Read time: 8 minutes  |  By: YUPSENI Team

1. I. The Second Container Problem: Why the Colour Shift Happens
2. II. Delta E: The Number That Decides "Close Enough"
3. III. What Drives Colour Variation on a PVC Extrusion Line
4. IV. From Sample to Batch: Setting a Colour Standard You Can Enforce
5. V. The Light Trap: Why the Same Panel Looks Different Indoors and Out
6. VI. Colour Without a Standard Is an Argument

Two batches of the same product, side by side. Whether the difference is acceptable or not depends entirely on the standard you set before production, not on the argument you have after delivery.

The first container is perfect. The second container is visibly different. Same product, same specification, same supplier - and yet the shade has shifted. This is the most common and the most avoidable complaint in building material procurement.

Colour variation is not a sign of a careless factory. It is a natural result of a continuous extrusion process, where raw material, temperature, and machine wear all drift slightly from one run to the next. The question is not whether variation happens - it always will. The question is whether you agreed in advance how much variation is acceptable. Setting that number is what separates a professional order from a test report that settles a claim.

You might think a factory that produced a perfect first batch should reproduce it exactly. That is not how extrusion works. Every production run starts with a new lot of virgin resin, a fresh batch of colour masterbatch, a machine that has worn slightly since the last run, and a different ambient temperature in the plant that day.

Each of these variables nudges the final shade by a small amount. Individually, the shifts are tiny. Added together, they can push a long profile or a wide panel into a shade that a customer notices the moment it is placed next to the first batch.

The absence of a defined tolerance is the real problem. If no one said "the difference must be smaller than X," then every difference is technically within spec - and every dispute is a matter of opinion.

Colour difference is measured with a single number called Delta E. It represents the distance between two colours in a three-dimensional colour space. The lower the value, the closer the two samples match. It replaces the subjective phrase "looks about the same" with a number an instrument can read.

Different applications tolerate different Delta E levels. An interior decorative PVC board installed in a single room needs a tighter tolerance than a PVC profile used across a long fence run, where the eye expects a subtle gradient rather than a perfect match. The tolerance belongs in the specification, not in the after-sales conversation.

The colour of a PVC product is set by a pigment masterbatch mixed into the compound. Getting an exact shade requires controlling several variables at once:

• Masterbatch lot variation - Each lot of colour concentrate is itself a batch, with its own small variation.
• Extrusion temperature - A hot barrel melts the pigment more evenly; a cold barrel can leave the colour streaked or washed out.
• Filler ratio - Shifts in the calcium-carbonate or wood-fibre content change the surface lightness.
• Screw wear and residence time - Worn screws and longer residence times affect mixing and can darken or lighten the melt.
• Cooling rate - Faster cooling can set a slightly different surface crystallinity, shifting the perceived shade.

This is why a factory with a disciplined process - one that records each variable and re-checks colour at set intervals - produces more consistent batches than a factory that simply runs the line and hopes.

The only reliable colour standard is a physical, signed reference sample that both parties hold. This is why the signed sample matters so much - it turns colour from a debate into a measurable benchmark.

A proper colour acceptance standard should specify three things. First, the reference: which physical sample is the master, and what Delta E tolerance it allows. Second, the light source: colour must be evaluated under a defined illuminant, because the same pair of colours can match under one light and separate under another. Third, the measurement: how the sample is taken and how often it is re-checked across a production run.

Without all three, the standard is incomplete. A Delta E of 2.0 is meaningless unless it is coupled with a specific light source and a defined measurement method. This is the same logic that applies to ordering a pre-production sample - you need the reference object, not just a description.

A panel can match its reference under a warm indoor light and visibly diverge under daylight. Specifying the light source avoids this metamerism trap.

One of the subtlest traps in colour matching is metamerism. Two samples can measure nearly identical under a fluorescent lamp, then separate noticeably under daylight or LED. The pigments reflect and absorb light differently across the spectrum, so the match is only valid under the light it was tested in.

For an exterior application like a fence that will always be seen in daylight, the standard must be set under daylight simulation. For an interior wall panel seen under LED or incandescent light, the standard should use that source. This is why a single Delta E number, without a light source attached, is dangerously incomplete.

Here is the uncomfortable truth about colour disputes. If there is no written standard, the argument is decided by whoever is more persuasive, not by who is right. The buyer says the panel is too dark. The factory says it is within tolerance. Both are true in the absence of a number.

The fix is to put the standard in the contract before production. Specify the signed reference sample, the Delta E threshold, the light source, and the re-check frequency. Then the delivery is either conforming or non-conforming on paper, and the claim resolves itself.

That is the entire point of a standard: it moves the argument from opinion to measurement. A factory that agrees to a defined colour tolerance is a factory confident in its process. One that avoids it may not be.