+86 13638632902
Jason@cleanroomequips.com
MENU
Every cleanroom project gets around to the same question sooner or later: how effective are HEPA filters once you strip away the marketing and look at test data? [1] Here is the honest version. A certified HEPA filter is one of the most effective mechanical barriers anyone has ever built for airborne particles — genuinely astonishing at what it does. But three things have to be true before that effectiveness shows up in your room.
- The filter has to be real, built to a written class.
- It has to be tested where it sits, not just at the factory.
- It has to be shielded by proper pre-filtration.
Fig.1 — A standard square HEPA filter element. Glass fiber media, 70 pleats, rated 99.97% at 0.3 μm.
Miss any one of those and you are running on faith, not filtration. Those three conditions are what keep HEPA filters effective outside the laboratory.
A quick word for readers who found this page while shopping for a living-room purifier. Everything below uses industrial H13/H14 data. Plenty of consumer boxes say "HEPA-type", a phrase EN 1822 simply does not define. If you are buying for a home or office, find a stated efficiency at 0.3 micron on the spec sheet — the word HEPA by itself means nothing. Engineers, read on. What follows is the same walk a validation engineer would take: efficiency classes, the MPPS paradox, pressure drop, dust loading, testing, the ULPA question, and a retrofit we can put real numbers behind.
The standard square element answers "how effective are HEPA filters" in one picture. A glass fiber medium folded into seventy-odd pleats, sealed in an aluminium frame, and rated at 99.97 percent efficiency for particles at 0.3 micron — the size class the media finds hardest to catch.
What "Effective" Means: The Standards Behind the Number
"Effective" is not a fuzzy compliment in this trade. It is a number, taken at a stated particle size, under a stated test method, written into a standard. So when someone asks how effective are HEPA filters, the complete answer starts with a class table — never with a percentage printed on a carton.
EN 1822 and the H13/H14 Classes
Most cleanroom specs lean on the European standard EN 1822. [1] It calls a filter HEPA at classes H13 and H14, and — this is the part people miss — it measures efficiency at the Most Penetrating Particle Size (MPPS), not at some convenient particle diameter.
| Class | Efficiency at MPPS (overall) | Local efficiency | Typical application |
|---|---|---|---|
| H13 | ≥ 99.95% | ≥ 99.75% | iso 7–8 cleanrooms, pharma Grade B/C |
| H14 | ≥ 99.995% | ≥ 99.975% | iso 5–6 cleanrooms, pharma Grade A/B |
| U15 | ≥ 99.9995% | ≥ 99.9975% | Semiconductor, optics |
| U16 | ≥ 99.99995% | ≥ 99.99975% | Wafer fab critical zones |
| U17 | ≥ 99.999995% | ≥ 99.9999% | Extreme nanotechnology |
| Class | Efficiency at MPPS | Typical use |
|---|---|---|
| H13 | ≥ 99.95% | ISO 7–8, pharma B/C |
| H14 | ≥ 99.995% | ISO 5–6, pharma A/B |
| U15 | ≥ 99.9995% | Semiconductor, optics |
| U16 | ≥ 99.99995% | Wafer fab zones |
| U17 | ≥ 99.999995% | Nanotechnology |
Two rows of fine print matter to buyers. First, overall efficiency and local efficiency are different numbers — a filter can pass overall and still leak through a pinhole, which is the whole reason scan testing exists. Second, Deiiang rates its H13 media at 99.97–99.99 percent at 0.3 micron and its H14 media at 99.995–99.999 percent [2], and it is the local numbers an auditor will want to see.
99.97% at 0.3 Micron — and the MPPS Paradox
That famous 99.97 percent at 0.3 micron comes from the US DOE standard, and buried inside it is a subtlety most buyers never notice. A HEPA filter is not at its weakest against the smallest particles. It is weakest at the Most Penetrating Particle Size — somewhere around 0.1 to 0.3 micron, depending on how fast the air moves through the media.
Below the MPPS, diffusion runs the show. Brownian motion knocks tiny particles across air streamlines until one of them bumps into a fiber and sticks. Above the MPPS, interception and inertial impaction take over — those particles are simply too heavy to swing around a fiber with the air. Right at the MPPS, both mechanisms are having an off day. That valley is the worst case the filter has to beat.
Velocity moves the valley. Push a filter past its rated airflow and the MPPS slides toward smaller sizes while penetration at the valley climbs. That is why EN 1822 demands testing at rated flow — a 610 × 610 × 120 mm element gets tested at its 1,000 m³/h rating, not at a flattering 700. [1]
Which answers a question we hear constantly: how effective are HEPA filters against particles far smaller than 0.3 micron? More effective, not less. The curve climbs again on the diffusion side.
How Effective Are HEPA Filters Across the Particle Size Range
How effective are HEPA filters at the extremes of the size range? The capture mechanisms trade places as particles grow, and that handover explains the U-shaped efficiency curve every manufacturer publishes.
Put some numbers behind it. Hit an H13 filter with a million particles at 0.3 micron and roughly 300 slip through. Hit the same media with a million particles at 0.01 or 3 micron and fewer than 100 make it. That is exactly what makes HEPA filters effective as a terminal barrier — the specification already sits at the media's worst point, so every other size is easy street.
Pressure Drop: The Hidden Cost of Effectiveness
How effective are HEPA filters in daily operation? Efficiency-wise, quietly excellent. But effectiveness has a running cost, and the fan collects it every hour the room runs.
Rated Airflow, Resistance and Energy
A Deiiang 610 × 610 × 120 mm HEPA filter is rated at 1,000 m³/h, with initial resistance somewhere around 220–250 Pa. Go to the 150 mm deep pleat and initial resistance drops to roughly 200 Pa at the same face velocity — same air, more media, less push.
Fan power follows one line of arithmetic:
P = Q × Δp / η
Q is flow in m³/s, Δp is pressure drop in Pa, and η is fan efficiency, usually 0.5–0.65. Run the numbers for one filter: 1,000 m³/h at 250 Pa with η = 0.6 works out to about 115 W. Now scale to a 40-filter bank — 4.6 kW of continuous filtration power, more than 4,000 USD a year in electricity at 0.10 USD/kWh, spent before the air has even reACHed the room.
So initial resistance is a genuine buying criterion, not spec-sheet decoration. Moving from a 120 mm to a 150 mm pleat, or to a lower-resistance media, buys back 20–40 Pa per stage. Multiply that by every filter, every hour, for the life of the system, and it stops looking small. How effective are HEPA filters per watt of fan power? Initial resistance decides that ratio before the room is even switched on.
What Happens as the Filter Loads
Three things happen as dust piles up:
- Pressure drop climbs toward the terminal value — typically 2–3× initial, around 500–600 Pa.
- Efficiency actually improves a little; the dust cake behaves like extra media.
- Fan energy rises, or airflow sags — one of the two, depending on fan control.
Managing that climb — not worrying about efficiency — is what keeps HEPA filters effective through the whole service interval.
Dust Holding Capacity and Real-World Degradation
How effective are HEPA filters after one year of service, or five? That comes down to dust holding capacity and to what you ask the filter to catch. A standard 610 × 610 × 292 mm separator-type element will swallow something on the order of 400–800 grams of particulate before terminal resistance, depending on the dust. Deeper pleats and graded-density media stretch that further.
Look at the side of a deep-pleat element and you are looking at dust holding capacity made visible. The 150 mm pleat pack carries roughly 25 percent more media than the 120 mm format at the same face area, which is why it starts life at a lower pressure drop and takes longer to reach the end of it.
Fig.2 — Deep-pleat HEPA element, side view. Pleat depth is the cheapest way to buy dust holding capacity, and it is what keeps HEPA filters effective between changeouts.
When filters age in the field, efficiency is almost never the thing that fails:
- Media efficiency stays at or above the certified value for the whole service life — a loaded filter cleans better, not worse.
- What degrades is everything around it: rising pressure drop, fan strain, gaskets taking a compression set, frames corroding in humid or chemical-laden ducts.
- Mechanical damage is the real killer. In Deiiang's after-sales records, filters dropped during installation, media torn by pressure pulses, and pinholes from careless tools outnumber every other failure cause combined.
That is why this industry treats effectiveness as a system property you verify in place, not a media property you assume from a certificate. Gentle handling and a correct installation are what keep HEPA filters effective in the field.
Pre-Filters: The Cheapest Way to Keep HEPA Filters Effective
How effective are HEPA filters over a full service life? That is decided upstream of the HEPA. The most cost-effective decision in any filtration train is pre-filtration, full stop. Deiiang's medium-effect bag filters in classes F5–F9 (EN 779) were built for exactly this job. [3]
The arithmetic could not be simpler. Atmospheric dust skews heavily toward coarse particles above 1 micron — the fraction a bag filter takes out at 80–95 percent efficiency, for a few dollars of media. Skip the pre-filter stage and all of that coarse dust lands on the HEPA, burning through its dust holding capacity on particles it never needed to see.
| Configuration | Added pre-filter Δp | Typical HEPA life |
|---|---|---|
| No pre-filter | 0 Pa | 12–18 months |
| G4 panel | +30 Pa | 18–24 months |
| F7 bag | +45 Pa | 30–42 months |
| F5–F9 staged | +60 Pa | 36–48 months |
| Pre-filter | Added Δp | HEPA life |
|---|---|---|
| None | 0 Pa | 12–18 mo |
| G4 panel | +30 Pa | 18–24 mo |
| F7 bag | +45 Pa | 30–42 mo |
| F5–F9 staged | +60 Pa | 36–48 mo |
Price it out on a 40-filter bank. At 120 USD per HEPA and 60 USD per position to change with downtime included, stretching the interval from 18 to 36 months saves 40 × 180 = 7,200 USD per cycle, against roughly 1,800 USD of extra bag filter spend. And the bags swap out in half an hour each — a HEPA changeout ends in re-scan testing.
One caveat before you over-specify. Every pre-filter stage adds its own pressure drop and fan energy. An F5–F9 bag stage — Deiiang's 592 × 592 × 381 mm format handles 2,050 m³/h — is the sweet spot for most HVAC-fed cleanrooms. Go further and you are buying pressure drop to save media you did not need to save. Getting that balance right is what keeps HEPA filters effective and economical in the same breath.
Testing and Certification: How Effectiveness Gets Verified
How effective are HEPA filters once they leave the factory? A factory certificate says the media passed a test. Only in-place verification says your installed bank is effective. Most cleanroom contamination problems are born in the gap between those two sentences — and certification exists to close it. Certification is what proves HEPA filters effective on site.
Test Methods Compared
| Method | Standard | What it measures | Where used |
|---|---|---|---|
| MPPS particle counting | EN 1822-1 / ISO 29463 | Penetration at the most penetrating particle size | Factory type testing |
| DOP/PAO scan test | EN 1822-4 | Local leaks across the whole filter face | In-place commissioning [4] |
| Counting scan with CNC probe | IEST-RP-CC034 | Local penetration mapping | cleanroom certification [5] |
| Counting at 0.3 μm (laser) | iso 14644-1 practice | Room cleanliness verification | room classification [6] |
| Method | Measures | Where used |
|---|---|---|
| MPPS counting | Penetration at MPPS | Factory type test |
| DOP/PAO scan | Local leaks, full face | In-place commissioning |
| CNC counting scan | Local penetration map | Certification |
| Laser count at 0.3 μm | Room cleanliness | Classification |
The DOP scan test is the in-place workhorse. You challenge the filter upstream with a PAO/DOP aerosol concentrated enough that any leak lights up immediately, then walk an isokinetic probe across the entire downstream face — typically 2–5 cm/s probe travel, full area, no skipped strips. Any local penetration above the class limit flags a repair or a rejection. Deiiang's DOP integrated housings carry a built-in test port for precisely this: upstream injection and downstream sampling with nothing dismantled.
On our factory rig, every element — not a sample, every single one — is scanned before it ships, and repairs are capped per EN 1822 rules (total repaired area under about 1.5 percent of face area depending on class). A heavily patched filter is a red flag even when it passes today.
Reading a Scan-Test Report
Three scores separate a genuinely effective installation from a hopeful one:
- Overall penetration — must meet the class value (99.95% or better for H13 at MPPS).
- Maximum local penetration — the pinhole detector; it must stay within the class local limit.
- Repair coverage — the standards cap it, so ask what was patched and where.
HEPA vs ULPA: When Is More Effective Worth Paying For
After "how effective are HEPA filters", the next question is almost always "do I need ULPA instead?" ULPA (U15–U17 under EN 1822) lifts the bar from 99.95 to 99.9995 percent and beyond, and moves the rated particle size down to the 0.12 micron range.
| Attribute | HEPA H13 | HEPA H14 | ULPA U15 | ULPA U16/U17 |
|---|---|---|---|---|
| Efficiency at MPPS | ≥ 99.95% | ≥ 99.995% | ≥ 99.9995% | ≥ 99.99995%+ |
| Rated particle size | 0.3 μm | 0.3 μm | 0.12 μm | 0.12 μm |
| Relative pressure drop | baseline | +10–20 Pa | +30–60 Pa | higher still |
| Media / separator | glass fiber, EVA hot-melt | same | finer HV glass fiber, baffle-free | finest grade |
| Typical application | ISO 7–8, general pharma | ISO 5–6, Grade A/B | semiconductor, optics | wafer fab, nano research |
| Class | Efficiency | Rated size | Typical use |
|---|---|---|---|
| H13 | ≥ 99.95% | 0.3 μm | ISO 7–8 pharma |
| H14 | ≥ 99.995% | 0.3 μm | ISO 5–6, Grade A/B |
| U15 | ≥ 99.9995% | 0.12 μm | Semiconductor, optics |
| U16/U17 | ≥ 99.99995%+ | 0.12 μm | Wafer fab, nano |
The trade-off is real. ULPA media is denser and finer, so resistance runs higher at the same face velocity — which means more fan energy or bigger banks. Deiiang's ULPA elements use American HV ultra-fine glass fiber media with baffle-free thermoplastic separators, rated from 0.12 micron up, and they mostly end up in semiconductor fabs, wafer processing and atomic research facilities.
Here is the decision rule we give clients. Specify ULPA when the process generates, or is damaged by, particles below 0.1–0.2 micron — photolithography, precision optics coating. For pharmaceutical Grade A zones, food cleanrooms and hospital isolation rooms, H13/H14 already clears the requirement with room to spare, and ULPA money spent there buys pressure drop instead of cleanliness. Matching class to the real ISO 14644-1 target [6] is what keeps HEPA filters effective and economical in the same sentence.
Two elements, same envelope, different jobs. Matching the media grade to the particle size that actually threatens your process — and running it at rated face velocity — is what keeps HEPA filters effective at the specified point instead of just at the certified one.
Fig.3 — Two standard HEPA elements side by side: same frame format, different class and pleat geometry.
Deiiang HEPA Product Data: Sizes, Airflow and Resistance
How effective are HEPA filters in your specific bank? Start from the rated airflow column. The tables below cover Deiiang's standard DOP integrated HEPA filter range: ultra-fine glass fiber media, PU two-component sealant, EVA hot-melt separators, EVA/neoprene gaskets, powder-coated diamond mesh guards, and your choice of aluminum, galvanized steel or stainless steel frames. Rated service conditions: 70 °C maximum, 80% RH maximum.
| Model size (mm) | Depth | Rated airflow (m³/h) | Initial resistance (Pa) | Class | Frame / sealant |
|---|---|---|---|---|---|
| 610 × 610 × 120 | 120 mm | 1,000 | ~220–240 | H13 / H14 | Alu / galv. / SS; PU two-part |
| 1170 × 570 × 120 | 120 mm | 1,500 | ~230–250 | H13 / H14 | Alu / galv. / SS; PU two-part |
| 1220 × 610 × 120 | 120 mm | 1,800 | ~230–250 | H13 / H14 | Alu / galv. / SS; PU two-part |
| 610 × 610 × 150 | 150 mm | 2,500 | ~200–220 | H13 / H14 | Alu / galv. / SS; PU two-part |
| 915 × 610 × 150 | 150 mm | 1,900 | ~200–220 | H13 / H14 | Alu / galv. / SS; PU two-part |
| Size (mm) | Airflow (m³/h) | Initial Δp |
|---|---|---|
| 610 × 610 × 120 | 1,000 | ~220–240 Pa |
| 1170 × 570 × 120 | 1,500 | ~230–250 Pa |
| 1220 × 610 × 120 | 1,800 | ~230–250 Pa |
| 610 × 610 × 150 | 2,500 | ~200–220 Pa |
| 915 × 610 × 150 | 1,900 | ~200–220 Pa |
Read that table with airflow discipline. Oversizing — running a 1,000 m³/h element at 700 — drops resistance and stretches life. Undersizing, pushing that same element to 1,300 to save one position, sends resistance up roughly with the square of media velocity, shortens life, and drags local efficiency toward the media's velocity limit. For most H13/H14 media the sweet spot sits at or just below rated flow.
Deiiang Field Case: An ISO 7 Retrofit Audited After Two Years
Background. A medical device manufacturer in Jiangsu ran a 480 m² ISO 7 assembly cleanroom whose filter bank was eleven years old and had never once been scan-tested in place.
The problems. The audit turned up what old installations usually hide, and it reframed the plant's question for them: how effective are HEPA filters after a decade of unverified service? Three of 62 terminal filters had gasket leaks at the knife-edge seal, with local penetration above 0.1 percent — a failed H13 by a factor of 200. The rest still passed overall efficiency on spot counts, but average bank resistance read 540 Pa against roughly 230 Pa initial. The fan was pinned at 92 percent speed just to hold design airflow, and the room lost its pressure cascade every time the door interlocks cycled.
Fig.4 — DOP integrated HEPA terminal with built-in test port: upstream challenge and downstream sampling without dismantling the installation.
The solution. Deiiang replaced all 62 elements with 610 × 610 × 150 mm H13 filters (2,500 m³/h rated, run at ~1,600 m³/h actual) behind a new F7 bag pre-filter stage, then scan-tested every position.
The results. Zero repairs; all 62 filters came in below 0.005 percent local penetration. particle counts at 0.3 micron fell 40 percent at the worst monitoring point. Fan speed dropped to 68 percent, recovering about 3.1 kW continuously. Two years on, average bank resistance has crept only from 205 Pa to 290 Pa — a loading rate tracking toward a six-to-seven-year change interval, against three years for the unprotected previous generation. That loading curve is the clearest field answer we have to "how effective are HEPA filters" once the bank is protected and verified.
Pre-filter spend runs about 900 USD a year; the deferred changeout alone is worth roughly 16,000 USD before counting avoided downtime and re-validation. Proper pre-filtration is what kept the new HEPA filters effective at two years, and it is what will keep them effective at five.
So, How Effective Are HEPA Filters? The Verdict
So, how effective are HEPA filters? At the rated test point, extremely effective — 99.97 percent at 0.3 micron is the media's worst case, not its best. Over time, effectiveness holds or even improves; what expires is pressure-drop life. In the field, effectiveness rides on installation quality and in-place testing, not on the media alone. And economically, effectiveness is a system property: pre-filtration, fan energy and test discipline decide the cost per clean cubic meter.
The same logic scales down to a living-room purifier. A true HEPA filter with a stated efficiency at 0.3 micron, run at rated airflow and changed on schedule, will do for a bedroom what an H13 terminal does for an ISO 7 room — capture essentially every particle that matters.
Before you sign a purchase order, run this checklist:
Deiiang™ manufactures DOP integrated HEPA filters (H13/H14) and ULPA elements (U15–U17) for cleanrooms from ISO 3 to iso 8. Product designer: Jason Peng.
FAQ: HEPA Filter Effectiveness
Do HEPA filters lose effectiveness over time?
No — media efficiency holds or slightly improves as dust loads the fibers, because the dust cake adds capture surface. What runs out is pressure-drop life. ｒｅｐｌａｃｅ when resistance reaches the terminal value, typically 2–3× initial (around 500–600 Pa).
How effective are HEPA filters at capturing particles smaller than 0.3 μm?
More effective than at 0.3 μm. The MPPS — typically 0.1–0.3 μm — is where a HEPA filter performs worst; below it, diffusion capture takes over and penetration falls. Quoting 99.97% at 0.3 μm therefore understates effectiveness at ultra-fine sizes.
Are HEPA filters effective against 0.1 μm particles, or do I need ULPA?
Yes — 0.1 μm sits at or near the MPPS where H13/H14 are still certified above 99.95 percent, so rated media is what makes HEPA filters effective at 0.1 μm as well. Specify ULPA (U15–U17, rated at 0.12 μm) only when the process is damaged by particles that size, such as photolithography.
Are HEPA filters effective in home air purifiers?
When the filter is a true HEPA — EN 1822/DOE-style media with a stated efficiency at 0.3 μm — yes. Beware "HEPA-type" or "HEPA-like" marketing labels, which carry no defined efficiency. Sealed body design and rated airflow matter as much as the media in portable units.
How often should a HEPA filter be replaced in a cleanroom?
When pressure drop reaches the terminal value, not on a fixed calendar — typically 2–3× initial resistance, around 500–600 Pa for most H13/H14 elements. With an F5–F9 pre-filter stage that usually lands at 3–6 years; without pre-filtration, expect 1–2 years.
What is the difference between HEPA and ULPA efficiency classes?
HEPA (H13/H14) is certified at ≥ 99.95 / 99.995 percent at an MPPS around 0.3 μm. ULPA (U15/U16/U17) is certified at ≥ 99.9995 percent and up at an MPPS around 0.12 μm. ULPA trades 30–60 Pa of extra resistance for that margin.
What does a DOP scan test actually verify?
That the installed filter, gasket and housing together meet local penetration limits across the entire face. A probe traverses the downstream side under an upstream PAO/DOP challenge; any reading above the class local limit marks a leak needing repair or replacement. It is the step that turns factory claims into proof and keeps HEPA filters effective in service.
Do pre-filters really extend HEPA filter life?
Yes — roughly doubling to tripling it in most HVAC-fed cleanrooms. An F5–F9 bag stage removes the coarse fraction that would otherwise consume HEPA dust holding capacity, typically returning several times its cost in deferred changeouts and avoided downtime.
References
- EN 1822: High efficiency air filters (EPA, HEPA and ULPA) — Classification, performance requirements, testing: en-standard.eu — EN 1822
- US DOE Technical Standard 3020: Specification for HEPA filters used by DOE contractors: energy.gov
- EN 779: Particulate air filters for general ventilation — Determination of the filtration performance: en-standard.eu — EN 779
- EN 1822-4: HEPA/ULPA filter elements — Determining leakage of filter elements (scan method): en-standard.eu — EN 1822-4
- IEST-RP-CC034: HEPA and ULPA filters, leak testing: iest.org
- iso 14644-1 / 14644-2: Cleanrooms and associated controlled environments — Classification and monitoring: iso.org
