Lighting Education

Your current location: home > Blog

Color Rendering Index and White Light LEDs

Time:2016-09-05 Views:544 Compile:SUNPER

In its 2007 report, CIE Technical Report 177:2007, the International Committee on Illumination (CIE), resolved that the color rendering index (CRI) was not an ideal standard for assessing white light LEDs.


The report “Color Rendering of White LED Light Sources”, states, The conclusion of the Technical Committee is that the CIE CRI is generally not applicable to predict the color rendering rank order of a set of light sources when white LED light sources are involved in this set. 


Before this time, the color rendering index (CRI) had been the yardstick for assessing fluorescent and high intensity discharge (HID) lamps for at least 40 years.


This resolution was a result of various academic studies that compared phosphor-coated white light LEDs and red-green-blue (RGB) LED clusters. In the studies, visual experiments were carried out with participants ranking the appearance of different scenes illuminated with lamps of varying CRIs. Overall, there was poor connection between the rankings and the resultant CRI values. It was observed that many of the RGB-based LED products ranked, have CRIs in the 20s, yet the color rendering of the light source appeared to be very acceptable.


The CIE report however, goes on to propose a new metric, still under development, that will effectively replace the CRI. Obviously, in the meantime there must be a method of using CRI for assessing LEDs which brings us to the focus of this article. Let’s look at what CRI measures, its relation to normal light sources and what makes LEDs so unique.


CRI and How it is Measured.


The
color rendering index (CRI) is basically a quantitative measure of the ability of a light source to reproduce colors and tones as realistically possible in comparison with reference ideal or natural light source. 


In measuring CRI, the procedure compares the appearance of eight CIE standardized color samples when illuminated by the light being tested and the selected reference illuminant. Only these eight colors are used in this test and they are pastels, not saturated colors. 


A CRI of 100 is the recognized maximum, so, any difference is deducted from 100 to get the CRI of the tested light. The smaller the difference, the higher the CRI score and vice-versa. Low CRI scores indicate unnatural appearance of certain colors. 


Incandescent lamps, for example, generally have a CRI above 95 while cool white fluorescent lamps rank at about 62 on the CRI scale. Some fluorescent lamps coated with rare-earth phosphors have CRI values of 80 and above.


For lamps with CCTs less than 5000 Kelvin (K), the reference illuminant sources are spectral power distribution (SPDs) or black body radiators which are very similar to incandescent lamps. For CCT sources above 5000 K, the reference is a specially defined SPD using a mathematical model from daylight. Illuminants that produce the closest resolution to incandescent light or daylight when tested against the eight color samples will score the highest CRI values. 


Correlation between Spectral Power Distribution and LEDs. 


Continuous improvements over the years have seen fluorescent phosphors being tuned and processed, to render the standard eight color samples as close as possible to incandescent or daylight references. An interesting development is the effect of changing these phosphors on the SPD for the fluorescent light source in Figure 1 below. Even slight changes shift the emission wavelengths and cause the CRI score to drop significantly. Yet, these changes are hardly discernable to the human eye. Phosphor-converted (PC) LEDs, for instance, use broadband phosphors to score as high as between 70 to over 90 on the CRI scale.


Comparing this with the effect on an RGB LED in Figure 2 above, the change in SPD is similar to the fluorescent lamp, with distinct blue, green, and red peaks. The RGB LED has a CRI score of 27 because there is a wide difference between those wavelengths when compared to incandescent lights on the CRI eight color samples. One advantage of commercial RGB LED clusters is that despite their low CRI, the white light they produce is usually appealing to the human eye. This is most likely due to the tendency to increase the Chroma of most colors illuminated without significant changes in hue and tone. 


Another example is neodymium coated incandescent lamps, which though manifesting low CRI, illuminate objects brighter as compared to unfiltered incandescent lamps. Some popular brands of these lights include GE Reveal, DaylightTM (Sylvania) and Philips Natural Light.


Recommendations for CRI in the Meantime.


Though CRI should not be used as a substitute for in-person and on-site evaluations in product selections, it can still serve as a baseline for evaluating white LED products and systems. This will likely stop once the revised color metric under research and development for all white light sources is released. 


Action points: 


1. On-site and in-person evaluations still remain the best way to evaluate LED systems when color fidelity and/or appearance are critical.

2. CRI comparisons for all light sources should be made only for illuminants of equivalent CCT levels, including LEDs. Little variances in CRI levels, e.g. 72 and 76, are insignificant and will be hardly noticeable.

3. CRI can still remain a valuable deciding factor for rating LED lights in some instances, especially where color accuracy is key. For example, when it’s necessary to compare color of items under both natural daylight and electric lighting. 

4. In some instances, color appearance is more desirable than color fidelity. In such cases, low CRI will obviously be a poor reason to exclude white light LEDs as some such LEDs still produce pleasant white light.


In all, the particular circumstances at hand and the importance of light fidelity vs light appearance will largely determine if CRI is used to determine the choice of LED lights or not.


User comments

Your current input 0 characters(Reply for at least 6 characters)。

Online Services

livechat

Mobile: +86 18938902515 (Mr.Allen)

Tel:+86 755 61272188

Fax:+86 755 61673151

E-mail: sales@sunper.net

zip code: 518108

Quick Messge