Why an Air Cleaner's Square Footage Is a Time

This page contains affiliate links. If you buy through one we may earn a commission, at no cost to you.

Before anyone answers with a model name, there is a problem with the question's units, and it is not the asker's fault. A purifier's square footage is not an area it covers. It is an airflow divided by an air change rate somebody at the manufacturer picked, and divided again by a ceiling height nobody prints. The two air change rates in common use differ by a factor of 4.80, and both of them are in use on our own pages.

For a 767 square foot apartment that factor decides everything. On one convention five of the five purifiers in our roundups publish a coverage figure big enough. On the other, none of them does. Same machines, same listings.

A coverage area is one line of algebra

Start from what the number has to mean. A machine delivers clean air at some rate. A room holds some volume. Divide the rate by the volume and you have air changes per hour. Turn that around and the room volume a machine can serve at a chosen air change rate is its delivery rate divided by that rate, and the floor area is the volume divided by the ceiling height.

With an eight foot ceiling, the whole thing collapses to:

coverage in square feet = 7.5 times the clean air delivery rate in cubic feet a minute, divided by the air changes per hour.

Two free parameters, neither printed. At one air change an hour a machine covers 7.5 times its delivery rate. At the 4.8 air changes an hour that the industry's own verification programme uses, it covers 1.5625 times its delivery rate. That is the factor of 4.80, and it is the entire reason coverage figures look incomparable.

The product that publishes both numbers at once

Our roundup of home air purifiers for smoke carries, in one specification block for one machine, a coverage of "Up to 1,882 sq ft in 1 hour" and a certification of "AHAM Verifide at 392 sq ft".

Divide them. One thousand eight hundred and eighty-two over 392 is 4.801.

That is the whole finding in a single division. The two figures are the same machine at the two conventions, and the arithmetic recovers the number neither of them states: 392 square feet times 8 feet times 4.8 air changes is 15,053 cubic feet an hour, which is 251 cubic feet a minute, and 1,882 square feet times 8 feet times 1 air change is 15,056, which is the same 251. A delivery rate the listing never prints is sitting inside two numbers it does.

And once you know to look, the pattern is everywhere on the same pages. A unit published as "674 sq ft in 12.5 minutes, 3,235 sq ft in one hour" has a ratio of 4.800, and 12.5 minutes is 60 divided by 4.8. Both figures imply 431 cubic feet a minute. Another, published as "222 sq. ft. 4.8 times per hour, 1,073 sq. ft. once per hour", has a ratio of 4.833 and a published CADR of 143 cubic feet a minute for smoke, and 143 times 1.5625 is 223.4 while 143 times 7.5 is 1,072.5. A third, "153 sq. ft. every 12.5 minutes, 743 sq. ft. every hour", gives 4.857 and about 98 cubic feet a minute.

The machine that leads both of those roundups publishes "Up to 900 sq ft in 30 minutes, 1,800 sq ft in 60 minutes" alongside a CADR of "Dust 242, Smoke 240, Pollen 249". Nine hundred square feet at eight feet is 7,200 cubic feet, and 240 cubic feet a minute for thirty minutes is 7,200 cubic feet. Exactly one air change. Then 1,800 square feet in sixty minutes is 14,400 cubic feet against 14,400 delivered. Exactly one air change again.

So that headline coverage figure is not a claim about cleaning a room. It is the statement that the machine can process the room's air once.

One air change is 63 percent

Here is why that distinction is not pedantry. In a well-mixed room the particle concentration falls exponentially with the number of air changes, so the fraction removed after n air changes is one minus e to the minus n.

One air change removes 63.21 percent and leaves 36.79. Two removes 86.47 percent. Three removes 95.02. Four point eight removes 99.18.

So "cleans 900 square feet in thirty minutes" and "clean air for 392 square feet" are not two phrasings of one capability. The first describes a room with better than a third of its particles still in the air, and the second describes a room with under one percent. Both sentences can be true of the same machine and they are 36 percentage points apart on the thing a buyer cares about.

The guide text on our own purifier page gets the direction right in words, and it deserves the credit for naming the right quantity: "Large rooms need enough CADR or stated coverage to keep air moving through the filter often." Often is the word doing the work, and often is an air change rate.

The 767 square foot apartment, answered

Seven hundred and sixty-seven square feet at eight feet is 6,136 cubic feet. Air changes an hour is the delivery rate times sixty divided by that, and the time to reach ninety percent clear is the natural log of ten divided by the air change rate, which is 2.3026 hours divided by it.

Run the five purifiers from our roundups, using the delivery rates recovered above.

The 431 cubic feet a minute machine manages 4.215 air changes an hour and reaches ninety percent clear in 32.8 minutes. The 251 manages 2.454 and 56.3 minutes. The 240 manages 2.347 and 58.9 minutes. The 143 manages 1.398 and 98.8 minutes. The 98 manages 0.958 and 144.2 minutes.

Now compare that with what the same five publish as coverage. On the one air change convention their figures are 3,235, 1,882, 1,800, 1,073 and 743 square feet, so four of the five sail past 767 and the fifth is within sight of it. On the 4.8 convention their figures are 674, 392, 375, 222 and 153, so not one of the five reaches 767.

Five machines, two sets of numbers from the same listings, and opposite verdicts about the same apartment. The only thing that changed is which convention the manufacturer wrote in, and only two of the five say.

Our roundup of the best air purifier for a large room and our general comparison of air purifiers are the places to look at the machines themselves. One thing to notice while you are there: that second page publishes "Coverage", "Large rooms" as a specification entry on two of its picks, which is a coverage field containing no number at all, and a third publishes "1,070 sq. ft." with no time attached, which cannot be placed on either convention.

The one product that publishes the honest unit

A TOOLBOOX 970CFM air scrubber is the only air mover in this catalogue that publishes an airflow, and that single decision makes it the only one whose figure works in a room you choose rather than a room the manufacturer imagined. It publishes "Moves up to 970CFM through three-stage filtration", a rotomolded polyethylene shell, lockable 360 degree wheels, a stackable design and UV-C, and it is clear about what it is: "This is professional-grade equipment", "large, heavy and genuinely noisy at high airflow", and explicitly "no substitute for proper professional containment" for hazardous materials.

Put 970 cubic feet a minute in the 767 square foot apartment and it is 9.485 air changes an hour and ninety percent clear in 14.6 minutes, which is 2.25 times the best purifier on our shelf and 6.8 times the 143. Converted into the language the purifiers use, its coverage would read 7,275 square feet on the one air change convention and 1,516 on the 4.8, and it publishes neither, which is the right choice.

There is one thing missing from it, and it is the same thing missing from the whole category. A CFM is a fan rating. A clean air delivery rate is a fan rating multiplied by the fraction of particles the filter removes on a single pass. The listing says "HEPA-grade" rather than HEPA and publishes no efficiency, so 970 is a ceiling on its delivery rate rather than the rate itself. Our page on why two 5 micron filters are not the same filter works through what a nominal rating does and does not promise, and the same distinction is the one at stake here.

The fan that proves the point by having a delivery rate of zero

If the difference between airflow and air cleaning still feels academic, there is a limiting case sitting in the same catalogue. A TALOYA 52 inch ceiling fan with light and remote publishes a reversible DC motor with six wind speeds, a stated level under 30 decibels, a 15 watt light dimmable from 10 to 100 percent across 3000K, 4000K and 6500K, and a 52 inch span.

A fan that size moves several thousand cubic feet a minute, far more than any purifier on this page. Its clean air delivery rate is exactly zero, because its single-pass efficiency is zero. It takes particles from one part of the room and puts them in another.

That is the clearest statement of the identity available: airflow is not air cleaning, and airflow times efficiency is. A 970 cubic feet a minute machine with a 70 percent single-pass filter delivers 679, and a 5,000 cubic feet a minute fan with a zero percent filter delivers nothing at all. Our page on why a bag is the filter and a cyclone is not makes the same argument about single-pass capture in a vacuum cleaner.

The fan is also worth noting for what it does not publish, because it is a pattern. It gives six speeds, a decibel figure, a light wattage and three colour temperatures, and no airflow whatsoever, on a device whose only job is moving air. Its own record is admirably careful about the status of all of it: treat the figures "as the manufacturer's specifications rather than independently measured".

A VZXAB 120mm RGB case fan is the same omission at the opposite scale. It publishes 120 by 120 by 25 millimetres, about 1650 revolutions a minute, 21.6 dBA, eight anti-vibration pads, a PBT frame and a draw of 0.30 amps, and no airflow. It is the most precisely specified air mover in this catalogue and the figure you would need to compare it with anything is absent.

A filter whose entire specification is the missing term

And then the product that makes the gap visible by being nothing but filtration. A G and F 9119 particulate respirator dust mask, 20 pack publishes a cup-style respirator with an exhale valve, a welded dual-point strap attachment, an "M" style nose clip over a molded nose bridge, and that it "meets ANSI/ISEA 110-2003 Section 7.11.1 for heat and flame resistance".

A respirator is a single-pass filter and nothing else. Its efficiency is not one of its specifications, it is the entire specification. And what the listing says about it is "high filtration efficiency against certain non-oil-based particles", with no percentage, no particle size and no test method, while the standard it does cite governs heat and flame.

So the one product here whose figure of merit is exactly the term that would turn an airflow into a clean air delivery rate publishes a qualifier instead of a number, and references a standard for a different property. Its own drawbacks are honest about the rest, including the genuinely important point that "a valved mask vents your own breath outward, so it protects the wearer, not the people nearby", and that the "Facts are maker-stated, not lab-verified". The honesty is real and the number is absent.

The ceiling height, and why the air feels different

Two more things the coverage figure quietly assumes.

The first is eight feet. Coverage goes as one over the ceiling height, so every figure on this page is 11.1 percent optimistic in a room with nine foot ceilings, 20 percent optimistic at ten feet and 33.3 percent optimistic under a twelve foot vault. Not one listing states the height its own number assumes, and anyone in a converted loft is reading a figure for somebody else's apartment.

The second is mixing. The exponential decay above assumes the room is well stirred, which is the best case and never the real one. A purifier in a corner recirculates its own output back into its own intake, and a machine behind a sofa processes the air behind the sofa. Every real departure from perfect mixing makes the times longer than the table, never shorter, which is exactly why people ask whether a unit will work on its side on a wire shelf. Nothing in the arithmetic answers that, and the direction of the error is known.

Which leaves the question of why a purifier changes how a room feels, because it is a different quantity entirely and the air change rate has nothing to do with it. What you feel is the face velocity at the outlet. Nine hundred and seventy cubic feet a minute through a 20 by 20 inch face, which is 2.778 square feet, is 349 feet a minute, or 1.77 metres a second. One hundred and forty-three cubic feet a minute through a 10 inch round grille, 0.545 square feet, is 262 feet a minute, or 1.33 metres a second. Perceptible air movement starts around 0.2 metres a second.

So both machines produce a felt draft at the outlet, and the draft falls away quickly as the jet spreads. The sensation is a fan, it is real, and it is independent of how often the room gets filtered. A machine with a small outlet and a modest flow can feel stronger than one with a big outlet and a large flow, which is the opposite of the ranking that matters.

The honest limit, and what to ask for instead

What is exact here is the algebra and the decay. A coverage area is a delivery rate times sixty, divided by an air change rate, divided by a ceiling height, so at eight feet it is 7.5 times the rate over the air changes. That identity reproduced five published pairs of coverage figures to within a rounding, which is the strongest evidence available that it is the formula the listings were generated from. The exponential decay is exact for a well-mixed room, and the 63.21 and 99.18 percent follow from one minus e to the minus n with nothing assumed.

What is representative is the eight foot ceiling, which is the convention the figures appear to use and which none of them states, and the perfect mixing, which flatters every time on this page. The 251, 431 and 98 cubic feet a minute were recovered by arithmetic from pairs of published coverage figures rather than read off a listing, so if a manufacturer used a 4.7 or a 5.0 in place of 4.8 those recovered rates move by a few percent. The outlet areas behind the face velocities are assumptions. And the single-pass efficiency, which is the difference between the airflow a product publishes and the delivery rate a room responds to, is published by nothing in this catalogue at all.

What this page deliberately does not do is say anything about what filtered air does for a person, and the reason is that it is not a question arithmetic on a volume can answer. The decay above is about particle concentration and nothing else. Our own purifier roundup draws the same line and draws it well: "This roundup describes what the filters capture, not any medical or health outcome." Nor has this page computed anything about gases or odours, because carbon adsorption has a capacity rather than an efficiency and no listing here publishes a carbon mass. And it has not tried to compare noise against airflow, which would need a sound power figure at a stated flow, and our page on why a 33 decibel fan is louder than a 30 decibel one explains why the decibel figures that are published do not support it.

And the practical residue is one substitution. Ignore the square footage and ask for the delivery rate in cubic feet a minute, because that is the number that survives contact with your own room. Multiply it by 60, divide by your room's volume in cubic feet, and you have air changes an hour. Three or four is a reasonable target and the industry's own verification convention is 4.8. Then the time to clear ninety percent of anything is 138 minutes divided by that air change figure. For the 767 square foot apartment that started this, reaching 4.8 air changes an hour needs 491 cubic feet a minute of delivery, which is more than the strongest purifier on our shelf and half of the air scrubber.