Quality Control and Test Standards

Which quality and test topics should you ask about when selecting a supplier?

In B2B textile sourcing, most disputes arise not from aesthetic preference but from unmeasured expectations. Complaints such as "the colour bleeds", "it shrank in the wash" or "it pilled on first wear" can all be agreed in advance through a test method and a tolerance value. For this reason a mature supplier with a coordinated contract network clarifies, before closing the sale, which standards, which specimen and which acceptance threshold will be used.

In the table below we have gathered the five topics buyers ask about most often, the question each answers and the standard family each rests on. These five topics also form the skeleton of the five sub-guides beneath this pillar page.

What do fastness tests measure and which standards do they rest on?

Colour fastness shows how well the dye is fixed on the fibre and where it may migrate in use. Wash fastness assesses colour bleeding and staining of an adjacent light ground; perspiration fastness assesses fading under acidic/alkaline perspiration conditions; rubbing (crocking) fastness assesses colour transfer in dry and wet rubbing; light fastness assesses fading under UV exposure. While reactive dyeing typically achieves high wet fastness on cellulosic fibres, in pigment and some garment-dye effects the rubbing fastness is managed carefully according to end use.

Using the standard numbers correctly makes the assessment objective. The table below summarises the most frequently cited ISO 105 sub-methods and their AATCC equivalents.

The interpretation of grades depends on end use: in babywear and underwear the wet fastness expectation is generally held higher, while in outerwear light fastness comes to the fore. Fixing the target grades in the contract removes any later debate over "is it adequate or not"; with KARCEM we can clarify these thresholds on a product-by-product basis.

How are shrinkage and dimensional stability measured?

Because knitted structure is more elastic than woven, it is more prone to dimensional change; this makes the shrinkage test one of the most critical physical tests in knits. The correct method is to choose a wash programme that mimics real care conditions, mark the specimen, and measure after multiple cycles. A negative value denotes shrinkage (contraction), a positive value extension. In fabrics containing elastane, the heat-set temperature and stenter setting directly affect recovery and final dimensions.

Alongside the dimensional results, spirality should also be reported; particularly in single-jersey structures such as single jersey, seam twisting is related to this value. The table below shows the basic steps to follow for knits and the corresponding standards.

The acceptable range of change is set according to fabric type, weight and the final product. Rather than committing to a specific percentage, we address the target values together with your fabric structure and care instructions in the dimensional stability guide, and clarify them together in production.

How are pilling, Martindale and abrasion resistance assessed?

Pilling is decisive for both aesthetics and perceived quality, and is seen more frequently especially in carded and open-end yarns, owing to the tendency of short fibres to rise to the surface. The use of combed and compact yarn reduces fuzzing and improves the pilling grade. Abrasion resistance comes to the fore in upholstery and heavy-use clothing; the Martindale abrasion test gives the number of cycles the surface withstands against a standard abradant.

Clarifying the test method and reporting format ties a "it pilled" complaint to a measurable criterion in advance. The table below summarises the relevant standards and the output format.

The pass grade and target cycle count vary with end use; underwear and outerwear are not assessed against the same threshold. We go deeper into product-specific targets in the pilling and abrasion guide.

How does the four-point inspection system score fabric quality?

Visual inspection is a step that complements laboratory tests but does not replace them: while fastness and shrinkage are measured on a specimen, four-point inspection scans the whole roll, counting surface defects such as holes, stains, missed stitches and lines. A single defect is given at most 4 points, and no more than 4 points are counted in a single metre; this prevents one area disproportionately inflating the score. When the result is assessed together with weight tolerance and width measurements, the commercial usability of the roll becomes clear.

Although the scoring calculation is common across the industry, the acceptable threshold varies from product to product; printing, dark grounds or critical end uses require a tighter limit. We explain the scoring logic and a worked example step by step in the four-point inspection and tolerance guide.

How does the lab-dip approval process make the ΔE<1 target measurable?

Colour management is the heart of the "registration" logic: standard, lab-dip and production batches are aligned around the same reference. The process typically proceeds as follows: the customer defines the standard, the laboratory prepares several lab-dips, the approved recipe is transferred to production, and production batches are measured against the standard with a spectrophotometer and the ΔE reported. To reduce the risk of metamerism, measurements are taken under standard light sources; consistency under different lighting is also checked separately.

The table below summarises the stages of the process and the measurable output produced at each stage.

The ΔE<1 target is meaningful when the light source, observer angle and formula (CMC l:c or CIEDE2000) are defined; promising a single figure on its own, without these parameters fixed, would be misleading. We go deeper into the process end to end in the lab-dip approval guide, and into the dyeing side on the colour fastness and ΔE page.

How are these tests combined into a quality agreement?

Individual tests are valuable, but the real strength lies in gathering them all into a consistent specification document. In a single-coordinator structure, because the process is coordinated under one point of contact from yarn to finishing — in-house knitting plus a vetted contract network — a deviation that surfaces in one test can be quickly traced back to the right stage (knitting, pre-finishing, dyeing, finishing treatment); this traceability speeds up corrective action.

The document can also cover sustainability and conformity expectations: when OEKO-TEX 100 limit values, ZDHC/MRSL chemical management and the certification chain can be tracked in the same file as the test results, audits become easier. Addressing this framework together with certificates and the dyeing/printing guide makes quality and conformity a single whole.

To keep this whole guide in a single file, download the PDF version of this guide.

To see the standard numbers for every test (ISO, AATCC, ASTM) in one filterable table, see the Textile Test Standards Reference.

Frequently asked questions

Which quality and testing topics should I ask about when evaluating a knit fabric supplier?

A solid evaluation rests on five axes: colour and crocking fastness, dimensional change after washing, surface durability (pilling/abrasion), defect-counting four-point inspection, and colour approval via lab-dip. When each is tied to a defined ISO or AATCC method and a numerical acceptance criterion, 'good fabric' turns from a subjective claim into a measurable contractual requirement.

What effects do fastness tests measure and which standard numbers do they rely on?

Fastness is the resistance of a colour to staying in place against washing, perspiration, rubbing (crocking) and light. Each effect has its own method: washing ISO 105-C06 / AATCC 61, perspiration ISO 105-E04 / AATCC 15, rubbing ISO 105-X12 / AATCC 8 / 116, light ISO 105-B02 / AATCC 16. Results are reported on the 1-5 grey scale for washing, perspiration and rubbing, and on the 1-8 blue scale for light.

How are shrinkage and dimensional stability measured in knit fabric?

The sample is washed per ISO 6330, and the change between marked dimensions is calculated as a percentage per ISO 5077; on the AATCC side, AATCC 135 is followed. A negative value indicates shrinkage, a positive value extension. Because knits are more elastic than wovens, this is one of the most critical physical tests for knits. Spirality (twisting) is also reported; sanforising and compacting bring these values under control.

How are pilling and abrasion resistance evaluated and which standards are used?

Pilling is measured by the Martindale method (ISO 12945-2) or the box method (ISO 12945-1) and reported on a 1-5 photographic scale, where 5 is best. Abrasion resistance is given per ISO 12947 (Martindale) as the number of cycles at which the surface is breached; on the AATCC side, ASTM D4970 is used. Pilling is more frequent with carded and open-end yarns, while combed and compact yarn improve the pilling grade.

How does the four-point inspection system score fabric quality?

Each defect is assigned 1-4 points according to its size: 7.5 cm and below 1 point, 7.5-15 cm 2 points, 15-23 cm 3 points, over 23 cm or a hole 4 points. The total points are normalised per 100 m² to produce a score; a lower score means cleaner fabric. A single defect is given a maximum of 4 points, and no more than 4 points are counted within one metre. The acceptance limit is agreed in advance between buyer and supplier.

How does the lab-dip approval process turn the ΔE<1 target into a measurable commitment?

A lab-dip is a small dye trial before production; the approved lab-dip becomes the standard against which the production colour is compared. The customer defines the standard, the laboratory prepares the lab-dip, the approved recipe is transferred to production, and lots are measured against the standard with a spectrophotometer to report ΔE. The CMC or CIEDE2000 formula is used; ΔE<1 denotes a difference the eye can barely distinguish. To be meaningful, the light source, observer angle and formula must be fixed.