Martindale fabric test explained for abrasion and pilling

What the Martindale fabric test measures

The Martindale fabric test is a laboratory method used to evaluate how a textile surface changes when it is rubbed under controlled conditions. In specifications and test reports, the term is often used for two related but different checks: abrasion resistance and pilling or surface change. Abrasion testing measures how long a fabric resists wear before a defined endpoint, such as yarn breakage, mass loss, specimen breakdown, or unacceptable appearance change. Pilling testing measures the amount of fuzzing, matting, or pill formation that appears after rubbing.

For fabric sourcing, quality control, upholstery selection, and product development, the Martindale method is useful because it gives buyers and suppliers a repeatable way to compare materials. It is not intended to predict exact service life. Results are most meaningful when the same standard, load, abradant, endpoint, and conditioning procedure are used. For more textile performance topics, see the Testing section.

A common mistake is to treat a Martindale number as a direct durability promise. A fabric tested to 40,000 cycles is not automatically twice as durable in use as one tested to 20,000 cycles. The result depends on the selected method, fabric type, abradant, pressure, stopping point, and evaluation criteria. Professional test reports should therefore be read as controlled comparative evidence, not as a guarantee of how many years a sofa, garment, automotive trim, or public seating fabric will last.

How the Martindale abrasion method works

In a Martindale abrasion test, circular fabric specimens are mounted in holders and rubbed against a specified abradant in a controlled motion. The motion spreads wear across the specimen surface rather than applying abrasion in one straight line. The test continues for a set number of cycles or until the specimen reaches a defined endpoint. That endpoint may be specimen breakdown, visible yarn breakage, unacceptable appearance change, or another criterion agreed by the buyer, supplier, laboratory, or applicable standard.

The ISO 12947 series is one of the main references for Martindale abrasion. ISO 12947-1 covers the Martindale abrasion testing apparatus and auxiliary materials. ISO 12947-2 covers determination of specimen breakdown. ISO 12947-3 covers determination of mass loss. ISO 12947-4 covers assessment of appearance change. ISO catalogue information identifies ISO 12947-1, ISO 12947-3, and ISO 12947-4 as 1998 editions reviewed and confirmed in 2023, while ISO 12947-2 is the 2016 edition, also reviewed and confirmed in 2023.

ASTM D4966 is the commonly cited ASTM method for abrasion resistance of textile fabrics using the Martindale abrasion tester. The ASTM listing for D4966-22(2026) is active and includes an important limitation: abrasion test results can vary within and between laboratories, and the method should not be used carelessly as the sole basis for commercial acceptance. This does not make the method unsuitable. It means the result is strongest when used to compare similar fabrics under the same conditions, with a clearly defined pass or fail point.

Abrasion, pilling, and appearance change are not the same result

The word Martindale often appears beside rub count, abrasion cycles, pilling grade, or appearance rating. These terms are related, but they answer different questions. Abrasion resistance focuses on material wear. Pilling resistance focuses on the tendency of loose fibers to form small balls or other surface defects. Appearance change can include flattening, color shift, fuzzing, matting, pile loss, or a general surface change that becomes unacceptable before structural breakdown occurs.

ISO 12945-2:2020 addresses the modified Martindale method for determining fabric propensity to surface pilling, fuzzing, or matting. ASTM D4970/D4970M-22(2026) addresses pilling resistance and related surface changes using the Martindale tester. ASTM describes pilling as a complex property affected by fiber type, fiber dimensions, yarn and fabric construction, finish, and conditions of use. It also reports observed pilling resistance on a visual scale from 5 to 1, where 5 indicates no pilling and 1 indicates very severe pilling.

This distinction is important in purchasing and product development. A fabric can perform well in abrasion and still develop fuzz or pills that make the surface visually unacceptable. The opposite can also happen: a fabric may look clean after a pilling test but reach an abrasion endpoint too early for a heavy-use application. A sourcing requirement should therefore state which Martindale result is needed, not simply ask for a high Martindale rating.

Key standards and what they tell you

As of November 2026, the following standards are among the most relevant references when reading Martindale fabric test reports. The required document should always be confirmed against the buyer specification, target market, product category, and laboratory scope.

One important boundary is coated or laminated fabric. ISO 12947-2 states that it is not applicable to coated fabrics, including laminated fabrics, and directs users to ISO 5470 methods when the abrasion behavior of a coated surface must be determined. This distinction matters in upholstery, automotive interiors, outdoor materials, and performance finishes, where the surface layer may behave differently from a woven or knitted textile base.

How to interpret Martindale results for fabric selection

The first rule is to compare like with like. A 30,000-cycle result under one standard, pressure, abradant, and endpoint should not be treated as identical to 30,000 cycles under a different setup. A report that lists only a number, without the standard, specimen condition, and evaluation criterion, is not complete enough for serious sourcing decisions.

For contract upholstery, industry guidance from the Association for Contract Textiles is often used as a practical benchmark. Its 2025 woven indoor abrasion guidance lists 20,000 Martindale cycles under ASTM D4966 at 12 kPa for low-traffic or private spaces and 40,000 Martindale cycles for high-traffic or public spaces. The same guidance also warns that abrasion results above the high-traffic guideline are not an indicator of product lifespan and that Wyzenbeek and Martindale results do not correlate. These figures are useful reference points, but they should not be copied into every fabric category without considering end use.

Acceptance levels can differ widely for apparel, home textiles, office seating, hotel furniture, public transport, luggage, and technical fabrics. A decorative drapery fabric may not need the same abrasion resistance as transit seating. A fashion garment may fail commercially because of pilling or color change long before it reaches a structural abrasion endpoint. A workwear fabric may need abrasion data alongside tear strength, tensile strength, seam performance, color fastness, dimensional stability, and laundering durability. See also: Applications.

Common mistakes when reading rub counts

The most common mistake is assuming that a higher number always means a better fabric. A higher Martindale result can indicate stronger abrasion resistance under the stated test conditions, but it does not automatically prove better comfort, color retention, cleanability, seam strength, flame performance, or long-term appearance. A stiff or heavily finished fabric may score well in one mechanical test while performing poorly in hand feel or aftercare.

A second mistake is comparing Martindale cycles with Wyzenbeek double rubs as if they were interchangeable. The two methods use different equipment and abrasion actions. ACT guidance explicitly states that there is no correlation between Wyzenbeek and Martindale results. If a specification allows either method, it should state the acceptable threshold for each method separately rather than converting one number into the other.

A third mistake is ignoring pretreatment. ASTM D4970/D4970M allows fabric to be laundered or dry cleaned before testing when agreed among interested parties. For pilling in particular, specimen condition matters because washing, dry cleaning, finishing, brushing, shearing, or coating can change the surface. If a report does not state whether the sample was tested as received, after laundering, or after another treatment, the result may be difficult to apply to real use.

A fourth mistake is overlooking laboratory variation. ASTM D4966 cautions that between-laboratory precision can be poor for abrasion testing. For that reason, a narrow difference between two fabrics may not be meaningful unless the comparison is controlled, repeated, and interpreted statistically. Large performance gaps are usually more useful for decision-making than small numerical differences.

A practical checklist for buyers and developers

Before using a Martindale result in a specification, request enough information to understand exactly what was tested. A complete report should identify the standard and version, laboratory, fabric construction, sample conditioning, abradant, pressure or load, number of specimens, test endpoint, inspection interval, result for each specimen, average result where relevant, and any visible surface changes. For pilling, it should also show the visual rating scale and whether photographic or fabric standards were used.

• Confirm whether the requirement is abrasion, pilling, fuzzing, matting, mass loss, or appearance change.
• Specify the exact method, such as ISO 12947-2, ISO 12947-4, ISO 12945-2, ASTM D4966, or ASTM D4970/D4970M.
• Do not compare results unless the standard, endpoint, abradant, and load are the same.
• Check whether the fabric was tested before or after washing, dry cleaning, finishing, or other pretreatment.
• Use Martindale data with other performance tests that match the final application.
• Avoid marketing language that turns a laboratory cycle count into a promise of product lifespan.

The Martindale fabric test is most useful as part of a broader evidence package. It can screen materials, support supplier comparisons, and help define minimum performance levels. It should not replace field knowledge, product-specific trials, user conditions, cleaning instructions, or other physical and chemical tests.

Frequently asked questions

Is the Martindale fabric test only for upholstery?

No. Martindale methods are widely associated with upholstery because rub counts are common in furniture specifications, but the relevant standards cover broader textile applications. Depending on the method, Martindale testing can be applied to woven, knitted, and some nonwoven fabrics. Suitability still depends on fabric thickness, surface type, construction, and the selected standard.

Does a higher Martindale number always mean longer fabric life?

No. A higher number under the same method and endpoint can indicate better abrasion resistance in that laboratory test. It does not guarantee longer real-world life. Actual performance also depends on use intensity, body movement, cleaning, fiber content, yarn structure, finishing, seam design, exposure to light, moisture, soil, and maintenance.

What is a good Martindale rating?

There is no universal good number. For contract upholstery, ACT guidance uses 20,000 Martindale cycles for low-traffic or private spaces and 40,000 cycles for high-traffic or public spaces under its stated ASTM D4966 conditions. Other textile products may require different levels, and some applications need pilling, colorfastness, tear, tensile, or seam tests more urgently than a higher abrasion figure.

Can Martindale and Wyzenbeek results be converted?

They should not be converted with a simple formula. The two tests use different abrasion actions and equipment. Industry guidance from ACT states that Wyzenbeek and Martindale results do not correlate. If both methods are accepted in a specification, each should have its own threshold.

Why do two labs get different Martindale results?

Differences can come from specimen preparation, conditioning, abradant wear, load, machine calibration, inspection interval, visual judgment, and endpoint interpretation. This is why the standard version and detailed test conditions matter. Small differences in rub count should be interpreted cautiously, especially when making high-volume sourcing or acceptance decisions.