Industrial Filtration Media — Technical Specification Overview

Filtration Media Substrate Comparison — The Parameters That Determine Field Performance #

The filter class printed on a datasheet tells you the minimum efficiency at the most-penetrating particle size. It does not tell you how fast the media will blind, how it behaves under cyclic loading, or whether it will hold its rated efficiency after the first pulse-jet cleaning cycle. Those are the parameters that drive maintenance intervals and total operating cost — and they are almost never requested at the purchasing stage.

For industrial filtration applications, the substrate material governs four variables that need to be specified simultaneously: initial filtration efficiency, clean pressure drop (Pa at rated face velocity), dust holding capacity (g/m²), and residual pressure drop after cleaning cycles. Specifying only filter class and omitting the latter two is the most common specification gap we see in incoming qualification files.

The table below compares five common industrial filtration media substrates across these four parameters plus two additional procurement-relevant values. Data is drawn from supplier COAs and independent incoming test results across 47 qualification lots evaluated over 18 months using our QC-F12 media verification protocol.

Efficiency values per ISO 16890-3; pressure drop measured at rated face velocity per ASHRAE 52.2; dust holding capacity per ISO 15957. Values represent ranges across multiple verified supplier lots.

The column most buyers skip: residual ΔP after 500 cleaning cycles. For pulse-jet systems running 6,000–8,000 hours annually, residual pressure drop accumulation determines fan energy consumption and bag replacement frequency more directly than initial efficiency does. A glass fiber media with 320 Pa residual ΔP will increase your fan energy draw by roughly 15–22% compared to a PTFE-laminated bag at 240 Pa — across a full filter population, that is not a marginal cost.

Supplier Qualification — What to Request and What the Response Tells You #

When we initiate qualification of a Chinese filtration media supplier, the first document request is not a product datasheet. It is three consecutive lot COAs covering efficiency, basis weight, and clean ΔP — all tested at the same face velocity, all referencing the same test standard. The response pattern tells you more about supplier capability than the numbers themselves.

Suppliers with genuine in-house testing infrastructure will return consistent COA formatting, matched test conditions, and lot-to-lot variation within ±5% on efficiency and ±8% on clean ΔP. Suppliers relying on periodic third-party testing — which is common among mid-tier Chinese media producers — will show gaps in lot coverage, variable test conditions between lots, or COAs that reference face velocities not matching your application. When we see efficiency data tested at 0.5 m/s face velocity on a COA for media specified at 1.0 m/s, we flag it immediately. The efficiency number is not wrong — it is just not the number you need.

Ask specifically: “Please provide test data for basis weight per ISO 9073-1, filtration efficiency per ISO 16890 at [your specified face velocity], and clean pressure drop at rated airflow — for the last three production lots.” A supplier who cannot provide this within five business days either does not run this testing routinely or does not maintain traceability between lots. Either answer is useful information at the qualification stage.

The second request: dimensional verification data. Filter bags cut from the same roll of media can show width variation of ±8 mm and length variation of ±12 mm if the supplier’s slitting and finishing process is not controlled. In caged filter systems, that dimensional spread translates directly to bypass leakage around the bag seat — which destroys system efficiency regardless of what the media itself is rated at. Request dimensional inspection reports for a minimum of 30 units per lot, not per SKU.

One practice that separates serious suppliers from commodity converters: ask for the media roll traceability record linking the finished filter bag to the raw media roll number and compounder batch. Roughly one in three Chinese suppliers we have qualified cannot provide this linkage. When it is absent and a batch fails incoming inspection, root cause analysis becomes guesswork.

Cost-Performance Trade-offs in Filtration Media Selection #

The price differential between standard needle-felt polyester and PTFE-laminated media is substantial — typically in the range of 3× to 4× on a per-bag basis at comparable sizing. For most procurement teams, that gap triggers an automatic preference for the lower-cost substrate. In some applications, that is the correct decision.

Specifically: in low-dust-loading environments with non-hygroscopic, coarse particulate (median particle diameter above 10 µm, inlet loading below 5 g/Nm³), needle-felt polyester at 400–550 g/m² basis weight will perform acceptably through 2–3 years of service without reaching the pressure drop ceiling. The PTFE-laminated option adds cost with no meaningful return in that scenario. I would not specify it there.

The calculus changes — sharply — when the dust is fine (sub-3 µm mass fraction above 30%), hygroscopic, or chemically aggressive. PTFE surface filtration prevents depth loading, which is the failure mechanism that drives premature blinding in those conditions. Bags that blind within 4–6 months in fine-dust service are almost never a media quality problem. They are a media selection problem. The cost of two replacement cycles on a 200-bag system at accelerated replacement frequency will exceed the initial premium of PTFE lamination within the first operating year.

Glass fiber media occupies a specific niche: continuous temperatures above 200°C combined with efficiency requirements above 90% ePM₁. Below 200°C, PTFE-laminated polyester matches the efficiency at lower cost and substantially better flex fatigue resistance. Glass fiber in pulse-jet systems below 180°C is a specification error, not a performance upgrade — fiber attrition from repeated flexing degrades the media and contaminates the downstream environment.

The counterargument worth making: cellulose-polyester blends are systematically underused in ambient-temperature applications with moderate dust loading. Basis weight in the 250–350 g/m² range, clean ΔP in the 140–200 Pa range, and unit cost typically 20–35% below equivalent needle-felt polyester. For HVAC pre-filtration stages or low-loading industrial recirculation, the performance-per-unit-cost ratio is favorable. The limitation is temperature: service above 80–90°C degrades cellulose fiber integrity, and that ceiling is lower than most datasheets state for continuous operation.

Particle Size Distribution vs. Filter Class — Why the Rating Doesn’t Tell the Whole Story #

This is the specification gap that causes the most field failures in Chinese-sourced filtration media, and it is worth examining in detail.

Filter class — whether expressed as MERV per ASHRAE 52.2, ePM category per ISO 16890, or EN classification per EN 1822 — is a single-point efficiency value at the most-penetrating particle size under standardized test conditions. It is measured at one velocity, one humidity, one dust type, one loading state. Production reality deviates from all four.

The parameter that filter class cannot encode is the relationship between your actual particle size distribution (PSD) and the media’s fiber diameter and pore structure. A media rated at 75% ePM₁ efficiency against the ISO 16890 test aerosol may perform at 60% or at 88% against your actual process dust — depending entirely on whether your sub-micron fraction is heavier or lighter than the test aerosol’s. Without site-specific PSD data, media selection by class alone is an approximation.

This matters most in two scenarios. The first is metal grinding and cutting operations, where the sub-micron fraction of the aerosol is often 40–60% of the total mass loading — substantially higher than ISO 16890 test conditions assume. Needle-felt polyester specified at 68–78% ePM₁ will underperform its rating in this environment. The second is food and pharmaceutical processing, where hygroscopic dusts alter the effective particle size in transit and cause the media to behave as though it is filtering a finer, stickier aerosol than the dry-state PSD suggests.

The approach we use for high-criticality applications is what our internal procedure calls the PSD alignment check: the buyer provides representative dust sampling data (volume-weighted PSD from laser diffraction, median D50 and D90 values), and we map the distribution against the media’s known efficiency curve across the 0.3–10 µm range. For Chinese suppliers, this curve is almost never included in standard documentation — it requires specific request and, often, third-party characterization. Of the 14 Chinese filtration media suppliers in our current approved vendor list (AVL), only four can provide full efficiency-vs-particle-size curves from their own laboratory. The remaining ten require third-party testing before we will approve their media for sub-5 µm applications.

This is also where the GB/T standard diverges meaningfully from ISO 16890. GB/T 14295 governs air filter performance in China and uses test conditions and efficiency metrics that do not map directly onto ISO 16890 ePM categories. A supplier quoting “GB/T 14295 compliant, 85% efficiency” is describing performance against a different test aerosol at different conditions than ISO 16890 ePM₁ 85% — and the two values are not interchangeable. For export-oriented procurement, always request ISO 16890 test data explicitly. If the supplier can only produce GB/T data, that is not disqualifying, but it requires recalibration of the efficiency expectation before specification sign-off.

One limitation we are still tracking: our PSD alignment check dataset covers mineral and metallic dusts well. Organic dusts — agricultural, pharmaceutical, food processing — are underrepresented in our reference library, and we are building that dataset through 2025. For buyers in those sectors, the mapping exercise is indicative rather than definitive, and third-party characterization is recommended.

Practical Guidance for Buyers #

When sourcing industrial filtration media from China, start the specification process with particle size distribution data from your process, not with a filter class number. The class tells you the minimum acceptable outcome — the PSD tells you whether a given media will actually reach it in your operating conditions.

The risk scenario that recurs in our qualification program: a buyer specifies PTFE-laminated media at 90%+ ePM₁, the Chinese supplier delivers bags that pass COA review on efficiency, and the system reaches maximum ΔP within 60 days of commissioning. The failure is almost never the PTFE membrane itself. In five of the seven cases we have investigated under this profile, the root cause was basis weight below specification (typically 460–480 g/m² delivered against a 520 g/m² spec), which reduces the structural support layer and causes the PTFE membrane to deform under pulse-jet pressure — progressively increasing residual ΔP after each cleaning cycle. A standard COA efficiency test does not load the media long enough to reveal this.

Before volume commitment, insist on the following: a 30-bag dimensional inspection report (width, length, seam integrity), a COA with basis weight per ISO 9073-1 tested at the correct face velocity, and a 500-cycle pulse-jet simulation test on 3 representative bags with residual ΔP recorded at cycles 100, 300, and 500. That test takes 10–14 days and catches the deformation failure mode before it reaches your production floor.

For dust and air filtration systems operating in high-temperature or chemically aggressive environments, add a thermal conditioning test: 24 hours at rated temperature, then re-measure efficiency and ΔP. Efficiency degradation above 5 percentage points after thermal conditioning is a disqualifying result in our QC-F12 protocol.

FAQ

What is the most commonly misspecified parameter when sourcing filtration media from China?
Basis weight tolerance. Buyers specify the nominal value but omit the acceptable range, and Chinese suppliers default to the widest tolerance their process produces — often ±15% on standard polyester media, which is wide enough to meaningfully affect both ΔP and mechanical performance.

Does ISO 16890 efficiency rating translate directly to GB/T 14295 efficiency rating?
No. The test aerosols, conditioning methods, and efficiency calculation protocols differ, and the two values are not numerically equivalent. A media tested to GB/T 14295 at 82% cannot be assumed to meet ISO 16890 ePM₁ 82% — in our experience the actual ISO 16890 result is typically 4–9 percentage points lower due to the more demanding test conditions.

At what face velocity should pressure drop be specified?
At your actual system face velocity, not the supplier’s default test velocity. This sounds obvious. In practice, roughly half the COAs we receive from Chinese suppliers use 1.27 m/s (the ASHRAE default) regardless of the application — and for systems running at 0.8 m/s or 1.5 m/s, those pressure drop values are not directly usable.

When is glass fiber media the right choice over PTFE-laminated polyester?
Continuous service above 200°C. Below that threshold, PTFE-laminated polyester matches efficiency at lower cost and handles pulse-jet flexing better. Specifying glass fiber for sub-180°C applications is a reliability risk, not a performance advantage.

How many lots should I evaluate before approving a new Chinese filtration media supplier?
Three consecutive production lots, not three samples from the same lot. Lot-to-lot consistency — specifically ΔP and basis weight variation across separate production runs — is the variable that predicts long-term supply reliability. One lot can be optimized for sample approval; three consecutive lots cannot.

Can cellulose-polyester blended media be used in outdoor industrial applications?
It depends on the humidity profile. Above 80% RH continuous exposure, cellulose fiber absorbs moisture and the ΔP climbs rapidly, sometimes reaching the blinding threshold within weeks. For outdoor applications with controlled moisture, it is usable up to approximately 80°C continuous. For humid equatorial environments or installations near water treatment, we would not approve it without a humidity conditioning test first.

What does residual pressure drop after cleaning cycles tell me that initial ΔP does not?
It tells you how much of the dust loading is irreversible — embedded in the media structure rather than released by the cleaning pulse. A media with low initial ΔP but high residual ΔP accumulation after 300–500 cycles will drive up fan energy consumption and force earlier replacement than a media with slightly higher initial ΔP and stable residual performance. That trade-off is invisible if you only test clean media.

Published by sinoraw.com Technical Team | Request a sourcing consultation