Shade Variation and Colour Approval: Light Sources, Tolerance Writing and Lot-to-Lot Control in Woven Polyester and Nylon

A shade that passed in the light booth fails in the shop, and nobody dyed it wrong. Woven polyester and nylon are coloured in high-temperature batches, and every batch is a small chemical experiment with its own endpoint. Colour difference is not an event but a distribution: what a buyer actually controls is the width of that distribution, the light it is judged under, and the paperwork that defines what counts as the same shade.

Where Shade Variation Actually Starts

Disperse dyeing of polyester is batch chemistry. Uptake depends on time, temperature, liquor ratio, circulation and the reduction clearing that follows, so two machines running one recipe land at slightly different endpoints. Heat setting then moves the tone again, which is why the last process step is part of the shade, not an afterthought.

The yarn matters as much as the bath. Lustre, heat history and lot number of the filament all change how the same dye depth reads on the face. The full chain from pretreatment through exhaust dyeing to heat setting is walked through in our article on how polyester woven fabric is dyed and set, and each stage is a place where the endpoint can drift.

The Light Source Decides the Answer

D65 approximates average daylight, TL84 and TL83 represent the triphosphor tubes common in European retail, CWF is the cool white tube still found in North American offices, and Illuminant A is warm incandescent. The same swatch reads bluer under D65 and yellower under A, without changing at all.

Naming the source in the approval is not pedantry. A match set under daylight says nothing about shop lighting, and a tolerance agreed under one tube does not transfer to another. Goods that sell under specific retail lighting should be approved under that lighting, alongside a second source agreed as a check.

Standard Viewing Conditions

ISO 105 and the AATCC evaluation procedures exist so that two people in two countries see the same thing: a booth fitted with the named sources, the specimen held at the standard geometry, a neutral grey surround, and no coloured walls or daylight leaking into the field. Judgement windows are short, because the eye adapts.

The grey scale is the shared language for the size of a difference. Grade 5 means no visible difference between swatches, and each step down marks a visibly larger gap, with 4-5 describing a just-noticeable one. It is used both to grade fastness after testing, as covered in our guide to reading a fabric test report, and to set shade tolerances before bulk.

Metamerism: The Match That Breaks

Two swatches that match under D65 and split under TL84 are metameric. The pairs differ in spectral shape, not in the light they reflect overall, so they only agree where the source lets them. Disperse dye combinations are full of such pairs, and the cheapest recipe is often the most metameric one.

The control is procedural. Approve under at least two sources, one of them the retail source, and treat a recipe that only matches under one as a rejected recipe. Where goods combine dyed panels with printed or coated panels, the risk multiplies, because ink, dye and pigment systems rarely share a spectral shape.

Numbers Behind the Eye: Delta E and Its Limits

A spectrophotometer reduces a shade difference to a number. CMC and CIEDE2000 weight the differences roughly as the eye does, so a small hue shift in a saturated shade is not treated the same as the same figure in a neutral one. Many commercial programs hold critical shades to a CMC difference of about one.

Instruments have blind spots. They read one flat geometry, so textured surfaces, peached faces and high-lustre satins measure differently from how they present to the eye, and directional fabrics must be cut consistently before reading. The number is a screen; the named light source and the retained master remain the decision.

Writing the Shade Tolerance Clause

A tolerance that survives a dispute is written down before bulk runs. The lines that matter:

1. A signed and dated master swatch, cut from the approved sample, retained by both sides with the order file.
2. The primary light source and one secondary source, named in the clause, judged in a compliant booth.
3. A grey scale minimum against the master, stated as a grade rather than left to negotiation after the fact.
4. An instrument clause where used: formula, geometry and whether the figure is a limit or an average.
5. Within-roll and lot-to-lot limits written separately, because they fail for different reasons.
6. Bulk head approval before the continuous run, so the bulk endpoint is signed off, not just the lab dip.
7. Roll numbering that supports linking, so adjacent garment panels are cut from one dye lot.

Lot-to-Lot and Within-Roll Difference

Two distributions wear the same word. Within one roll, edge-centre and head-tail difference come from temperature gradients and circulation across the machine; between rolls, lot difference comes from separate batches reaching separate endpoints. The first is fixed in finishing, the second in scheduling and linking.

The practical defence is to treat the two limits separately in the specification and to link rolls of one lot into one marker. Where shade sits on the edge of tolerance, inspection is the backstop, and our article on grading fabric defects and the four-point system shows how an appearance tolerance is written into the acceptance clause.

The Approval Workflow That Works

The cheap gates come first. A lab dip confirms the recipe direction, a corrected dip confirms it is repeatable, a bulk head piece confirms the recipe at machine scale, and a shade blanket taken across the run confirms it holds. Each gate costs little and catches drift before it becomes a claim.

On long runs, approve the head and the tail, because a run that drifts slowly looks fine at every single point checked and wrong as a whole. Shade matching for both polyester and nylon programmes runs at long-term partner mills, and the achievable lot tolerance for a given shade and construction is confirmed per order.

What Else Moves Perceived Shade

The dye is only one input. Before assuming a dyeing error, check the rest of the specification:

• Weight. A heavier cloth of the same recipe reads deeper; the relationship is set out in our guide to fabric weight and end use.
• Yarn lustre. Bright and semi-dull filament scatter light differently at one dye depth, as covered in our article on reading a filament yarn specification.
• Finish and coating. Peach finishes, calenders and pigment coatings each shift the face, per our articles on hand-feel finishing routes and on choosing a coating.
• Functional finishes. Some treatments carry their own colour change, as discussed in choosing a functional finish.
• Printed goods. Ink systems match differently from dye systems, and the strike-off route is covered in our comparison of printing methods on polyester.

Four Misreads That Cost Real Money

• Approving under one light source. A single-source approval is a coin toss on metamerism; name the retail source and a second check.
• No retained master. Without a signed, dated swatch on both sides, each party approves against a memory, and memories drift in opposite directions.
• Instrument number treated as the pass. A textured or high-lustre face games a flat reading; the booth decision under the named source stays primary.
• Lab dip treated as bulk approval. Scale changes circulation, loading and clearing; the bulk head piece is a separate gate, not a formality.

Frequently Asked Questions

Can two dye lots be made identical?

Not reliably, and honest mills do not promise it. Lot-to-lot variation is managed, not eliminated: tight tolerances, head-piece approval, roll linking and a master swatch keep the distribution narrow enough that cut panels sit together. Nylon and polyester differ in how much a recipe can be corrected, as our polyester versus nylon comparison explains.

Which light source should we approve under?

The one your goods will be sold under, plus a daylight source as the secondary. A European retail programme commonly approves under TL84 with D65 as the check; an outdoor programme adds the UV component, because brighteners change under it. Name both in the clause.

What grey scale grade is normal between garment panels?

Adjacent panels in one garment are usually held to 4-5 or better against the master, so any difference stays below notice at arm's length. Separately cut components that never sit side by side can carry a wider limit, which is exactly why the clause should say which is which.

How is shade approved for printed fabrics?

Through the strike-off: a production-method proof on the production cloth, approved under the same sources as the ground shade, then retained like any master. Colour fastness of the print is graded separately on the grey scale, following the approach in our test-report guide, and the method choices are covered in the printing methods article.

Next Step

Write the clause before the bulk runs: a retained master, a named source, a grey scale minimum, separate within-roll and lot limits, and a head-piece gate. Send the shade, the construction and the tolerance you need and we will confirm the route, the approval points and what the lot control looks like for that programme. If the tolerance is still open, send the garment panel plan and we will propose the clause that matches it.