Best HVAC air filter

How to choose

Start with your slot

Copy the size printed on the frame of your current filter. You need one filter for each return-air slot, and some homes have several. A filter should fit snugly without being bent or crushed.² Air that goes around a filter gets no cleaning, so ASHRAE suggests a piece of tape for a slot that's a bit loose.¹⁴

Use a thick cabinet if you have one

Thicker filters usually clean better, last longer and restrict airflow less, because deep pleats spread the dirt over more media.^{4 20} In Project Farm's test, a 4-inch Honeywell read 0.23 inches of water new and didn't rise under flour.⁷ One 2009 test found depth mattered less than expected on clean filters, so media area counts for more than thickness alone, and a well-pleated 1-inch filter can come close to a thicker one.²¹

A system built for 1-inch filters needs a technician to change the slot before it can take a deeper one.¹⁹

Aim for MERV 13, but don't overpay for the last few steps

MERV 13 is the lowest rating that must remove at least half of particles from 0.3 to 1 micron, which is why EPA, and ASHRAE in its COVID-era guidance from 2020, point to it when the system allows.^{2 5 22} DOE's Building America program recommends MERV 8 or higher in new furnaces built for 2- to 4-inch filters, and says that in older 1-inch slots, a lower MERV changed more often may be the better choice.³

The charge in the media may matter more than the last few MERV steps. An electrostatic, or charged, filter holds a static charge that helps it catch fine particles. In decay tests in 21 Toronto homes, median PM2.5 removal was 36 percent for a charged MERV 8 and 16 percent for an uncharged one. PM2.5 is the standard measure of fine particles in air. Going from a charged MERV 8 to a MERV 14 added only 5 to 9 points, and the same filter varied more between homes than different filters did within a home.¹³ If you have an uncharged MERV 8, that's the one to move away from.

Charged filters generally lost efficiency over three months in those homes, so the number on the box is a best case.²³ A study of 50 new filters also found that same-rated filters from different makers often performed differently, so treat MERV as a rough class.²⁴

Find out what your system can take

EPA says most systems can take a MERV 13, but also that systems are rarely labeled with a filter recommendation, so ask a technician to confirm.² They can measure static pressure, the resistance the blower pushes against, with the new filter in place. DOE says most home systems run well at or below 0.4 to 0.5 inches of water in total, with about 0.1 across the filter as the optimum.¹⁹

In 17 home and light-commercial systems in Austin, MERV 11-12 filters cut median airflow by about 4 percent with the fan alone and about 10 percent in cooling, compared with MERV 2, and the energy effect was small.²⁵ In Berkeley Lab's simulations of a new California home, MERV 10 to 13 changed blower energy by under 5 percent against MERV 5.²⁶ So MERV 13 doesn't wreck systems, but one short of return air has less margin. Older PSC blower motors lose airflow as resistance rises, which can freeze an air conditioner coil, and newer ECM motors use more power instead.^{19 26}

Don't put a MERV 16 in a slot not built for it. In the Berkeley Lab study, MERV 16 filters brought noise complaints and about 20 percent more blower energy in some cases, and in two homes they came out within days because of the noise.²⁶ Don't jump to MERV 14 to 16 or HEPA furnace filters in a 1-inch slot without checking with a technician either. If you hear new whistling or the furnace starts shutting off after any filter change, go back to the old filter and call someone.

Run the fan on bad air days

A furnace filter cleans only while the blower runs, and in more than 7,000 homes with smart thermostats that was a median of 18 percent of the time. At those run times, the filter's efficiency matters much less.²⁷ Setting the fan to On raises filtration, and EPA advises it during smoke. It also warns that longer run times raise electricity costs and can make humidity control less reliable in cooling season.^{2 28}

This is also why HEPA-labeled furnace filters are poor value. EPA says most MERV 13 filters are likely to be almost as effective as HEPA filters in these systems, though field studies haven't confirmed that central systems run long enough to improve health.⁵

Set a schedule you'll keep

ENERGY STAR says to inspect, clean or change filters monthly.²⁹ EPA says makers typically recommend replacing them every 60 to 90 days, sooner if the old filter comes out heavily soiled, which works out to about four to six 1-inch filters a year.² Berkeley Lab found that dirt loading varied more from house to house than by MERV, so let pets and dust set your schedule.²⁶ During heavy smoke, EPA says to change filters more often than the maker recommends.²⁸

For smoke or asthma, add a portable cleaner

Better filtration brings modest health gains. A 2013 review of 16 intervention studies found typical improvements of 7 to 25 percent, particularly in people with allergies or asthma.³⁰ EPA's review found that portable HEPA cleaners cut particle exposure by roughly half or more, while central filtration results were less consistent.⁵ Use both if you can, with the portable cleaner where people sleep.^{5 28}