongside the humidity. Relative humidity is meaningless without it, and self-heating from the electronics is a real source of error in poorly designed units.
• Where it is put. Placement routinely causes more error than the instrument does — direct sun, an exterior wall, a vent or a draught will each move a reading further than the difference between a good and a mediocre sensor.

What matters less than it appears to: a decimal place on the display, which is resolution rather than accuracy; a very wide stated measuring range, when the space you are monitoring occupies twenty points of it; and “factory calibrated”, which describes an event before shipping. The underlying sensing technologies, and why some are inherently steadier than others, are covered in what a hygrometer is and how it works, and the routine for taking a reading you can trust is in how to measure humidity.

How to Test the Accuracy of Your Own Hygrometer

There is exactly one way to know how accurate your hygrometer is, and it costs nothing.

Seal the instrument in a bag or sealed container alongside a bottle cap of damp table salt — salt wetted to a slush, not dissolved in water — and leave it for eight to twelve hours. A saturated salt solution holds the air above it at a precisely known humidity. Greenspan’s 1977 study for the US National Bureau of Standards puts sodium chloride at 75.47% ± 0.14 at 20 °C, 75.29% ± 0.12 at 25 °C and 75.09% ± 0.11 at 30 °C. At room temperature, expect 75%.

Whatever it reads is your offset. Units with an adjustment screw or a calibration mode get corrected directly; the rest get a piece of tape reading “+4” and a mental correction for the rest of their lives. Both work. The full procedure and its failure cases are in how to calibrate a hygrometer.

Comparing two hygrometers against each other is not a test. It tells you they disagree, which you already knew, and gives no way to decide which is closer. Test each against salt instead.

Is There a “Most Accurate” Hygrometer?

In the laboratory sense, yes: chilled-mirror instruments determine dew point by directly detecting condensation on a cooled surface, and they are what national metrology institutes build their services around. NIST calibrates dew-point and mole-fraction hygrometers and thermohygrometers to an expanded relative uncertainty of under 0.3% for relative humidity. Nothing sold for a bookshelf approaches that, and nothing on a bookshelf needs to.

For household purposes the honest answer is that the most accurate hygrometer is a mid-priced digital unit that has been salt-tested and put in the right place. Calibration and placement move the real-world error more than any purchasing decision does. A ±2% instrument in direct sunlight on an exterior wall is worse than a ±5% instrument sitting where the air actually is.

If you are deciding what to buy rather than how far to trust what you own, our comparison of digital vs analog hygrometers covers the sensor trade-off, and hygrometer types and their uses covers the wider range of instruments.

The Bottom Line

Treat every accuracy figure on a box as a claim about a batch of sensors under laboratory conditions when new, and treat your own unit as unknown until you have salt-tested it. For general household use ±5 %RH is plenty, because the target band is twenty points wide. For a humidor, an instrument case or anything with a narrow tolerance, buy something calibratable, calibrate it, and retest it once a year.

Accuracy you have verified is worth more than accuracy you have paid for.

References

• Sensirion. Datasheet SHT4x — Humidity and Temperature Sensor.
• National Institute of Standards and Technology. Humidity Calibration Services at NIST.
• Greenspan, L. (1977). Humidity Fixed Points of Binary Saturated Aqueous Solutions. Journal of Research of the National Bureau of Standards — A. Physics and Chemistry, 81A(1).
• US Environmental Protection Agency. A Brief Guide to Mold, Moisture and Your Home.