Two international filter classification systems are in common use and they rest on different bases, so no exact conversion between them is defensible. This page explains the difference, what a specifier should do when documents and products disagree, and how the Al Sa'fat minimum applies in Dubai.
ASHRAE Standard 52.2, published by ASHRAE, a registered mark of that society, is currently issued as Standard 52.2-2025, Method of Testing General Ventilation Air-Cleaning Devices for Removal Efficiency by Particle Size. Its MERV classification is widely used in project specifications, while ISO 16890 is a separate international filter classification system published by ISO. The systems use different test and reporting bases, so neither should be described as an equivalent label for the other.
Filter classification developed through different standards communities and market practices. Standard 52.2 characterises performance across defined particle-size groupings and reduces the prescribed test results to a single minimum efficiency reporting value. ISO 16890 classifies general ventilation filters in relation to particulate-matter fractions used in ambient-air reporting and considers the filter's efficiency against those fractions.
Both systems are intended to make product performance more comparable, but they organise the information differently. MERV asks how the product's worst measured efficiency fits within the categories of its method. ISO 16890 asks how the measured efficiency curve contributes to removal of particulate-matter fractions represented by the international system.
The existence of two systems does not mean that one is globally correct and the other obsolete. A project may encounter either because of the design consultant's practice, the jurisdiction, the manufacturer's market or the source of the equipment. The technical task is to preserve the intended filtration performance when documents and products use different languages.
A MERV value is derived from the lower performance observed across the relevant size groupings and loading stages prescribed by Standard 52.2. The result is a single category. The category does not state one universal removal percentage and cannot be interpreted without understanding the method behind it.
ISO 16890 uses measured fractional efficiency to estimate performance against ambient particulate-matter fractions. That basis relates the laboratory efficiency curve to particle-size distributions used to describe outdoor particulate matter. Its group label therefore communicates a different type of result from MERV, even where the same physical filter has been tested under both systems.
Because the calculations, conditioning, aerosols, reporting conventions and classification rules differ, two products that appear close under one system may not retain the same order under the other. A cross-reference can be useful for screening, but it cannot reproduce the test outcome that the missing classification would have provided.
An exact conversion would require a one-to-one mathematical relationship between the classification rules. That relationship does not exist. Each system processes the underlying efficiency curve differently, and a single category hides details that may affect the result under the other method.
Published conversion tables usually group classifications into broad approximate bands. They can help a project team understand the likely region of performance, but they should not be used as proof that a product satisfies a specification written in the other system. The apparent precision of a table can exceed the evidence behind it.
The problem is especially clear near category boundaries. Two filters may share one MERV classification while having different efficiency curves. When assessed under ISO 16890, those differences can lead to different particulate-matter classifications. The reverse can also occur. An approximation may therefore be suitable for early design discussion but not for final product acceptance.
Where the project document names MERV and a manufacturer quotes only ISO 16890, the first step is to request a test report for the offered product under Standard 52.2. Where the project names ISO 16890 and only MERV evidence is available, the corresponding ISO report should be requested. Dual-tested products remove ambiguity because each classification is supported directly.
If direct evidence is unavailable, the design team should identify the performance objective rather than silently substituting an approximate conversion. Relevant questions include which particle fraction matters, what pressure drop the system can accommodate, whether the filter dimensions and media construction match the equipment, and whether the specification permits an alternative classification route.
Any accepted alternative should be recorded as a project decision. The documentation should explain the evidence reviewed, the limitations of any approximation and the authority responsible for accepting the substitution. This is a specification-management exercise, not a declaration that the systems are equivalent.
A clear specification can avoid the problem by naming the required test system and classification, requiring the underlying report, and stating whether equivalent proposals will be considered. Where both systems are commonly encountered, the specification may request results under both without attempting to create an exact conversion.
Higher particle-removal efficiency commonly requires denser media, greater media area, different fibre characteristics or a combination of these features. These choices can increase resistance to airflow. The filter's pressure drop therefore needs to be considered alongside its classification rather than after the product has been selected.
A fan must provide enough pressure to move the required airflow through the filter and the rest of the system. Increasing filter resistance can increase fan power, reduce delivered airflow where the fan cannot compensate, or move the operating point away from the intended condition. The consequence depends on the fan, controls, available reserve and system curve.
The initial pressure drop is only part of the picture. Resistance generally changes as material accumulates. A product with a low initial value may rise differently from another product with similar classification. The selection should therefore consider the manufacturer's tested resistance data and the system's allowable operating range rather than comparing efficiency labels alone.
Energy efficiency and filtration should not be treated as opposing absolutes. The design objective is to obtain the required particle-removal performance with a filter and system arrangement that can deliver the necessary airflow efficiently. Greater media area, suitable filter depth and appropriate fan control may reduce the energy penalty compared with forcing air through a smaller, more restrictive product.
Al Sa'fat, the Dubai Green Building System, 2nd edition, January 2023, clause 401.02, requires minimum MERV 8 for temporary return air filters when the heating, ventilation and air-conditioning system is used during construction or renovation and minimum MERV 8 for the permanent replacement return air filters installed before occupancy. This requirement captures both the construction-phase filter and the operational filter rather than applying only after handover.
The clause does not turn MERV into a building-performance guarantee. It establishes a minimum product classification for the stated return-air applications. The project still needs to select filters that fit the equipment duty, provide acceptable resistance and are supported by evidence for the supplied product.
The construction-phase requirement also has a distinct purpose. Temporary filters are part of the measures used when the air-conditioning system operates during works, while the permanent filters are installed for occupancy. A project cannot satisfy the permanent requirement by leaving temporary filters in place, because the clause expressly distinguishes removal and replacement before occupancy.
Al Sa'fat's mandatory tier is Silver, and Dubai Municipality describes the system as requiring new buildings to meet the Silver Sa'fa before optional higher tiers are pursued. The compliance route, project scope, certificate process and verification responsibilities belong to a separate UAE compliance resource and are not explained on this page.
A filter submittal should identify the manufacturer, exact model, dimensions, media construction, classification and the test standard used. The attached report should correspond to the offered product. A catalogue statement that a product is "equivalent" to another system is not the same as a direct test result.
Pressure-drop information should be reviewed at the airflow relevant to the design. Where several filter sizes or depths are offered under one family name, the reviewer should confirm that the evidence applies to the selected configuration. Small changes in construction can alter resistance and efficiency even when the marketing description remains similar.
The specification should also address replacement availability. Selecting a product that cannot be sourced in the required dimensions may encourage later substitution with poorly documented alternatives. Standardised frame sizes, clear model references and retained test evidence support consistent procurement without converting the rating into a maintenance instruction.
Filter condition, fit, bypass, change intervals, duct and air-handling-unit hygiene, and the air-quality effects of dirty filters are treated by a separate duct-hygiene resource and should not be inferred from either classification label.
One error is to write "MERV or equivalent ISO" without defining how equivalence will be assessed. That phrase transfers a technical decision to the supplier and may lead to acceptance based on a generic conversion chart. A better approach is to require direct evidence under the named system or a documented engineering review of an alternative.
Another error is to select the highest available efficiency without checking fan capability. A higher classification can be beneficial, but only when the system can maintain its intended airflow and energy performance. The filter and fan need to be considered as interacting components.
A third error is to accept a classification for a media type rather than for the exact product. Ratings are established through tests on defined articles. The same general media description does not guarantee identical construction, depth, pleat geometry or resistance.
The defensible selection process is therefore evidence-led. It starts with the project requirement, confirms the applicable classification system, reviews the direct test report, checks pressure drop against the system duty and records any approved alternative. Conversion tables can inform discussion, but they should not replace product evidence.
Selecting the right class is a design question; keeping that performance once the plant is running is a separate maintenance one, covered in how filter condition, bypass and change criteria are handled in an operating system.
Edition years matter here more than they usually do, because UAE instruments reference these standards in two incompatible ways. Al Sa'fat, Dubai's green building system, whose Silver Sa'fa requirements are mandatory for new buildings, requires the latest edition of Standards 62.1, 62.2 and 170 and deliberately attaches no year, so the applicable requirements move as the standards are revised. Estidama's Pearl Building Rating System, Version 1.0 of April 2010, does the opposite: it names specific editions — the 2007 edition of Standard 62.1, the 2007 edition of Standard 62.2, the 2004 edition of Standard 55 and the 2007 edition of Standard 90.1 — so the requirements it imposes are frozen at those editions regardless of what has been published since. Dubai Municipality's Technical Guidelines for Indoor Air Quality for Healthy Life, Version 4 of 11 December 2024, references Standard 62.1 without giving an edition year. Anyone reading a requirement should therefore establish which instrument imposes it and whether that instrument names a year, before establishing what the current edition says.
Al Sa'fat §401.01 (2nd edition, January 2023) requires the latest edition and attaches no year — Estidama PBRS Version 1.0 (April 2010) names fixed editions
No. The systems use different test, calculation and reporting bases. Approximate comparison bands may help with preliminary discussion, but direct test evidence is needed to demonstrate a classification under the system named by the project.
The project should request a report under the specified system or obtain a documented engineering review of the proposed alternative. Silent substitution using a conversion chart is not technically robust.
Not automatically. Higher efficiency may increase pressure drop, fan energy or the risk of reduced airflow where the system lacks capacity. Selection must consider efficiency and system resistance together.
No. Clause 401.02 addresses temporary return-air filtration when the system operates during construction or renovation and requires replacement with permanent return-air filters before occupancy, with the same stated minimum classification applying to both stages.
This is an independent information resource. It is not affiliated with, endorsed by, or connected to ASHRAE. ASHRAE and the names of its standards are the trademarks of their respective owners and are used here only to identify the standards described.