Nine vans, one colour, two print runs
The job was nine service vans in a single brand red, booked as two runs six weeks apart because that is how the fleet came off the road. Every panel matched its proof. The proofs matched the file. The file matched the brand guide. On delivery day the fleet manager walked the yard, stopped at van six, and asked why it was orange.
It was not orange. It was ΔE 4.2 away from vans one to five, which is enough for anyone to see in daylight and enough for a brand team to reject. The file had not changed. The film lot had, the ink lot had, and the laminate had gone from gloss on run one to matte on run two because the gloss was out of stock.
Four numbers that decide whether a printed panel passes
A printed panel is a car film plus ink plus laminate plus light, and each of those four moves the final number.
1. Name the instrument, the formula and the illuminant, or you have not specified a colour
ΔE is not one number. ΔE76 and ΔE00 give different values for the same pair of samples, sometimes 3 against 1.8, so a contract that says “within 3 Delta E” without naming the formula is a contract that will be argued about. Write ΔE00, ASTM D2244, measured with a spectrophotometer at D65/10° or D50/2°, on the laminated panel, 24 hours after lamination.
Then set the thresholds the way a brand team actually reads them. Below 1.0 is effectively invisible and is where brand-critical solid colours should sit. 1.0 to 2.0 holds up across a matched panel set and is the normal production tolerance. 2.0 to 3.0 is where a trained eye finds it side by side. Above 3.0 it shows under most lighting; above 5.0 it reads as a different colour and the panel gets reprinted. Measure five points per panel, at least 50 mm in from any edge, and record them. An adjective is not a measurement.
2. The laminate is part of the colour, not a protective afterthought
A gloss overlaminate sits the ink under glass: darks deepen, saturation lifts, and the panel reads richer than the unlaminated proof. A matte laminate scatters light in every direction, which lifts blacks toward grey and pulls saturation down. On a saturated red or a deep blue that shift alone is commonly 1.5 to 3 ΔE, and it is the single most common reason a second run stops matching the first.
So the laminate is a specification item with a lot number, not a line item chosen by what is on the shelf. If run two cannot get the same laminate, the tolerance has already moved before the first panel is printed. That is why the laminate, not the ink, is where most fleet wraps fail after two winters, and why the same layer decides colour on day one.
3. The white point under the ink moves everything above it
Ink is translucent. What sits under it — the white base of the Car Wrap Vinyl, its opacity, its slight tint batch to batch — shifts the whole printed image. A base that reads a fraction warmer than the previous lot will make every colour on the panel read warmer, and no ICC profile fully compensates for a substrate shift of that kind.
Ask for the base film’s L* value and opacity tolerance, not just “gloss white”. Ask for the film lot number and keep one roll of the previous lot as a physical reference. When the new lot arrives, print a small target on both and measure them against each other before you commit a fleet.
4. Metamerism decides the argument you will have in the car park
Two samples can match under D65 daylight and split under the warm-white fluorescents in a depot, or under LED street lighting at night. Warm palettes — golds, creams, brand reds and off-whites — are the usual victims. Approve under the illuminant the customer will judge in, then check under two more: north-facing daylight, cool-white shop light, and the light in the yard where the vans park.
Optical brighteners in white vinyl add a second trap. Measured with UV included (M1) and with UV excluded (M2), the same panel can differ by a visible margin. State the measurement condition, or two people with two instruments will produce two truths.
| What to ask | Why it matters | Pass range |
|---|---|---|
| ΔE00 tolerance, formula and instrument named | ΔE76 and ΔE00 disagree; unnamed numbers get argued | ΔE00 ≤1.0 brand-critical, ≤2.0 standard, >3 reject |
| Measurement point and timing | Ink and laminate keep settling for a day | Laminated panel, 24 h after lamination, 5 points ≥50 mm from edges |
| Base film L* and opacity tolerance | Translucent ink amplifies any white-point shift | L* 95–97, opacity tolerance stated per lot |
| Laminate type, gloss level and lot | Gloss deepens, matte desaturates; a swap alone is 1.5–3 ΔE | Same product and lot across the whole fleet order |
| ICC profile per film + ink + laminate | One profile cannot cover three variables | Profile dated and tied to that combination |
| Control panel per print run | The only way to catch drift mid-run | One panel filed per run; first vs last panel ≤1.5 ΔE |
| Illuminants for approval | Metamerism only shows under the second light | D65 or D50 plus daylight and shop light |
| Ink limit and pass count | Over-inking shifts hue and slows cure | 240–280% total ink, per the media profile |
The five-step colour control that survives a second run
Step 1. Build a spectral target, not a Pantone number. Measure the brand’s physical reference — a painted panel, an approved swatch, a vehicle — and store Lab*. A Pantone code is a starting point; the number is the target.
Step 2. Print three conditions on the real materials. One strip unlaminated, one with gloss laminate, one with matte, all on the production film. Wait 24 hours, then measure each against the target. Now you know what the laminate will cost you before you price the job.
Step 3. Put the number in the quote. ΔE00 ≤2.0 panel to master, ≤1.2 panel to panel within one vehicle, instrument and illuminant named, measurement points named. Sign it. A dispute about a number is short; a dispute about memory is not.
Step 4. Lock the variables on the job sheet. Film lot, ink lot, profile, laminate lot, operator, printer. If any one of them changes between runs, the run is a new colour job.
Step 5. Print one control panel per run and measure the ends. First panel and last panel off the roll, same day. Drift above 1.5 ΔE means re-profile before the next vehicle, not after the fleet is finished.
What it costs when you skip it. Nine vans in two runs. Run two lost the gloss laminate and picked up a new ink lot, and the rear quarters measured ΔE00 4.2 against the master. Three panels reprinted: about 65 USD of film and laminate each, 2.5 hours of install at 75 USD each, roughly 760 USD and two days of bay time, plus a conversation with a client who now measures things. The control panel that would have caught it costs one square metre and ten minutes.
If the brand colour has to hold for eight summers rather than three, the cheaper route may not be print at all: a stock shade of Color TPU PPF removes ink, laminate and profile from the equation entirely, at a higher material cost and a shorter colour list. That trade is worth pricing honestly before you promise a fleet a printed match.
The line to hold
Colour on a wrap is a process output, not a file. If you cannot name the instrument, the illuminant and the number, you have not specified the colour — you have described it.











