It’s not just about long life, low voltage and high efficiency, although there’s nothing to complain about with any of these characteristics. But the rapid development of high output LED technology has also created the opportunity for new lighting paradigms where the spectral characteristics of the light source can be tailored or “sculpted” to best suit the objectives of the application.
So, how is this done?
Well. It’s not done by ordering LEDs out of a catalog. The metrics most commonly used by the lighting industry, CCT (correlated color temperature) and CRI (color rendering index), which can be useful for general lighting applications provide very little information about the output spectrum of the light source. To produce a customized output you must start with the light producing elements themselves, the LEDs.
In the “early days” of LED lighting (basically the late 1990’s to around 2005) it wasn’t clear what the best way of using LEDs for lighting would be. Current LED technology has the capability of generating light from about 1080nm (the near-infrared) to about 210nm (the deep-ultraviolet) which, of course, includes the entire visible spectrum. So, what’s the problem with making white light? It’s not an intrinsic issue but more of a practical one. Every LED device emits in a fairly narrow wavelength band. Therefore, to produce white light typically requires 3 different LEDs. Conventionally, these are red, green and blue.
In order to control the color of the combined output spectrum requires 3 independent electronic control channels. That’s starting to sound complicated and, more importantly, expensive. Then add to that the fact that these LEDs age at different rates which forces the use of separate compensation circuits if a constant color is to be maintained over time.
This is getting messy and even more expensive.
Fortunately, some practical solutions to the multichannel white problem came along very soon after the first high output blue LEDs were developed. These devices typically start with a blue LED as the light generating source and then cover this LED chip with a phosphor mixed into a silicone goop which is precision dispensed on top. The phosphor absorbs some of the blue light and re-emits it at a longer wavelength which, together with the unabsorbed portion of the blue light combines to produce some kind of white light output.
So, where’s the change in the paradigm, you ask?
Well, up to this point, producing quality and cost effective lighting involved big companies and lots of manufacturing equipment. The economy of scale was obtained by one factory producing the light bulb in large volume. There was not much opportunity for customization with incandescent or fluorescent lights and any customization resulted in a very expensive light source. But, with LED lighting, the “heavy lifting” of producing the blue LED chip can be done by a rapidly growing number of chip manufacturers. Now, turning this blue light into a customized white light spectrum can be done cost effectively with readily available fluid dispensing equipment and widely available silicone encapsulants.
Of course, you still need to know what you’re doing with this back-end process and you particularly need to understand the phosphor technologies in order to do your own spectrum “sculpting”. But, you don’t need to build a computer chip factory to obtain cost effective, customized LEDs. You can do it by really understanding the world-wide LED manufacturing infrastructure to establish strong supplier relationships with the industry. Then work at the chip packaging level to blend your own phosphors which are used to make the finished LED light sources as per your design.