As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
A radon detector is not a chemical sensor. It is a particle counter with a calibrated chamber. Radon gas diffuses into the chamber, decays, and fires off alpha particles that ionise air or strike a detector. The electronics count those events and convert them to a concentration in becquerels per cubic metre or picocuries per litre. Because radioactive decay is random, the count rate is noisy until many events accumulate, which is why long term averages matter more than any single reading. Products such as the Airthings Corentium Pro and the Safety Siren Pro4 are continuous monitors that do this counting automatically. The key limitation is that all of them estimate radon from what decays inside a small volume, so placement, ventilation and time are as important as the sensor.
What the chamber is actually counting: alpha particles, not radon molecules
Radon-222 is a noble gas produced by the decay of radium in soil and rock. It is chemically inert, which is why it migrates out of the ground and into buildings rather than binding to surfaces. The gas itself is not what damages lung tissue directly. The hazard comes from its short lived decay products, polonium, lead and bismuth isotopes, which attach to dust and can deposit in the airways and emit alpha particles there.
A radon detector works by letting ambient air diffuse into a small enclosed volume, often called an ionisation chamber or a detection cell. Inside that volume, some of the radon atoms decay. Each decay releases an alpha particle, a helium nucleus carrying two protons and two neutrons. In an ionisation chamber, the alpha particle knocks electrons off air molecules along its track, creating a small pulse of charge that is collected by an electrode and counted. In a scintillation cell, the alpha particle excites a phosphor that emits light, and a photodetector counts the flashes.
The important point is that the instrument is not measuring radon concentration directly. It is measuring a count rate of decay events inside a known volume, then applying a calibration factor to estimate the concentration in the surrounding air. That calibration is what separates a real monitor from a toy. It also means the reading is statistical: radioactive decay is random, so the count in any short interval fluctuates around a mean.
The GZAIR listing for the GZAIR 10-Min states a sensitivity of 30 counts per hour per picocurie per litre and describes a pulse ionisation chamber. That figure is a statement about how many counts the chamber produces for a given concentration, which is a measure of how efficiently it converts decays into a signal. Whether that number is meaningful to a buyer depends on the calibration behind it, and the listing does not publish a calibration uncertainty.
Why the first number is not the answer: statistics of a very small signal
Radon concentrations in homes are low in absolute terms. A typical indoor level might be tens of becquerels per cubic metre, which corresponds to only a few decays per second in a chamber of a litre or so. Counting a handful of events per minute means the random variation is large. If you count ten events in a minute, the statistical uncertainty is roughly the square root of ten, about three, so the reading could plausibly be anywhere from seven to thirteen. That is a thirty percent swing from randomness alone.
This is why every credible radon protocol asks for a long measurement. The uncertainty of a count falls as the square root of the number of counts, so to halve the uncertainty you need four times the counting time. A device that reports a number after ten minutes is reporting a very noisy estimate. The X-Sense listing for the X-Sense XR0B-SR claims results in 24 hours and accuracy reaching 95 percent after seven days or more. That is consistent with the physics: a week of counting gives enough events for a useful average, but a single day is still a short sample. The listing does not publish the confidence interval behind the 95 percent figure, so treat it as a manufacturer claim rather than a measured specification.
The GZAIR listing for the GZAIR 10-Min claims reliable data after one hour and a refresh every ten minutes. A one hour average can be useful for spotting a gross problem, such as a spike when a sump pump runs or a window is opened, but it is not a substitute for a multi day measurement. The listing does not state the statistical uncertainty of its one hour result.
| Measurement period | What it can tell you | What it cannot tell you |
|---|---|---|
| Minutes to one hour | Whether there is a large, obvious source nearby; short term trends from ventilation or weather | Your long term exposure; whether the home needs mitigation |
| One day to one week | A rough indication of the level; useful for screening and for checking a mitigation system | A reliable annual average; compliance with a long term action level |
| One to three months | The basis of most national protocols for deciding whether to mitigate | Seasonal variation, which needs a full year of data |
The table is not a product specification. It is the arithmetic of counting. Any instrument, including the Airthings Corentium Pro or the Safety Siren Pro4, is subject to the same statistics. The difference between models is how well they manage the chamber and the calibration, not whether they can beat the randomness of decay.
The chamber is only half the instrument: diffusion, humidity and the calibration nobody sees
For the detector to work, radon must get into the chamber. Most continuous monitors rely on passive diffusion through a filter or a membrane that keeps dust and radon progeny out while letting the gas through. That filter matters. If it clogs with dust, the response time slows and the reading lags. If it is missing, the chamber can become contaminated with the decay products themselves, which changes the calibration.
Humidity is the second hidden variable. An ionisation chamber measures a tiny current, and water vapour can affect the mobility of ions and the leakage currents on insulators. A well designed instrument compensates for this, but the listing rarely tells you how. The Airthings Corentium Pro listing states that it is EMI proof and tamper resistant, which speaks to electrical noise and to the integrity of the measurement, but it does not publish a humidity specification. The X-Sense listing does not publish one either. Where a manufacturer does not publish a figure, the honest position is that the listing does not state it.
Calibration is the part buyers cannot see. A radon chamber can be calibrated against a reference atmosphere, and the calibration factor is what turns counts into becquerels per cubic metre. The GZAIR listing says the sensors have been calibrated and that the device is lab free, meaning no lab fee for the user. It does not state the calibration uncertainty or the interval between recalibrations. The Airthings listing states that the unit is fully AARST/NRPP certified, which is a statement about the device and the organisation behind it meeting a recognised professional standard for radon measurement. That is a stronger claim than a calibration statement alone, because it implies a documented quality system.
For a homeowner, the practical consequence is simple. A cheap uncalibrated sensor can be wrong by a factor that matters, and you will not know. A certified device with a documented calibration gives you a number you can act on. The difference in cost is real, but it buys you the ability to trust the number.
What the display hides: averages, alarms and the difference between a monitor and a test
Most home radon devices show a short term average and a long term average. The short term average is usually a rolling window of a day or a week. The long term average is a rolling window of weeks or months, or a cumulative average since the device was reset. The distinction matters because radon varies with weather, season, heating and ventilation. A closed house in winter can read several times higher than the same house in summer with windows open.
The Safety Siren Pro4 listing states that it provides real time readings plus short term and long term averages, and that it has three selectable audible alarm modes plus a visual alert. That is a monitor with an alarm function, not a substitute for a formal test. An alarm is useful for catching a sustained elevation, but it is triggered by a threshold you or the manufacturer set, and the threshold may not match the action level used in your country. The listing does not publish the default alarm threshold.
The X-Sense listing describes sound and light warnings when radon exceeds a threshold, and low power notifications. Again, the threshold is not published in the bullets. The GZAIR listing describes real time, one day and long term averages on screen and a multi dimensional view in the app, with over 508 days of data storage. That is a data logging monitor, which is useful for seeing patterns, but the listing does not state whether the stored data can be exported in a form that a mitigation contractor would accept.
The distinction that matters is between a monitor and a test. A test is a defined measurement protocol with a start and end date, a specified placement, and a result that is interpreted against a national or international reference level. A monitor is a continuous instrument that helps you understand your home. You can use a monitor to decide whether to order a formal test, and you can use it to check that a mitigation system is working. You should not use a monitor as the sole basis for a legal or contractual decision, because the protocol and the calibration chain are not the same.
Where the detector stops being the right instrument
A radon detector tells you the concentration in the air where it sits. It does not tell you where the radon is coming from. If you have a high reading, the next step is usually a diagnostic test by a qualified contractor, who will use a series of measurements and a pressure field extension test to find the entry route and design a mitigation system. No consumer monitor does that.
A radon detector also does not measure radon in water. If your water supply comes from a private well, radon can dissolve in water and be released when you shower or run a tap. That requires a water test, which is a different sample and a different laboratory method. The listing bullets for the products here do not mention water testing, and none of them is a water radon test.
Finally, a radon detector is not a general air quality monitor. It does not measure carbon dioxide, particulate matter, volatile organic compounds or humidity. Some multi sensor units combine radon with other sensors, but the radon channel is still a chamber counting decays. If your concern is general indoor air quality, a radon monitor is the wrong instrument, and the site has separate guidance on radon detectors for home use and on how to choose a radon monitor. If your concern is a basement, where radon often concentrates, the guidance on radon monitors for basements is more specific.
The honest answer to the question in the title is that a radon detector works by counting radioactive decays in a small chamber and converting the count rate into a concentration using a calibration. It is a good instrument for what it does. It is not a radon source finder, not a water test, and not a substitute for a formal measurement protocol when a decision with legal or financial consequences is being made.
The number is an estimate, and that is the point
Radon detection is one of the few areas of home measurement where the physics is genuinely simple and the interpretation is genuinely subtle. A detector counts alpha particles from decays inside a chamber. The count rate is low, so the reading is noisy until enough time has passed. Calibration and certification are what make the number trustworthy, and no listing here publishes a calibration uncertainty. Use a monitor to understand your home and to check a mitigation system. Use a formal test, with a defined protocol, when you need a result you can rely on. The instruments named here, the Airthings Corentium Pro, the X-Sense XR0B-SR, the GZAIR unit and the Safety Siren Pro4, are all continuous monitors of the same basic type. None of them changes the underlying statistics.
Frequently Asked Questions
Do radon detectors actually detect radon gas, or something else?
They detect the alpha particles produced when radon and its short lived decay products decay inside the detection chamber. The gas diffuses in passively. The instrument counts decay events and converts the count rate to a concentration using a calibration factor. So the sensor is a particle counter, not a chemical gas sensor.
Why does my detector show a high reading for an hour and then a lower one?
Radioactive decay is random, and the number of counts in a short interval is small. A one hour reading has a large statistical uncertainty, so it will jump around. A window being opened, a fan running or a change in weather can also change the true concentration quickly. The long term average is the number to watch.
Is a home radon monitor as good as a professional test?
Not for a decision that needs to be defensible. A professional test follows a defined protocol for placement, duration and calibration, and produces a result with a documented uncertainty. A consumer monitor is useful for screening and for checking a mitigation system, but the listing bullets here do not publish calibration uncertainty or a protocol statement, so it should not be the sole basis for a legal or contractual decision.
Can a radon detector tell me where the radon is coming from?
No. It measures concentration at one point. Finding the entry route requires a diagnostic visit by a qualified contractor, who uses pressure field measurements and a suite of tests. No consumer monitor in this list does that.
Does a radon detector measure radon in drinking water?
No. Radon in water requires a water sample and a laboratory method. The products listed here are air monitors. If you have a private well, ask for a water radon test separately.
What does AARST/NRPP certification mean on a radon monitor?
It means the device and the organisation behind it meet a recognised professional standard for radon measurement, including documented calibration and quality procedures. The Airthings Corentium Pro listing states that it is fully AARST/NRPP certified. That is a stronger claim than a bare calibration statement, but the listing does not publish the calibration uncertainty, so the certification tells you the process exists rather than giving you a number.