Explaining an Unknown CRI Grade in a Released LED Lot

Separate source color from illuminated-object appearance

The color seen when looking at an illuminated surface depends on the light reflected from that surface. Nichia uses this relationship to introduce color rendering: the light source participates in how the object's colors appear. This is a different subject from the color assigned to the LED itself. A description of the source and a description of an object under that source occupy different positions in the optical comparison.

The reference illuminant gives the rendering comparison its meaning. The question is how the object's appearance under the test source differs from its appearance under that reference. Without that comparison, a statement that the source looks white or belongs to a familiar color category leaves object-color fidelity undescribed. The source's own visible color does not replace the reference-based subject.

This distinction explains why source color sorting and rendering information can coexist. They describe different attributes rather than competing names for one grade. Keep the illuminated object, test light and reference light identifiable in the explanation. A bin-field discussion can identify a source selection; rendering asks what happens to the appearance of colors illuminated by that source.

Distinguish the average from a particular sample

Nichia distinguishes the average color rendering index, Ra, from individual special color rendering indices, Ri. The average concerns a specified group of low-saturation test colors. An individual index concerns a particular test color. These descriptions retain both the subject being evaluated and the way the result is combined, rather than treating every CRI entry as an interchangeable grade.

A group average compresses several sample comparisons into one description. That makes it useful for discussing the general colors represented by the group, while leaving the individual comparisons with their own meaning. Reading an average as though it were the result for every sample would remove information about what the averaging operation includes. The same distinction applies even before any numerical result is considered.

An individual Ri preserves the question associated with its named sample. It is therefore useful to ask whether a statement describes the general low-saturation group or one particular color. The answer determines the metric's coverage, not the present measured quality of a reel. Keep the averaging subject explicit so that a familiar Ra label does not silently absorb every other rendering description.

Retain the red-sample question separately

The strong-red sample provides a concrete reason to retain individual rendering information alongside the average. Nichia explains that vivid colors can be difficult to evaluate through the low-saturation group used for Ra. Its discussion singles out strong red as particularly difficult for typical white LEDs for general lighting to reproduce. The separate red-sample subject consequently adds information about coverage.

This is a distinction between the colors represented by different evaluations. A favorable description of the general group does not become an individual strong-red result merely by being called color rendering. Likewise, discussing the red sample does not reconstruct the entire spectrum of the source. The named sample connects one aspect of object-color reproduction to its particular evaluation context.

Keeping this distinction visible helps a reader interpret two rendering descriptions without assuming they are redundant. One concerns an average over the defined general group; the other retains a saturated-color question that the average does not fully express. The source's explanation supports preserving both subjects. It supplies no basis here for assigning a numerical grade, ranking an offered product or asserting what any unmeasured LED currently reproduces.

Keep spectral tradeoffs and observed resin separate

Nichia links rendering behavior with the distribution of emitted light across wavelength bands. Its discussion explains that changes associated with rendering can also affect apparent brightness because the eye's sensitivity is not uniform across the spectrum. Rendering and luminous flux therefore describe different optical outcomes. A brighter-looking source is not automatically a more faithful illuminant for the colors of an object.

The note also describes the use of different phosphors for different rendering selections and observes that encapsulating resin can consequently look different. That observation concerns the relationship between construction choices and appearance. It does not turn resin color into a rendering measurement. Looking at the package and comparing illuminated test colors are different observations with different subjects.

Retain source brightness, encapsulant appearance and object-color fidelity as separate descriptions. Spectral choices help explain why these attributes can vary together without becoming equivalent. This keeps the technical discussion centered on what each attribute represents rather than using a visible package difference as a grade decoder. The resulting distinction is qualitative: it identifies the optical questions that need their own evidence, without deriving a spectrum or a verified rendering rank from appearance.

Sources and further reading

• Nichia — Color Rendering Evaluation for LEDs for General Lighting (2021)

Nichia 2021 application note: bounded explanatory prose on physical pages 2, 3, 5 and 6 concerning reflected-object color, Ra/Ri sample coverage, strong red and phosphor appearance. Numerical grades, the ambiguous same-wavelength statement, figures and current-standard claims are excluded.