Three numbers dominate every LED product datasheet's colour section — CRI, CCT, and spectral power distribution. They get treated as a package, but they measure genuinely different things, and a high score on one doesn't guarantee a good result on the others.
CCT: What Colour Is the Light, Broadly Speaking
Correlated colour temperature describes the apparent warmth or coolness of a light source in kelvins. Lower CCT (2700–3000K) reads warm; higher CCT (5000–6500K) reads cool. It's the figure most directly tied to how a space feels, but it says nothing about colour accuracy — a source can have an attractive "correct" CCT and still render colours poorly, which is the gap CRI is designed to measure.
CRI: How Accurately Does It Render Colour
Colour rendering index quantifies how naturally a source reveals true colours, compared against a reference illuminant of the same CCT, across eight standard colour samples. CRI is only meaningful when comparing sources of the same CCT — a 3000K source with CRI 90 and a 5000K source with CRI 90 aren't measured against the same reference.
CRI also has known limitations: because the eight test samples are relatively desaturated, a source's spectrum can be tuned to score well on them while still rendering saturated colours — particularly deep reds — poorly. This is why CRI is increasingly supplemented with newer metrics like TM-30 (99 colour samples, reporting fidelity and gamut separately), though CRI remains the figure most commonly required today.
Spectral Power Distribution: The Data Both Numbers Are Calculated From
SPD is the underlying measurement CCT and CRI are both derived from — a graph of optical power at each wavelength across 380–780nm, showing the actual shape of the spectrum rather than a single summary number.
Two sources can produce identical CCT and similar CRI while having visibly different spectral shapes — phosphor-converted LEDs commonly show characteristic spikes and gaps that don't hurt CRI score but can still affect how saturated colours appear under careful comparison. For applications where colour accuracy genuinely matters — retail display, art and museum lighting, photography — reviewing the actual SPD curve is the only way to confirm a source does what the application needs.
How These Are Measured
| Parameter | Measurement method | Typical instrument |
|---|---|---|
| CCT | Calculated from measured chromaticity coordinates | Spectroradiometer + integrating sphere |
| CRI | Calculated from spectral comparison against reference | Spectroradiometer |
| SPD | Direct spectral measurement | Spectroradiometer |
All three are generated from the same underlying spectral measurement, which is why a quality spectroradiometer system — rather than a simple colorimeter reporting only summary numbers — is the right equipment choice for any lab that needs to inspect spectral shape rather than just read off figures.
Reading a Datasheet Correctly
CCT 4000K and CRI 90 tells you the light reads as neutral white and renders most everyday colours reasonably faithfully. It doesn't tell you how the source handles deep saturated reds, says nothing about flicker, and doesn't let you compare colour quality against a product with a different CCT on a like-for-like basis. For anything where colour fidelity is the actual point of the purchase, pulling the SPD curve and considering TM-30 alongside the headline numbers gives a materially more complete picture.