CADR, Sizing, and Particle Capture
Most air purifier guides optimise the wrong variable. They obsess over filter grades and ignore the thing that actually determines whether your air gets cleaner: whether the unit is switched on.
A purifier running on low all day removes vastly more particulate than a better purifier running on high for twenty minutes before you get sick of the noise and turn it off. Nearly every practical recommendation below follows from that.
What a filter can and cannot do
An air purifier is a fan pulling air through a filter. That is the whole device. It is excellent at one job and mediocre at another.
It works well for particles. Fine particulate matter, wildfire smoke, pollen, dust, mould spores, pet dander. If your problem is PM2.5, filtration is the correct tool and it works.
It works poorly for gases. Volatile organic compounds, nitrogen dioxide, carbon dioxide. A HEPA filter does nothing to these at all, and the carbon stage in most consumer units is thin enough that it saturates within weeks. For gases, the effective interventions are removing the source and opening a window.
Worth being blunt about this because purifier marketing blurs the line constantly. If your kitchen smells of frying, the answer is an extractor fan. If the room feels stuffy in a meeting, that is carbon dioxide and the answer is ventilation. A purifier solves neither.
Decoding HEPA labels
Two different standards are in play, and marketing copy routinely mashes them together.
The US standard. “True HEPA” generally refers to the Department of Energy specification: at least 99.97% capture of particles at 0.3 micrometres. It is a marketing term rather than a certification, and no one audits its use.
The European standard. EN 1822, now largely superseded by ISO 29463, grades filters E10 through H14 and tests each one at its most penetrating particle size, determined per filter rather than fixed. H13 requires at least 99.95% at MPPS. H14 requires at least 99.995%. Anything below H13 is not HEPA under this standard, which is why E11 and E12 media get sold as “HEPA-type.”
Why 0.3 µm shows up in both: particles around that size are the hardest to catch. Larger ones get intercepted or hit fibres directly. Smaller ones move erratically enough through Brownian diffusion that they collide with fibres anyway. So the efficiency figure at the worst case is a floor, and real performance is better above and below it.
What this means for buying. Look for H13 or a stated 99.97% at 0.3 µm. Treat “HEPA-type,” “HEPA-like” and “99% HEPA” as absent. H14 is unnecessary at home and actively counterproductive, because tighter media means higher pressure drop, which means less airflow or a louder fan for the same airflow.
One thing that matters more than the grade: whether the housing is sealed. Air that slips around the filter edge is unfiltered air, and a leaky H14 will underperform a well-sealed H13.
CADR, and the correction nobody makes
Clean Air Delivery Rate is the useful metric, because it combines filter efficiency with fan output. A perfect filter behind a weak fan cleans a room slowly.
Here is what the box does not tell you. CADR is measured at the highest fan speed with a brand-new filter. Most units at full speed sit somewhere around 55 to 65 dB, comparable to a running dishwasher. Almost nobody lives with that, so almost everybody runs the unit on low or sleep mode, where output is roughly half the rated figure. The filter then loads with particles over months and output drops further.
So the standard sizing advice, the AHAM two-thirds rule, is a floor rather than a target.
The two-thirds rule (imperial). Required CADR in cfm = floor area in square feet × 0.67. At 8-foot ceilings this works out to exactly 5 air changes per hour.
The metric equivalent. Required CADR in m³/h = room volume in m³ × 5. Use volume, not floor area, because a room with 3-metre ceilings holds far more air than the packaging assumes.
Then double it. Buy roughly twice the calculated figure so you can run the unit on low or medium permanently and still hit five air changes. If you have asthma, allergies, pets, or live somewhere with wildfire smoke or heavy winter wood smoke, aim higher still, in the range of 6 to 12 air changes.
A worked example. A bedroom of 4 m × 3 m with a 2.6 m ceiling is 31 m³. Five air changes needs a CADR of about 156 m³/h. Buy something rated around 300 m³/h and run it on the quietest setting. It will be inaudible and it will do the job.
Sizing up is the single most effective noise strategy available, and it is cheaper than buying a premium quiet unit at the wrong size.
The cheap option that genuinely works
If cost is the barrier, a box fan with a furnace filter taped to it is not a compromise. It is competitive.
Researchers measured DIY units built from MERV 13 to 16 filters against commercial purifiers. A single 2-inch MERV-13 filter on a box fan at its lowest speed produced an estimated CADR of 293 cfm for about $35 in parts. The best-in-class low-noise HEPA purifier they tested managed 282 cfm at its maximum speed for $549. Building a four-filter cube, the design known as a Corsi-Rosenthal box, adds up to about 80% more output over the single-filter version, though running a second single-filter fan gains more for the same money.
Separate chamber testing against simulated wildfire smoke found a basic box fan and MERV-13 filter delivered 111 cfm, and that a Corsi-Rosenthal configuration raised that by 261%.
The honest caveats. These units get loud at high speed, with some measured at 67 dB. Filter loading during a smoke event cuts output sharply, so change filters far more often than you would think. Use a modern box fan in good condition, and do not leave a DIY unit running unattended overnight if the fan is old.
MERV 13 is less efficient per pass than H13. It compensates with far more airflow for the money, and air changes per hour is what actually drives the concentration down in your room.
What to avoid
Anything that generates ozone. Ozone generators, and many ionisers, plasma and “active” purification devices, produce ozone as an intentional output or a byproduct. Ozone is a lung irritant, which makes this a device that trades one pollutant for another. California’s Air Resources Board maintains a certification list for air cleaners sold there, and it is a reasonable reference even outside the US.
Anything without a published CADR. If a manufacturer promotes coverage area, filter stages and proprietary technology names but will not state a CADR figure, the omission is the answer.
Carbon as a headline feature. A thin carbon-coated fabric sheet is close to decorative. Meaningful gas adsorption needs a deep bed of pelletised carbon measured in kilograms, and even then it saturates and needs replacing every three to six months. Carbon also adds pressure drop, reducing your particle CADR. Plain activated carbon handles odours and many solvents reasonably and handles nitrogen dioxide poorly without chemical impregnation.
Placement and running it
Simple rules, in descending order of importance:
- Run it continuously. Reactive use does not work. Particle concentrations rebuild within an hour of switching off.
- Put it in the room you occupy. One unit in the bedroom running all night beats one unit in the hallway. Close the bedroom door and you are cleaning 30 m³ rather than the whole flat.
- Leave clearance around the intake. Roughly 30 to 45 cm from walls, curtains and furniture. Blocking the intake is the most common installation mistake.
- Close windows when outdoor levels are high, open them when they are not. This is where knowing the outdoor number pays off.
Ignore advice about elevating the unit to catch particles before they settle. Fine particles do not settle on any useful timescale, which is precisely why they are a problem.
Maintenance: vacuum or rinse the pre-filter monthly, since a clogged pre-filter strangles airflow and costs you CADR. Replace HEPA media every 6 to 12 months. Replace carbon every 3 to 6 months if you have a real carbon bed. During a smoke episode, expect to replace filters several times faster than the manual suggests.
Where SharedSky fits
A purifier is a purchase you make once and then have no feedback on. SharedSky closes that loop from both ends.
Before. The current air, pollen, pressure and heat reading tells you when to shut the windows and turn the unit up, and when the outdoor air is better than your indoor air and you should open up instead.
After. Log how you feel in one tap, including the days you feel fine. If you log for two weeks before installing a purifier and two weeks after, you have a before-and-after comparison rather than an impression. That matters because the placebo effect on a €300 purchase is substantial, and because it tells you when performance has drifted. If congestion or fatigue creeps back after eight months of quiet running, saturated filters are the first thing to check.
One limitation stated plainly: SharedSky reports outdoor conditions, drawn from the Copernicus Atmosphere Monitoring Service via Open-Meteo. It does not measure the air in your living room. Your symptom log is the indoor half of the picture, and an inexpensive indoor PM2.5 monitor is a worthwhile companion if you want the actual number.
SharedSky is not a medical device and does not provide medical advice. See our Medical Disclaimer.
Sources
Illinois Institute of Technology, Portable Air Cleaner Test Report: Corsi-Rosenthal Box with MERV 13 filters (2021). https://www.built-envi.com/wp-content/uploads/IIT-CADR-Testing-C-R-Box-September-2021.pdf
Srikrishna, Can 10× cheaper, lower-efficiency particulate air filters and box fans complement HEPA purifiers?, Science of the Total Environment (2022). https://www.sciencedirect.com/science/article/pii/S0048969722029813
Dal Porto et al., Characterizing the performance of a do-it-yourself (DIY) box fan air filter, Aerosol Science and Technology (2022). https://www.tandfonline.com/doi/full/10.1080/02786826.2022.2054674
Impact of do-it-yourself air cleaner design on the reduction of simulated wildfire smoke in a controlled chamber environment, Indoor Air (2022). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9828579/
Want to see how your indoor air improvements impact your physical health? Log your daily energy, focus, and sinus symptoms in SharedSky to track your baseline gains.