LED Lamp - Technology Overview

Technology Overview

General-purpose lighting needs white light. LEDs emit light in a very small band of wavelengths, emitting light of a color characteristic of the energy bandgap of the semiconductor material used to make the LED. To emit white light from LEDs requires either mixing light from red, green, and blue LEDs, or using a phosphor to convert some of the light to other colors.

The first method (RGB- or trichromatic white LEDs) uses multiple LED chips, each emitting a different wavelength, in close proximity to generate the broad spectrum of white light. The advantage of this method is that the intensity of each LED can be adjusted to "tune" the character of the light emitted. The major disadvantage is high production cost. The character of the light can be changed dynamically by adjusting the power supplied to the different LEDs.

The color rendering of RGB LEDs, however, is worse than one would expect; the wavelength gap between red and green is much larger than that between green and blue, resulting in an uneven spectral density. An orange fruit, for example, does reflect some red and it does reflect some green, but not in a ratio that the human retina interprets as orange. Neglecting to poll the orange line makes most orange objects appear reddish. RGB LEDs are therefore suitable for display purposes, but less so for illumination, which prompted some manufacterers to add a fourth, amber LED, marketing the product as RGBA LED (not to be confused with the RGBA color space) or tetrachromatic white LED. It can be expected that the number of colors will be further increased to six or more, equally-tempered wavelengths.

The second method, phosphor converted LEDs (pcLEDs) uses one short-wavelength LED (usually blue, sometimes ultraviolet) in combination with a phosphor which absorbs a portion of the blue light and emits a broader spectrum of white light. (The same mechanism—the Stokes shift—is used in a fluorescent lamp emitting white light from a UV-illuminated phosphor.) The major advantage is the low production cost. The CRI (color rendering index) value can range from less than 70 to over 90, and color temperatures in the range of 2700 K (matching incandescent lamps) up to 7000 K are available. The character of the light cannot be changed dynamically. The phosphor conversion absorbs some energy, but most of the electrical energy is still wasted as heat within the LED chip itself. The low cost and adequate performance makes this the most widely used LED technology for general lighting today.

A single LED is a low-voltage solid-state device and cannot be directly operated on AC power without circuitry to control the current flow through the lamp. In principle a series diode and resistor could be used to limit the current and to control its direction, but this would be very inefficient since most of the applied power would be dissipated by the resistor. A series string of LEDs would minimize dropped-voltage losses, but one LED failure would extinguish the whole string. Paralleled strings increase reliability by providing redundancy. In practice, three or more strings are usually used. To be useful for illumination, a number of LEDs must be placed close together in a lamp to combine their illuminating effects. As of 2012, white LED assemblies emitting 10,000 lm are available. When using the color-mixing method, a uniform color distribution can be difficult to achieve, while the arrangement of white LEDs is not critical for color balance. Further, degradation of different LEDs at various times in a color-mixed lamp can lead to an uneven color output. LED lamps usually consist of clusters of LEDs in a housing with driver electronics, a heat sink, and optics.