How to check your air purifier is actually working
A purifier’s CADR was measured in a sealed test chamber at top speed. Your room has doors, a fan setting you can sleep through and a filter that has been running for months. A PM2.5 monitor and half an hour show what it delivers where you actually use it.
What a purifier changes, and what it does not
EPA’s technical summary on residential air cleaners describes the clean air delivery rate, or CADR, as the product of two things: how much air the unit moves and how well its filter removes particles from that air. So the reading that answers whether it works is a particle reading — PM2.5 on any home monitor.
CO2 is a different matter. It is a gas, a particle filter does not capture it, and it falls when a room is ventilated. A monitor that shows CO2 next to PM2.5 is useful for knowing when to open a window, but only the PM2.5 number tells you about the purifier.
The number to expect
In a closed room with the particle source stopped, a purifier removes the same share of what is left every minute, so the level halves at a steady pace. The halving time in minutes is about 0.69 × the room volume in cubic feet ÷ the CADR in cubic feet per minute.
According to EPA, AHAM’s room-size recommendation assumes clean air equal to four to five times the room’s volume every hour. At that rate the halving time works out at 8 to 10 minutes, whatever the size of the machine, as long as the room matches its rating.
A worked example: a 12 by 15 foot bedroom with an 8 foot ceiling holds 1,440 cubic feet. A unit rated at 141 cfm of smoke CADR should halve PM2.5 there in about 0.69 × 1,440 ÷ 141, or 7 minutes, at top speed.
Run the test
You need a PM2.5 monitor that updates at least once a minute, a room you can close and about half an hour. Note the readings on paper or use the monitor’s own history.
- Close doors and windows, switch off fans or central air that mix this room with the rest of the home, and put the monitor across the room from the purifier, off the floor.
- With the purifier off, wait until PM2.5 is well above normal for the room — straight after cooking often does it. Aim for at least 30 µg/m³ on the monitor.
- Leave the purifier off for 20 minutes and note the reading every 5 minutes. This is how fast the room clears on its own, through settling and air leaks.
- Switch the purifier to its highest setting and keep noting the reading every 5 minutes until it reaches the low teens.
- For each run, find how many minutes the reading took to halve. Use the stretch above about 15 µg/m³: near the bottom, a monitor’s own offset makes the fall look slower.
Working out what the purifier delivered
Turn each halving time into a removal rate by dividing 0.69 by the minutes. Subtract the purifier-off rate from the purifier-on rate and multiply by the room volume. The result is the clean air, in cubic feet per minute, that the purifier delivered in your room.
Example: in the 1,440 cubic foot bedroom, the reading halved in 40 minutes with the purifier off and in 6 minutes with it on. That gives (0.69 ÷ 6 − 0.69 ÷ 40) × 1,440, about 140 cfm, close to a 141 cfm rating. Had the purifier run taken 15 minutes, the same sum gives about 41 cfm.
Treat the result as approximate: no room mixes perfectly and particles keep arriving from outside. A figure near the rating means the unit is doing its job where you use it. One far below it points to the checks further down.
Why your monitor does not need to be exact
What this test needs from a monitor is readings that move in proportion to the real level. A monitor that reads a third low reads a third low at the start and at the halfway point too, so the halving time comes out the same.
That property is what AQ-SPEC, South Coast AQMD’s sensor evaluation program, measures when it runs monitors beside reference instruments. In its chamber the Qingping Air Monitor Lite tracked the reference at R² 0.99 while reading about a quarter to a third low; the Temtop M2000 2nd Generation tracked at above 0.99 while reading 30–45% high.
A fixed offset is different. A monitor that never reads below a few µg/m³ makes the last part of the fall look slower than it is, which is why the timing stops in the low teens.
If the reading falls slower than it should
Check the fan speed first: CADR is measured at the highest setting, and an auto mode may have throttled down. Then make sure the room is really closed, because an open door adds the volume of the rest of the home to the arithmetic.
Look at the filter. EPA notes that a filter clogged enough to restrict airflow cuts what the unit delivers. Make sure nothing blocks the intake or the outlet, and that the particle source has really stopped.
Questions people ask about this
Can I use the purifier’s own air quality sensor for this test?
Its reading can show the trend, but the sensor sits in the machine and samples the air right around it rather than the room as a whole. EPA’s technical summary also notes it found no studies of how those built-in sensors perform over time. A separate monitor across the room gives a fairer picture.
How often should I repeat the test?
Whenever something changes: after replacing the filter, after moving the purifier to another room, or when the filter indicator says all is well but you doubt it. A run with a fresh filter also gives you a baseline to compare against months later.
Does the test work in an open-plan space?
Only roughly. The arithmetic assumes the air mixes evenly and does not escape into other rooms. In an open plan, use the volume of the whole connected space and expect slower halving times than in a closed bedroom.
Where this comes from
Written from published standards and manufacturer documentation. We do not bench-test units, and nothing here is a health claim — where the honest answer is that something is not documented or not measured, the page says so.