Lighting design begins with identifying the lighting needs for the space and then deciding what surfaces you want to light and at what intensities. Once you make these decisions, you can begin selecting equipment.

A typical lighting system includes the light source; ballast or driver, if the source needs one; lighting fixture; and controls. The light source is at the heart of this system. All other components are designed to support it. The ballast starts the lamp and regulates its operation. The fixture directs and controls the intensity of its light output. And the control devices turn it on/off or allow its intensity to be raised/lowered (dimming).

There are thousands of choices in four major families named for how they produce light: incandescent/halogen, fluorescent, high-intensity discharge (HID), and solid-state lighting (SSL), which includes light-emitting diodes (LEDs). Light is produced by passing a current through an element (incandescence), gas (fluorescence), plasma (HID) or chemical solid (SSL). These light sources are in constant competition. Each source type’s unique characteristics will make it particularly competitive for certain types of applications.

Obviously, picking the right light source is vital to the success of a lighting project. As electrical contractors are often in a position to make this choice, your understanding of key metrics is essential. To evaluate light sources, we must answer questions about each source based on specific criteria:

• What is the distribution of the light? Distribution is -measured in candelas (cd).
• How long does the lamp last? Service life is measured in hours.
• How much light does it produce? Light output is -measured in lumens (Lm).
• How much electric power does the system require? Power is measured in watts (W).
• How efficient is it compared to others? Efficacy is measured in lumens per watt (LPW).
• What is the color appearance of the source? Color tone is measured in kelvins (K).
• How well does the source render colors? Color rendering is expressed on the color rendering index (CRI).

Of course, there are many other characteristics of light sources to consider, but the above list establishes a good basis from which to design.

Distribution
Knowing what distribution you desire is important when starting out. Most light sources have a particular standardized size and shape that affects how objects and surfaces will be illuminated.
Point sources are small lamps, often featuring a clear outer glass bulb revealing the arc tube or bare incandescent filament, used to produce dramatic highlights and pronounced shadows through contrast between light and dark (see Figure 1).

Linear sources, such as linear fluorescent lamps, emit diffuse output from the surface of the lamp, softening shadows (see Figure 2).

Area sources are not lamps but instead large surfaces that emit diffuse light, such as ceilings reflecting illumination from an indirect lighting fixture or the surface of a luminous bowl pendant (see Figure 3).

Service life
Lamp service life is expressed in operating hours, or “burn time,” as rated by its manufacturer. You can find a given lamp’s rated life in its manufacturer’s catalog.

Rated life provides an expectation but not a precise prediction of actual life achieved in the field. The manufacturer rating is based on a standardized test method while field conditions influence actual life. For example, high line voltage and connecting the lamp to the wrong type of ballast can shorten lamp life.

Fluorescent and HID rated lamp life is an average that is predictable for a large lamp population. At 100 percent of rated life, 50 percent of a large lamp group can be expected to have failed. The rate of failure is shown on the lamp’s mortality curve (see Figure 4). Lamp life is based on hours per start, with a cycle being on for a period of time and then off for 15–20 minutes. Fluorescent lamp life is typically rated at three hours per start and at 10 hours per start for HID. The shorter the cycle, the shorter the lamp life.

Some lamps are built to last. The latest generation of “extended life” fluorescent lamps, for example, can achieve a rated life of up to 55,000 hours at 12 hours per start. The selection of ballasting and controls also can affect lamp life. Instant-start ballast operation typically produces shorter life while programmed-start operation maximizes it, with the latter being particularly useful in applications where the lamps may be frequently switched, such as installations with occupancy sensors.

A critical question to ask related to service life is what is the source’s failure mode. Aside from burnouts, another failure mode is lumen depreciation, used to determine LED and induction lamp life. For these sources, rated life is determined at the point in time at which light output is expected to decline below an acceptable limit—typically 70 percent of initial rated light output for LED products, called the L70 rating.
Other failure modes include unacceptable color shift at end of life, lamp efficacy falling to a point where the lamp is no longer economical to continue operating, lamp cycling (high-pressure sodium lamps) or other instabilities.

Light output
The initial light output of a given lamp can be found in its manufacturer’s catalog. Fluorescent and HID lamps are operated on ballasts, which may affect light output through the application of ballast factor. Fluorescent electronic ballasts for 4-foot T8 lamps provide a choice of low (0.74–0.78, typically 0.77 or 0.78); standard (0.85–0.90, typically 0.87 or 0.88); and high (>1.0, typically 1.15–1.18) ballast factors.
As the lamp ages and nears end of life, however, it produces less light—a process called lamp lumen depreciation (LLD). Various lamps suffer different rates of LLD expressed on their lumen maintenance curve, which plots light output versus time. LLD must be factored into lighting design calculations to ensure that the brand new lighting system continues to produce target light levels in the future.

Input watts
Lamp input wattage is rated in manufacturer’s catalogs, expressing the amount of power the lamp needs to operate at any given instant of time. If the lamp is fluorescent or HID, however, the rating is nominal since it requires a ballast to operate. As a result, the rated system wattage for the given lamp-ballast combination, which can be found in the ballast manufacturer’s catalog, offers the greater practical value.

Relative efficiency
The relative efficiency of lamp-ballast system combinations can be compared using a simple metric called efficacy (see Figure 5). Efficacy, analogous to vehicle miles-per-gallon ratings, expresses the ratio of light output per unit of electrical input, or LPW.

Efficacy declines over time as light output is affected by LLD over time while wattage stays the same. As a result, maintained efficacy has more practical value in decision-making compared to initial efficacy. Similarly, note that any field condition that affects light output but not wattage, such as ambient temperature, will affect efficacy. Also, note that if the lamp is to be dimmed, efficacy may gradually decline over the lamp’s dimming range. LEDs are an exception because they maintain efficacy until the low end of the dimming range where efficacy actually increases.

Note that efficacy is only truly useful when comparing light sources with similar characteristics and ability to achieve a design target. It also should not be overemphasized in decision-making because it does not account for the overall product effectiveness.

Color quality
Lamps produce light that has a color appearance, which can affect the color of objects when the light strikes them, and this affects the color appearance of the lamp itself.

The color temperature of a light source indicates the color appearance of the light source itself and the light it emits. Light sources are generally classified as “cool” (>4,000 K), which appear bluish-white; “neutral” (3,000 K–4,000 K), which appear white; or “warm” (

Color temperature describes the “whiteness,” “bluishness,” etc., of a light source, e.g., its warmth or coolness. However, it does not define how naturally colored objects will appear when illuminated by the source. Two light sources can have the same color temperature but render colors differently. The CRI, a scale with a maximum rating of 100, offers a separate metric to address this.

In many applications, the higher the CRI, the better, with 80–100 being optimal for rendering colors more “naturally”—that is, how most people would expect them to appear. A CRI of 90-plus is recommended for color-critical applications. For a valid comparison of CRI ratings of two light sources, however, they must have the same color temperature.

While these metrics can be useful, the most accurate picture of the color characteristics of a light source is its spectral power distribution curve, available from the manufacturer, which can provide an important reference.

Light, then light source
By answering these basic questions about a light source, the designer of a lighting system gains tools to use to control and predict how people, objects and surfaces appear in a space lighted by that source within a given lighting fixture and how efficiently illuminated can be delivered to its purpose.

Remember, however, that light is more important than the devices that deliver it. We must determine our lighting needs first, then work backward to what devices will satisfy these needs. Good design starts with goals, not equipment, and those goals should start with satisfying human needs, not energy efficiency.


DILOUIE, L.C., a lighting industry journalist, analyst and marketing consultant, is principal of ZING Communications. He can be reached at www.zinginc.com.