Color Match Approval at Mold Trial — How Is Delta E Measured and Specified for Molded Parts, and What Drives Color Drift Shot to Shot?

Last updated: 2026-09-04 · Reading time: about 6 min

Color on a molded part is approved against a measured Delta E (ΔE) in CIELAB space, not by eye. A spectrophotometer reads master and part, computes ΔE from three axes. ΔE < 1.0 indistinguishable; 1–2 slight; 2–4 typical consumer limit. But ΔE alone does not close the approval — gloss mismatch and metamerism can fail a part that passes the number, so the spec carries a signed visual limit sample and defined measurement conditions.
See also related article.

Key facts

<>

  • Delta E is the total color difference between a molded part and a master in CIELAB space (L = lightness, a = red-green, b = blue-yellow). A spectrophotometer reads both and reports the number.
  • ΔE is necessary but not sufficient — a part can pass the number and fail on gloss, viewing direction, or metamerism. The full spec adds a signed visual limit sample and defined instrument geometry.
  • Shot-to-shot drift traces to five process variables — resin lot, masterbatch dose, melt and mold temperature, regrind % — not just the pigment. Color control is process control.

    • Cavity — The hollow space inside the mold that forms the part shape. Multi-cavity molds produce multiple parts per shot. Cavity count determines production output but also mold cost and complexity. Each cavity must be balanced for uniform fill. See also related guide.

    What causes color to drift on the same mold?

    Five sources move the spectrophotometer reading even when the masterbatch has not changed. Each has a different mechanism and a different fix.

    Drift Source How It Shifts Color Control
    Resin lot variation Base resin tint differs between lots (natural PC yellows, filled grades shift), moving L and b even with the same colorant Reserve one lot for the program; check each incoming lot against master before compounding
    Masterbatch dose A ±0.2 % change in let-down ratio shifts color measurably; poor metering causes batch-to-batch ΔE Fixed let-down ratio, gravimetric dosing, blend to homogeneity
    Melt temperature & residence Heat-sensitive pigments (blues, some reds) darken or shift when overheated or held too long Hold melt temp to a narrow window; right-size shot for short residence
    Mold temperature & gloss Colder mold raises apparent gloss, reading as lighter color even with identical pigment Fix mold temperature; measure gloss (GU) alongside color on textured faces
    Regrind & drying Regrind has degraded pigment and different flow; moisture causes splay that scatters light Cap regrind %, dry to spec, keep regrind from same color only

    Delta E (ΔE). The total color difference between a sample and a master, computed from three CIELAB axes: ΔL (lightness), Δa (red-green), Δb (blue-yellow). ΔE = √(ΔL² + Δa² + Δb²). A spectrophotometer reads both surfaces under defined illuminant and geometry. It is the quantitative half of a color approval; the qualitative half is the visual limit sample.

    The subtlety is that several drifts change perceived color without changing the pigment at all. A gloss shift from a colder mold, a wall-thickness change altering light transmission, or unbalanced fill orienting pigment differently at the surface can move the spectrophotometer reading while the material recipe is unchanged. The correct model is not “the color is in the masterbatch” but “the color is the part as molded” — every appearance variable is in scope.

    How do you build a color approval that survives production?

    Four steps. The goal is an approval that holds from T1 sign-off through full production, not just on the day of the trial.

    1. Set the master and measurement geometry before trial. Agree a physical master chip — or a defined Lab value with stated instrument and aperture — and fix conditions on the drawing: geometry (D/8 or 0°/45°), illuminant (D65 plus TL84 or A), substrate thickness, surface finish, and read locations. Color is only comparable when instrument, illuminant and surface match. Put all of this in the spec before T1.
    2. Specify the pass rule as ΔE limit plus visual boundary. Set a numeric ceiling (ΔE ≤ 1.5 on A-surface, ≤ 3.0 hidden) and add two reinforcements: a signed visual limit sample — the worst acceptable part kept at the line — and directional language where the eye is sensitive (never darker on a pale part, tighter Δb for beige). A part “within ΔE 1.5” but biased yellow can look wrong on a mating set even though it passes.
    3. Check metamerism before signing off. Two parts can match under daylight and diverge under store lighting — metamerism, from different spectral reflectance curves agreeing under one illuminant. Guard by reading under at least two illuminants (D65 plus fluorescent or incandescent) with a metamerism index limit. On multi-material assemblies, check parts together in real light, not just against loose chips.
    4. Lock the process and monitor as a control chart. Once a matched part is approved, freeze the recipe: resin and colorant lots, melt and mold set points, regrind %, cycle. In production, read color at defined frequency — first-off, then intervals — and trend ΔE over time. A control chart catches slow drift while still inside the limit.

    “Color approval is four things: a physical master with defined measurement conditions, a numeric limit paired with a visual boundary, a metamerism check under two lights, and then statistical monitoring so drift is caught before a shipment is off-color.”

    — Process Engineer, aCatMould

    Frequently asked questions

  • Key parameter identified — critical variable confirmed against spec.
  • Tolerance within achievable range — commercial ±0.15mm or tighter as needed.
  • Material matched to application — resin grade selected for end-use environment.
  • Process window validated at T1 — parameters locked after first trial.
  • Inspection method defined — CMM, visual, or functional test specified.

    • What Delta E value should I specify?

    Choose from how the part is seen and mated, not a generic number.
    Mating parts side by side (bezel and body) need ΔE under 1.0; a stand-alone housing can accept 2–4; a hidden bracket may need no limit at all.

    Set per surface — A-face tight, hidden faces looser — and consider directional limits. A useful starting point: ΔE ≤ 1.5–2.0 on the A-face with a signed visual limit, tightened only if a real mate demands it. An unnecessarily tight universal ΔE raises scrap without benefit.

    Why do mating parts not match each other even though each matches the master?

    Because each was approved against the same loose master, and small errors that each read “OK versus master” can differ by twice the individual tolerance. A part at ΔE 1.2 one way and its cover at ΔE 1.2 the other can sit 2.4 apart even though both pass 1.5.

    Fixes: set tolerance tight enough that two worst-case parts still match; approve mating parts together in the actual assembly under real light; control the process so parts cluster around the master. For mated color, spec the spread between parts, not each part’s distance from a chip.

    Is color the same issue as gloss?

    No — related but distinct. Color (L-a-b from the spectrophotometer) is hue and lightness of the pigment. Gloss (gloss units at 20/60/85°) is how mirror-like the surface is, set by mold texture and melt replication. A part can be dead-on in pigment but read as a different color because its gloss differs.

    A glossier surface looks deeper and darker; matte reads lighter and chalkier. On cosmetic parts the spec carries a gloss range alongside the ΔE limit, both under defined geometry, and a color pass at the wrong gloss is still a reject. Treat them as two required readings on the same acceptance.

    Key takeaways

    <>

  • ΔE is necessary but not sufficient — color approval also requires a signed visual limit sample, defined spectrophotometer geometry and a metamerism check under multiple illuminants.
  • Most color drift comes from process variables — resin lot, masterbatch dosing, melt and mold temperature, regrind percentage — not from the pigment itself.
    Color control is process control.
  • For mating parts, spec the maximum spread between parts and approve them together in the actual assembly under real light, not just each against a loose master chip.

    • Color on a molded part is approved against ΔE in CIELAB — below 1.0 indistinguishable, 1–2 slight, 2–4 typical consumer limit.
    But ΔE alone does not close the approval: gloss mismatch and metamerism can fail a part that passes the number, so the full spec carries a signed visual limit sample and defined conditions. Shot-to-shot drift traces to five process variables — resin lot, masterbatch dose, melt and mold temperature, regrind % — so controlling color means controlling the molding process, not just the colorant.

    Sources

    1. ASTM International — Standards for measurement methods, including color difference evaluation. www.astm.org
    2. ISO 20457:2018 — Plastics moulded parts: Tolerances and acceptance conditions. www.iso.org
    3. Plastics Technology — Molding process data and appearance defect guidance. www.ptonline.com
    4. BYK Instruments — Gloss measurement practice and instrument geometry. www.byk-instruments.com
    5. Plastics Industry Association — Industry resources on plastics processing practice. www.plasticsindustry.org

    Sources: Plastics Technology · Plastics Technology Online · Society of Plastics Engineers (SPE) · ThomasNet