Comparison
HEPA Filters Explained
Four labels, only some of which mean anything. Here is which is which, and the one figure that lets you compare units honestly.
By Scooter M. · Published · Last updated
How this page is funded: the buy links below are affiliate links. If you buy through one, we earn a commission at no extra cost to you. It does not change the order of the picks or what we say about them. Full disclosure.
The short answer
HEPA is a filtration specification. True HEPA is a marketing phrase used to distinguish a filter that meets it from “HEPA-type” filters that do not. MERV is a different rating scale used for HVAC filters, where MERV 13 is a good whole-house choice but is well below HEPA. The figure to actually compare purifiers on is not the filter label at all - it is CADR.
| Label | Is it a standard? | What it means | Trust it? |
|---|---|---|---|
| HEPA | Yes - a defined filtration specification | A dense fiber filter meeting a defined particle-capture efficiency | Yes |
| True HEPA | No - a marketing phrase | Used to signal a genuine HEPA filter rather than a HEPA-like one | Usually a positive signal, but it adds nothing to HEPA itself |
| HEPA-type / HEPA-style / 99% HEPA | No | Explicitly not a HEPA filter. The qualifier is doing the work. | Treat as a non-HEPA filter |
| H13 / H14 | Yes - European EN 1822 classes | Higher efficiency classes under a European standard | Real, but not directly comparable with the US convention |
| MERV 8 / 11 / 13 | Yes - ASHRAE 52.2 | The HVAC filter scale. MERV 13 is a strong furnace filter, still below HEPA. | Yes, for HVAC filters. Not comparable to purifier ratings. |
How a HEPA filter actually works
It is not a sieve with holes smaller than the particles, which is the common mental model and it is wrong. A HEPA filter is a randomly arranged mat of fine fibers, and it captures particles by three mechanisms at once:
- Interception - a particle following the airflow passes close enough to a fiber to touch it and stick.
- Impaction - a larger, heavier particle cannot follow the air as it curves around a fiber, and hits it.
- Diffusion - very small particles move erratically due to collisions with air molecules, which makes them wander into fibers.
Impaction works best on large particles and diffusion works best on very small ones, which produces a counter-intuitive result: filters are least efficient in the middle, around 0.3 microns. That size is the hardest case, which is why the specification is defined there - and why particles both larger and smaller are captured more efficiently, not less.
Why the filter label is the wrong thing to compare on
A HEPA filter with a weak fan behind it cleans very little air. A slightly less efficient filter with a strong fan may clean far more. The filter label describes one component; what you care about is the output of the whole machine.
That is exactly what CADR measures - it combines filter efficiency and airflow into one comparable number, tested under AHAM’s AC-1 standard. Two units both advertising “True HEPA” can differ by a factor of two in CADR, and the label will not tell you.
MERV 13 and your furnace
If you have forced-air heating, the furnace filter is already a whole-house air cleaner, and upgrading it is often the cheapest air-quality improvement available. MERV 13 is a common recommendation.
The caveat is real: a higher-MERV filter restricts airflow more, and some systems are not specified for it. Check what your system supports before fitting one - a restricted system is a worse outcome than a lower-rated filter.
What HEPA does nothing about
Gases, odors and VOCs. HEPA is mechanical particle capture; odor molecules are orders of magnitude smaller than anything a fiber can intercept. That is the activated carbon stage’s job entirely - HEPA vs carbon filters.
What we would buy for this
Chosen from the same registry as our roundups, on published specifications and computed running cost.
Levoit Core 400S
Living rooms and open-plan spaces
The one to buy if you want a purifier that genuinely covers a living room, and one of the few spec sheets in this bracket that publishes coverage at a useful air-change rate rather than only the flattering one.
#ad We do not hold a current price for this product, so we do not show one. How we make money.
Published specifications
| Specification | Value | Source |
|---|---|---|
| CADR | 231 CFM | Levoit published specification |
| Coverage | 358 sq ft at 4.8 air changes/hr (1,718 sq ft at 1 ACH) | Levoit published specification |
| Rated power | 38 W | Levoit published specification |
| Noise | 22-52 dB | Levoit published specification |
| Filtration | 3-stage: pre-filter, HEPA, activated carbon | Levoit published specification |
| Dimensions | 10.8 x 10.8 x 20.5 in, 11 lb | Levoit published specification |
What it costs to run
About $25-30 a year in electricity at 12 hours a day, before filters.
38 W / 1000 x 12 hours x 365 days = 166.4 kWh. At 18.34 cents/kWh that is $30.52 a year at rated power - and the unit only draws its full 38 W on the top fan speed, so a mid setting lands nearer $25. Add the replacement filter, which you should price before you buy rather than after.
What is good about it
- Levoit publishes coverage at both 1 ACH and 4.8 ACH, which is unusually straight - most brands quote only the flattering 1 ACH number
- 38 W is low for this CADR, which is what keeps the running cost down
- 22 dB on its lowest setting is quiet enough to sleep beside
- Carbon, pet-allergy and smoke-specific replacement filters are sold for the same body
What is not
- We could not find a published filter replacement interval in Levoit's specification, so we are not quoting one
- The 1,718 sq ft headline is a single air change per hour, which is not a useful rate for allergies
Who should skip this
Skip this if your room is under 200 sq ft. The Core 300S covers that at lower power and lower cost, and the extra CADR buys you nothing in a small bedroom.
How it scored
- Clean air per watt9.0
- Honest coverage claim9.5
- Quiet operation8.5
- Filter economics7.5
Overall 8.6 / 10. This is our editorial score from published specifications and computed running cost. It is not a test result, and it is deliberately not marked up as a rating in this page’s structured data.
Coway Airmega AP-1512HH Mighty
Running continuously in a medium room
The most efficient unit here at low speed - 3.7 W on speed one - which matters enormously if you leave a purifier running all day and almost not at all if you do not.
#ad We do not hold a current price for this product, so we do not show one. How we make money.
Published specifications
| Specification | Value | Source |
|---|---|---|
| CADR (smoke) | 233 CFM | Coway published specification |
| CADR (dust) | 246 CFM | Coway published specification |
| CADR (pollen) | 240 CFM | Coway published specification |
| Coverage | 361 sq ft at 4.8 air changes/hr (1,748 sq ft at 1 ACH) | Coway published specification |
| Power draw by speed | 0.3 W standby / 3.7 W speed 1 / 8.7 W speed 2 / 75.3 W speed 3 | Coway published specification |
| Filtration | Pre-filter, odor filter, Green True HEPA, bipolar device | Coway published specification |
What it costs to run
About $6 a year running 24/7 on speed one, or $121 a year running 24/7 on speed three.
Speed 1: 3.7 W / 1000 x 24 x 365 = 32.4 kWh = $5.94 a year. Speed 3: 75.3 W / 1000 x 24 x 365 = 659.6 kWh = $120.97 a year. That twenty-fold spread is the most important number on this page, and it is why we chase per-speed figures instead of accepting a single rated wattage.
What is good about it
- Coway publishes power draw per fan speed, which almost nobody does and which is the only way to compute a real running cost
- 3.7 W on speed one is remarkable - cheaper to run continuously than most nightlights
- Coverage published at 1, 2 and 4.8 air changes per hour, so you can size it honestly
What is not
- 75.3 W on speed three is the highest draw here; how the auto mode behaves decides your bill
- The bipolar device is a stage we would rather do without; check it can be disabled
Who should skip this
Skip this if you only run a purifier for an hour or two at a time. Its whole advantage is low-speed efficiency over long hours, and short bursts on high speed give that straight back.
How it scored
- Clean air per watt9.5
- Honest coverage claim9.5
- Quiet operation8.0
- Filter economics7.5
Overall 8.7 / 10. This is our editorial score from published specifications and computed running cost. It is not a test result, and it is deliberately not marked up as a rating in this page’s structured data.
Winix 5500-2
Smoke, cooking smells and pet odors
The one with a genuine granular carbon stage rather than a carbon-dusted sheet, which is why it belongs on a smoke and odor list and why it costs more to run.
#ad We do not hold a current price for this product, so we do not show one. How we make money.
Published specifications
| Specification | Value | Source |
|---|---|---|
| CADR (smoke) | 232 CFM | AHAM-verified figure published by Winix |
| CADR (dust) | 243 CFM | AHAM-verified figure published by Winix |
| CADR (pollen) | 246 CFM | AHAM-verified figure published by Winix |
| Coverage | 360 sq ft | Winix published specification |
| Rated power | 70 W | Winix published specification |
| Filtration | 4-stage: washable pre-filter, activated carbon, True HEPA, PlasmaWave | Winix published specification |
| Certification | Energy Star, CARB certified | Winix published specification |
What it costs to run
About $56 a year in electricity at 12 hours a day, before filters.
70 W / 1000 x 12 hours x 365 days = 306.6 kWh. At 18.34 cents/kWh that is $56.23 a year. Over three years that is roughly $95 more than the Levoit Core 400S at the same hours - typically more than the price gap between the two units at purchase.
What is good about it
- AHAM-verified CADR, which means the figure was measured to the AC-1 standard rather than asserted by marketing
- A real granular carbon filter, which is what actually adsorbs gases and odors
- The washable pre-filter slows down how fast the expensive filter loads up
What is not
- 70 W is roughly double the Levoit Core 400S for a similar CADR, and you pay that difference every year
- PlasmaWave is an ionizing stage; it can be switched off, and we would switch it off
Who should skip this
Skip this if odor is not your problem. Buying for dust and pollen alone means paying a 70 W electricity premium for a carbon stage you will not use.
How it scored
- Clean air per watt6.5
- Honest coverage claim9.0
- Odor and gas removal9.5
- Filter economics8.0
Overall 8.4 / 10. This is our editorial score from published specifications and computed running cost. It is not a test result, and it is deliberately not marked up as a rating in this page’s structured data.
How we ranked these
We have not tested these units. Nobody here owns one. What we did was compile each manufacturer’s published specifications, check the coverage claims against AHAM’s room-sizing rule, and compute what each one costs to run for a year at the current US average electricity price. Every figure on this page is attributed to whoever published it, and the arithmetic is shown so you can redo it.
How we rank these sets out the full method, including what would change a ranking.
Questions people ask
+−What is the difference between HEPA and True HEPA?
“True HEPA” is a marketing phrase rather than a separate specification. It exists to distinguish genuine HEPA filters from “HEPA-type” products that do not meet the specification. A filter labeled simply HEPA is not inferior to one labeled True HEPA.
+−Is HEPA better than MERV 13?
HEPA is a higher filtration efficiency, yes. They are also used in different places - HEPA in portable purifiers, MERV in HVAC systems that need airflow a HEPA filter would choke. MERV 13 in a furnace treating the whole house can be more useful overall than HEPA in one room.
+−Do HEPA filters catch viruses?
That is a health question rather than an environmental one, and it depends on factors well beyond filter efficiency. What we can state is the mechanism: filtration efficiency improves below 0.3 microns due to diffusion, so the “too small to catch” intuition does not hold. Anything about disease transmission belongs with a public health source.
+−Can I wash a HEPA filter?
No, unless it is explicitly sold as washable - and washable ones are generally not true HEPA. Washing a HEPA filter damages the fiber structure the whole mechanism depends on. The washable component is the pre-filter, and that one you should clean regularly.
+−What does H13 mean?
It is a class under the European EN 1822 standard, indicating a high-efficiency filter. It is a real classification, but it is a different standard from the US convention and the two are not directly interchangeable in a specification comparison.
+−Is a pre-filter necessary?
Not necessary, but valuable - it catches hair and large debris before they reach the expensive stage, which keeps the unit closer to its rated CADR for longer. In a house with pets it is close to essential.