7# 4' ffiikkm/m/m/m?:my:mm`nxm/n n~*~i nNn'u8| P~B~p ~(~~~~This file contains Appendix A (Lighting Technology Characteristics) from Vorsatz, Diana, Leslie Shown, Jonathan G. Koomey, Mithra Moezzi, Andrea Denver, and Barbara Atkinson. 1997. Lighting Market Sourcebook. Ernest Orlando Lawrence Berkeley National Laboratory. LBNL-39102. December. Contact Leslie Shown (LJShown@lbl.gov) with questions on this report, or go to the web page http://enduse.lbl.gov/Projects/LMS.html. LAMPS Defining The Technical Characteristics of Lamps Various types of lamps differ from one another in numerous ways such as how energy-efficient they are, what color of light they produce, whether or not the level of light they produce dims over time, and whether or not they can be dimmed by the user. Below, we define some of the important characteristics by which lamps are often assessed and compared. Lamp Wattage: Lamp wattage is a measure of the power input to a lamp, measured in watts (W). Efficacy: The energy efficiency of lighting is referred to as "efficacy". Efficacy is calculated by dividing the quantity of the light emitted by the lamp (in lumens) by the power input to the lamp (in watts):  Different lighting mechanisms have different efficacies. The theoretical maximum efficacy is 683 lumens/watt (lm/W) for a yellowish-green light. The efficacy of a pure white light with equal energy at every wavelength of the visible spectrum is about 220 lm/W, but a light that is white in appearance can have efficacies of over 350 lm/W. The lighting technologies available today have maximum efficacies over 100 lm/W for white lights, and up to 140 or 150 lm/W for yellow lights. Rated Lifetime: The average rated lamp life of a given lamp type is the number of operating hours after which only half of a large group of lamps are still operating; this definition allows for the lifetimes of individual lamps to vary significantly from the average (IES 1993). Commonly used incandescent lamps have relatively short rated lifetimes (750-3000 hrs); compact fluorescent lamps have rated lifetimes of about 10,000 hours; full-size fluorescents have rated lifetimes ranging from 12,000-20,000 hours; and general-use high-intensity discharge lamps have rated lifetimes ranging from 3500-29,000 hours. Color Temperature: A lamp's color temperature is a measure of the color appearance of the lamp's light, expressed in degrees Kelvin (K). Conceptually, color temperature is based on the fact that the emitted radiation spectrum of a blackbody radiator depends only on temperature. A given lamp's "correlated" color temperature is the temperature of the blackbody closest in temperature to the light source. "Warm" white light that appears yellowish or reddish in color is emitted by lamps with low color temperatures (3000 K and below). "Cool" white light appearing bluish in color is emitted by lamps with high color temperatures (4000 K and above). Table A.1 provides the approximate color temperature of common light sources. Color Rendering: Color rendering refers to the effect of a light source on the color appearance of objects in conscious or subconscious comparison with their color appearance under a reference or standard light source of the same correlated color temperature. The color rendering properties of a lamp are expressed in terms of a color rendering index (CRI), which has a value of up to one hundred. The higher a lamp's CRI (the closer to 100), the less a color shift occurs compared to the reference source. In general, lamps with CRIs of 70100 are considered to render color excellently, a CRI of 6075 is considered good, 5060 is considered fair, and less than 50 is considered poor (Ontario Hydro 1992). Some lamp types, such as low-pressure sodium, have CRIs of less than zero (Clear 1996). Table A.2 provides approximate CRIs for common light sources. Lumen Maintenance: Typically, lamps continue to draw approximately the same amount of power and yet produce fewer lumens as they age. A lamp's lumen maintenance refers to the extent to which the lamp maintains its lumen output, and therefore efficacy, over time. Dimmability: Whether or not a lamp is dimmable refers to the user's ability to vary the lumens that it emits. Lamp dimming is important for two reasons: aesthetic lighting effects and energy conservation. Incandescent lamps can be easily dimmed using a simple device to lower the voltage across the lamp filament. Fluorescent lamps can be dimmed using dimming ballasts; almost all dimming ballasts used today are electronic. An electronic dimming ballast alters the output power to a lamp by sending a low-voltage signal to the output circuit (Eley Associates 1993). Some high-intensity discharge (HID) lamps are dimmable with specialized ballasts. Table A.1. The Color Temperature of Common Light Sources Source of LightColor Temp. (K)DescriptionSky - extremely blue25,000coolSky - overcast6500coolSunlight at noon5000coolRare earth fluorescent2700-5000warm/coolCool-white fluorescent 4300coolMetal halide 3000-4200warm/coolWarm-white fluorescent 3000warmIncandescent (100 W)2900warmHigh-pressure sodium1900-2100warmCandle flame1800warmLow-pressure sodium1740warmSource: Ontario Hydro (1992) and lamp manufacturers catalogs for General Electric (1995), Osram Sylvania (1996), and Philips (1996) Table A.2. The Color Rendering Indexes of Common Light Sources Source of LightCRIColor RenderingTungsten-halogen99excellentStandard incandescent97excellentRare-earth fluorescent72-84good/excellentCompact Fluorescent82excellentMetal halide (400 W, clear)65goodCool-white fluorescent 62goodWarm-white fluorescent 52fairMercury vapor (phosphor-coated)45-50poorHigh-pressure sodium (400 W, diffuse-coated)22poorMercury Vapor (clear)15poorSource: Ontario Hydro (1992) and lamp manufacturer catalogs for General Electric (1995), Osram Sylvania (1996), and Philips (1996) Introduction to Incandescent, Fluorescent, and High-Intensity Discharge Lamps The primary categories of lamps that we address in this report include incandescent, fluorescent, and high-intensity discharge. We describe the basic operating principles of these different lamp types briefly below. Tables A.3-A.6 present primary physical characteristics for the lamp types discussed below. In addition to physical lamp characteristics, we also include lamp price in Tables A.3 through A.6. Lamp prices are sensitive to demand; lamps in higher demand tend to be less expensive. In general, commercial customers buy lighting products in large quantities and thus pay the prices at the lower end of the price ranges provided in this report; residential customers typically buy lamps one or two at a time and thus pay prices at the higher end of the range. Lamp prices also vary depending on where they are purchased - for example, lamps purchased from lighting design stores are likely to be more expensive than those purchased from a do-it-yourself store such as Home Depot. Incandescent Filament Lamps In the late 1800s, the incandescent lamp was invented independently by Thomas Edison in the United States and Joseph Swan in England (Atkinson et al. 1995). Today, incandescent lamps provide most of the light in households and are also used widely for lighting commercial buildings. Because about 90-95% of an incandescent lamp's emissions are in the infrared (thermal), rather than visible, range of the electromagnetic spectrum, incandescent lamps are much less efficacious than other lamp types. However, as discussed in Atkinson et al. (1995), aside from energy-efficiency, incandescent lamps have many advantages: Although the prevalence of incandescent lamps in the residential sector may be partially due to historical precedence and inertia, these lamps do have advantages that, to some extent, counterbalance their relatively poor efficacies: they have excellent CRIs and a warm color; they are easily dimmed, inexpensive, small, lightweight, and can be used with inexpensive fixtures; and, in a properly designed fixture, they permit excellent optical control...They are simple to install, maintain, and dispose of. General service and reflector/PAR (parabolic aluminized reflector) lamps are the most common types of incandescents. General service lamps (also called "A-lamps") are the pear-shaped light bulbs that are regularly used in households. Reflector lamps are typically used to highlight indoor retail displays and artwork and to illuminate outdoor areas. Modern incandescent lamps use filaments that are made of tungsten. When electricity is used to heat a lamp filament to the point of incandescence, light is produced. The efficacy of the light production depends on the filament temperature. The higher the filament temperature, the greater the portion of radiated energy that falls into the visible part of the irradiated spectrum. Consequently, when designing an incandescent filament lamp, it is important to keep the temperature of the filament as high as possible while still maintaining a satisfactory lamp life. See Table A.3 for more information on standard incandescent lamps. A tungsten-halogen lamp, which uses the halogen regenerative cycle, is a variation of an incandescent filament lamp. The tungsten-halogen bulb has a quartz envelope that is located close to the filament so that the envelope can reach temperatures of 260 C or more in normal operation. At this temperature, the halogen gas fill in the lamp reacts with any tungsten that evaporates from the filament and deposits on the lamp wall. The resulting gaseous tungsten-halogen compound circulates inside the bulb until it comes in contact with the incandescent filament. Here, sufficient heat breaks down the compound into tungsten and redeposits it on the filament. Tungsten-halogen lamps improve on regular incandescent sources because of their excellent lumen maintenance, long lifetime, and compactness. Although they are not as efficacious or long-lived as fluorescent or HID lamps, tungsten-halogens offer excellent color, brilliance, and control characteristics at a relatively low unit price (Eley Associates 1993). These lamps are most often combined with a reflector housing, and are available in a wide variety of tubular forms, and in spotlights and floodlights. The tungsten-halogen infrared-reflecting (HIR) lamp is even more efficacious than the standard tungsten-halogen lamp. As mentioned above, 90-95% of the energy radiated by incandescent lamps, including tungsten-halogen lamps, is in the form of heat In an HIR lamp, a multi-layer interference film-coating technology is applied to a tungsten-halogen lamp to reflect the emitted heat back to the filament; consequently, the required power input to reach the operating temperature for the tungsten-halogen cycle is reduced. HIR lamps have been available for a number of years as high-wattage double-ended quartz lamps, and HIR PAR lamps only recently (1994) became widely available (Atkinson et al. 1995). HIR lamps have been promoted to residential- and commercial-sector customers primarily as low-wattage reflector lamps; general service HIR lamps have been developed as prototypes but are not yet commercially available (Atkinson et al. 1995). See Table A.3 and Table A.4 for more information on different types of tungsten-halogen lamps. Reflector lamps are standard incandescent or tungsten-halogen lamps made in special or standard bulb shapes and with a reflective coating applied to part of the bulb surface. Both silver and aluminum coatings are used. In reflector lamps, better optical control directs the illuminance to a specific area; thus, reflector lamps can be energy-efficient alternatives to general service incandescents in applications where illumination requirements are direction-specific. In spotlight and floodlight applications, the filaments are concentrated and are accurately positioned with respect to the base. When the filament is placed at the focal point of a reflector or lens system, a precisely controlled beam is obtained. See Table A.4 for more information on reflector lamps. Fluorescent Lamps The first practical fluorescent lamps were produced in the United States in the late 1930s, and fluorescent lamps came into general use in the 1950s (Atkinson et al. 1995). Fluorescent lamps are used to illuminate most commercial buildings, and are also common in the industrial sector. Only a small amount of fluorescent lighting is found in homes, primarily in kitchens, bathrooms, and utility areas. The most common fluorescent lamps are tubular and have a length of four feet. Tubular lamps that have a diameter of 1.5 inches (38 mm) are called T12s and tubes that have a diameter of one inch (26 mm) are called T8s; the "8" and "12" refer to the number of eighths of an inch in the diameter of the lamp tube. Lamp tubes are available in other diameters as well. Like most discharge lamps, fluorescent lamps must be operated using a "ballast" to limit the current to the value for which each lamp is designed and provide the starting and operating lamp voltages. Typically, the ballast adds another 10-20% to the power draw, thus decreasing system efficacy. A fluorescent lamp system's efficacy depends on lamp length and diameter; the type of phosphor used to coat the lamp; the type of ballast used with the lamp; the number of lamps per ballast; the temperature of the lamp (which depends on the fixture and its environment); as well as a number of lesser factors (Atkinson et al. 1995). Ballasts are discussed in greater detail below. Technically, a fluorescent lamp is a low-pressure gas discharge source in which light is produced when UV energy generated by a mercury arc activates fluorescent powders that coat the inside of the lamp tube. Fluorescent lamps are usually long and tubular with an electrode sealed into each end; they contain mercury vapor at low pressure and a small amount of inert gas for starting. Standard fluorescent lamps are filled with argon gas. The interior of the bulb wall is coated with fluorescent powders that are usually referred to as 'phosphors'. When a suitably high voltage is applied across the electrodes, an electric arc discharge is initiated and the resulting current ionizes the vaporized mercury in the tube. The ionized mercury emits mostly invisible UV radiation, which strikes and excites the phosphor tube coating, causing a glow or 'fluorescence' and producing visible light. The blend of phosphors used to coat a fluorescent lamp's inner wall determines the color of light produced by the lamp. In the past, the most frequently used lamps have been the halophosphate ("standard phosphor") cool-white and warm-white lamps. In a newer type of fluorescent lamp, the inside of the lamp tube is coated with a combination of rare-earth (RE) phosphors that produce visible light at wavelengths to which the red, green, and blue retinal sensors of the human eye are most sensitive. Lamps using RE phosphors can withstand a higher loading (arc power per unit of phosphor area) and thus provide better lumen maintenance than standard-phosphor lamps. The arc power per unit of phosphor area increases as lamp diameter decreases, and lumen degradation in standard-phosphor lamps of small diameter is too severe to make lamp production practical. The introduction of RE phosphor coatings for lamp tubes, however, has made it possible to develop fluorescent lamps with smaller diameters such as the T-8 and T-5. All fluorescent lamps with diameters of one-inch or less use the new RE phosphors. Rare earth coatings can also be used for lamps of larger diameter. Although the use of RE phosphors increases the price of a lamp, RE phosphor lamps provide improved lumen maintenance, color rendering, and lamp efficacy. See Table A.5 for more information on T12 and T8 fluorescent lamps. Compact fluorescent lamps (CFLs), which are significantly smaller than standard fluorescent lamps, were introduced in the early 1980s as an energy-efficient alternative to incandescent lamps. As mentioned above, the introduction of RE phosphor coatings for fluorescent lamps made it possible to develop fluorescents with smaller diameters. In a CFL, the small-diameter tube (T4 or T5) is bent into two to six sections. Originally, CFLs were designed to be interchangeable with conventional 25100 W incandescent lamps, but they are now available in various sizes, colors, wattages, and bases. Typically, a CFL produces three to four times more lumens per watt than an incandescent A-lamp; efficiency increases with lamp wattage. In addition, the rated lifetime of a CFL is about 10 times longer than that of an incandescent A-lamp. However, factors such as ambient temperature, switching, mounting position, lumen depreciation, and fixture size may alter the laboratory-determined efficacies and lifetimes of CFLs. For example, a CFL operating in a base-down position may produce 15-20% fewer lumens than a CFL operating in a base-up position (Siminovitch and Mills 1994). There are three different types of compact fluorescent lamp-ballast systems (Eley Associates 1993): Integral systems are self-ballasted packages and are made up of a one-piece, disposable lamp, ballast, and socket adapter combination. Integral systems are designed to replace incandescent lamps in fixtures fitted for incandescents. A disadvantage of the integral system is that the ballast (which would otherwise have a life of 45,000 hours) must be disposed of when the lamp fails (normally, CFL lamp life is about 10,000 hours). Modular systems are self-ballasted packages as described for integral systems except that the lamp is replaceable. Like integral systems, modular systems are designed for incandescent retrofit situations, but are more cost-effective in the long run because the ballast does not need to be replaced every time a lamp fails. Dedicated (hardwired) systems are new or retrofitted fixtures that are hardwired for CFL ballasts. These systems do not use socket adaptors; instead, they use a pin socket for the lamp. See Table A.5 for more information on compact fluorescent lamps. High-Intensity Discharge Lamps HID lamps are most widely used in the commercial and industrial sectors and, for many commercial and industrial applications, provide the most-cost-effective illumination. Low-wattage HID lamps can be used effectively for outdoor security, corridor, and landscape lighting in the residential sector, particularly in timer-controlled functions. Like fluorescent lamps, HID lamps produce light by discharging a well-stabilized arc discharge through a mixture of gases in a refractory envelope. Unlike fluorescent lamps, HID lamps use a compact "arc tube" in which the pressure and temperature are very high. Because the arc tube is small, it permits compact reflector designs with good optical controllability. Like fluorescent lamps, HID lamps require a ballast to supply the correct voltage and control the current. The three primary types of HID lamps in use today are mercury vapor (MV), metal halide (MH), and high-pressure sodium (HPS). Mercury vapor lamps were the first HID lamps to be developed. In MV lamps, light is produced by the passage of an electric current through an arc tube filled with mercury vapor; a small amount of argon is added to facilitate starting IES (1993). Metal halide lamps are similar in construction to MV lamps, and produce light by passing a current through an arc tube containing various metallic halides in addition to mercury and argon (IES 1993); compared to MV lamps, metal halide lamps have shorter rated lamp lives, but offer improved efficacy, color rendering, and lumen maintenance In high-pressure sodium lamps, light is produced by the passage of current through an arc tube containing sodium vapor (IES 1993); HPS lamps are even more efficacious and have better lumen maintenance than MH lamps. HID lamps are most effectively used for applications in which switching (turning lamps off and on) is limited. One reason for this is the amount of time they require for starting (cold start) and restriking (hot start). A mercury vapor lamp, once started, requires several minutes to achieve full light output; restrike time (cooling time required before the lamp will restart) is 3-7 minutes, depending on lamp type (IES 1993). A metal halide lamp, once started, requires about 2-10 minutes to achieve full light output and equilibrium color, depending on the lamp type; restrike time can be as long as 15 minutes because of the high operating temperature (IES 1993). A high-pressure sodium lamp, once started, requires about 10 minutes to achieve full light output, during which time the color of the light changes; restrike time is less than a minute and full warm-up takes 3-4 minutes (IES 1993). For industrial and outdoor applications where color was not a priority, MV lamps were the most efficient lamp type for many years. With the introduction of higher-efficacy HPS and MH lamps, MV lamps are now the least efficacious of the three primary HID lamp types and have lost a significant portion of their original market share. However, although many building owners have now replaced MV lamps with more efficient MH and HPS lamps, many MV lamps are still in use because they are relatively inexpensive and conversion to MH and HPS often requires installation of new ballasts and sometimes requires installation of new fixtures (Atkinson et al. 1995, Clear 1997a). See Table A.6 for additional information on HID lamps. Lamp table - page 1 Lamp table - page 2 Lamp table - page 3 Lamp table - page 4 Lamp table - page 5 Lamp table - page 6 Lamp table - page 7 Lamp table - page 8 Lamp table - page 9 BALLASTS Defining The Technical Characteristics of Ballasts All discharge lamps must be operated with a current-limiting device referred to as a "ballast". A lamp ballast is an electrical device that controls the current provided to the lamp and provides the high voltage necessary to start most discharge lamps. In addition, ballasts can provide power quality correction and control features such as dimming or compensation for lumen depreciation. Ballasts differ from one another in numerous ways such as how energy-efficient they are, how much distortion they cause in a power wave, and how much light a lamp produces when using them. Below, we define the primary physical characteristics by which ballasts are most often assessed and compared. Ballast Factor (BF): The ballast factor provides a relative measure of how much light is produced using a specific ballast. A meaningful comparison can only be made between ballasts that are used to operate the same type of lamps.  For most ballasts, the BF is less than one; for some of the new electronic ballasts, however, the BF is greater than one (Koomey et al. 1994) . Ballast Efficacy Factor (BEF): The ballast efficacy factor is used to determine which ballast supplies more light for a given wattage. . As with BF, BEF can be used to meaningfully compare different ballasts only when they operate the same number and type of lamps. System Efficacy: Like lamps, ballasts consume power. Consequently, the only meaningful measure of the efficiency of a lighting system is the efficacy of the lamp-ballast system. Typically, a fluorescent ballast consumes from a few to a dozen watts. HID ballasts consume from 1020% of nominal lamp watts; this percentage is usually higher for lower-wattage lamps. System efficacy refers to the efficacy of the lamp-ballast combination, and is calculated as follows:  Power Factor (PF) Ratio: The power factor ratio represents, for a given ballast, the amount of power that a customer is actually using as a fraction of what the utility must supply. This ratio is used to determine how efficiently a ballast uses total input power. To calculate the PF ratio, the power (watts) is divided by the root mean square of the ballast volt-amps (Eley Associates 1993). Utilities may penalize customers whose electric load has a low PF. Ideally, lighting equipment should have a PF greater than 0.9 and as close to 1.0 as possible. PFs of less than 1.0 occur when the voltage and current are out of phase or when the sinusoidal shape is distorted. Total Harmonic Distortion (THD): Ballasts, especially electronic ones, affect power quality by generating harmonic distortion. Total harmonic distortion refers to the amount of distortion that a ballast causes in the power wave form. Utilities often require a THD of less than 20%, but the electric industry is considering a standard that permits a THD up to 32% (Audin et al. 1994). High THD can disrupt powerline carrier controls and create unacceptably high currents in three-phase systems. THD values are typically calculated for ballasts based on their operation of a 4-foot fluorescent lamp. Introduction to Ballasts for Fluorescent and High-Intensity Discharge Lamps The light output of a lamp depends on the ballast that operates the lamp. Ballasts are often designed to operate a unique lamp type; some ballasts, however, can be used to operate more than one type of lamp. It is important to use the ballast specified by the manufacturers because improper lamp-ballast combinations can result in reduced light output, efficacy, and lifetime. For all types of ballasts, rated lifetimes are in the range of 45,000 hours. Ballast life is rated for 12 hours of use per day. Ballast life is very dependent on operating temperature an increase of 10C over the rated ballast operating temperature of 90C can translate into as much as a 50% reduction in ballast life (National Lighting Product Information Program 1994). Because manufacturers who specify longer-than-usual ballast lifetimes may also prescribe lower operating temperatures, it is advisable to check the ballast specifications for the temperature to which a specific ballast lifetime corresponds. Typically, fluorescent lamps are operated using magnetic core-coil or electronic high-frequency ballasts; both magnetic and electronic ballasts are available for most types of fluorescent lamps. Hybrid ballasts are also available for rapid-start lamps. Typically, HID lamps cannot be operated using fluorescent ballasts. The three primary types of HID ballasts are magnetic ballasts: reactor ballasts, high-reactance autotransformers, and constant-wattage autotransformers (Audin et al. 1994). We briefly describe these different ballast types below. Ballasts for Fluorescent Lamps Magnetic core-coil ballasts use a capacitor and a transformer with a magnetic core coiled in copper or aluminum wire in order to control the current provided to a lamp. A thermal cutoff switch protects the ballast from overheating. Magnetic ballasts operate at an input frequency of 60 hertz (Hz) and also operate lamps at 60 Hz. Electronic ballasts use integrated electronic circuitry rather than magnetic components to control voltage and current. Like magnetic ballasts, electronic ballasts use standard 60 Hz power; however, electronic ballasts operate lamps at a much higher frequency (20,00060,000 Hz), which increases lamp efficacy. Lamp efficacy is also improved because electronic ballasts are less sensitive to ambient (room) temperature than magnetic ballasts. Although dimming magnetic ballasts are available, almost all dimming fluorescent ballasts are electronic (IES 1993). A third type of fluorescent ballast is the hybrid ballast, which is also referred to as a cathode cut-out or heater cut-out ballast. In the hybrid ballast, which is a modified version of the magnetic ballast and operates at a low frequency, electronic circuitry is used to control power to the lamp's cathodes and magnetic components drive the main arc (Audin et al. 1994). Energy consumption is reduced because the electronic circuitry removes the power that is used to heat the lamp filaments once the lamp has started. Typically, hybrid ballasts use 510% less energy than energy-efficient magnetic ballasts; however, they can only be used with rapid-start lamps and are not dimmable. Hybrid ballasts account for only a small share of the ballast market. Table A.7 provides a comparison of magnetic, hybrid, and electronic ballasts. Ballasts for HID Lamps Typically, HID lamps cannot be operated using fluorescent ballasts. The three primary types of HID ballasts are magnetic ballasts: reactor ballasts, high-reactance autotransformers, and constant-wattage autotransformers (Audin et al. 1994): Reactor ballasts consist mainly of an inductor coil. They are small, inexpensive, simple, and have low losses; however, their use leads to more rapid lumen depreciation than the use of other ballast types. Reactor ballasts have a low power factor, and can cause flicker or turn-off if power is unstable. High-reactance autotransformers are more expensive and consume more power than reactor ballasts, but also have a more sophisticated design. Although they are similar to reactor ballasts, they are capable of boosting line voltage when it is insufficient to start a lamp. Constant-wattage autotransformers are the most expensive of these three ballast types, but are also the most commonly used. Of these ballast types, constant-wattage autotransformers regulate power the best and their use thus reduces flicker and shutoffs when power is unstable. There are several disadvantages associated with magnetic HID ballasts, such as high internal losses (an especially high percentage in the case of low-wattage lamps), audible noise, and bulkiness. Electronic ballasts are now available for some low-wattage MH and HPS lamps, but they are uncommon. Electronic ballasts for HID lamps do not operate on the same principles as those for fluorescent lamps. The primary benefits of an electronic HID ballast are reduced size and weight, quieter operation, and increased control of the arc tube wattage during the lamp's life. More precise arc tube wattage management improves the color consistency over the lamp life, and can lengthen expected life. Unlike electronic fluorescent ballasts, with few exceptions, electronic HID ballasts do not significantly improve lamp efficacy. Ballast table - page 1 Ballast table - page 2 FIXTURES A lighting fixture provides physical support for lamp(s), ballast(s), and wiring. The function of the fixture is to efficiently direct and distribute light to the desired area without causing glare or discomfort. The geometric design of a fixture, as well as the material of which the reflector and/or lens is made, determines how the light of a lamp is distributed as well as the overall efficiency of the lighting system. We use the term "fixture" to refer to the physical housing for a lamp, including: sockets, lamp holders, and fittings to attach the lamp to the fixture; reflectors to direct light in the desired direction; shielding and diffusion components (such as lenses, diffusers, and louvers) to shield the light from non-desired directions, reduce visual discomfort, prevent glare, and distribute light evenly; and, for certain types of lamps, ballasts to start lamp and control electric characteristics during lamp operation. An efficient fixture optimizes the system performance of each of its components. If installed in the wrong fixture, even the most efficacious lamp can be inefficient and provide light of poor quality. There are more fixtures on the market today than any other type of lighting equipment. Consequently, it is important that fixtures be selected carefully, based on factors including the user's specific lighting needs, lamp requirements, and environmental conditions. Various types of fixtures differ from one another in numerous ways such as reflector design, operating position of the lamp, ease of lamp insertion and removal, thermal characteristics, and fixture life time. Fixtures also differ from one another in terms of their energy use and light distribution characteristics. The performance of a fixture is assessed by evaluating its performance as part of a "luminaire"; in this appendix, the term "luminaire" refers to a complete lighting system including lamp(s), ballast(s), and fixture. Below, we define some of the energy-efficiency and light distribution characteristics by which fixture performance can be assessed and compared. Luminaire Efficiency: Luminaire efficiency is the ratio of the lumens leaving a luminaire to the total number of lumens produced by a lamp (IES 1993). Luminaire Efficacy Rating (LER): As mentioned in the main body of this report, in response to EPAct's call for a voluntary national testing and information program for luminaires, a program has been created by the National Lighting Collaborative (1996). Members of the Collaborative include the National Electrical Manufacturers Association, the American Lighting Association, and other interested parties. The working group has introduced a new tool for comparing luminaires, the LER, which is based on NEMA's LE5 standard for fluorescent luminaires. The LER is a single number expressing luminaire efficacy in lumens per watt, and is calculated using the following equation: LER = F(luminaire efficiency * total rated lamp lumens * ballast factor,luminaire input watts) Coefficient of Utilization (CU): The coefficient of utilization expresses the ratio of the lumens from a luminaire that are received on a room's workplane to the total number of lumens produced by the lamps within the luminaire (IES 1993). LIGHTING CONTROLS A large variety of technologies are available for controlling the way that lights are used in a building. These technologies can be mechanical and/or electronic and range from a basic timer that turns the lights off or on at a given hour of the day to a complex energy management system (EMS) that controls not only the lighting in a building but also the space conditioning system. Both simple and highly complex lighting control systems are used in commercial buildings. In homes, lighting control systems tend to be simple; however, the control systems installed in the recently introduced smart houses are quite complicated. Common lighting control strategies and tools are summarized in Table A.8 and Table A.9, respectively. The choice of a lighting control strategy and tools, which can be a combination of the options described in Tables A.8 and A.9, depends on numerous factors including the type of lamp one wishes to control. Not all lighting controls are appropriate for all lamp types. For example, HID and fluorescent lamps may not be ideal for applications where a motion sensor frequently switches the lights on and off, because the lifetime of these lamps is very sensitive to frequent switching. In addition, HID lamps may take too long to start up. HID and fluorescent lamps are the ideal choice in time-controlled applications where relatively long burning cycles are needed. The type of controls one selects will also depend on whether they are being installed as a retrofit, renovation, or for new construction. As described in Koomey et al. (1994), the electrical wiring configuration is the major constraint in installing controls in buildings. Most often, it is not cost-effective to substantially re-wire the ceiling electric lighting system in an existing building in order to install lighting controls. Consequently, lighting control systems for retrofits in existing buildings tend to be simpler than the lighting control systems designed for new buildings. In new construction, it can be cost-effective to install more advanced lighting control systems. As described in Atkinson et al. (1995), integrated workstation sensors and energy management systems are two highly promising efficiency options: Of the control systems available today, integrated workstation sensors and energy management systems are two of the most promising efficiency options. An integrated workstation sensor allows users to control lighting, electric heating and cooling equipment, and other electrical equipment (such as plug loads) for individual workstations or spaces. For example, user lighting controls might include dimmer switches for area and task lighting as well as daylight sensors. From their workspace, users can adjust lighting and HVAC controls according to their preference. Occupancy sensors automatically shut down electrical equipment when the space is unoccupied, and system memory allows the equipment to come back on at the same level when the occupant returns. Comprehensive, automated, building energy management systems are user-programmable and can control equipment for several energy end-uses including lighting, HVAC, security, and safety systems. A well-designed energy management system may offer greater energy savings than individual controls for single end uses; the "systems approach" is becoming more common in both new construction and retrofits of existing buildings. For a clear and practical guide to the strategies and tools used in designing lighting control systems for commercial buildings, see Rundquist et al. (1996). Table A.8. Common Lighting Control Strategies Scheduling Scheduling is a lighting control strategy based on turning lamps off and on according to the need for illumination. Predictable scheduling regulates illumination levels in a predetermined way, with the use of equipment such as timed controls, and can be effective for buildings in which activities follow a similar routine from day to day. Unpredictable scheduling controls lighting levels based on whether or not someone is present; for example, occupancy/motion sensors extinguish or dim the lights when a space is unoccupied and turn the lights back on when someone enters the space. Task-Tuning Normally, spaces are illuminated uniformly. Using a task-tuning control strategy and dimming devices, however, the lighting levels of different spaces can be adjusted to meet the needs of the different people using those spaces. For example, workers performing visually detailed tasks are likely to require more illumination than workers who are primarily looking at their computer monitors throughout the day. Additionally, lighting levels can be reduced in spaces that are not oriented towards visual tasks (e.g., hallways and reception areas).DaylightingIn many buildings, the daylight coming in through windows and skylights can provide a significant amount of the light necessary for many visual tasks. After decades of overdependence on artificial light, many lighting designers are once again thinking in terms of using sunlight to illuminate interior spaces. For designers, the challenge of daylighting is to admit only the required amount of daylight, distribute the light evenly, and avoid glare. When daylight is used within a building, the electric lighting levels can be reduced. Using a control photocell, a dimmable lighting system can be connected to the ambient light levels within a room; in this way, electric light levels can be reduced during the times when natural light is available and supply most or all of the light needed when natural light levels are low and when it is dark outside.Lumen Maintenance Typically, electric lighting systems are designed to produce light levels that are 20-35% higher than the design minimum so that, as lamps age and the amount of light delivered by the lamp-luminaire system diminishes, the illuminance level will always meet or exceed the minimum light requirement. Light losses over time are the result of lamp lumen depreciation as well as the accumulation of dirt on the luminaire and room surfaces. Lumen maintenance is a control strategy that uses photocells and dimmers to sense the actual illuminance level in a space and reduce system power input to maintain only the desired light level. In this way, a lighting system can be designed with lower initial lighting power densities and design-specified illuminance levels are maintained at all times, rather than only at the end of the maintenance cycle. Load SheddingIn order to avoid brownouts and blackouts, many utilities charge their larger customers based on peak power demand. Selective reduction of lighting levels in less critical areas of a building is an effective way of reducing lighting power demand for short periods of time. Typically, lighting levels can be reduced by 10% or more with only minimal impact on the occupant's visual performance or productivity. Automatic dimming controls allow the reduction in light level to occur without occupant awareness.Adaptation CompensationIn places that are illuminated both during the day and throughout the night (e.g., 24-hour supermarkets or entry foyers), the level of electric lighting needs to be higher during the daytime because a person whose eyes are adapted to daylight will need more light to see in areas that are less bright. When a person's eyes are adapted to the lack of light at night, however, they do not require as much light to see indoors. An adaptation compensation control strategy uses dimming devices or switching relays in combination with automatic timers to vary the lighting level accordingly. Sources: IES (1993), Eley Associates (1993) Table A.9. Common Lighting Control Tools* Programmable TimersProgrammable timers are used to implement time-based control of electric lights. The usual method of implementation is a system of low-voltage controlled relays that are controlled by a programmable time clock. These systems are primarily used to efficiently schedule the operation of a lighting system in areas where the occupant schedule is relatively predictable. To accommodate lighting needs during off-hours, these systems are typically equipped with overrides so that building occupants can control the lights using a low-voltage switch or a telephone override system.Occupancy Sensors Occupancy sensors are switches that are activated by detecting the presence or absence of people in the sensor's field of view. There are two basic types of occupant sensor: passive infrared sensing and ultrasonic (some sensors combine these two methods). These sensors are most effective in locations where occupancy is not easily predicted (e.g., conference rooms, restrooms, and storerooms).Photo-Switches Photo-switches are photo-electrically controlled switches that can be used to switch off lights in building zones receiving daylight from adjacent windows. These devices are usually installed in one of three ways: on each fixture; on groups of fixtures using intermediate relays; or as inputs to low-voltage programmable relay systems. Dynamic Controls Dynamic controls are devices that allow standard lighting equipment (including both fluorescent and HID sources) to be continuously dimmed to an intermediate level. These systems can control a single lamp or entire branch circuits. Although these controls can typically provide any light level within the control range, they rarely permit dimming below 40% of maximum. They generally accept an input from a photocell and/or an input from an energy management system.Static Controls Static controls are devices that allow the light output of standard lighting equipment to be reduced to one intermediate level. These systems can control a single lamp or entire branch circuits. The larger systems generally accept an input from an EMS system for scheduling control. The smaller systems generally control only a single lamp or ballast - their sole function is to reduce input power (and light output). The primary application of these systems is in areas that are overlit.Dimmable Ballasts With the use of dimmable ballasts, fluorescent lamps can be dimmed over a wide range, and represent the state-of-the-art in controllable lighting. Although dimming magnetic ballasts are also available, almost all dimming ballasts in use today are electronic (Clear and Rubinstein 1997). Typically, electronic ballasts can be controlled using a low-voltage wiring network that allows them to respond to inputs from a photocell, occupancy sensor, or input from an energy management system.* Except where otherwise noted, the descriptions of these lighting control tools were obtained from Koomey et al. (1994). Appendix A references Atkinson, Barbara, Andrea Denver, James E. McMahon, Leslie Shown, and Robert Clear. 1995. "Energy-Efficient Lighting Technologies and Their Applications in the Commercial and Residential Sectors" in the CRC Handbook of Energy Efficiency. Boca Raton, FL: CRC Press, pp. 399-427. Audin, L., D. Houghton, M. Shephard, and W. Hawthorne. 1994. Lighting Technology Atlas. Boulder, Colorado: E Source, Inc. Brown, Richard, and Barbara Atkinson. 1994. Incandescent Reflector Lamp Energy Efficiency Standard Analysis: Draft Report. Berkeley, CA: Lawrence Berkeley National Laboratory. Draft LBL-36223. September. Calwell, Chris. 1996. Halogen Torchieres: Cold Facts and Hot Ceilings. Boulder, CO: E-Source. September. Calwell, Chris. 1997. Personal Communication: Telephone conversation with Jonathan Koomey regarding review of April 1997 DRAFT Sourcebook. May. Census Bureau. 1994. Current Industrial Reports: Electric Lamps - MQ36B (93)-5, Summary 1993. Washington, D.C.: U.S. Department of Commerce. November. Clear, Robert (Lawrence Berkeley National Laboratory). 1994. Personal Communication: Memo to co-workers entitled "Summary Statement on Economics of IR-Halogen Lamps." May. Clear, Robert (Lawrence Berkeley National Laboratory). 1996. Personal Communication: E-mail to Jonathan Koomey regarding review of October 1996 DRAFT Sourcebook. October 21. Clear, Robert (Lawrence Berkeley National Laboratory). 1997a. Personal Communication: Written comments provided to Leslie Shown regarding review of April 3, 1997 DRAFT Sourcebook. April. Clear, Robert (Lawrence Berkeley National Laboratory). 1997b. Personal Communication: E-mail to Leslie Shown regarding technical characteristics of an HIR A-lamp. March 17. Clear, Robert, and Francis Rubinstein (Lawrence Berkeley National Laboratory). 1996. Evaluating the Relative Efficacy of Incandescent Lamp Technologies. Unpublished report. Clear, Robert, and Francis Rubinstein (Lawrence Berkeley National Laboratory). 1997. Personal Communication: Conversation with Leslie Shown regarding the dimmability of fluorescent lamps. March 10. Denver, Andrea (Lawrence Berkeley National Laboratory). 1996. Personal Communication: Conversation with Leslie Shown regarding lamp prices (based on catalogs and survey of shelf prices at Bay Area stores). Electric Power Research Institute (EPRI). 1993a. Compact Fluorescent Lamps: High-Efficiency Electric Technology Fact Sheet. Palo Alto, CA: EPRI. Electric Power Research Institute (EPRI). 1993b. Electronic Ballasts: High-Efficiency Electric Technology Fact Sheet. Palo Alto, CA: EPRI. Eley Associates. 1993. Advanced Lighting Guidelines: 1993. San Francisco, CA: Eley Associates. Prepared by Eley Associates, Luminae Souter Lighting Design, and Lawrence Berkeley National Laboratory for the U.S. Department of Energy, California Energy Commission, and Electric Power Research Institute. DOE/EE-0008. General Electric Company. 1995. GE Lighting: Spectrum 9200 Lamp Catalog. U.S.: General Electric Company. Illuminating Engineering Society of North America (IES). 1993.Lighting Handbook: Reference & Application . 8th Edition. New York, NY: IES. Koomey, J.G., A.H. Sanstad, and L.J. Shown. 1995. Magnetic Fluorescent Ballasts: Market Data, Market Imperfections, and Policy Success. Berkeley, CA: Lawrence Berkeley National Laboratory. LBL-37702. December. Koomey, Jonathan G., Francis X. Johnson, Jennifer Schuman, Ellen Franconi, Steve Greenberg, Jim D. Lutz, Brent T. Griffith, Dariush Aresteh, Celina Atkinson, Kristin Heinemeier, Y. Joe Huang, Lynn Price, Greg Rosenquist, Francis M. Rubinstein, Steve Selkowitz, Haider Taha, and Isaac Turiel. 1994. Buildings Sector Demand-Side Efficiency Technology Summaries. Berkeley, CA: Lawrence Berkeley National Laboratory. LBL-33887. March. Lamp Manufacturer Catalogs for General Electric (1995), Osram Sylvania (1996), and Philips (1996). Leslie, Russell, and Kathryn Conway. 1993. Research for "The Lighting Pattern Book for Homes", Final Report. Troy, NY: Lighting Research Center, Rensselaer Polytechnic Institute. Moezzi, Mithra. 1996-97. Personal Communication: Memos to Diana Vorsatz and Leslie Shown regarding statistical distribution of residential lighting fixtures and lamps based on data obtained from the TPU analysis. National Lighting Collaborative. 1996. What is LER?: Announcing the New Luminaire Energy Information Program. Rosslyn, VA: National Lighting Collaborative. July. National Lighting Product Information Program. 1993. Lighting Answers: T8 Fluorescent Lamps. Troy, NY: Lighting Research Center, Rensselaer Polytechnic Institute. April. National Lighting Product Information Program. 1994. Specifier Reports: Electronic Ballasts. Troy, NY: Lighting Research Center, Rensselaer Polytechnic Institute. May. Ontario Hydro. 1992. Lighting Reference Guide, 5th Edition. Ontario, Canada: Ontario Hydro. Osram Sylvania Inc. 1996. Osram Sylvania Product Catalog: 1996 Lighting Technology in the Age of EPAct. Danvers, MA: Osram Sylvania Inc. Page, Erik (Lawrence Berkeley National Laboratory). 1997. Personal Communication: Conversation with Leslie Shown regarding efficacies of halogen torchieres. Philips Lighting Company. 1996. Lamp Specification and Application Guide. Somerset, NJ: Philips Lighting Company. Rundquist, R.A., T. McDougal, and J. Benya. 1996. Lighting Controls: Patterns for Design. Palo Alto, CA: Electric Power Research Institute. Prepared by R.A. Rundquist Associates for Empire State Electric Energy Research Corporation and the Electric Power Research Institute. EPRI TR-107230. December. Sezgen, A. Osman, Y. Joe Huang, Barbara A. Atkinson, and Jonathan G. Koomey. 1994. Technology Data Characterizing Lighting in Commercial Buildings: Application to End-Use Forecasting with COMMEND 4.0. Berkeley, CA: Lawrence Berkeley National Laboratory. LBL-34243. May. Siminovitch, Michael, and Evan Mills. 1994. " Fixing the Fixtures." Home Energy 11(6): 47-49. U.S. House of Representatives. 1992. Energy Policy Act of 1992. Washington DC: U.S. Government Printing Office.  There are some exceptions to this generalization: for example, some HID lamps draw increasing amounts of power as they age (Clear 1996) and some low-pressure sodium lamps maintain constant lumen output over time (Philips Lighting Company 1996)  An incandescent source that emits the theoretical maximum amount of energy is called a blackbody radiator. A blackbody radiator emits energy at all wavelengths, but the amount and proportion of the energy that is potentially visible increases rapidly with temperature. A blackbody at 600 C will probably be visible under normal lighting. Objects below about 300 C are not visibly brighter than their surroundings even for the dark adapted eye. A-  xŀhIE)'' 'dxpr MSMT 'Grphbj '"' currentpoint ",Times .+$luminous efficacy of a light source ( lumens+watt" , Symbol(  *  ( (  *  ( ( )=) ( total luminous flux(total lamp power input" \Q/MTsave save def 40 dict begin currentpoint 3 -1 roll sub neg 3 1 roll sub 9440 div 864 3 -1 roll exch div scale currentpoint translate 64 58 translate /thick 0 def /th { dup setlinewidth /thick exch def } def 16 th 4675 371 moveto 949 0 rlineto stroke 6367 371 moveto 2960 0 rlineto stroke /cat { dup length 2 index length add string dup dup 5 -1 roll exch copy length 4 -1 roll putinterval } def /ff { dup FontDirectory exch known not { dup dup length string cvs (|______) exch cat dup FontDirectory exch known {exch} if pop } if findfont } def /fs 0 def /cf 0 def /sf {exch dup /fs exch def dup neg matrix scale makefont setfont} def /f1 {ff dup /cf exch def sf} def /ns {cf sf} def /sh {moveto show} def 320 /Times-Roman f1 (luminous efficacy of a light source ) -6 454 sh ( ) 5783 454 sh ( ) 6223 454 sh (total luminous flux) 6635 251 sh (total lamp power input) 6395 699 sh 320 /Times-Italic f1 (lumens) 4694 251 sh (watt) 4871 699 sh 320 /Symbol f1 (\351) 4541 296 sh (\353) 4541 717 sh (\352) 4541 601 sh (\371) 5643 296 sh (\373) 5643 717 sh (\372) 5643 601 sh (=) 5956 454 sh end MTsave restore dBMATH6S> luminous efficacy of a light source lumenswatt[] = total luminous fluxtotal lamp power inputgtdoENICAf (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENICAfENR-t adENIC3, m(Ontario Hydro 1992) 1a. Sourcebook.libzOntario Hydro, 1992 #122ENRFE7'dENIC3, mENR-es dtENICL0}T (Clear 1996) 1a. Sourcebook.libClear, 1996 #160zENRFE7'dENICL0}TENR-(19d{ENIC{٥8(Eley Associates 1993) 1a. Sourcebook.lib DOE, 1993 #13ENRFE7'dENIC{٥8ENR-dENICK0}(1992) 1a. Sourcebook.libzOntario Hydro, 1992 #122ENRFE7'dENICK0}ENR-@dENIC (1995) 1a. Sourcebook.lib#General Electric Company, 1995 #204ENRFnE7'dENIC ENR-agedENIC 7(1996) 1a. Sourcebook.libOsram Sylvania Inc, 1996 #203ENRFInfE7'dENIC 7ENR-d RdENIC* (1996) 1a. Sourcebook.lib#Philips Lighting Company, 1996 #205ENRF)E7'dENIC* ENR-dENICL (1992) 1a. Sourcebook.libzOntario Hydro, 1992 #122ENRFE7'dENICL ENR-dENICL (1995) 1a. Sourcebook.lib#General Electric Company, 1995 #204ENRFE7'dENICL ENR-dENICL (1996) 1a. Sourcebook.libOsram Sylvania Inc, 1996 #203ENRFE7'dENICL ENR-sdENICM (1996) 1a. Sourcebook.lib#Philips Lighting Company, 1996 #205ENRFE7'dENICM ENR-&dENICLZ$ x(Atkinson et al. 1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICLZ$ xENR-dzENICLY (1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRF E7'dENICLY ENR-h d{ENIC{D(Eley Associates 1993) 1a. Sourcebook.lib DOE, 1993 #13ENRFE7'dENIC{DENR-dENICM(Atkinson et al. 1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICMENR-eKdENICM(Atkinson et al. 1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICMENR-dENICLQ *(Atkinson et al. 1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICLQ *ENR-dENICLW (Atkinson et al. 1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICLW ENR-dENICM>i(Siminovitch and Mills 1994) 1a. Sourcebook.lib"Siminovitch, 1994 #34ENRFjE7'dENICM>iENR-o3d{ENIC{TA(Eley Associates 1993) 1a. Sourcebook.lib DOE, 1993 #13ENRFE7'dENIC{TAENR-doENICTB (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENICTBENR-PdoENICTB: (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRF@E7'dENICTB:ENR-doENICTBN (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFxE7'dENICTBNENR-\tdoENICTBb (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENICTBbENR-doENICTBv (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENICTBvENR-NdoENICTB (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENICTBENR- dENICwSz"(Atkinson et al. 1995, Clear 1997) 1a. Sourcebook.lib$Atkinson, 1995 #24; Clear, 1997 #191ENRFE7'dENICwSzENR-?dxpr MSMT Grphbj " currentpoint ",Times .+BF)  )=) (!/actual lumen output of lamp operated by ballast+"rated lumen output of the lamp ߡ/MTsave save def 40 dict begin currentpoint 3 -1 roll sub neg 3 1 roll sub 7232 div 832 3 -1 roll exch div scale currentpoint translate 64 59 translate /thick 0 def /th { dup setlinewidth /thick exch def } def 16 th 975 338 moveto 6146 0 rlineto stroke /cat { dup length 2 index length add string dup dup 5 -1 roll exch copy length 4 -1 roll putinterval } def /ff { dup FontDirectory exch known not { dup dup length string cvs (|______) exch cat dup FontDirectory exch known {exch} if pop } if findfont } def /fs 0 def /cf 0 def /sf {exch dup /fs exch def dup neg matrix scale makefont setfont} def /f1 {ff dup /cf exch def sf} def /ns {cf sf} def /sh {moveto show} def 320 /Times-Italic f1 (BF) 8 421 sh 320 /Times-Roman f1 ( ) 398 421 sh (=) 565 421 sh ( ) 831 421 sh (actual lumen output of lamp operated by ) 996 218 sh (ballast) show (rated lumen output of the lamp) 2080 666 sh end MTsave restore dMATHH# BF = actual lumen output of lamp operated by ballastrated lumen output of the lampndd}ENIC G(Koomey et al. 1994) 1a. Sourcebook.libuKoomey, 1994 #117ENRFE7'dENIC GENR-dxpr  " currentpoint ",Times .+BEF) )=)  (GBF)  , Symbol)) )100('lamp)+)ballast input power" &lY30 dict begin currentpoint 3 -1 roll sub neg 3 1 roll sub 4768 div 864 3 -1 roll exch div scale currentpoint translate 64 53 translate 8 427 moveto /fs 0 def /sf {exch dup /fs exch def dup neg matrix scale makefont setfont} def /f1 {findfont dup /cf exch def sf} def /ns {cf sf} def 320 /Times-Italic f1 (BEF) show 593 427 moveto 320 /Times-Roman f1 ( ) show 760 427 moveto (=) show 1026 427 moveto ( ) show 2228 218 moveto 320 /Times-Italic f1 (BF) show 2618 218 moveto 320 /Times-Roman f1 ( ) show 2789 218 moveto 320 /Symbol f1 (\264) show 3055 218 moveto 320 /Times-Roman f1 ( ) show 3132 218 moveto 320 /Times-Roman f1 (100) show 1196 680 moveto 320 /Times-Roman f1 (lamp) show 1911 680 moveto (+) show 2175 680 moveto (ballast input power) show /thick 0 def /th { dup setlinewidth /thick exch def } def 16 th 1170 344 moveto 3485 0 rlineto stroke end dMATH d BEF = BF  100lamp+ballast input powerovM///dxpr MSMT /Grphbj /"/ currentpoint ",Times .+System Efficacy ( Ilm+ W"Y , Symbol (F())(c )=) (vrated lamp lumens+input power (W)" uJ( )*) number of lamps )K*) BF:/MTsave save def 40 dict begin currentpoint 3 -1 roll sub neg 3 1 roll sub 9696 div 832 3 -1 roll exch div scale currentpoint translate 64 59 translate /thick 0 def /th { dup setlinewidth /thick exch def } def 16 th /stb { newpath moveto 0 setlinewidth 2 copy rlineto } def /enb { rlineto neg exch neg exch rlineto closepath fill } def /hb { stb 0 thick enb } def /vb { stb thick 0 enb } def -313 440 2803 120 vb 16 th 3691 338 moveto 2408 0 rlineto stroke /cat { dup length 2 index length add string dup dup 5 -1 roll exch copy length 4 -1 roll putinterval } def /ff { dup FontDirectory exch known not { dup dup length string cvs (|______) exch cat dup FontDirectory exch known {exch} if pop } if findfont } def /fs 0 def /cf 0 def /sf {exch dup /fs exch def dup neg matrix scale makefont setfont} def /f1 {ff dup /cf exch def sf} def /ns {cf sf} def /sh {moveto show} def 320 /Times-Roman f1 (System Efficacy ) -13 421 sh (lm) 2291 338 sh (W) 2686 556 sh ( ) 3114 421 sh (=) 3281 421 sh ( ) 3547 421 sh (rated lamp lumens) 3722 218 sh (input power \(W\)) 3835 666 sh ( ) 6163 421 sh (*) 6303 421 sh ( number of lamps ) 6522 421 sh (*) 8906 421 sh ( BF) 9125 421 sh /f3 {ff 3 -1 roll .001 mul 3 -1 roll .001 mul matrix scale makefont dup /cf exch def sf} def 320 1000 1837 /Symbol f3 (\() 2176 480 sh (\)) 3001 480 sh end MTsave restore dMATHC System Efficacy )lmW() = rated lamp lumensinput power (W) * number of lamps * BFrld{ENIC{F](Eley Associates 1993) 1a. Sourcebook.lib DOE, 1993 #13ENRFE7'dENIC{F]ENR-rmadzENIC U(Audin et al. 1994) 1a. Sourcebook.libAudin, 1994 #28ENRFE7'dENIC UENR- dENIC &4(National Lighting Product Information Program 1994) 1a. Sourcebook.libProgram, 1993 #29ENRFNRFE7'dENIC &ENR-(19dzENIC;g\(Audin et al. 1994) 1a. Sourcebook.libAudin, 1994 #28ENRFE7'dENIC;g\ENR-doENIC;g\ (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFE7'dENIC;g\ENR-dzENIC;g\(Audin et al. 1994) 1a. Sourcebook.libAudin, 1994 #28ENRF'E7'dENIC;g\ENR-dzENICM/[&(Audin et al. 1994) 1a. Sourcebook.libAudin, 1994 #28ENRFE7'dENICM/[&ENR-doENICM8 (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRF4E7'dENICM8ENR-EdENICM'(1996) 1a. Sourcebook.lib*National Lighting Collaborative, 1996 #140ENRFE7'dENICM'ENR-doENICMq (IES 1993) 1a. Sourcebook.lib IES, 1993 #21uENRFbE7'dENICMqENR-dyENICRڕ1(1994) 1a. Sourcebook.libuKoomey, 1994 #117ENRFE7'dENICRڕ1ENR-dzENICM!(1995) 1a. Sourcebook.libAtkinson, 1995 #24ENRFE7'dENICM!ENR-d|ENIC\% o5(1996) 1a. Sourcebook.libRundquist, 1996 #184ENRFE7'dENIC\% o5ENR-duENICw(1993) 1a. Sourcebook.lib IES, 1993 #21{ENRFE7'dENICwENR-on duENICw(1993) 1a. Sourcebook.lib  DOE, 1993 #13{ENRFE7'dENICwENR-dENICw/(Clear and Rubinstein 1997) 1a. Sourcebook.libClear, 1997 #158ENRF'E7'dENICw/ENR-dyENICwW(1994) 1a. Sourcebook.libuKoomey, 1994 #117ENRFE7'dENICwWENR-dENIC@\ -Atkinson, Barbara, Andrea Denver, James E. McMahon, Leslie Shown, and Robert Clear. 1995. Energy-Efficient Lighting Technologies and Their Applications in the Commercial and Residential Sectors. Boca Raton, FL: CRC Press. Pages pp. 399-427. Audin, L., D. Houghton, M. Shephard, and W. Hawthorne. 1994. Lighting Technology Atlas. Boulder, Colorado: E Source, Inc. Prepared by Prepared for Clear, Robert. April, 1997. Personal Communication: Written comments provided to Leslie Shown regarding review of April 3, 1997 DRAFT Sourcebook. Clear, Robert, and Francis Rubinstein. March 10, 1997. Personal Communication: Conversation with Leslie Shown regarding the dimmability of fluorescent lamps. Clear, Robert (Lawrence Berkeley National Laboratory). October 21, 1996. Personal Communication: E-mail to Jonathan Koomey regarding review of October 1996 DRAFT Sourcebook. Eley Associates. 1993. Advanced Lighting Guidelines: 1993. San Francisco, CA: Eley Associates. Prepared by Eley Associates, Luminae Souter Lighting Design, and Lawrence Berkeley National Laboratory. Prepared for U.S. Department of Energy, California Energy Commission, Electric Power Research Institute. DOE/EE-0008. General Electric Company. 1995. GE Lighting: Spectrum 9200 Lamp Catalog. U.S.: General Electric Company. Prepared by Prepared for IES. 1993. Lighting Handbook: Reference & Application. 8th Edition. New York, NY: Illuminating Engineering Society of North America. Koomey, Jonathan G., Francis X. Johnson, Jennifer Schuman, Ellen Franconi, Steve Greenberg, Jim D. Lutz, Brent T. Griffith, Dariush Aresteh, Celina Atkinson, Kristin Heinemeier, Y. Joe Huang, Lynn Price, Greg Rosenquist, Francis M. Rubinstein, Steve Selkowitz, Haider Taha, and Isaac Turiel. 1994. Buildings Sector Demand-Side Efficiency Technology Summaries. Berkeley, CA: Lawrence Berkeley National Laboratory. Prepared by Prepared for LBL-33887. March. National Lighting Collaborative. 1996. What is LER?: Announcing the New Luminaire Energy Information Program. Rosslyn, VA: National Lighting Collaborative. Prepared by Prepared for July. National Lighting Product Information Program. 1994. Specifier Reports: Electronic Ballasts. Troy, NY: Lighting Research Center, Rensselaer Polytechnic Institute. Prepared by Prepared for May. Ontario Hydro. 1992. Lighting Reference Guide, 5th Edition. Ontario, Canada: Ontario Hydro. Osram Sylvania Inc. 1996. Osram Sylvania Product Catalog: 1996 Lighting Technology in the Age of EPAct. Danvers, MA: Osram Sylvania Inc. Prepared by Prepared for Philips Lighting Company. 1996. Lamp Specification and Application Guide. Somerset, NJ: Philips Lighting Company. Prepared by Prepared for Rundquist, R.A., T. McDougal, and J. Benya. 1996. Lighting Controls: Patterns for Design. Palo Alto, CA: Electric Power Research Institute. Prepared by R.A. Rundquist Associates. Prepared for Empire State Electric Energy Research Corporation and the Electric Power Research Institute. EPRI TR-107230. December. Siminovitch, Michael, and Evan Mills. 1994. " Fixing the Fixtures." Home Energy 11(6): 47-49.! Z  1 J   W o   !     ? i     w     @  }    )  Q         0ENBBE7'dENIC@\ENB-EdpENICL2 (Clear 1996) Sourcebook.libClear, 1996 #160vENRFE7'dENICL2ENR-RFdENICB 3~(Philips Lighting Company 1996) 1a. Sourcebook.lib#Philips Lighting Company, 1996 #205ENRFE7'dENICB 3~ENR-N Equation6 luminous efficacy of a light source lumenswatt[] = total luminous fluxtotal lamp power inputf hV2ENRFMAf (IES 1993)  IES, 1993 #21" hV2ENR- Af{ hV2ENRFb3, m(Ontario Hydro 1992) zOntario Hydro, 1992 #122" hV2ENR- 3, mk hV2ENRFRL0}T (Clear 1996) Clear, 1996 #160" hV2ENR- L0}Tr hV2ENRFY{٥8(Eley Associates 1993)  DOE, 1993 #13" hV2ENR- {٥8s hV2ENRFZK0}(1992) zOntario Hydro, 1992 #122" hV2ENR- K0}~ hV2ENRFe (1995) #General Electric Company, 1995 #204" hV2ENR- x hV2ENRF_ 7(1996) Osram Sylvania Inc, 1996 #203" hV2ENR- 7~ hV2ENRFe* (1996) #Philips Lighting Company, 1996 #205" hV2ENR- * s hV2ENRFZL (1992) zOntario Hydro, 1992 #122" hV2ENR- L ~ hV2ENRFeL (1995) #General Electric Company, 1995 #204" hV2ENR- L x hV2ENRF_L (1996) Osram Sylvania Inc, 1996 #203" hV2ENR- L ~ hV2ENRFeM (1996) #Philips Lighting Company, 1996 #205" hV2ENR- M w hV2ENRF^LZ$ x(Atkinson et al. 1995) Atkinson, 1995 #24" hV2ENR- LZ$ xm hV2ENRFTLY (1995) Atkinson, 1995 #24" hV2ENR- LY r hV2ENRFY{D(Eley Associates 1993)  DOE, 1993 #13" hV2ENR- {Dw hV2ENRF^M(Atkinson et al. 1995) Atkinson, 1995 #24" hV2ENR- Mw hV2ENRF^M(Atkinson et al. 1995) Atkinson, 1995 #24" hV2ENR- Mw hV2ENRF^LQ *(Atkinson et al. 1995) Atkinson, 1995 #24" hV2ENR- LQ *w hV2ENRF^LW (Atkinson et al. 1995) Atkinson, 1995 #24" hV2ENR- LW  hV2ENRFgM>i(Siminovitch and Mills 1994) "Siminovitch, 1994 #34" hV2ENR- M>ir hV2ENRFY{TA(Eley Associates 1993)  DOE, 1993 #13" hV2ENR- {TAf hV2ENRFMTB (IES 1993)  IES, 1993 #21" hV2ENR- TBf hV2ENRFMTB: (IES 1993)  IES, 1993 #21" hV2ENR- TB:f hV2ENRFMTBN (IES 1993)  IES, 1993 #21" hV2ENR- TBNf hV2ENRFMTBb (IES 1993)  IES, 1993 #21" hV2ENR- TBbf hV2ENRFMTBv (IES 1993)  IES, 1993 #21" hV2ENR- TBvf hV2ENRFMTB (IES 1993)  IES, 1993 #21" hV2ENR- TB hV2ENRFwSz"(Atkinson et al. 1995, Clear 1997) $Atkinson, 1995 #24; Clear, 1997 #191" hV2ENR- wSz Equation BF = actual lumen output of lamp operated by ballastrated lumen output of the lampt hV2ENRF[ G(Koomey et al. 1994) uKoomey, 1994 #117" hV2ENR-  G Equation BEF = BF  100lamp+ballast input power' Equation System Efficacy )lmW() = rated lamp lumensinput power (W) * number of lamps * BFr hV2ENRFY{F](Eley Associates 1993)  DOE, 1993 #13" hV2ENR- {F]q hV2ENRFX U(Audin et al. 1994) Audin, 1994 #28" hV2ENR-  U hV2ENRF{ &4(National Lighting Product Information Program 1994) Program, 1993 #29" hV2ENR- &q hV2ENRFX;g\(Audin et al. 1994) Audin, 1994 #28" hV2ENR- ;g\f hV2ENRFM;g\ (IES 1993)  IES, 1993 #21" hV2ENR- ;g\q hV2ENRFX;g\(Audin et al. 1994) Audin, 1994 #28" hV2ENR- ;g\q hV2ENRFXM/[&(Audin et al. 1994) Audin, 1994 #28" hV2ENR- M/[&f hV2ENRFMM8 (IES 1993)  IES, 1993 #21" hV2ENR- M8 hV2ENRFlM'(1996) *National Lighting Collaborative, 1996 #140" hV2ENR- M'f hV2ENRFMMq (IES 1993)  IES, 1993 #21" hV2ENR- Mql hV2ENRFSRڕ1(1994) uKoomey, 1994 #117" hV2ENR- Rڕ1m hV2ENRFTM!(1995) Atkinson, 1995 #24" hV2ENR- M!o hV2ENRFV\% o5(1996) Rundquist, 1996 #184" hV2ENR- \% o5h hV2ENRFOw(1993)  IES, 1993 #21" hV2ENR- wh hV2ENRFOw(1993)   DOE, 1993 #13" hV2ENR- wz hV2ENRFaw/(Clear and Rubinstein 1997) Clear, 1997 #158" hV2ENR- w/l hV2ENRFSwW(1994) uKoomey, 1994 #117" hV2ENR- wW4 hV2ENBB@\" hV2ENB- @\ hV2ENRFxB 3~(Philips Lighting Company 1996) #Philips Lighting Company, 1996 #205" hV2ENR- B 3~ustrial sectors and are rarely used for new lighting systems (2). For industrial and outdoor applications where color was not a priority, MV ?MopXgdeop Q Z QRfg  01GH()CD`a=>V @Ԩ @ J @Ӵ @ @ @ @ @, B@DVWvwCDh >?EFWX^_mntuvKLbc~ @ݠ @ @ܨ @ @ۤ @ @ڤ @ @٠ @ @ب @ @פ @ @֤ @ @ՠ @0$$%%/())))),^,_,u,v-R-S-i-j-k-l--1|1}116L6M6c6d>q>r??@@DDD.D/D}D~DDDDFNF]GG @` @ @l @ @h @ @p @ @x @ @ @ @߈ @@ J @ޔ @5GHOHQMQMRM\M]NNN'N(O.O/O9O:Q2Q3Q=Q>R%R&R0R1SSSSUUUUVVVVYYZZ[ [ [[ [#[B[[\-{ B @ @ B4 @ @ @ @0 @ @H @ @` @ @x @0 @ @H @/\-\.\?^^^^_x_y__``bbb+b,ccReMeNfffDfEiii(i)idifiijjllmmm m m5mCnknlnnppphpiqGqHq[ @ @d @ @p @@ @ @ @ @ @| @ @ @ B5q[q\q`qprrrsstt|8|9i$ .?@FGnouv~ ! @  @ d @  @ p @ ( @ | @ 4 @  @L @ @@ @@ @X9!$R]GRJK\$"0178JKQRSTw34:;T @ @4 @ @8@ @ @ H @ @ X  @  BTU Ac8P8Pk3 cc6f~J!/T‘>VIoŘ ƙƤƲƳ'(ǤǥDZDz !" @!8 @ l @ $ @ J @<@ @(RAs @ @?MoqXi  ᥝwmPP(H!T@@@@@@@@@@@@  (  (( ($  $ "$" $ $ $ $ $$ $ $ $ $#()8>CDU[`ax .8=>KQVWkqvwDTXhɵ  8 (  (h  ($ l $ l( (P(H!T@@@@@@@@@@@@  (h  (h1hiz~#$<@EF^bghv89!5㭨 $ h $ $ $ $ $ $ l  (  8 (  (h  (P(T$,@@@@@@@@@@@@1!5"%'))++-l-0002~2336:?P?AD1DFLGHQHHJ KOS¼ܨܟ뼅wi^XR$ $  $ hh $ hLh $ hLh $ hLh$  $ $ $ $$ $ $ $$ $ $ $  SUUV V!V6VKV`VuVVVVVYYZZ[#[B[[\.\?^^^``ccRdg:ifiǹǒwiic]W$ $ $  $ hh" $ hh" $hh" $ hh" $ hh" $hh" $ hh" $ hh" $ "$h$ h$ "ijm ppRpjq^rstuwxx-x7y|9}~i$ .ý藏臭|tltdtt$  $ $  $ $ $$ $ $ h$ $ $ hLh" $ h"$$ $ $ $ ##R]^JK^"#S~vwӼݲݨݲӼzcZcQl (4 (P$@@@@@@@@4 (l ( "$   (  (  (P$@@@@@@@@  (  ("$  $ =Tn!Kg\|%3L#qwâEཷz $ h $ h$  h $ h $ h$ l (4 (l (P$@@@@@@@@4 (,EUƴ&'!$  $ h $ h $ h Normal PTitle referencestable table title Word4095NullTOC 6TOC 4TOC 3TOC 2 Page Number@ @   @  X!@  X!@8 X!@ X!@ X!@    x   hhh0h 0  h X!@ X!@8 X!@ X!@"#C"## a 6->PKTU U"U7ULUaUvUUU_n_vww.$T>}P?bD      ! .*=. VG\-q[!TAsh!5SiE  &p&r?HH(FG(HH(d@=/R@H-:LaserWriter 8 New YorkTimes HelveticaNEmbedded_Object_280 ICEmbdObject2 ICEmbdObject3ICEmbdObject49_285ICEmbdObject145_286 ICEmbdObject6 ICEmbdObject7 ICEmbdObject4 ICEmbdObject5ICEmbdObject219_283ICEmbdObject220_284ICEmbdObject41ICEmbdObject46ICEmbdObject47ICEmbdObject48ICEmbdObject49ICEmbdObject50ICEmbdObject219_358ICEmbdObject220_359ICEmbdObject41_360ICEmbdObject46_361ICEmbdObject47_362ICEmbdObject48_363ICEmbdObject49_364ICEmbdObject50_365ICEmbdObject14ICEmbdObject15ICEmbdObject4_331ICEmbdObject5_332ICEmbdObject42ICEmbdObject43ICEmbdObject22ICEmbdObject23ICEmbdObject24ICEmbdObject25ICEmbdObject10ICEmbdObject11ICEmbdObject12ICEmbdObject13ICEmbdObject2_87ICEmbdObject3_88ICEmbdObject28ICEmbdObject29ICEmbdObject0_337ICEmbdObject1_338ICEmbdObject2_339ICEmbdObject3_340ICEmbdObject4_341ICEmbdObject5_342ICEmbdObject6_343ICEmbdObject7_344 ICEmbdObject8 ICEmbdObject9ICEmbdObject10_345ICEmbdObject11_346 ICEmbdObject0ICEmbdObject34Embedded_Object_4ICEmbdObject79ICEmbdObject152Embedded_Object_3Embedded_Object_2ICEmbdObject44ICEmbdObject45ICEmbdObject16_43ICEmbdObject17_44ICEmbdObject24_61ICEmbdObject25_62ICEmbdObject26_59ICEmbdObject27_60ICEmbdObject126ICEmbdObject127ICEmbdObject153ICEmbdObject154ICEmbdObject16ICEmbdObject17ICEmbdObject18ICEmbdObject19ICEmbdObject126_333ICEmbdObject127_334ICEmbdObject20ICEmbdObject21ICEmbdObject30ICEmbdObject31ICEmbdObject26ICEmbdObject27ICEmbdObject32ICEmbdObject33ICEmbdObject35ICEmbdObject36ICEmbdObject37ICEmbdObject38ICEmbdObject0_353ICEmbdObject1_354ICEmbdObject39ICEmbdObject40ICEmbdObject1_367ICEmbdObject53odoQf0G>EW^mtKb'(+^+u,R,i0|05L5cCC.C}CLQL\MM'N.N9P2P=Q%Q0RRTTYZ ZZ]^x^aa+eeDhh(klmkmpGp[?Fnu 07JQ3:TŲH`!}"#1#S##$[$}$ % % % &$ &F&&'^''((()-)O))*f**+++ ,"!,D",#,$-T%-v&-'.(.).*/#+/E,/-/.03/0U001021C31e415162S72u839H`30:4P;46?7@7=A7B7C8dD8E8F9G9H9I:J:4K:L:M;-N;OO;P;Q<\R<~S<T= U=yV=W> X>,Y>Z>[?\?@]?^?_@H`@ja@pepRg1H ?FX_nuLc((+_+v,S,j0}05M5dCC/C~CLRL]MM(N/N:P3P>Q&Q1RRTTYZ Z Z]^y^aa,eeEhh)klmlmpHp\@Gov!18KR4;UųHHHH ICEmbdObject51ICEmbdObject52@AQBB BE E BComprehnsv Lighting Mkt AssmtVorsatz Diana Vorsatz Diana