Fabric Testing Standards: Color Fastness, Martindale, Pilling, Shrinkage and Strength Tests

Who asks for fabric tests, and what is covered here?

A fabric test comes up in three places: in the buyer's spec, in an independent laboratory's report and in the producer's own internal control. This guide covers physical and color tests. Chemical safety tests and OEKO-TEX-type certificates are in the textile certificates guide, and sampling of a garment lot is in the AQL quality control guide.

The method has to be a standard one, because the result depends on the conditions. ISO 12947-2 says that only results obtained in the same test atmosphere can be compared, and ISO 13934-1 asks that the result of one method not be compared with the result of another.

Fabric testing standards at a glance: which test measures what

The standard years below are the editions listed in EVS records in October 2026. "EVS" marks the record page of Estonia's standards body; "ISO text" marks the published preview text of the standard.

Last checked: October 2026.

How is color fastness tested?

Color fastness is the degree to which a dyed or printed fabric keeps its color when it meets an effect such as washing, rubbing, perspiration or light. Every test grades two things: the color change in the specimen (ISO 105-A02 gray scale) and the dye that stains the adjacent fabric or the rubbing cloth (ISO 105-A03 staining scale). On the change scale, 5 is given only when no difference from the original fabric can be seen and 1 is the largest difference; half grades such as 4-5 are used (ISO 105-A02:1993).

Washing fastness test (ISO 105-C06)

This test measures how well color survives domestic or commercial laundering, using a reference detergent. The standard defines S tests, which give a result close to a single wash, and M tests, which have a higher mechanical action; in some cases the result of an M test approaches that of up to five washes at no more than 70 °C. The specimen is joined to an adjacent fabric (usually a multifiber strip) and washed. Temperature and conditions are given by a code, and the code must appear in the report.

Rubbing fastness test (ISO 105-X12)

Two separate tests are run, dry and wet. A rubbing finger travels 20 times in total (10 forward, 10 back) over a 104 mm track on the specimen with a force of 9 N, and the staining on the cotton cloth is graded on the A03 scale. The standard describes the water pick-up of the wet cloth as 95–100%, says other pick-ups may be used, and notes that the pick-up can change the result a great deal. The pick-up used must be written in the report.

Perspiration fastness test (ISO 105-E04)

The specimen and an adjacent fabric are wetted with two solutions that contain histidine: an alkaline one (pH 8) and an acid one (pH 5.5). Each is held between two plates under 12.5 kPa at 37 °C for 4 hours. Color change and staining are graded with the gray scales, and each solution gets its own grade.

Light fastness test (ISO 105-B02)

The specimen is exposed in a xenon-lamp device that imitates daylight, together with blue wool references. The standard allows two different reference sets, and the results from the two sets may not be the same. Q-Lab's summary describes sets numbered 1 to 8 and says the standard contains five methods: some give a grade against the references, and one compares the specimen with an agreed reference specimen. In a report, look for the method number and the reference set.

How is the Martindale abrasion test done?

The Martindale test (ISO 12947-2) measures how well a fabric resists wear by rubbing. A round specimen is rubbed against a standard abradant fabric under load in a Lissajous-figure motion and is checked at intervals. It counts as broken down when a yarn breaks or the worn-area criterion is met. The result is the last number of rubs (cycles) at which no breakdown had been seen.

The load depends on end use: 12 kPa (795 ± 7 g) for workwear, upholstery, bed linen and technical fabrics, and 9 kPa (595 ± 7 g) for apparel and home textiles other than upholstery and bed linen. A spec for upholstery fabric or workwear (see workwear standards) must state the load as well. The standard gives no acceptance limit for the number of rubs, and there is no single universal threshold. Coated and laminated fabrics are out of scope (ISO 5470).

How is the pilling test done?

Pilling is the entangling of surface fibers into small balls. ISO 12945-4 separates it from fuzzing and matting, and the three are graded separately. In the modified Martindale method (ISO 12945-2), a round specimen is rubbed against the fabric itself or, where needed, against a wool abradant, with a 155 g carrier (415 g in total with the optional extra weight), and is assessed by eye at defined stages.

Grades follow ISO 12945-4: 5 is no change, 4 is partial pill formation, 3 moderate, 2 distinct and 1 severe pilling; intermediate values such as 3-4 are written. The assessment is done in a dark room under D65 light of at least 600 lux, from a distance of 30–50 cm. The standard gives no acceptance grade, so the contract must say which grade at which stage is acceptable.

According to the introduction to ISO 12945-2, three processes decide the outcome: fibers entangling into pills, more fibers working their way to the surface, and fibers and pills wearing off through abrasion. With durable fibers, pill formation is faster than wear-off, so pills build up. Hossain and colleagues (2021) treat pilling as a surface defect in a polyester-cotton three-thread fleece. In their pilling box test (ISO 12945-1:2000), applying singeing and heat setting together raised the grade from 1 to 4.

How is the fabric shrinkage test done?

A dimensional change (shrinkage) test measures how much length and width change after washing and drying. The specimen is marked and measured (ISO 3759), processed with an ISO 6330 washing and drying procedure that the parties choose, and measured again under ISO 5077. Where possible, ISO 5077 calls for three specimens.

Dimensional change (%) = (later measurement − original measurement) ÷ original measurement × 100. It is calculated separately for length and width and rounded to the nearest 0.5%; shrinkage is written with a minus sign and extension with a plus sign. A different machine, detergent or dryer can change the result (ISO 6330). For the cutting allowance, use the shrinkage allowance calculator.

How is fabric strength tested? Tensile, tear and bursting

Three tests load the fabric in different ways: pulling, tearing and inflating. Their results cannot stand in for one another. For help matching a woven or knitted structure to a method, see the fabric types and properties guide.

Tensile (breaking) strength test (ISO 13934-1)

In the strip method, a specimen 50 mm wide is pulled at a constant rate until it breaks, over a 200 mm gauge length (100 mm if more than 75% elongation is expected). The maximum force and the elongation at that force are recorded. Separate sets are cut for warp and weft, with at least five specimens in each. The method is mainly for woven fabric; it is not normally applied to nonwovens and coated fabrics.

Tear strength test (ISO 13937-2)

In the trouser method, a 100 mm slit is cut in the middle of a 200 × 50 mm strip, the two legs are clamped in the jaws and the tear is propagated at 100 mm/min. The result is the force needed to keep the tear going, in N, and it is named for the yarn set that is torn (warp or weft). The method is mainly for woven fabric; it is generally not suitable for knitted fabric and stretch woven fabric.

Bursting strength test (ISO 13938-2)

The specimen is clamped over a flexible diaphragm with a ring, and air pressure from below is raised until the specimen bursts; the time to burst is 20 ± 5 seconds. Bursting strength is the bursting pressure minus the diaphragm pressure. The method suits knitted, woven, nonwoven and laminated fabrics; for high-pressure technical textiles, the hydraulic method (ISO 13938-1) is more suitable.

How do sampling and conditioning affect test results?

ISO 139 defines the standard atmosphere as 20 °C and 65% relative humidity (±2 °C, ±4%). If the parties agree, an alternative atmosphere of 23 °C and 50% can be used, and ISO 12947-2 and ISO 105-X12 then require the atmosphere to be stated in the report. Specimens rest in this atmosphere first: at least 18 hours under ISO 12947-2 and at least 4 hours under ISO 105-X12.

The specimen must represent the fabric. ISO 12947-2 asks for a piece cut from the full fabric width where possible, containing at least two complete pattern repeats; ISO 13934-1 and ISO 13937-2 ask for cutting at least 150 mm in from the edge of the fabric. How GSM changes with the state of measurement is covered in the T-shirt, hoodie and jogger GSM guide; for your own fabric, use the fabric weight (GSM) calculator.

How do you read a fabric test report?

When a mill or laboratory sends you a report, check these fields. The table gives the basis for each.

Last checked: October 2026.

How do you write fabric tests into a spec?

Every spec line says four things: the test, the standard and year, the conditions and the acceptance limit. A test standard defines the method, not the limit; the limit is set by the buyer or, where one exists, by the relevant product standard. Put these lines into the spec you send to the mill, and send the same lines to every supplier you compare. The table below has a blank template and two hypothetical examples; the letters A and N stand for values the buyer sets and are not recommendations.

TÜBİTAK BUTAL's 2026 laboratory list cites these methods by "TS EN ISO" number, without the year. Before you write a number into a spec, confirm the year and number in the catalog of TSE, the Turkish Standards Institution.

Which test for which problem?

The table is our own assessment, not a standard, and it contains no limit values.

How do you choose a testing lab and send samples?

The first question is whether the test falls within the laboratory's accreditation scope. ISO/IEC 17025:2017 sets out the requirements for a laboratory to work competently, impartially and consistently. In Turkey, TÜRKAK, the Turkish Accreditation Agency, accredits testing laboratories and publishes current information on accreditation, suspension and changes of scope on its website (Regulation on the Accreditation of Conformity Assessment Bodies, art. 5; Official Gazette, November 3, 2017). You can also ask the laboratory for the scope annex of its accreditation certificate.

Scope is defined by method and range. In TÜBİTAK BUTAL's 2026 list, tests within the accreditation scope are marked (A): the strip tensile test on woven fabric is marked up to 7,500 N, and the 7,500–50,000 N range is not. In the same list, the customer sets the conditions of the washing fastness test.

To find a laboratory, see our directory of textile testing laboratories in Turkey and garment quality control and inspection services. When you send samples:

1. Write each test with its standard number and year.
2. Write the conditions for each test: washing procedure, load, method number.
3. Take the specimen so that it represents the fabric, and mark the face and the warp direction.
4. Ask whether the test is within the accreditation scope and in which language the report will be issued.
5. Read the report against the fields in the table above.

Common mistakes in fabric test requests

• Writing the standard number without its year.
• Writing only "fastness 4"; it is not clear whether it means washing or rubbing, dry or wet.
• Leaving the load (12 or 9 kPa) out of a Martindale line.
• Combining the length change and the width change in one number.
• Not saying which grade at which stage is acceptable in a pilling test.
• Asking for the strip tensile method on knitted fabric; the standard mainly covers woven fabric.
• Reading only the word "pass" in the report.

Summary checklist

• Is the standard number and year written for every test?
• Are the conditions (washing procedure, load, method number, wet-cloth pick-up) written?
• Is the acceptance limit set for each test, by the buyer?
• Is the test within the laboratory's accreditation scope?