Ask almost any electrical engineer with even modest analog and power expertise to assess the efficiency of a power supply, and he or she will know how to do it, or at least how to begin the process. You get a dummy load or (even better) an active load, measure the voltage and current into the supply, measure the same parameters at the supply's output, do some basic calculations, and you have the initial answer. If needed, run the test over a range of input values and loads, as well as under static and dynamic situations. It can get a little complicated, but the basic setup and process are there to do it.
Now ask that same engineer to measure the efficiency of an optical source, such as an LED or CFL, and there's a lot of confusion. Why? Because measuring optical parameters brings in a whole new set of issues, many of which are far more complicated and having subtleties beyond basic electrical measurements. For example, do you want optical output power across all wavelengths, only the visible spectrum, or a selected band (such as IR, or a slice within visible)? Do you want the power in all directions, or just for a specific direction and solid angle?
Optical measurements have a very long history, going back hundreds of years. Some of the terms and concepts are unchanged over the years, while some have been updated to SI-based units (International System of Units), so there is occasional confusion as old and new terms are used, or old terms have been formally redefined (what is a "candlepower" unit, anyway?).
Before you can get the right instrumentation and set-up, you have to be familiar with basic optical parameters and terminology of photometry (measurement of visible light) and colorimetry (measurement of colors), so you can be sure you are measuring what you really need to know. Also keep in mind this critical and basic difference between electrical measurements and optical ones: in most cases (but not all, RF is an exception) when you measure electrical power, you measure voltage and current, and then calculate power via analog circuitry or digital processing. However, when you measure optical power, you really are measuring the power itself via a transducer with output voltage or current which has a well-defined correlation to the impinging optical power.
Key optical parameters
Color (or chromaticity)
: this is an obvious parameter, which we describe with words such as red, yellow, and green, but there are two ways to describe it technically.
First is the wavelength of the light, usually expressed in nanometers (nm). The visible spectrum spans red (620–750 nm) to violet (380–450 nm); beyond red, infrared (IR) ranges from 700 to 1,000 nm (1 mm), while beyond violet is ultraviolet (UV, 10–400 nm). Note that these range borders do not have sharply delineated, consistent boundaries, since spectrum is a continuum of color mixed with human judgment1
(Figure 1). There are multiple but different standardized divisions of near, medium, and far IR. One is defined by International Commission on Illumination (CIE), one from the International Standards Organization (ISO) 20473 standard, and there are a few variations specified by astronomy organizations and societies.
Figure 1: Visible light is only a small slice of the electromagnetic spectrum, with IR on one side and UV on the other; the demarcations within the visible and broader spectrum are not “hard borders” but are defined slightly differently by different organizations.
Of course, most light sources are not strictly monochromatic but include a range of colors and wavelengths. Therefore, the equivalent blackbody temperature is a better way to define chromaticity of collection of light wavelengths of varying intensity. Blackbody radiation is the electromagnetic radiation (which light is) which is emitted by an opaque, non-reflective body kept at a constant temperature; as a result, the radiation has a specific spectrum and intensity that depends only on the temperature of the blackbody.
The CIE chromaticity chart (1931)2
, shown in Figure 2, is based on the equivalent blackbody radiation temperature, and is a standard which can be re-created in a lab environment. Note that sunlight above the atmosphere is 5,900 K, while daylight at ground level it is between 5,500 K and 6,000 K, depending on time of day, cloudiness, and other factors. Incandescent bulbs are between 2,700 K and 3,300 K, and fluorescent bulbs are between 3,500 K and 5,000 K (all numbers are approximate; there is a wide range in each). Important note: using blackbody radiation, light which is perceived as more "blue" is at higher, warmer temperatures, while more reddish light occurs at lower, cooler, temperatures; this is a source of confusion since it is the opposite of our normal pairing of blue as cool and red as hot.
Figure 2: The CIE chromaticity chart relates color to wavelength and also to blackbody temperature; note wavelengths associated with colors marked around the periphery, and blackbody temperatures marked on the arc within the figure.
Radiant and luminous flux
: Radiant flux
is a measure of the total power of emitted light (IR, UV, and visible), while luminous flux
is a measure of the perceived power of the visible light only. Unlike radiant flux, luminous flux takes into account the varying sensitivity of the human eye to different colors, while radiant flux measurements indicate the total power of all electromagnetic waves emitted, independent of the eye's ability to perceive it. To measure the luminous flux, a device called an integrating sphere (Figure 3, also see the video3
) is used to capture and "even out" all the light emitted in all directions; then the intensity of the sphere's surface illumination is measured via an optical power sensor (see referenced video).
Figure 3: Integrating spheres come in many sizes, from about ½ meter diameter to several meters; the size needed depends on the size and power of the light source being evaluated.
Lumens, lux, and candela
: the basic SI unit of luminous flux is the lumen (lm), which is the luminous flux of visible light of a source which produces one candela (cd) of luminous intensity over a solid angle of one steradian (recall your solid geometry: a full sphere has a solid angle of 4π steradians (Figure 4)). The candela is the luminous intensity, in a given direction, of a source that emits monochromatic radiation at a frequency of 540 × 1012
Hz and that has a radiant intensity in that direction of 1/683 W per steradian. A source that uniformly radiates one candela in all directions has a total luminous flux of 1 cd × 4π steradians ≈ 12.6 lumens. A related parameter is the lux, equal to one lumen per square meter.
Figure 4: Many optical parameters and optical measurements are not limited to a two-dimensional plane, so familiarity and comfort with solid angles and geometry is critical.
is the amount of optical radiation passing through (or emitted from) a specified solid angle, and in a specified direction. It is a useful parameter because it indicates how much of the power will be received by an optical system looking at the surface from some angle of view, which may be off-axis.
Efficiency and efficacy
: Luminous efficiency
is the measure of how much optical output power is produced from the source (here, electric power), regardless of wavelength. Luminous efficacy
, in lumens per watt, measures the ability of the light source to produce visible light, and how well the emitted radiation is detected by the human eye. The maximum possible efficacy is 683 lm/W at 555 nm, corresponding to an efficiency of 100 percent.4
Engineers should note that efficiency and efficacy are often confused in discussions, either due to basic misunderstandings or the casual nature of the discussion. Therefore, it is important to make sure that you are clear about what is being discussed, either in person or via the datasheet, and which term is meant.
Let’s now look at a few real-world examples (all parts are available from Digi-Key). Delivering high lumen output and efficacy in the XM package, the Cree XLamp XM-L2
single-die LED is built on the company’s SC³ Technology Platform and is said to deliver up to 20 percent more lumens and lumens-per-watt and double the lumens-per-dollar of the original XM-L. The XM-L2 LED offers the attractive combination of high efficacy and high lumen output at high drive currents, delivering 1,198 lumens at 116 lm/W efficacy at 3 A, 25°, according to Cree.
XLamp XM-L2 LEDs are a good choice for lighting applications where high light output and maximum efficacy are required, such as LED light bulbs and outdoor, portable, indoor, and solar-powered lighting.
Multifunctional devices such as tablets, ultrabooks, and even smartphones need high-brightness levels and good-color rendering without draining the batteries too quickly. For these and other applications, the Osram MicroSideled
series is said to be extremely efficient in white and also in blue, offering constant brightness throughout its lifetime of 15,000 hours and able to withstand high temperatures and currents.
The white version of MicroSideled 3806 achieves a high efficacy of 150 lumen/watt (lm/W). The efficiency of the blue MicroSideled is 55 percent, measured as the external quantum efficiency (EQE), in other words, the ratio of the electrical power used to the emitted optical power. According to Osram, efficacy is 10 to 15 percent greater than with classic white LED solutions. Both versions are compact, measuring only 3.8 x 1.0 x 0.6 mm (length x width x height), and have good thermal conductivity with a thermal resistance of 66 K/W.
series from Seoul Semiconductor
has been optimized to deliver maximum efficacy and luminous flux with an industry-standard 3535 surface-mount package. The device delivers up to 132 lm/W at 350 mA, (85°C junction temperature, 80 CRI minimum) in warm white (3,000 K). In cool white (6,000 K), the Z5M1 delivers up to 150 lm/W at 350 mA (85°C junction temperature, 70 CRI minimum). The Z5M1 series is available in the full Correlated Color Temperature (CCT) range of 2,600 to 7,000 K and two CRI options to provide lamp and luminaire manufacturers with maximum flexibility. The improved efficacy helps lighting manufacturers use fewer LEDs in their system designs which translate to lower system costs. These LEDs can be operated over a broad range of drive currents, allowing SSL product developers to trade off the number of LEDs used in a design with efficacy and rated life. The Z5M1 can be operated from 150 mA to 1.5 A, making them suitable for a wide range of general illumination applications from street and area lighting to replacement lamps to high-output flashlights.
So, what is the LED's efficiency?
Now go back to the original question of "how efficient is that source?" The answer is, "it depends." Before you can answer this simple question, you have to clarify the answer to: "in what way efficient?" This means you consider factors such as the light source spectrum of interest (all, just visible, or some defined slice of the visible/not visible spectrum), the radiated direction (over the full sphere, or just along one axis), and if just in one direction, over how wide-angle a solid cone along the radiation axis (note that the "cone" does not even have to be circular, it may well be elliptical). A source which generates light in a wide spectrum and over a wide solid angle may not be as useful as one which is both narrow in spectrum and solid-angle radiance, although it may be more efficient in the basic power sense.
These are just some of the top-level parameters for optical measurement; others include factors such as dispersion and loss. Due to the non-intuitiveness, subtleties, and hands-on difficulties in measuring some of these parameters, engineers working with optical components and subsystems need to be prepared to do enough up-front research, and work with knowledgeable vendors of test equipment and applications specialists, to make meaningful, accurate measurements.
For engineers familiar with measuring volts, current, power, and other standard, basic electrical parameters, the world of optical measurements can come as a big challenge with many new parameters: issues such as color, chromaticity, optical power, dispersion, solid angle, and more are the key concerns. For example, it is relatively easy to measure the efficiency of a power supply, but how do you measure the efficiency of an LED? What do you even mean by that question: efficiency in total output, visible output, or for a specific color output? In which direction, and over what solid angle? This article has provided an overview of basic optical parameters related to lighting, for the electrical engineer who does not have substantial experience with, or exposure to, optical measurement.
- CIE chart
- Integrating sphere video
- Luminous efficacy