Dye Fastness Testing: How to Read the Report Before You Release a Bulk Dye Order

Somewhere in your quality file is a fastness report with a row of numbers on it — 4, 4–5, 4, 3. It was accepted at goods-in, filed, and never looked at again. Then a customer returns a shipment because the black crocks onto a white lining, and the report turns out to have been run on a different substrate, to a method nobody wrote down.

Fastness numbers are not measurements the way a pH reading is. They are the output of a specific method, on a specific sample, judged against a specific scale — change any of the three and the number moves. This guide covers what the tests do, how to read a report line by line, and how to build a sampling and retention protocol that survives an audit.

A fastness grade describes a method, a sample and a judgement — not an intrinsic property of the dye.

What Does a Fastness Test Actually Measure?

Fastness is resistance to change. The dye is fixed in the substrate; the test asks what happens when that substrate meets whatever moves it — rubbing, water, light, sweat, washing, heat. Every method follows the same structure: condition the sample, expose it under controlled parameters, let it recover, then compare against a reference scale.

That is why two labs can report different numbers for the same drum and both be right: crock pressure, conditioning humidity, bath temperature and adjacent fabric all move the result.

The two scales you will see

Colour change and staining are graded on the grey scale, 1 to 5 in half steps, where 5 is no perceptible difference and 1 is severe. Light fastness uses the blue wool scale, 1 to 8. A "6" on a light test and a "4" on a rub test are not on the same ruler — and tables that mix the two without labelling are one of the commonest errors in supplier literature.

Visual versus instrumental assessment

Grading by eye is permitted by the standards and remains the reference method, but it depends on the assessor, the light cabinet and the viewing geometry. A spectrophotometer repeats better across lots; the eye settles customer disputes better.

Technical Specifications: The Tests That Matter

You do not need every test in the catalogue — only the four or five that match how the article will be used, abused and cleaned.

Property	Common Methods	Scale	What It Predicts in the Field
Rub / crocking	ISO 105-X12; AATCC 8	Grey scale, dry and wet separately	Colour transfer to lining, upholstery, hands, packaging
Water	ISO 105-E01; AATCC 107	Grey scale, change + staining	Rain spots, damp storage, sweat-adjacent staining
Water spotting	ISO 105-E07	Grey scale, change + staining	The ring mark left by a single drop — the classic aniline complaint
Perspiration	ISO 105-E04; AATCC 15	Grey scale, acidic and alkaline	Collars, cuffs, glove linings, shoe interiors
Light	ISO 105-B02; AATCC 16	Blue wool 1–8	Fading in shop windows, car interiors, display
Wash / laundering	ISO 105-C06; AATCC 61	Grey scale, change + staining	Garment leathers, washable textiles, re-tan goods
Migration / bleed	Supplier or brand-specific	Visual or grey scale	PVC contact staining, interleaving transfer in storage

Sampling position matters: a swatch from the top of the drum is not a swatch from the batch.

Comparison of Options: Where the Evidence Comes From

Three sources of fastness data reach a buyer, and they carry very different weight. Knowing which one you hold decides what you can do with it.

Source	Strength	Weakness	Best Used For
Supplier's own laboratory report	Fast, cheap, tied to a drum number	Self-assessment; method may be undocumented	Screening and trend tracking
Independent third-party laboratory	Impartial, method-controlled, defensible	Cost per test and lead time	Pre-shipment release and audit
Your own in-house bench	Your substrate, your water, fast turnaround	Needs equipment and trained assessors	Receiving checks and complaint triage
Brand or customer test house	Accepted by definition	You learn late, often on finished goods	Final verification on high-risk programs

The practical arrangement is a triangle: supplier data for screening, in-house for routine release, third party for anything disputable. A single-source file is a single point of failure.

How to Read a Report in Five Steps

1. Identify the sample before you read the result. Substrate and tannage, drum or lot number, production date, crust or finished article. A report that cannot be tied to a drum cannot be tied to a shipment.
2. Check the method, revision year and conditions. ISO or AATCC number, dry or wet, temperature, contact time, adjacent fabric. Then confirm the method matches the end use you sell into.
3. Read both directions of the scale. Colour change tells you the article faded or shifted; staining tells you what it did to whatever touched it. A report showing only one is incomplete, and staining generates most claims.
4. Look for the light source. Visual grading should name the light cabinet and observer conditions. "Assessed visually" with no light source is indicative only.
5. Compare against the spec, not the last report. A 4 against a 4 minimum is a pass even if the previous lot was 4–5. But a downward trend across three lots is a signal to re-test before the fourth ships.

Conditioning space, light cabinet and a trained assessor are what separate a number from evidence.

Quality-Control Evidence: Building a File That Holds Up

No supplier can promise a grade on your substrate; any that does is describing a reference crust. What a supplier can offer is traceable, method-labelled data on the batch you are buying, plus a sealed sample of the material that produced it.

Four documents make a defensible file: a batch-specific certificate of analysis giving colour strength, moisture, solution pH and solubility, tied to the drum numbers on the packing list; a fastness report naming method, revision and conditions; a sealed retention sample of the approved lot; and a change-control note whenever the formulation or site changes, because a grade belongs to a formulation, not a trade name.

On a brand audit, the fastness file sits alongside the restricted-substance documentation, not instead of it. A dye can be conformant on chemistry and still fail on crocking.

Packaging, Sampling & Shipping the Evidence

Swatches travel badly if nobody plans for it. Cut samples large enough to run the test twice, label them with lot, drum and date in waterproof ink, and bag them individually — a swatch that has rubbed against a darker one in transit is ruined, not a failed dye. For third-party testing, ship the sample with the written method instruction, or the lab applies its default. Retention samples follow the same discipline: sealed, dry, and held for the contract period.

Five Mistakes Buyers Make with Fastness Data

Frequently Asked Questions

Conclusion

Fastness testing is not a box to tick at goods-in. It converts a supplier's claim into something you can defend to a customer, a brand auditor or a claims handler — but only if the report names its method, identifies its sample and points to a drum you actually received.

Build the file around three habits: method plus grade rather than a bare number, test the article as sold, and keep a sealed sample of every approved lot. Those three settle most arguments before they start.

---