4. Energy

Issues

For many applications, especially where directional light sources are required, such as signals, displays, and accent or task lighting, the potential for LEDs appears to be promising. Whereas in the 1980s and before, LEDs had relatively low luminous output and were suitable only for applications such as indicator lights, in the 1990s, prototype high-brightness LEDs using new light-generating materials such as AlInGaP and InGaN (Maranowski et al., 1997; Nakamura, 1997) are being manufactured; they have luminous efficacies exceeding those of incandescent lamps and rivaling mercury or fluorescent lamp technologies. LEDs that are currently available on the market, however, are only about half as efficacious as these prototypes, but as the techniques for developing the substrate materials improve, so will the resulting efficacies of LEDs (Kish and Fletcher, 1997).

Luminous efficacy may not be the most useful metric of efficiency for LEDs in many applications, particularly those in which the directionality of illumination is important. In these cases, metrics which consider the directionality of the light source, such as luminance or luminous intensity per watt, what might be properly termed "system efficacy," might more effectively characterize LEDs than lumens per watt, the conventional measure of luminous efficacy.

Regardless of the metric used to measure them, however, LEDs remain relatively low-power solid state devices (IESNA, 1993). For example, one red LED traffic signal head that contains 196 individual LEDs operates on a 120 V circuit, and has a wattage of less than 10 W, in comparison with a 150 W incandescent lamp it is intended to replace. Published estimates of the energy savings that can be achieved due to the replacement of a red signal head with an LED unit range from 82% to 93% (Vargas, 1994; Delean, 1996; Snel, 1996; Haussler, 1997; Stahl, 1997).

Because the red signal head is on an estimated 50% of the time throughout the day for a typical traffic cycle (Suozzo, 1998), estimates of the total traffic signal energy that could be saved by installing red LED signal heads alone range from 35% to 40% (Pollack, 1996; Deemer, 1997). In order to determine the added capital cost of purchasing LED traffic signal heads for installation, several cities and U.S. states have estimated the "payback" period for LED traffic signal heads - the time required to recoup capital costs through savings (ignoring the time effect on monetary values due to inflation). These times can vary widely. Payback periods of 1 to 1.5 years (Delean, 1996), 1.5 to 3 years (Lundberg, 1997b), 4.5 years (Haussler, 1997), and 6 to 7 years (Vargas, 1994) have all been reported. The actual payback period will be a function of the utility energy cost, the actual cost of the unit, and possible financial incentives offered by utility or government organizations.

Potential consequences

With various possible ways to characterize the energy efficiency of LED products, the actual efficiency of these sources relative to others could be under- or over-estimated. The reduced energy use of LED traffic signals, for example, is only a benefit if the resulting signals provide equal (or superior) visibility to conventional signals. On the other hand, if luminous efficacy in lumens per watt is not a relevant measure of a signal's visibility, it does not make sense to compare different light sources on a lumen per watt basis.

Options and recommendations

The continued widespread use of LEDs in applications such as exit signs, and traffic, automotive, and railroad signals appears to warrant agreement on measures to objectively compare light sources used for these applications. Previously, only incandescent lamps had the optical and luminous characteristics required to create visible signals. Now, the advent of LED technology, and its recent gains in popularity, as well as advances in other technologies such as compact high-intensity discharge lamps, fiber optics, and even improved incandescent lamps makes energy comparisons more difficult.

A theoretical framework for the concept of system efficacy, in some way measuring the lumens per watt per area produced by a system in a particular direction, should be developed. With such a framework in place, tests of LED products and equivalent products using other technologies can eventually be conducted. The results of such tests will in turn permit cost/payback calculations that are more realistic than currently possible.

TABLE OF CONTENTS

1. INTRODUCTION
2. TRAFFIC SIGNALS
3. CODES AND SPECIFICATIONS
4. ENERGY
5. COST
6. VISIBILITY
7. OPERATION: POWER AND ENVIRONMENT
8. MARKET ISSUES: SUPPLY AND DEMAND
9. OTHER APPLICATIONS
10. PRELIMINARY CONCLUSIONS AND RECOMMENDATIONS
11. REFERENCES