Understanding CADR and room size: the sizing maths, shown

CADR stands for clean air delivery rate. It is the single most useful number on an air purifier, and it is routinely misread as a filter rating. It is not. It is a rate: the volume of air per hour that the machine delivers back to the room with the target pollutant removed from it.

The standard behind it is AHAM AC-1. The test is deliberately blunt. A sealed chamber of about 28.5 m³ is loaded with a known concentration of a test contaminant. The decay of that concentration is measured twice: once with the chamber sealed and the machine off, to capture how fast the particles settle and stick to surfaces on their own, and once with the machine running. CADR is the difference between those two decay rates, expressed as a volume flow.

Two consequences fall straight out of that definition. First, natural decay is subtracted, so CADR is the machine’s own contribution and nothing else. Second, because it is airflow multiplied by single-pass efficiency, a big fan behind a mediocre filter and a small fan behind a superb filter can produce the same CADR. The number tells you throughput, not quality of filtration — which is why filter grade is a separate question, covered in HEPA grades explained .

AHAM AC-1 does not produce one CADR. It produces three, because it runs the same test with three different contaminants of three different particle sizes.

The smoke figure is always the lowest of the three, and for a good physical reason: those are the hardest particles to catch and the slowest to settle out on their own. The pollen figure is the highest, because large particles are both easy to filter and quick to fall out of the air unaided — which flatters the machine.

So the smoke CADR is the conservative one, and it is the one to size from if smog or PM2.5 is your reason for buying. A brand quoting a single unlabelled CADR is usually quoting the pollen figure, which can be a third higher than the smoke figure on the same machine. If a specification sheet gives you three numbers, take the smallest.

Chinese machines are tested to GB/T 18801 instead, which reports a particulate CADR and — separately — a formaldehyde CADR, because the two are genuinely different jobs. It also grades cumulative clean mass, CCM: how much pollutant the filter can absorb before its performance falls away. Those grades run P1 to P4 for particulate and F1 to F4 for formaldehyde, highest being best. A high CADR with a low CCM means a machine that starts well and fades, which shows up in your wallet rather than in the spec sheet. Air purifier running costs follows that through.

CADR on its own means nothing without a room attached to it. 400 m³/h is generous in a bedroom and inadequate in an open-plan floor. The bridge between the two is air changes per hour, ACH: how many times over the machine could process a volume of air equal to the whole room, in one hour.

That gives the definition everything else rests on:

1. ACH = CADR ÷ room volume
2. Room volume = floor area × ceiling height
3. So: CADR needed = floor area × ceiling height × the ACH you want

Why does the target sit at 4 to 5 rather than 1 or 2? Because a room is not a sealed box. Outdoor air leaks in continuously through gaps, and people indoors keep generating particles by cooking, moving, and walking dust off carpets. The purifier is not clearing a fixed quantity of pollution once; it is competing, permanently, with a supply. The steady-state concentration you end up with depends on the ratio between the machine’s clean-air rate and the rate at which dirty air arrives. One air change an hour barely shifts that balance. Four or five pulls it down substantially, and past about six the returns flatten while the noise does not.

Two to three ACH is a reasonable target for general freshening in a low-pollution setting. Four to five is the figure usually quoted for smog and for allergy sufferers, and it is the one to design around if either is why you are buying.

AHAM’s own sizing advice agrees, and you can check it. Its rule of thumb is that CADR in CFM should be at least two-thirds of the room’s floor area in square feet. Take a 300 ft² room: that asks for 200 CFM, which is 12,000 ft³ of clean air an hour. At a standard 8 ft ceiling the room holds 2,400 ft³. 12,000 ÷ 2,400 = 5.0 air changes per hour, exactly. The rule of thumb is an ACH target in disguise.

Convert that into metric and it becomes a single multiplier. At a 2.5 m ceiling, five air changes an hour needs floor area × 2.5 × 5 = floor area × 12.5. Four air changes needs floor area × 10. Those two numbers will size almost any room you own.

Read that last row carefully, because it is the honest answer to a common question. The largest CADR in this catalogue is 600 m³/h, shared by two machines. In a 55 m² open space that is 600 ÷ 137.5 = 4.4 air changes an hour — good, but short of five, and that is the ceiling of what any single machine here can do. For a space that size the real options are to accept 4.4 ACH, or to run two machines, which also gets the clean air into two parts of the room instead of one. Where to put an air purifier explains why that second point matters more than it sounds.

If your ceilings are higher than 2.5 m, redo the multiplication rather than adjusting by eye. A 35 m² room with a 3.2 m ceiling holds 112 m³, not 87.5 — 28% more air, and 28% more CADR to move it. The room-sizing planner does the multiplication for your room and checks every machine here against it.

Every machine here also states a coverage area in square metres, from 28 m² to 85 m². It is tempting to skip the arithmetic and trust that. Work out what it is claiming and you will trust it less.

Eight of the twelve air rows in this catalogue sit at almost exactly the same ratio: coverage ≈ CADR ÷ 8.3. Take the Xiaomi Smart Air Purifier 4 — CADR 400 m³/h, coverage 48 m². At a 2.5 m ceiling that room holds 48 × 2.5 = 120 m³. So the implied air-change rate is 400 ÷ 120 = 3.3 ACH. Do the same for the 352 X60 Plus at 600 m³/h and 72 m²: 72 × 2.5 = 180 m³, and 600 ÷ 180 = 3.3 again. The ratio is deliberate.

And 3.3 air changes an hour is below the 4 to 5 you want for smog or for allergies. It is a perfectly defensible number for general freshening. It is not the number to buy on if you are buying because of January.

The four rows that break the pattern do so in both directions, and the two extremes are instructive. The Philips AC1215/20 Series 1000 claims only 28 m² from 270 m³/h — 270 ÷ (28 × 2.5) = 3.9 ACH, the most conservative claim on the list. The IQAir HealthPro 100 claims 85 m² from 470 m³/h, which is 470 ÷ (85 × 2.5) = 2.2 ACH: the most expensive machine in the catalogue in both markets carries its most optimistic coverage figure. That is not a fault in the machine, which is an excellent one. It is a demonstration that coverage figures are a marketing convention and CADR is a measurement.

The rule to take away is simple: size up from stated coverage. If a box says 48 m² and you have a 35 m² living room with a smog problem, that machine is a reasonable fit rather than a comfortable one, and the next size up is the better buy.

There is one more correction, and it is the one that undoes most real-world sizing. CADR is measured on the machine’s highest fan setting. That is the setting almost nobody uses.

Sleep-mode noise across the twelve air machines here runs from 22 dB to 35 dB, with a median of 31 dB. Those figures describe the lowest fan speed, and on that speed the machine is not delivering its rated CADR — it is delivering some fraction of it, often a third or less, because airflow through a fixed filter scales with fan speed. The headline CADR and the sleep-mode decibel figure are two descriptions of two different operating points, and no machine is at both simultaneously.

That has a clean practical consequence. If a machine will spend most of its life on a low or medium setting — which is what happens in any room people sleep, work or talk in — then sizing it exactly to its rated CADR sizes it to a mode you will rarely run. Buying deliberately larger than the arithmetic demands is how you get 4 to 5 air changes an hour at a fan speed you can live beside. It is the same reason a slightly oversized machine tends to have longer filter life: it spends less of its time working hard.

Auto mode changes this usefully, where it exists — seven of the twelve machines here are smart, most with an onboard particle sensor. An auto machine lifts its own speed when the room needs it and drops back when it does not, so the average over a day lands closer to the rated figure than a manually set machine ever does.

1. 1

   Measure the room, not the flat

   Floor area in m², and the ceiling height. Measure the room the machine will actually stand in with the door usually shut. Do not add the hallway.

2. 2

   Multiply out the volume

   Area × height. A 35 m² living room with a 2.5 m ceiling is 87.5 m³. Write that number down; everything else divides into it.

3. 3

   Pick your air-change target

   4 to 5 ACH for smog, PM2.5 or allergies. 2 to 3 ACH if you simply want the air fresher. Higher than 6 buys little and costs noise.

4. 4

   Multiply for the CADR you need

   Volume × target ACH. 87.5 m³ × 5 = 438 m³/h. At a 2.5 m ceiling the shortcut is floor area × 12.5 for 5 ACH, or × 10 for 4 ACH.

5. 5

   Compare against smoke CADR, then add headroom

   Use the smoke figure if three are published. Then size up again, because you will run the machine below its maximum. Ignore the coverage figure on the box — in this catalogue it implies only about 3.3 ACH.

6. 6

   Sanity-check the noise

   If the machine has to be quiet, the one that just meets your CADR on maximum will not be quiet enough. The one a size up will.

All of that sits inside the wider decision — pollutant first, then size, then filters, then cost, then noise — which is laid out in how to choose an air purifier .