A result such as “30,000 Martindale rubs” looks precise, but it is not a complete test result. Before a QA manager accepts it, the report must identify the test method, endpoint, loading condition, abradant, inspection interval, specimen details and any deviations. Without that context, two laboratories can report different numbers for the same fabric and both numbers may appear plausible.
Method-control note
“Martindale” is not a single universal result. State the applicable ISO 12947 part, edition, specimen and endpoint. The product or buyer specification supplies the acceptance requirement. Use a licensed standard for apparatus settings, specimen preparation and detailed procedure.
The practical rule is simple: use the buyer or end-use specification to choose the method, then compare results only when the test conditions and endpoint are equivalent. A higher rub count is not automatically proof of longer product life.
What Does the Martindale Abrasion Test Measure?
The Martindale method subjects a textile specimen to controlled multidirectional rubbing. The apparatus produces a Lissajous movement so that wear is applied in changing directions rather than along one straight path.
The test can answer different questions depending on the selected method:
- When does the specimen reach a defined breakdown endpoint?
- How much mass does it lose after a stated abrasion exposure?
- How much does its appearance change after a stated exposure?
Those are different outputs. A laboratory should not reduce all three to a generic “Martindale score.”
Martindale data is best used to compare materials tested under the same conditions, check conformity with a named specification and investigate how construction or finishing changes affect abrasion behaviour. It should not be presented as a direct prediction of years of service. ASTM’s Martindale method similarly cautions that laboratory abrasion is only one contributor to real durability and that the relationship changes with end use.
ISO 12947 Parts 1–4: Which One Do You Need?
ISO 12947 is a series, not a single pass/fail test. The required part depends on the decision the laboratory must make.
| ISO 12947 part | Main purpose | Typical result | Use it when the question is… |
|---|---|---|---|
| Part 1 | Requirements for the Martindale apparatus | Equipment and test setup framework | Is the apparatus suitable for the specified Martindale method? |
| Part 2 | Determination of specimen breakdown | Abrasion exposure at the defined endpoint | When does the fabric reach the specified breakdown condition? |
| Part 3 | Determination of mass loss | Mass loss after stated abrasion exposure | How much material is lost under the specified test conditions? |
| Part 4 | Assessment of appearance change | Appearance assessment after stated exposure | How visibly has the surface changed? |
The current ISO catalogue lists ISO 12947-2:2016 for specimen breakdown, ISO 12947-3:1998 for mass loss and ISO 12947-4:1998 for appearance change. These catalogue entries were checked on 5 August 2026. Always confirm the current edition and any corrigenda before testing or publishing a specification.
ISO states that Part 2 is not applicable to coated or laminated fabrics when the coated surface itself is being evaluated; its catalogue directs users to the relevant parts of ISO 5470 for that situation. This is an important method-selection check for coated workwear, protective fabrics and technical textiles.
Standards note: This guide explains how to make the testing decision. It does not reproduce the licensed standard. The laboratory must use the applicable licensed edition, the buyer specification and its approved standard operating procedure for specimen preparation, conditioning, loading, consumables, inspection intervals, endpoints and reporting.
Martindale Abrasion and Martindale Pilling Are Not the Same Test
The same machine family may support both abrasion and pilling configurations, but the results answer different questions.
| Test | Relevant ISO series | What it evaluates | Result type |
|---|---|---|---|
| Martindale abrasion | ISO 12947 | Resistance to wear under the specified abrasion method | Breakdown exposure, mass loss or appearance change |
| Modified Martindale pilling | ISO 12945-2 | Propensity to pill, fuzz or matt | Visual surface assessment under the specified method |
A pilling grade must not be reported as an abrasion result, and an abrasion rub count does not prove that the fabric will resist pilling. If surface pilling is the problem, use the separate ISO 12945 pilling test guide.
How the Martindale Test Works: The Decision Sequence
1. Define the performance question
Start with the product, material and failure being controlled. Apparel, upholstery, a nonwoven and a coated technical textile may require different methods or acceptance criteria. Record the customer specification and its revision before cutting specimens.
2. Select the correct standard and result
Do not choose Part 2 merely because a buyer asked for a “rub count.” Confirm whether the contract requires breakdown, mass loss, appearance change or another named test method. If the specification is unclear, obtain written clarification before testing.
3. Take representative specimens and condition them
Sampling must represent the submitted lot and the relevant fabric directions, pattern areas, face and finishing condition. Physical textile results can be affected by conditioning; ISO 139 defines standard atmospheres for conditioning and testing textile materials. The report should identify the conditioning method rather than simply stating “laboratory conditions.”
4. Set up the apparatus and consumables
Use the abradant, backing materials, specimen holder, applied loading condition and other consumables required by the selected method. Record their identity and condition. Do not substitute a convenient material merely because it fits the instrument.
The apparatus must also be controlled under the laboratory quality system. Manufacturer calibration literature identifies items such as the Lissajous motion, spindle-bearing behaviour, specimen-holder geometry and ambient conditions as relevant checks. The appropriate verification or calibration interval should be based on the method, manufacturer instructions, use, risk and the laboratory system—not an unsupported universal “once per year” rule.
5. Inspect and report the defined endpoint
Inspection must follow the selected method and approved procedure. Record individual specimen results, the summary calculation required by the method, observations and any deviation. If the endpoint lies between inspection stages, the reporting resolution must be understood before comparing a borderline result with a specification limit.
How to Interpret a Martindale Rub Count
Use this order before calling a result “good,” “poor,” “pass” or “fail.”
Check 1: Is the result tied to a named method?
“Martindale: 30,000” is incomplete. At minimum, identify the standard, part, edition and result endpoint.
Check 2: Are the test conditions comparable?
Compare the specimen type, fabric face and direction, conditioning, abradant, applied loading condition, inspection interval and endpoint. A difference in any of these can invalidate a direct comparison.
Check 3: Where did the acceptance limit come from?
ISO 12947 describes test methods; it does not create one universal commercial pass level for every garment, upholstery fabric or technical textile. Acceptance limits should come from a named buyer specification, product standard, contract or validated internal requirement.
Generic online tables that label a particular number as “light duty,” “heavy duty” or “contract grade” may be useful for retail orientation, but they are not sufficient evidence for shipment release unless the applicable specification adopts those categories.
Check 4: Is the difference large enough to matter?
Abrasion testing is sensitive to test conditions and can show operator and laboratory variation. ASTM D4966 specifically discusses poor between-laboratory precision and recommends comparative testing when purchaser and supplier laboratories disagree. A small numerical difference should not be converted into a sweeping durability claim without considering the method’s precision, laboratory bias and the decision limit.
Check 5: Does the product need other durability tests?
Abrasion resistance does not replace tensile, tear, seam, pilling, snagging, colour-fastness or dimensional-stability testing. Select the test package from the actual failure risks of the product. For instrument selection, see the textile tensile tester guide and tear strength tester guide.
Why Martindale Results Vary Between Specimens or Laboratories
| Possible source of variation | What to check | Why it changes the decision |
|---|---|---|
| Non-representative sampling | Lot, roll position, fabric direction, pattern area, face and finish | The specimens may not represent the same material population. |
| Conditioning difference | Conditioning method, exposure and recorded laboratory atmosphere | Moisture and atmosphere can influence physical textile behaviour. |
| Wrong method or edition | Standard identifier, part, edition and customer method | The laboratories may be measuring different endpoints. |
| Specimen mounting or movement | Alignment, tension, clamping, slippage and holder condition | Movement at the holder can change the wear mechanism. |
| Consumable condition | Correct abradant and backing, lot identity, contamination and replacement control | Consumable changes can alter abrasion severity. |
| Apparatus condition | Lissajous movement, spindle behaviour, loading components and service records | The applied motion or force may not match the approved setup. |
| Inspection and endpoint judgement | Inspection interval, lighting, assessor training and endpoint record | Borderline endpoints can be classified differently. |
| Data handling | Individual results, excluded values, rounding and summary calculation | A reporting or transcription problem can look like a material failure. |
What to Do When the Mill and Buyer Laboratory Disagree
Do not immediately change the process or reject the lot. First determine whether the laboratories produced comparable data.
- Hold the decision and retain evidence. Preserve the submitted sample, tested specimens, photographs, raw readings and test report.
- Compare method identities. Confirm the exact standard, part, edition, buyer method and endpoint used by each laboratory.
- Compare specimen histories. Establish whether both laboratories tested material from the same lot, roll, position, face, direction and finishing condition.
- Compare test conditions. Review conditioning, consumables, loading condition, inspection interval, apparatus checks and deviations.
- Check laboratory competence. Where accreditation is required, confirm that the laboratory’s current scope covers the applicable method. ISO/IEC 17025 addresses competence, impartiality and consistent operation of testing and calibration laboratories.
- Plan a controlled comparison. Divide the most homogeneous retained material available, randomise specimens between laboratories and agree the method and decision rule before testing. Use competent statistical support where the commercial consequence is material.
- Correct the cause, not the number. If bias or a procedural difference is found, document it and agree how future results will be interpreted.
This sequence separates a genuine material problem from a method, sampling or laboratory problem before production changes create new variation.
Martindale Test Report Checklist
A professional report should make the result reproducible and interpretable. Check for:
- Customer or buyer specification, including revision
- Test standard, part and edition
- Product and specimen identification
- Fibre composition, construction, mass per unit area and finish where relevant
- Lot, roll, sampling location, face and direction
- Conditioning standard and recorded test atmosphere
- Apparatus identification and relevant verification or calibration status
- Abradant, backing and consumable identification
- Applied loading condition required by the method
- Number of specimens and inspection interval
- Endpoint or assessment method
- Individual results and required summary
- Photographs or observations where useful
- Deviations from the method
- Acceptance limit and the document that defines it
- Test date, laboratory and authorised report signatory
If the report contains only a single rub-count number, request the missing information before making a shipment or claim decision.
Choosing a Laboratory or Martindale Tester
When outsourcing, verify that the laboratory can perform the exact method required—not merely that it owns a Martindale instrument. Ask for the applicable accreditation scope where required and request a sample report before routine submissions.
When buying an in-house tester, evaluate:
- Supported abrasion and pilling standards as separate configurations
- Number of test stations and whether each station can be managed safely
- Availability and traceability of compliant consumables
- Method for verifying the abrasion and pilling motion
- Loading components, holders and accessories for the intended materials
- Service, calibration and local technical support
- Operator training and documented setup guidance
- Data recording, export, user permissions and audit trail
- Laboratory space and environmental requirements
Define these requirements before requesting quotations. A touchscreen or high station count cannot compensate for weak method control, unavailable consumables or poor service support.
Frequently Asked Questions
Is a higher Martindale rub count always better?
Only when the materials were tested using equivalent methods, conditions and endpoints. A higher value does not automatically predict longer service life or prove suitability for a different end use.
What is a good Martindale result for upholstery or apparel?
There is no universal value in ISO 12947 for every product. Use the applicable buyer specification, contract, product standard or validated internal requirement. Do not release a shipment against an unsourced internet benchmark.
Is the Martindale abrasion test the same as a pilling test?
No. ISO 12947 addresses abrasion resistance. ISO 12945-2 uses a modified Martindale method for pilling, fuzzing and matting. The machine may be adaptable to both, but the setup and result are different.
Can Martindale and Wyzenbeek numbers be converted?
They are produced by different apparatus and test actions. ASTM D4157 covers the oscillatory-cylinder method commonly associated with Wyzenbeek testing, while ISO 12947 and ASTM D4966 cover Martindale approaches. Treat the values as method-specific; do not use a generic conversion factor for acceptance decisions.
Final Takeaway
The most useful Martindale result is not the largest number. It is the result that answers a defined product question, follows the correct method and carries enough test information to be compared and defended.
Before accepting a fabric, confirm three things: the correct ISO 12947 part was used, the report identifies the test conditions and the pass limit comes from an authorised specification. That discipline prevents laboratory disagreement from becoming an unnecessary production or shipment loss.
Sources and Review
- ISO 12947-2:2016 — determination of specimen breakdown, accessed 16 August 2026.
- ISO/TC 38/SC 24 catalogue — current abrasion and pilling standards, accessed 16 August 2026.
Confirm the exact part required by the product specification. Have all descriptions of endpoints, exclusions and laboratory procedure reviewed by an accredited textile laboratory.
