PRELIMINARY TECHNICAL SUPPORT DOCUMENT: ENERGY EFFICIENCY PROGRAM FOR COMMERCIAL AND INDUSTRIAL EQUIPMENT:

High-Intensity Discharge Lamps Analysis of Potential Energy Savings

Docket #: EE-DET-03-001

April 2010

U.S. Department of Energy

Assistant Secretary Office of Energy Efficiency and Renewable Energy Building Technologies Program Appliances and Commercial Equipment Standards Washington, DC 20585

TABLE OF CONTENTS LIST OF ACRONYMS AND ABBREVIATIONS .............................................................. V 1.1 LEGAL AUTHORITY ................................................................................................. 1-1 1.2 HIGH-INTENSITY DISCHARGE LAMP DEFINITION ........................................... 1-2 2.1 SHIPMENT ESTIMATES ........................................................................................... 2-1 2.1.1 Energy Use by Sector..................................................................................... 2-1 2.1.2 High-Intensity Discharge Lamp Shipment Estimates .................................... 2-3 2.1.3 Manufacturers of High-Intensity Discharge Lamps, Luminaires, and Ballasts ........................................................................................................... 2-7 2.1.4 Distribution Channels .................................................................................... 2-7 2.1.5 Decision Makers ............................................................................................ 2-9 2.1.5.1 Residential Decision Makers ............................................................... 2-9 2.1.5.2 Commercial, Industrial, and Public Sector Lighting Decision Makers ............................................................................................... 2-10 2.2 APPLICATIONS FOR HIGH-INTENSITY DISCHARGE LAMPS .......................... 2-10 2.2.1 Residential Applications ................................................................................ 2-10 2.2.2 Commercial Applications .............................................................................. 2-11 2.2.3 Industrial Applications ................................................................................... 2-11 2.2.4 Other (Outdoor Public Sector) Applications.................................................. 2-11 2.3 CLASSIFICATION OF HIGH-INTENSITY DISCHARGE LUMINAIRES ............. 2-12 2.3.1 Proposed High-Intensity Discharge Luminaire Classification System .......... 2-12 2.3.1.1 Product Applications: Indoor ............................................................ 2-13 2.3.1.2 Product Applications: Outdoor .......................................................... 2-14 2.3.1.3 Product Applications: Specialty ........................................................ 2-16 2.4 REGULATORY AND NON-REGULATORY PROGRAMS..................................... 2-16 2.4.1 Regulatory Programs and Legislation Impacting High-Intensity Discharge Lamps ............................................................................................................. 2-16 2.4.1.1 Federal Regulatory Programs and Legislation .................................. 2-16 2.4.1.2 Draft Legislation................................................................................ 2-17 2.4.1.3 DOE Background Studies ................................................................. 2-19 2.4.1.4 State Regulatory Programs ................................................................ 2-20 2.4.2 Non-Regulatory Programs Affecting High-Intensity Discharge Lamps ....... 2-23 3.1 SPECIFICATIONS, TEST STANDARDS, AND PROCEDURES............................. 3-1 3.2 LAMP METRICS ......................................................................................................... 3-3 3.2.1 Lumens ........................................................................................................... 3-3 3.2.2 Mean Lumens/Lamp Lumen Depreciation/Lumen Maintenance .................. 3-3 3.2.3 Efficacy .......................................................................................................... 3-4 3.2.3.1 Visual Efficiency ................................................................................. 3-5 3.2.3.2 Moving Beyond the Limits.................................................................. 3-7 3.2.4 Color Rendering Index ................................................................................... 3-8 3.2.5 Correlated Color Temperature ....................................................................... 3-8 3.2.6 Lamp Life....................................................................................................... 3-8 3.3 MERCURY VAPOR (MV) LAMPS ............................................................................ 3-8 3.3.1 Lamp Construction......................................................................................... 3-9 3.3.2 Lamp Performance ......................................................................................... 3-10

1-i

3.4

3.5

3.6 4.1

4.2 4.3 5.1

5.2

5.3

6.1

3.3.3 Electrical Characteristics ............................................................................... 3-13 3.3.4 Other Issues .................................................................................................... 3-14 HIGH-PRESSURE SODIUM LAMPS ........................................................................ 3-14 3.4.1 Lamp Construction......................................................................................... 3-14 3.4.2 Lamp Performance ......................................................................................... 3-15 3.4.3 Electrical Characteristics ............................................................................... 3-17 3.4.4 Other Issues .................................................................................................... 3-20 METAL HALIDE LAMPS........................................................................................... 3-20 3.5.1 Construction ................................................................................................... 3-20 3.5.1.1 Probe-Start Quartz Metal Halide Lamps ........................................... 3-24 3.5.1.2 Pulse-Start Quartz Metal Halide Lamps ............................................ 3-25 3.5.1.3 Pulse-Start Ceramic Metal Halide Lamps ......................................... 3-25 3.5.2 Lamp Performance ......................................................................................... 3-26 3.5.2.1 Probe-Start Quartz Metal Halide Lamps ........................................... 3-28 3.5.2.2 Pulse-Start Quartz Metal Halide Lamps ............................................ 3-29 3.5.2.3 Pulse-Start Ceramic Metal Halide Lamps ......................................... 3-30 3.5.3 Electrical Characteristics ............................................................................... 3-31 3.5.4 Other Issues .................................................................................................... 3-32 HIGH-INTENSITY DISCHARGE LAMP BALLASTS ............................................. 3-33 ENGINEERING ANALYSIS FRAMEWORK ............................................................ 4-1 4.1.1 Representative Equipment Categories ........................................................... 4-2 4.1.2 Baseline Lamps .............................................................................................. 4-2 4.1.3 Substitute Lamps............................................................................................ 4-2 EQUIPMENT PRICE DETERMINATION ................................................................. 4-3 ENGINEERING ANALYSIS RESULTS .................................................................... 4-4 METHOD AND DATA ................................................................................................ 5-1 5.1.1 Annual Operating Hours ................................................................................ 5-1 5.1.2 Lamp Prices ................................................................................................... 5-2 5.1.3 Ballast Prices .................................................................................................. 5-4 5.1.4 Fixture Prices ................................................................................................. 5-5 5.1.5 Lamp Lifetime ............................................................................................... 5-7 5.1.6 Ballast Lifetime .............................................................................................. 5-8 5.1.7 Contractor Mark-up ....................................................................................... 5-8 5.1.8 Sales Tax ........................................................................................................ 5-9 5.1.9 Maintenance Costs ......................................................................................... 5-9 5.1.10 Electricity Price .............................................................................................. 5-10 5.1.11 Discount Rate ................................................................................................. 5-10 RESULTS FOR MEDIUM-WATTAGE MERCURY VAPOR DESIGN OPTIONS ...................................................................................................................... 5-11 5.2.1 Event 1 Lamp-Only Replacement ............................................................... 5-11 RESULTS FOR MEDIUM-WATTAGE PROBE-START METAL HALIDE DESIGN OPTIONS ...................................................................................................... 5-12 5.3.1 Event 1 Lamp-Only Replacement ............................................................... 5-12 5.3.2 Event 2 New Construction / Fixture Replacement ..................................... 5-14 METHODOLOGY ....................................................................................................... 6-1 6.1.1 Shipment Forecasts ........................................................................................ 6-1

1-ii

6.2

6.1.2 National Energy Savings Analysis................................................................. 6.1.3 Net Present Value Analysis ........................................................................... ESTIMATES OF POTENTIAL ENERGY AND CONSUMER IMPACTS ............... 6.2.1 National Energy Savings................................................................................ 6.2.2 Net Present Value ..........................................................................................

6-2 6-2 6-3 6-3 6-3

LIST OF TABLES Table 2.1.1. 2008 Total Lighting Technology (High-Intensity Discharge) Electricity Consumption ........................................................................................................ 2-1 Table 2.1.2. Installed Base of High-Intensity Discharge Luminaires in the United States ......... 2-3 Table 2.1.3. Distribution of Electricity Consumption by Sector of High-Intensity Discharge Lighting ................................................................................................................ 2-3 Table 2.1.4. Portion of Total NEMA High-Intensity Discharge Lamp Shipments ..................... 2-3 Table 2.1.4. Portion (continued) .................................................................................................. 2-4 Table 2.1.5. Freedonia Group High-Intensity Discharge Lamp Shipments ................................ 2-5 Table 2.1.6. 2008 NEMA Shipments by Wattage Grouping ....................................................... 2-6 Table 2.3.1. Classification for Indoor Applications ................................................................... 2-13 Table 2.3.2. Classification for Outdoor Applications ................................................................ 2-14 Table 2.3.3. Classification for Specialty Applications .............................................................. 2-16 Table 2.4.1. California Metal Halide Fixture Requirements ..................................................... 2-20 Table 3.1.1 Documents Pertaining to High-Intensity Discharge Lamps ..................................... 3-1 Table 3.2.1. Visual Efficiencies and Efficacies of Various Sources ........................................... 3-5 Table 3.2.2. Sample Energy Output for High Intensity Discharge Lamps .................................. 3-6 Table 3.2.3. Chromaticities and Maximum Spectral Efficiencies of Various Sources................ 3-6 Table 3.2.4. Color Rendering Indexes and Spectral Efficiencies of High-Intensity Discharge Sources ................................................................................................................. 3-7 Table 3.2.5. Probable Efficacy Limits by Color Correlated Temperature (CCT) for Conventional High-Intensity Discharge Lamps................................................... 3-7 Table 3.3.1. Performance Summary of Mercury Vapor Lamps ................................................ 3-12 Table 3.4.1. Performance Summary of High-Pressure Sodium Lamps ..................................... 3-16 Table 3.5.1. Performance Summary of Probe-Start Metal Halide Lamps ................................. 3-28 Table 3.5.2. Performance Summary of Quartz Pulse-Start Metal Halide Lamps ...................... 3-29 Table 3.5.3. Performance Summary of Pulse-Start Ceramic Metal Halide Lamps ................... 3-31 Table 3.5.4. Total Mercury Sold in Lamps ................................................................................ 3-32 Table 3.5.5. Total Mercury Sold by HID Lamp Type ............................................................... 3-32 Table 4.3.1. HID Lamps Determination Engineering Analysis Results ...................................... 4-4 Table 5.1.1. Lamp Prices from RS-Means ................................................................................... 5-2 Table 5.1.2. Municipal Lamp Prices ............................................................................................ 5-2 Table 5.1.3. Lamp Price derived from Online Distributor Web Sites ......................................... 5-3 Table 5.1.4. Lamp Price derived from Online Distributor Web Sites ......................................... 5-5 Table 5.1.5. Prices from RS-Means for HID Fixtures ................................................................. 5-5 Table 5.1.6. Lamp Price derived from Online Distributor Web Sites ......................................... 5-6 Table 5.1.7. Lamp Life Data (175W MV Baseline) .................................................................... 5-7 Table 5.1.8. Lamp Life Data (250W MV Baseline) .................................................................... 5-7

1-iii

Table 5.1.9. Lamp Life Data (400W MV Baseline) .................................................................... 5-7 Table 5.1.10. Lamp Life Data (175W Baseline) .......................................................................... 5-7 Table 5.1.11. Lamp Life Data (250W MH Baseline) .................................................................. 5-8 Table 5.1.12. Lamp Life Data (360W Baseline) .......................................................................... 5-8 Table 5.1.13 Lamp Life Data (400W Baseline) ........................................................................... 5-8 Table 5.2.1. LCC-PBP Analysis for 175W MV Baseline.......................................................... 5-11 Table 5.2.2. LCC-PBP Analysis for 250W MV Baseline.......................................................... 5-11 Table 5.2.3. LCC-PBP Analysis for 400W MV Baseline.......................................................... 5-12 Table 5.3.1. LCC-PBP Analysis for 175W MH Baseline.......................................................... 5-12 Table 5.3.2. LCC-PBP Analysis for 250W MH Baseline.......................................................... 5-13 Table 5.3.3. LCC-PBP Analysis for 360W MH Baseline.......................................................... 5-13 Table 5.3.4. LCC-PBP Analysis for 400W MH Baseline.......................................................... 5-13 Table 5.3.5. LCC-PBP Analysis for 175W MH Baseline.......................................................... 5-14 Table 5.3.6. LCC-PBP Analysis for 250W MH Baseline.......................................................... 5-14 Table 5.3.7. LCC-PBP Analysis for 360W MH Baseline.......................................................... 5-15 Table 5.3.8. LCC-PBP Analysis for 400W MH Baseline.......................................................... 5-15 Table 6.2.1. Cumulative National Energy Savings for HID Lamps (20172046) ...................... 6-3 Table 6.2.2. Cumulative NPV Results for HID Lamps (20172046) .......................................... 6-4

LIST OF FIGURES Figure 2.1.1. Shares of Sectoral Energy Use by Lighting Technology ....................................... 2-2 Figure 2.1.2. Proportion of Lamp Type in High-Intensity Discharge Market, 19902008 ......... 2-5 Figure 2.1.3. High-Intensity Discharge Lamp Distribution Channels ......................................... 2-8 Figure 3.3.1. Mercury Vapor Lamp Construction ..................................................................... 3-10 Figure 3.3.2. Spectral Power Distribution of Mercury Vapor Lamps (a) Without Phosphors and (b) With Phosphors ..................................................................................... 3-11 Figure 3.4.1. High-Pressure Sodium Lamp Construction .......................................................... 3-15 Figure 3.4.2. Spectral Power Distribution* of High-Pressure Sodium Lamps........................... 3-17 Figure 3.4.3. Sample Power Flow of a 400 W High-Pressure Sodium Lamp ........................... 3-18 Figure 3.4.4. Trapezoid Boundary for High-Pressure Sodium Ballast Design .......................... 3-19 Figure 3.5.1. Tin Halide ............................................................................................................. 3-21 Figure 3.5.2. Dysprosium Iodide ............................................................................................... 3-21 Figure 3.5.3. Formed Arc-Tube Configuration for Horizontal MH Lamps............................... 3-22 Figure 3.5.4. Offset Electrode Configuration for Horizontal MH Lamps ................................. 3-23 Figure 3.5.5. Metal Halide Arc Tube with a Shroud ................................................................. 3-23 Figure 3.5.6. Metal Halide Lamp Construction ......................................................................... 3-24 Figure 3.5.7. Common Arc Tubes ............................................................................................. 3-25 Figure 3.5.8. Spectral Power Distribution of Metal Halide Lamps (a) Without Phosphors and (b) With Phosphors ..................................................................................... 3-27 Figure 5.1.1. Lamp Wattage/Price Comparison........................................................................... 5-3 Figure 5.1.2. Ballast Prices .......................................................................................................... 5-4 Figure 5.1.3. Lamp Wattage/Price Comparison........................................................................... 5-6

1-iv

LIST OF ACRONYMS AND ABBREVIATIONS

ACEEE AEO AK ANSI Ar ASHRAE atm At B Ba Be BBL Br BR BTU C Ca CA Caltrans CCT Ce CFL CFR CIE Cl CMH CRI CW CWA D DOC DOE Dy DyI EIA EISA EPA EPCA EPAct Er American Council for an Energy-Efficient Economy Annual Energy Outlook Alaska American National Standards Institute argon American Society of Heating, Refrigerating and Air Conditioning Engineers atmosphere astatine self-ballasted mercury barium beryllium blackbody locus bromine bulged reflector British thermal unit Celsius calcium California California Department of Transportation correlated color temperature cerium compact fluorescent lamp Code of Federal Regulations Commission Internationale de lEclairage (International Commission on Illumination) chlorine ceramic metal halide color-rendering index constant wattage constant wattage autotransformer Democrat Department of Commerce Department of Energy dysprosium dysprosium iodide Energy Information Administration Energy Independence Security Act Environmental Protection Agency Energy Policy and Conservation Act Energy Policy Act erbium

ER Eu F FL fc Gd He Hg HID HIF Ho HPS HX I IC IDA IESNA IL In K kg kPA Kr kV kWh La LA lbs LCC LCCF LEDs LLD LLF lm Lm/W LPS LPD LPW LRC Lu Med mg

elliptical reflector europium fluorine fluorescent footcandle gadodinium helium mercury high-intensity discharge high-intensity fluorescent holmium high-pressure sodium high-reactance autotransformer iodine integrated circuit International Dark-Sky Association Illuminating Engineering Society of North America induction lamp indium Kelvin kilograms kiloPascal krypton kilovolt kilowatt-hour lanthanum Louisiana pounds life-cycle cost lamp current crest factor light-emitting diodes lamp lumen depreciation light loss factor lumen lumens per watt low-pressure sodium lighting power density lumens per watt Lighting Research Center lutetium Medium milligrams

vi

Mg MH MHLF MI MLO Mog MR MS MV Nd ND Ne NEMA NES NJ nm NM NPV Pa PAR PBP Pin Pm PMH Pout Pr quads R Ra Ra RE Rn RX S/MH Sc Sm Sn SnI Sr SOx SPD Tb

magnesium metal halide, probe-start metal halide metal halide lamp fixture Michigan Model Lighting Ordinance Mogul Multi-faceted reflector Mississippi mercury vapor neodymium North Dakota neon National Electrical Manufacturers Association national energy savings New Jersey nanometer New Mexico net present value Pascals parabolic aluminized reflector payback period power in promethium pulse-start metal halide power out praseodymium Quadrillion BTU Republican color-rendering index radium rare earth radon reactor spacing-to-mounting height scandium samarium tin tin iodide strontium Sulfur dioxide spectral power distribution terbium

vii

TCLP Tm torr TSD TWhr UL UV V W Xe Y Yb

toxicity characteristic leaching procedure thullium Torr Technical support document terawatt-hour Underwriter's Laboratory ultraviolet volt watt(s) xenon yttrium ytterbium

viii

CHAPTER 1.

INTRODUCTION

The Department of Energy (the Department or DOE) is preparing an analysis to determine the potential energy savings for high-intensity discharge (HID) lamps. The purpose of this analysis is to provide a basis for the Secretary of Energy to make a determination if mandatory energy conservation standards are technologically feasible and economically justified, and would result in significant energy savings. If the Secretary makes a determination that standards are warranted, the Department will develop a test procedure and then initiate a standards rulemaking. This preliminary technical support document contains the market assessment; technology assessment; engineering analysis; life-cycle cost (LCC); payback period (PBP); the national energy savings (NES); and net present value (NPV) analyses. Each of these sections addresses questions such as: Market assessment How large is the HID lighting market? How are HID lamps specified (i.e., which products will service a particular application)? What are the distribution channels for HID lamps? What are the product applications the Department will consider in this analysis? Technology assessment What are the HID lamp technologies and their performance characteristics? What are the ballasts that operate them? Engineering analysis If energy conservation standard results in the removal of certain HID lamps from the market, which technologies might consumers choose instead? Which high-volume HID markets should the Department consider when it conducts its LCC analysis? LCC and PBP analysis What LCC and PBP calculation methods does the Departments LCC model for HID lamps use? What are the inputs to the model? How did the Department gather pricing and other input data for the model? What are the LCC and PBP results? NES analysis What NES and NPV calculation methods does the Department use for HID lamps? What are the inputs to the NES and NPV model? How did the Department gather input data for this model? What are the NES and NPV results? LEGAL AUTHORITY

1.1

The Department conducted this analysis of potential energy savings under the authority of Section 346 of the Energy Policy and Conservation Act (EPCA) (42 U.S.C. 6317), which states: Sec. 346. Energy conservation standards for high-intensity discharge lamps, distribution transformers, and small electric motors a. High-intensity discharge lamps (and distribution transformers) 1. The Secretary shall, within 30 months after October 24, 1992, prescribe testing requirements for those high-intensity discharge lamps (and distribution transformers) for which the Secretary makes

1-1

a determination that energy conservation standards would be technologically feasible and economically justified, and would result in significant energy savings. 2. The Secretary shall, within 18 months after the date on which testing requirements are prescribed by the Secretary pursuant to paragraph (1), prescribe, by rule, energy conservation standards for those highintensity discharge lamps (and distribution transformers) for which the Secretary prescribed testing requirements under paragraph (1). 3. Any standard prescribed under paragraph (2) with respect to highintensity discharge lamps shall apply to such lamps manufactured 36 months after the date such rule is published. Thus, the Secretary of Energy will make a determination whether energy conservation standards for HID lamps are technologically feasible and economically justified, and would result in significant energy savings. If the Secretary finds that energy conservation standards are warranted, the Department will commence work on a test procedure. Following the issuance of the test procedure, the Department will initiate a standards rulemaking process to establish energy conservation standards for HID lamps, based on the Departments annual priority setting process. 1.2 HIGH-INTENSITY DISCHARGE LAMP DEFINITION

The Code of Federal Regulations (CFR) currently contains definitions for HID (10 CFR 431.282), MV (10 CFR 431.282), and MH (10 CFR 431.322) lamps. The definitions are as follows: High intensity discharge lamp means an electric-discharge lamp in whichThe light-producing arc is stabilized by the arc tube wall temperature; and The arc tube wall loading is in excess of 3 Watts/cm2, including such lamps that are mercury vapor, metal halide, and high-pressure sodium lamps. Mercury vapor lamp means a high intensity discharge lamp, including clear, phosphorcoated, and self-ballasted screw base lamps, in which the major portion of the light is produced by radiation from mercury typically operating at a partial vapor pressure in excess of 100,000 Pa (approximately 1 atm). Metal halide lamp means a high intensity discharge lamp in which the major portion of the light is produced by radiation of metal halides and their products of dissociation, possibly in combination with metallic vapors. No definition for HPS lamps is currently listed in the CFR. Therefore, the Department reviewed the American National Standards Institute (ANSI) and the Illuminating Engineering Society of North America (IESNA) definitions for HPS lamps.

1-2

ANSI C82.9-1996 defines high-pressure sodium lamps as a high-intensity-discharge lamp in which the major portion of the light is produced by radiation from sodium vapor operating at a partial pressure of about 6.67 x 103 pascals (50 torr) or greater. The IESNA LIGHTING HANDBOOK Reference and Application 9th Edition defines high-pressure sodium lamp as a high-intensity discharge (HID) lamp in which light is produced by radiation from sodium vapor operating at a partial pressure of about 1.33 x 104 Pa (100 Torr). Includes clear and diffuse-coated lamps. Low-Pressure Sodium (LPS) lamps are often considered HID lamps, but this distinction is incorrect. ANSI defines LPS lamps as A discharge lamp in which light is produced by radiation from sodium vapor operating at a partial pressure of 0.13 to 1.3 pascals (10-3 to 10-2 torr). The IESNA defines LPS lamps as a discharge lamp in which light is produced by radiation from sodium vapor operating at a partial pressure of 0.1 to 1.5 Pa (approximately 10-3 to 10-2 Torr). Due to the lower pressure, the arc tube wall loading for LPS lamps is lower than the three watts per square centimeter defined for HID lamps. The full title of ANSI C82.9-1996, American National Standard for High-Intensity Discharge and Low-Pressure Sodium Lamps, Ballasts and Transformers Definitions, highlights that LPS lamps are not HID lamps. Therefore, this analysis does not incorporate LPS lamps in its scope. However, it is still important to note that many within the lighting industry generally treat LPS lamps as HID lamps.

1-3

CHAPTER 2.

MARKET ASSESSMENT

For the market assessment of high-intensity discharge (HID) lighting, the U.S. Department of Energy (DOE) gathered information on the state of the industry and the market characteristics of HID lighting. DOE uses these data as some of the inputs to the life-cycle cost analysis and the national energy savings estimate. This chapter provides the quantitative and qualitative findings of the market assessment, including shipment estimates, market structure, installations and applications, and non-regulatory initiatives to improve HID lamp efficiency. 2.1 SHIPMENT ESTIMATES

The HID lamp market in the United States is an approximate $390 million per year industry, with annual sales exceeding 30 million units. 1,2 In 2001, HID lamps consumed 125 gigawatt-hours nationally, which is approximately 17 percent of the electricity consumed by all light sources and approximately 3.6 percent of total U.S. electricity generated in 2000. 3 Three light sources characterize the HID lighting market: mercury vapor (MV), high-pressure sodium (HPS), and metal halide (MH). Technical data on these HID lamp technologies, including details about their performance, light characteristics, operating life, and other information, are provided in chapter 3, Technology Assessment. 2.1.1 Ener gy Use by Sector

Table 2.1.1 presents the energy consumption for lighting in four general sectors: three building-related (residential, commercial, and industrial) and other, which includes street and area lighting, parking garages, airport runway systems, and traffic signals. The other category does not, however, represent all outdoor stationary HID luminaires. The three building-related sectors also include some outdoor stationary lighting, where the circuit for the outdoor luminaires is tied to the electric meter for a building (e.g., architectural lighting, parking lot lighting, residential security lighting). Table 2.1.1. 2008 Total Lighting Technology (High-Intensity Discharge) Electricity Consumption
Consumption by Sector terawatt-hr (TWh) Residential Commercial Industrial Other Mercury Vapor 1 7 3 12 High-Pressure Sodium 0 6 5 30 Metal Halide N/A 34 25 4 Total for HID 1 47 33 46 Total 202 391 108 56 Source: DOE, Buildings Energy Data Book 2008, table 5.6.4. * Note: Due to rounding, the sum of the columns may not equal the total provided. HID Lamp Total* 22 41 62 125 756

Table 2.1.1 provides the electricity consumed (TWh) per lamp type per sector. The estimate of total lighting electricity consumption in the United States is 756 TWh (2.6 quads) of primary energy. In a broader context, the total energy consumption in the United States was approximately 98.3 quads in 2001, of which more than one-third (about 37 quads) was used to

2-1

generate electricity.3 Thus, lighting accounted for about 7 percent of national primary energy consumption, or about 22 percent of the electricity generated in 2001. Table 2.1.1 also provides additional details on specific HID technology energy consumption by sector. For residential, where 90 percent of lighting energy use is consumed by incandescent technologies, HID sources account for less than 1 percent (0.66 TWh/year [yr] or 0.002 quads) of the energy consumption. The commercial sector has a large share of HID sources, which constitute approximately 12 percent (47 TWh/yr or 0.16 quads) of the annual 391 TWh/yr of electricity use in this sector. The industrial sector also operates a large share of HID sources, which constitute 31 percent (33 TWh/yr or 0.11 quads) of this sectors 108 TWh/yr lighting-related energy usage. The primary lighting energy consumer in the other sector is HID, which accounts for 82 percent (46 TWh/yr or 0.16 quads) of the outdoor stationarys 56 TWh/yr lighting electricity use. Across all four sectors, HID sources consumed 125 TWh/yr (0.42 quads) in 2001. This amounts to approximately 17 percent of the annual electricity consumption for lighting in the United States. Figure 2.1.1 shows the total lighting energy consumption for the four sectors and illustrates the amounts of energy used by the various lighting technologies. From this figure, the significance of HID technology and its dominance in the other sector becomes further evident.

Other

Industrial Fluorescent HID Incandescent Commercial

Residential

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Source: Navigant Consulting, Inc. (NCI), U.S. Lighting Market Characterization Volume I: National Lighting Inventory and Energy Consumption Estimate, 2002.

Figure 2.1.1. Shares of Sectoral Energy Use by Lighting Technology Table 2.1.2 presents the estimated installed base of HID luminaires in the United States by technology and sector. HPS is the leading source in the outdoor stationary sector, while MH dominates in the commercial and industrial sectors. In the residential sector, MV is the most prevalent technology with nearly three times more installations than HPS. The total estimated

2-2

installed base of HID luminaires in the United States for 2001 was approximately 101 million units. Table 2.1.2. Installed Base of High-Intensity Discharge Luminaires in the United States
Sector number of units Residential Commercial Industrial Other Total Mercury Vapor 3,103,000 5,401,000 1,750,000 12,146,000 22,400,000 High-Pressure Sodium 1,204,000 5,814,000 2,694,000 34,347,000 44,058,000 Metal Halide N/A 19,378,000 10,706,000 4,726,000 34,809,000 Total 4,307,000 30,593,000 15,150,000 51,219,000 101,267,000 Source: NCI, U.S. Lighting Market Characterization Volume I: National Lighting Inventory and Energy Consumption Estimate, 2002. HID Light Source

Table 2.1.3 shows the electricity consumption of each lamp type as a percentage of total HID electricity consumption. HPS is prominent in the outdoor stationary sector, and MH is the largest electricity consumer in the commercial and industrial sectors. Overall, the residential sector accounts for only one-half of 1 percent of HID electricity consumption. Table 2.1.3. Distribution of Electricity Consumption by Sector of High-Intensity Discharge Lighting
HID Lamp Residential Commercial Sector % Industrial 3 4 20 27 Outdoor Stationary 9 24 4 36 Total 17 33 50 100

Mercury Vapor 0 5 High-Pressure Sodium 0 5 Metal Halide N/A 27 Total 0 37 Source: DOE, Buildings Energy Data Book 2008, table 5.6.4.

2.1.2

High-Intensity Dischar ge Lamp Shipment Estimates

Table 2.1.4 presents estimates of annual HID lamp shipments to the U.S. lighting market, broken down into the three primary HID lamp types. These data were provided by the National Electrical Manufacturers Association (NEMA), whose members are responsible for 100 percent of domestic HID lamp production and over 90 percent of national HID lamp sales.2 Table 2.1.4. Portion of Total NEMA High-Intensity Discharge Lamp Shipments
Year Mercury Vapor 32.12 24.73 23.62 21.30 19.92 17.44 15.77 HID Lamp % High-Pressure Sodium 38.34 45.05 44.22 44.91 43.98 41.86 42.65 Metal Halide 29.53 30.22 32.16 33.80 36.10 40.70 41.58

1990 1991 1992 1993 1994 1995 1996

2-3

Table 2.1.5. Portion (continued)

Year Mercury Vapor 12.24 9.21 8.08 5.73 8.02 7.29 6.83 7.35 7.01 7.37 6.37 5.49 HID Lamp % High-Pressure Sodium 41.61 40.13 37.72 36.62 35.49 35.56 34.85 33.59 36.69 32.50 31.60 32.85 Metal Halide 46.15 50.66 54.19 57.64 56.48 57.14 58.32 59.06 56.29 60.13 62.03 61.66

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

From 1990 to 2008, NEMA shipments of HID lamps nearly doubled with the largest shipment in 2005. In 1990, MV lamps represented 32 percent of the total HID lamp shipments. In 2008, MV only represented a mere 5 percent of total lamp shipments. Although there was a significant decrease in the proportion of lamp shipments, the actual amount of lamps shipped only reduced by 70 percent. This is due to the overall volume of HID lamps shipped increased from 1990 to 2008 by almost a factor of two. In 2003, DOE released the draft report, Draft Framework for Determination Analysis of Energy Conversation Standards for High-Intensity Discharge Lamps, and received comments. The DOE draft report stated that in 1990, the MV lamp shipments were 6.7 million units and in 2002, 2.9 million units were shipped. This represents a 57 percent decline in MV lamp shipments. Also in 2003, NEMA commented that [a]bsent any newly created markets, we do not agree with the presumption that the trend of declining mercury vapor lamp shipments has bottomed out. 4 During the near 20-year period of data analyzed, HPS lamps accounted on average for 38 percent of the total HID lamp shipments. In 1991, HPS represented the plurality of HID lamps shipped. Since that year, HPS has been on a decline in representing the total amount of HID lamps shipped. In 2008, HPS only represented 33 percent of the total HID lamps shipped. In terms of absolute shipments, HPS peaked in 1999 at 12.6 million units shipped. In 2008, only 10.9 million lamps were shipped. Of the three lamps in the HID family, MH lamps have grown in both percentage of total HID lamp shipments as well as absolute amount of lamps shipped. In 1990, MH lamps only represented 30 percent of the HID lamps shipped, slightly less than the percentage that MV lamps represented. Eighteen years later, MH lamps represent 62 percent of all the HID lamps shipped. This growth is further demonstrated in the volume of lamps shipped. In 1990, 5.7 million MH lamps were shipped compared with 2008, when 20.4 million lamps were shipped. This significant growth over the last 18 years stems from the commercialization of both the pulse-start quartz (PMH) and the ceramic (CMH) metal halide lamps.

2-4

DOE purchased a research report from the Freedonia Group regarding lamps. Section V of the report focuses on Electric Discharge Lamps. The Freedonia report provides both historical analysis as well as future analysis. Table 2.1.5 summarizes the estimates and projections in the Freedonia Group report. Table 2.1.6. Freedonia Group High-Intensity Discharge Lamp Shipments
HID Lamp 1998 2003 2008 2013 2018 Million Portion Million Portion Million Portion Million Portion Million Portion Units of Units of Units of Units of Units of Total Total Total Total Total % % % % % 17 30.36 16 19.28 19 16.67 10 6.67 5 2.44 24 42.86 39 46.99 50 43.86 70 46.67 100 48.78 15 56 26.79 100 28 83 33.73 100 45 114 39.47 100 70 150 46.67 100 100 205 48.78 100

Mercury Vapor High-Pressure Sodium Metal Halide Total

Although the absolute values (and thus the portions) are different from the NEMA data, the Freedonia Group data do reinforce the NEMA data. Both sets of data suggest that MV shipments are significantly curtailing, the MH has grown in considerable size since the mid1990s, and HPS represents a sizable yet relatively flat portion of the market. Figure 2.1.2 illustrates the annual shipments and market share by HID lamp type (MV, HPS, and MH). a It shows the obvious decline in MV shipments, contrasting with sales of HPS and MH. It also presents annual shipment data as a percent of total shipments in a given year. The decline in the market share of MV lamps is even more evident, as the MV lamp proportion of total shipments contrasts sharply with shipments of MH.

Sources: NEMA data provided to DOE 2009.

Figure 2.1.2. Proportion of Lamp Type in High-Intensity Discharge Market, 19902008

a

DOE prepared Figure 2.1.2 and Figure 2.1.3 from the data in Table 2.1.5.

2-5

National shipments of HPS lamps appear to have leveled off after a strong growth period in the early 1990s. HPS lamps are a mature technology, having been commercially available for more than 40 years. Similar to HPS lamps, MH lamp technology is mature, although this is the only type of HID lamp that has increased market share in the last 10 years. Industry continues to invest in improving MH technology, enabling it to continue to compete strongly in both the HID lamp markets and other lighting markets, such as those typically serviced by fluorescent sources (e.g., retail stores). Examples of these performance improvements (discussed in chapter 3 of this technical support document [TSD]) include pulse-start technology and the development of a ceramic arc tube. These and other improvements have lengthened the operational life and performance characteristics of MH technology, making it an attractive light source in many sectors and applications. Table 2.1.6 lists the domestic and import shipments of HID lamps, HPS represents 33 percent of the total shipments, MH represents 60 percent of the shipments, and MV represents 7 percent of the shipments. The data in Table 2.1.6 confirm the historic data shown in Table 2.1.4 and Figure 2.1.2. For the last year, 2002, of historical national sales data (Table 2.1.4), HPS represented 35 percent of national sales; MH represented 57 percent of national sales; and MV represented 8 percent of sales. Figure 2.1.2 depicted the historical shipment data for HID lamps. For the last year of available data (2008), MV represented about 5 percent (0 to 5 percent, Figure 2.1.2); HPS represented about 33 percent (5 to 38 percent, Figure 2.1.2); and MH represented 62 percent (38 to 100 percent, Figure 2.1.2) of the market. Table 2.1.7. 2008 NEMA Shipments by Wattage Grouping
Lamp Type 1149 W Mercury Vapor 37.68 High-Pressure Sodium 49.45 Probe-Start Quartz Metal Halide 0.48 Pulse-Start Quartz Metal Halide 8.28 Pulse-Start Ceramic Metal Halide 14.06 Source: NEMA Correspondence with DOE 2009 Wattage Range % 150500 W 58.75 44.49 52.11 9.93 3.11 > 501 W 3.58 6.06 11.41 0.62 0.00

Table 2.1.6 shows the lamp shipments by wattage bin for HID lamps. The bins are the standard wattage groupings of low (less than or equal to 149 W), medium (150500 W), and high (greater than or equal to 501 W). A majority of the MV lamps shipped were of medium wattage and a sizable amount of low-wattage MV lamps were shipped. The MV lamps commercially available affect the MV distribution; there are many lamps available at 175, 250, 400, and 450 W. In contrast, fewer options are available in the other wattages. An almost equal amount of low and medium HPS lamps were shipped in 2008. Virtually all of the HPS lamps shipped were 500 W or less. The typical applications HPS lamps are used dictate the distribution in the low and medium groups. Of the entire metal halide lamp family (probe, pulse, and ceramic) a majority of lamps shipped in 2008 are medium wattage probe-start lamps. Low and medium wattage pulse-start both quartz and ceramic comprise roughly 18 and 17 percent, respectively, of the entire MH market. The other significant amount of market share is the high-

2-6

wattage probe-start market. These values are for 2008; in the future, the probe-start MH values will probably not be as sizable because of current and pending legislation (Section 2.4). 2.1.3 Manufactur er s of High-Intensity Dischar ge Lamps, Luminair es, and Ballasts

NEMAs website lists manufacturers that constitute the NEMA membership for HID lamp manufacturing who are responsible for the entire domestic (U.S.) production of HID lamps.1 Like many of NEMAs members, these manufacturers may have non-domestic manufacturing facilities and thus bring their products into the United States as imports. NEMA companies also include members that manufacture HID ballasts. 5 Electronic ballasts for HID lamps are relatively new to the marketplace; 6 consequently, the list of manufacturers is shorter than the list established magnetic ballast technology. 7. All these manufacturers make luminaires for MH sources. Thirty-nine companies manufacture luminaires solely for MH sources, 42 companies manufacture luminaires for both HPS and MH sources, and 27 companies manufacture luminaires for all HID sources (HPS, MV, and MH). Some of these luminaires are similar in shape, design, or construction but have specific features to accommodate the various light sources. Manufacturers that make products for different light sources may not manufacture the same luminaire type for each source. The four largest lighting conglomerates in the United States are Acuity Lighting Brands, Cooper Lighting, Hubbell, and Philips. These manufacturers encompass numerous other companies and brands, and represent the majority of the domestic production of HID luminaires. 2.1.4 Distr ibution Channels

Figure 2.1.3 shows the distribution channels for the HID lamp market. The structure of this figure is relevant for all end-use sectors (residential, commercial, industrial, and other); however, the volume of product going through the various channels varies. The dashed lines in the figure represent distribution channels that contractors infrequently use. DOE adapted this illustration of HID lamp distribution channels from Sardinsky, 8 updating it using manufacturers and lighting specifiers literature to reflect the current HID lamp market.

2-7

HID Lamp Manufacturer

Manufacturer Representative
(1) Specialty Retailer (2) Local Retailer
Contractor

(3) Regional Retailer

Contractor

(4) Mail Order Distributor

Internet

(5) Electrical Distributor

Contractor

(6) National Accounts

Contractor

Consumer
Figure 2.1.3. High-Intensity Discharge Lamp Distribution Channels HID lamp manufacturers sell products through six categories of distributors. Specialty, local, and regional retailers primarily serve the residential sector. Mail-order and electrical distributor networks are the principal outlets used by the commercial, industrial, and public sector lighting markets. The national accounts serve all sectors. The specialty retailer is a broad distribution category typified by independently owned specialty lamp retailers. However, as an HID lamp distribution channel, it only represents roughly 10 percent of the total market for HID lamps. Local retailers are relatively large retail outlets, such as local hardware stores. This limited outlet for HID products primarily serves the residential sector. Although product availability is limited, some local and small contractors may opt to purchase HID lamps through these channels. Regional retailers are the larger chain stores that have multiple locations. Similar to the local retailers, they usually offer a limited selection of HID lamps tailored to the residential sector. Contractors have limited contact with this channel. Mail-order distribution is another major distribution channel. Although these distributors stock products used in all sectors, lamp availability through this channel is typically limited to high-volume items. Smaller businesses and contractors are the primary end-users for these distributors. However, the Internet is expanding the end-user base for these outlets and enabling them to adapt their offerings more quickly to end-user needs.

2-8

Large electrical-equipment distributors handle the majority of the HID lamp shipments, effectively operating as a wholesale clearinghouse for the HID lamp industry. Often, contractors operate as middlemen between end-users and distributors. Building-management companies, maintenance contractors, developers, municipalities, utilities, and related entities purchase lamps either directly from electrical-equipment distributors or through a contractor. The sixth major channel of distribution is through the national accounts. Lamp manufacturers sell some products through large national retailers, such as Wal-Mart, Sears, and big-box retailers, such as Home Depot and Lowes. Large maintenance companies and other national businesses may also purchase lamps directly from the manufacturers. However, product availability through these outlets is limited to high-volume (commodity) HID products. The advent of the Internet created new opportunities that could significantly impact all distribution channels and sectors. In response to the growing base of Internet users, manufacturers created new online databases that list product offerings and links to various distribution channels for purchase. Although the references primarily link to mail-order distribution houses, other links to walk-in retailers are also available. For end-users interested in accessing products, these new online sources and distributors offer an alternative to the traditional channels of large retailers and specialty stores. Many of these websites incorporate design centers that help end-users match lamps with their needs, much like a knowledgeable specifier or sales representative would do. The total impact on shipments for this industry is small at this time. 2.1.5 Decision Maker s

After initial installation of a luminaire, the purchase of replacement lamps becomes a maintenance issue. Therefore, in most cases, the luminaire purchasing decision dictates the subsequent replacement lamps purchased because there is very little interoperability of different lamp wattages or lamp types for the various HID ballasts. Specifiers who determine which light source to install are the electrical contractors, architects, lighting designers, electrical engineers, building owners, homeowners, building maintenance personnel, municipal officials, state officials (highway departments), and corporate officials who select HID luminaires for their properties. In addition, electrical distributors and manufacturers may also influence the choice of light sources. 2.1.5.1 Residential Decision Maker s

For the residential sector, the interaction between the builder (or homeowner) and the other stakeholders (i.e., designer, contractor, engineer, retailer, and electrical distributor) may change based on the budget and type of installation: new or retrofit. For both new and retrofit installations, the builder or homeowner typically acts as the primary specifier. The budget of the project generally determines the level of involvement of the various stakeholders. The role and decision-making power of the designer and contractor typically grows with the escalating cost of the project. However, as project budgets become tighter, direct involvement with retailers, electrical distributors, and the Internet gains prominence.

2-9

2.1.5.2

Commer cial, Industr ial, and Public Sector Lighting Decision Maker s

For the commercial, industrial, and public sectors, the interaction among decision makers is more complex than in the residential sector. The key decision makers in the commercial and industrial sectors are: (1) the owner (or developer); (2) the building manager; (3) the design team (i.e., architect, engineer, lighting designer); and (4) electrical contractor. Local utilities, municipalities, and state highway agencies install and maintain many of the luminaires used in the public sector. Thus, these owners of HID lighting systems represent the most influential figures in the public sector decision-making process. 2.2 APPLICATIONS FOR HIGH-INTENSITY DISCHARGE LAMPS

The following sections discuss where and how the residential, commercial, industrial, and public sectors use HID lamps. 2.2.1 Residential Applications

In comparison to other lamp technologies, such as incandescent and fluorescent, HID light sources occupy a very small niche in the residential sector. Of the lighting electricity consumed in this sector in 2001, 90 percent was consumed by incandescent sources, over 9 percent was consumed by fluorescent sources, and less than 1 percent was consumed by HID sources.3 Due to their electrical (long warm-up time, lack of quick re-strike) and photometric (produce too much light, poor color quality compared to incandescent lamps) characteristics, HID lamps are not well suited for most residential lighting applications. In this sector, HID lighting is used primarily in landscape and security-lighting applications. For security applications, MV technology is popular because both the lamps and luminaires have low initial cost. The extremely long life of the MV lamp is an added bonus. These applications are typically areas in which lighting is needed for safety and visual assessment of the property to make the homeowner feel secure. The oldest of the three HID source types, MV was the first HID light source to enter the residential market. Many residential customers associate MV with residential security products, so the other HID sources have gained little penetration in the market. Initial price is also a major factor in the residential market decision-making process; the high initial costs of HPS and MH are another reason the MV dominates this application. In some rural areas, home and farm owners may lease a security light from an electric utility or a rural cooperative. This lease may carry a fixed tariff that does not measure actual electricity consumption. These luminaires have been primarily MV sources, but utilities now offer a wider array of sources, including HPS and MH. For landscape lighting, the capability of MV lamps to enhance the color of the foliage has created a niche application. The spectral power distribution (SPD) of MV light results in greater saturation of the shorter wavelength colors (from green to blue). Since most foliage is green, the light from the MV lamp gives it a richer appearance than the lumen output of the lamp would suggest, while creating greater contrast with its surrounding objects by muting the longer-wavelength colors (from red to yellow). MH lamps offer comparable performance while
2-10

maintaining a much higher efficacy level; however, they have a higher initial cost and shorter operating life than MV lamps. MH landscape lighting has increased in recent years with the commercialization of numerous reflector (parabolic aluminized reflector [PAR], multi-faceted reflector [MR]) lamp types, offered in a variety of wattage and light output ratings. 2.2.2 Commer cial Applications

Of the lighting electricity consumed in this sector in 2001, 32 percent was consumed by incandescent sources, 56 percent by fluorescent sources, and less than 12 percent by HID sources.3 In spite of their potential for significant energy savings, HID lamps are not widely used in interior applications in the commercial sector for several reasons. The primary reason for HPS, MV, and non-ceramic MH is low color-rendering index (CRI). Other limiting factors include long warm-up and re-strike times, ballast noise, lamp size, and limited compatibility with lighting controls. Recent improvements in MH technology incorporating pulse-start systems and ceramic arc tube variants made the MH lamp more attractive than incandescent and fluorescent sources in certain applications (e.g., retail and hospitality spaces). Despite these improvements, end-of-life color shift is a problem associated with MH technology. Reflectorized MH (PAR, T-shaped, AR-type) lamps are used in accent lighting applications (e.g., track lighting) often to meet energy codes. Most MH lamps are large in size, so these lamps are typically used in downlights or low-bay style luminaires lighting commercial spaces. 2.2.3 Industr ial Applications

Of the lighting electricity consumed in this sector in 2008, 2 percent was consumed by incandescent sources, 67 percent by fluorescent sources, and less than 31 percent by HID sources.3 HID lamps are popular in the industrial sector because the lamps offer a large lumen package (i.e., very high light output) with a relatively compact form factor. The maximum wattage for high-output fluorescent lamps is 200 W, where HID lamps in this application can be 1000 W or greater, with light ouput exceeding 100,000 lumens. Therefore, multiple fluorescent lamps are needed in a luminaire to equal the flux of the high-wattage HID sources. Unlike linear fluorescent lamps, HID lamps are point sources (i.e., the actual light source is a comparatively small plasma arc within the lamp envelope). Point sources like HID lamps allow for luminaires with good optical control, which can more efficiently deliver light to the work area from greater mounting heights. HID sources are best operated on long operating cycles (e.g., over 12 hours per start), which are typical for industrial applications. MH sources are typically specified in industrial applications where white light with adequate color rendering (CRI greater than 50) is required. 2.2.4 Other (Outdoor Public Sector) Applications

HID lamps dominate public-sector lighting installations. Public-sector installations account for roughly 8 percent of the total electricity consumed by all lighting in the United States, and HID sources represent 87 percent, incandescent 12 percent, and fluorescent only 1 percent.3

2-11

For street and roadway lighting, HPS represents the vast majority of this market, having replaced most MV installations. A small percentage of installations use MH, MV, or LPS (though LPS is not an HID source). Often, luminaires used in these applications are equipped with photosensors that turn the luminaires on at dusk and off at dawn. MH lamps are now replacing some HPS lamps that previously replaced MV lamps. Although MH sources are less efficacious than HPS, they offer superior color rendition and appear as a white-light source compared to HPS. In some area and street lighting applications, utilities turned down requests by municipalities for MH sources because maintenance costs were too high compared to those of HPS sources. The pulse-start metal halide (PMH) lamp addresses this problem, increasing lamp life as compared to the standard probe-start MH. However, the life for pulse-start MH is still shorter than typical HPS and MV lamps. Meanwhile, lamp manufacturers are addressing HPS color issues by trying to broaden the lamp SPD. The whitelight HPS products available today offer better CRI and CCT, but these improvements come at the cost of shorter lamp life and lower efficacy. MH and white-light HPS may be considered competitors in some applications, but lighting specifiers will typically specify MH for any application where color rendering or appearance are important. 2.3 CLASSIFICATION OF HIGH-INTENSITY DISCHARGE LUMINAIRES

Lamps are intended to be used inside fixtures, also referred to as luminaires. To describe how the lamps are to be used, the luminaires need to be classified. Luminaire classifications help describe, organize, and catalog the HID lamp market for easy identification by specifiers, manufacturers, and consumers. Virtually all lighting industry luminaire classifications focus on the distribution of the luminaire, which varies with lamp type and fixture design. For instance, a classification system developed by the Commission Internationale de lEclairage (CIE) classifies luminaires by the proportion of upward and downward light emission. This system most aptly applies to suspended luminaires where fluorescent dominates the market. A system developed by NEMA classifies luminaires by circular or oval distribution defined in terms of field angle. This system most aptly applies to floodlight luminaires where HID sources dominate the market. A third system developed by the Illuminating Engineering Society of North America (IESNA) classifies luminaires by distribution for desired patterns of illumination. This system is for area/roadway luminaires where HID sources dominate. Since the existing classifications systems focus on luminaire distribution, DOE developed a system that takes into account the application, construction, and photometrics of luminaires that most typically use HID sources (non-HID sources typical of this luminaire type will be addressed as needed). 2.3.1 Pr oposed High-Intensity Dischar ge Luminair e Classification System

In the following sections, DOE categorizes luminaires by selecting their most appropriate and distinguishing feature, whether in terms of their source, mounting, construction, or application. The three primary groups are indoor (includes both industrial and commercial), outdoor, and specialty (can be indoor or outdoor) applications. For applications, DOE adapted the classifications and their descriptions from luminaire manufacturers literature, industry standard practice, and the IESNA Lighting Handbook. 9 Although specialty-application luminaires do not account for a large portion of the general-service HID market, DOE presents
2-12

these to complete the classification of HID light-source applications. This classification system constitutes the product applications that DOE used in this analysis. 2.3.1.1 Pr oduct Applications: Indoor

Table 2.3.1 summarizes the indoor application classifications, following which are detailed descriptions on each of these product applications. Table 2.3.1. Classification for Indoor Applications
Summary Description Luminaires providing directional lighting to emphasize a particular object or surface feature or to draw attention to a part of the field of view. Down-Lighting Small direct-lighting units that direct the light downward. They may be recessed or surface-mounted. Suspended Luminaires that are pendant-mounted via chain or cables or suspended from a post (stem). They may be direct, indirect, or direct/indirect combination luminaires. High-Bay Luminaires providing general illumination, primarily in industrial applications, where the floor-to-ceiling height is greater than 25 feet. Low-Bay Luminaires providing general illumination, primarily in industrial applications, where floor-to-ceiling height is equal to or less than 25 feet. Source: IESNA, The IESNA Lighting Handbook: Reference and Application, 2000. Classification Accent

Accent Lighting Luminaires that provide light from a preferred direction (directional lighting) to emphasize a particular object, feature, or to draw attention to a part of the field of view. A point-source technology (e.g., incandescent) is needed to create the directional lighting, and large-source technologies (e.g., fluorescent) that create diffuse lighting are the least effective for this application. The luminaires can be either recessed or surface-mounted. Track lighting, a non-fixed system with track heads (housing for the lamp and typically the ballast), is often used for accent lighting. Other non-track luminaires utilize some means of articulation for proper aiming. Although the small point source and high CRI of the new ceramic MH make it a competitive addition to this class, the light sources in this class are typically incandescent (e.g., halogen MR and PAR lamps). Ceramic MH sources are often used in applications where high CRI, significant illumination, small luminaires are needed, and are more often being installed to comply with stricter energy codes. Down-Lighting Luminaires that provide general lighting in residential and commercial spaces. These luminaires either have a reflector designed around a point source or house a directional light source (i.e., a PAR or MR lamp). HID down-lights have not penetrated the residential market. HID down-lights are commonly used in applications in which the luminaire will operate for long periods of time, and color appearance/rendering is not a high priority. Spaces with a mounting height over 9 feet are another potential application for down-lights with HID sources. Canopy lightsluminaires typically found under gas station canopies or used in parking garagesalmost always are one of the three HID sources. Ceramic MH and pulse-start ballasts enabled the use of low to mid-wattage MH lamps in interior applications; however, incandescent and fluorescent sources still dominate interior down-light applications. Suspended Lighting Luminaires range from decorative pendants/chandeliers to pendants used in office lighting. These luminaires emit light directly, indirectly, or a combination. Ornamental or decorative chandeliers typically operate incandescent sources. A
2-13

small exception exists for large ornamental luminaires that typically have the option for incandescent, fluorescent, or MH sources. Less decorative and more functional luminaires typically use fluorescent or compact fluorescent lamp (CFL) sources. High-Bay Lighting Luminaires in spaces with high ceilings (greater than 25 feet) that tend to be industrial applications. Typically, the luminaire spacing-to-mounting-height (S/MH) is less than or equal to 1.0 for these applications. Depending on the ceiling type and construction, these luminaires are either surface or pendant-mounted. HID sources previously dominated this market; in recent years, fluorescent and CFL high-bay luminaires (known as high-intensity fluorescent [HIF]) are becoming more common. Unlike HID, fluorescent technology allows for more control flexibility via easier dimming or switching groups of lamps. Low-Bay Lighting Luminaires used for lighting spaces with ceiling heights less than 25 feet. Applications include industrial, commercial, and retail spaces. The luminaire S/MH is typically greater than 1.0. These luminaires often appear similar to high-bay luminaires, but usually have different refractors and reflectors. HID sources represent a majority of the luminaires in this category, with fluorescent sources filling in the rest of the market. 2.3.1.2 Pr oduct Applications: Outdoor

Table 2.3.2 summarizes the classifications for outdoor applications. The IESNA defines seven types of classifications for outdoor luminaires: floodlight, sports light, area/roadway lighting, pathway, parking garage, security, and landscape. Detailed descriptions and further information on each of these product applications follow Table 2.3.2. Table 2.3.2. Classification for Outdoor Applications
Summary Description Luminaires for building lighting and other special applications. Luminaires designed to light playing fields. Luminaires designed to produce reasonably uniform illuminance on streets and roadways. Pathway Luminaires designed for lighting pedestrian pathways. Parking Garage Luminaires designed to light parking garages and lots. Security Outdoor luminaires typically used visually to assess an area. Landscape Luminaires designed to light buildings, planting, water features, and walkways. Source: IESNA, The IESNA Lighting Handbook: Reference and Application, 2000. Classification Floodlight Sports Lighting Area/Roadway Lighting

Floodlight Enclosed lensed luminaire with permanent reflector. These luminaires often light the exterior facades of buildings or large parking lots from a pole. It is not uncommon for the light distribution to be a square or rectangular pattern for floodlightsHID luminaires completely dominate this market. Sports Lighting Luminaires with very narrow light distributions for lighting playing fields/surfaces from a great distance (often 100+ feet) either from the side or from above. This category is entirely comprised of HID sources. Incandescent sources could not be used to provide the amount of light needed from the distance at which these luminaires are often located. Fluorescent (linear or compact) sources cannot be used because reflector design makes it impractical. These luminaires have devices to prevent glare and stray light. This is a specialty

2-14

category of outdoor luminaires with a small number of applicable light sources and only a handful of luminaire manufacturers making products. Area/Roadway Lighting Luminaires designed for illuminating major driving surfaces (e.g., streets, roadways, parking lots). HPS and MH are the main sources for this application. A very small percentage of the luminaires installed across the country use LPS (not an HID source) lamps. These luminaires are classified by photometric distribution and by appearance/mounting method. The two classifications are neither exclusive nor concurrent. A luminaire might have a specific appearance and mounting and yet be capable of numerous photometric distributions. Luminaires in this category are typically mounted above 15 feet and up to about 50 feet. Luminaires mounted above 50 feet are known as high-mast and produce the same photometric distributions as luminaires mounted at a lower height. The most typical luminaire in this category is known as the cobra head. The luminaire resembles the head of a snake and, more often than not, the arm that supports the head curves over the roadway. This allows the pole to be located in a right of way or off the road, while positioning the luminaire directly over the surface being lighted. A majority of the cobra head installations across the country contain HPS sources. Architectural roadway luminaires incorporate aesthetic design features, materials and finishes not typically used in more common cobra head fixtures. Pathway Luminaires designed to illuminate surfaces used by pedestrians. These luminaires are typically of two types: overhead or ground level. The overhead luminaires are often similar to street and roadway luminaires but on a pedestrian-level scale with shorter pole heights. Typically, the overall height of the luminaire and pole does not exceed 15 feet. Luminaires mounted closer to ground level use low-wattage (under 150 W) sources. The stereotypical pathway source is the bollard. This is a round or square luminaire between 30 and 60 inches tall, made of concrete, aluminum, or steel. HID sources represent a large portion of the pathway luminaire market, but luminaires for CFL and incandescent sources are also used. Recently, luminaires with light emitting diodes (LEDs) have entered this market as well. Pathway luminaires are most prone to vandalism or accidental breakage from snowplows or other vehicles because the luminaire is located near the ground. Parking Garage Luminaires are typically either surface-mounted or suspended near the ceiling. Gas station canopy luminaires often fall in this category. These produce light over a very wide area compared to the distance they are mounted above the surface. HID sources comprise a majority of the installations, followed by fluorescent sources. Recently, luminaires with LEDs have entered this market as well. Security Luminaires specifically installed to provide a sense of safety or to illuminate a field of view for safety/security reasons. Luminaires in this category include recessed step lights for lighting steps, wall-mounted luminaires at entrances to facility access into the building, and wall packs. Step lights are typically recessed luminaires mounted approximately 2 feet above a stair or walkway. Typical sources include CFL or incandescent sources; LEDs have entered this market as well. HID sources are sometimes used, but this luminaire typically needs both a lowwattage source and a small-size source. Wall packs are the most common security light. These luminaires are mounted on the exterior facades of buildings with the purpose of lighting the area for safety. HID sources are prevalent for wall packs. For most other security luminaires, one source does not dominate the market.

2-15

Landscape Luminaires light foliage, parts of building facades, water features, exterior statues, etc. Unlike floodlights, landscape luminaires are used to illuminate discrete landscape and architectural features. Low-voltage halogen represents the bulk of this market. HID sources excluding HPS are also used. Of the HID sources, MH is most often used because its compact lamp types (T6, PAR, MR) allow good optical control. For specific types of foliage, MV is the ideal source because of the spectrum of light produced by the source. The largest drawback for HID sources in landscape lighting is lack of interchangeability. HID sources use wattage-specific ballasts, which precludes using lamps with different wattages and light output. If lighting requirements change over time (e.g., the lighted object grows larger), the original fixture and lamp type may no longer meet the application needs. 2.3.1.3 Pr oduct Applications: Specialty

HID luminaires in specialty applications often employ lamp shapes and bases that are not typically associated with general-service HID lamps discussed in the indoor and outdoor application groupings. Therefore, some lamps used in specialty applications would be exempt from standards. Although the impact from this product application would be very small, DOE included them in this report for completeness. Table 2.3.3 presents the of specialty-product applications. Table 2.3.3. Classification for Specialty Applications
Classification Summary Description

Theatrical Luminaires Reprographics

Luminaires designed with precise optical control and maximum flexibility. These are niche products used in the stage, studio, and other related fields. Lamps used in reprography, such as photography and photocopying.

2.4

REGULATORY AND NON-REGULATORY PROGRAMS

As part of this analysis, DOE reviews all voluntary and compulsory programs that relate to HID technology. This review provides DOE with a greater understanding of the marketplace in which HID lamp, ballast, and luminaire manufacturers operate as well as what programs and trends may influence the future of the market. 2.4.1 Regulator y Pr ogr ams and Legislation Impacting High-Intensity Dischar ge Lamps

DOE reviewed Federal and State regulatory programs and legislation that impact HID technology, and applications where HID has a large market share. DOE acknowledges that not all programs and legislation affecting HID technology are discussed below but that its review addresses the major federal examples and, where possible, representative state programs are listed as well. 2.4.1.1 Feder al Regulator y Pr ogr ams and Legislation

In 2005, the president signed into law the Energy Policy Act of 2005 (EPAct 2005), which amended and expanded parts of the EPAct of 1992. This regulation included the

2-16

requirement that ballasts for MV lamps b cannot be imported or manufactured effective January 1, 2008. 10 Previously, the Energy Policy Act of 1992 (EPAct 1992) required states to adopt an energy code for buildings, which includes energy consumption guidelines for lighting. On July 15, 2002, DOE issued a Determination that American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) Standard 90.1-1999 was more stringent than 90.1-1989. 11 As a result, each state must certify that it reviewed and updated the provisions of its commercial building code regarding energy efficiency to meet or exceed Standard 90.1-1999 by July 15, 2004, for any building under the means of Section 303(2) of EPAct, as amended. EPAct 2005 provides substantial property tax credits for commercial properties that exceed the allowed lighting power density (LPD) in ASHRAE Standard 90.1-2001 by at least 25 percent. Further, the EPAct 2005 required new federal buildings constructed to be designed 30 percent less than the current version of ASHRAE Standard 90.1 (90.1-2004 at time of enactment). Both the current 90.1-2010 Standard as well as Californias Energy Efficiency Standards for Residential and Nonresidential Buildings, Title 24, Part 6 (Title 24) have stated minimum efficacy values for lighting exterior building grounds. All luminaires for exterior building grounds that operate at greater than 100 W must contain lamps having a minimum efficacy of 60 lumens per watt (LPW), unless the luminaire is controlled by a motion sensor or qualifies for one of the exceptions. This minimum efficacy value affects HID sources and, in some cases, prevents them from being used in non-residential applications. In addition to the minimum efficacy values, energy codes regulate energy use via the power density (W per square or linear foot, depending on the application) of luminaire installations. Prescriptive design requirements allow for a variety of energy efficient sources to be used without banning or restricting technologies outright. As the power density values decrease, the demand for more efficient lighting increases. A number of states and the Federal government set requirements for MH ballasts of specific wattages. The Energy Independence Security Act of 2007 (EISA 2007) requires ballast efficiency (Pin/Pout) values for MH ballasts between 150 and 500 W of 88 percent for pulse-start, 94 percent for probe-start, 90 percent for nonpulse-start electronic ballasts operating lamps between 150 and 250 W; and 92 percent for nonpulse-start electronic ballasts operating lamps between 250 and 500 W. EISA 2007 allows a one-time exception to California. In 1999, the U.S. Environmental Protection Agency (EPA) required recycling (based on volume of use) and proper disposal of lamps containing mercury. All HID lamps contain some amount of mercury. 2.4.1.2 Dr aft Legislation

During 2009, different members of Congress drafted legislation focusing on energy efficiency, with aspects of this draft legislation affecting HID lamps and luminaires primarily using HID lamps. Rep. Jane Harman (D-CA36) introduced H.R. 1732: Outdoor Lighting Efficiency Act that was referred to Committee. The requirements in H.R. 1732 were later
b

Some ballasts for special applications are exempted from the ban of importation or manufacture.

2-17

adopted in H.R. 2454: American Clean Energy and Security Act of 2009, sponsored by by Rep. Henry Waxman (D-CA30), and passed by the House on June 26, 2009. This bill H.R. 2454 had requirements for outdoor lighting: 1. Each outdoor luminaire manufactured on or after January 1, 2016 shall: a. have an initial luminaire efficacy of at least 50 LPW; and b. be designed to use a light source with a lumen maintenance, calculated as mean rated lumens divided by initial lumens, of at least 0.6. 2. Each outdoor luminaire manufactured on or after January 1, 2018 shall: a. have an initial luminaire efficacy of at least 70 LPW; and b. be designed to use a light source with a lumen maintenance, calculated as mean rated lumens divided by initial lumens, of at least 0.6. 3. In addition to the requirements of paragraphs (1) through (3), each outdoor luminaire manufactured on or after January 1, 2016, shall have the capability of producing at least two different light levels, including 100 percent and 60 percent of full lamp output as tested with the maximum rated lamp power UL1598 or the manufacturers maximum specified for the luminaire under test. Outdoor luminaires used for roadway lighting applications shall be exempt the two light level requirement. Other requirements in the bill focus on outdoor high light output lamps: c Each outdoor high light output lamp manufactured on or after January 1, 2017 shall have a lighting efficiency of at least 45 lumens per watt. On March 2, 2010, Senator Jeff Bingaman (D-NM) introduced the National Energy Efficiency Enhancement Act of 2010. The bill was cosponsored by Evan Bayh (D-IN), Thad Cochran (R-MS), Byron Dorgan (D-ND), Mary Landrieu (D-LA), Robert Menendez (D-NJ), Lisa Murkowski (R-AK), Debbie Ann Stabenow (D-MI). u The main focus of the bill was to improvThe draft legislation includes requirements for outdoor pole-mounted lighting and MV lamps - A general purpose mercury vapor lamp shall not be manufactured on or after January 1, 2016. DOEs review of legislation affecting HID technology indicated three major approaches: regulating lamp efficacy, regulating lumen maintenance, and requiring multi-state operation of a light source. The proposed outdoor lighting bill bans MV lamps outright because they are the most inefficient HID source. H.R. 2454 sets the bar for a class of lamps at 45 LPW, effectively eliminating most MV sources. The outdoor lighting bill set minimum luminaire efficacy values, calculated as the amount of light leaving the luminaire (measured in lumens) divided by the input power (watts) supplied to the luminaire. For HID lighting, fixture efficacy can be calculated by

This is not an industry term. The bill defines the term as a lamp that (a) has a rated lumen output not less than 2601 lumens; (b) is capable of being operated at a voltage of not less than 110 volts and not greater than 300 volts, or driven at a constant current of 6.6 amperes; (c) is not a Parabolic Aluminized Reflector Lamp; and (d) is not a Jtype double-ended (T-3) halogen quartz lamp, utilizing R-7S bases that is manufactured before January 1, 2015.

2-18

multiplying the fixture efficiency by the lamp/ballast efficacy. d Chapter 3 will include more information about lamp/ballast efficacy. However, typical parking lot and area luminaires (polemounted luminaires that comprise much of exterior lighting) typically have downward fixture efficiencies e between 54.8 to 88.7 percent with an average of 75.3 percent. 12 The minimum lamp/ballast efficiacy can be calculated by dividing the required 50 LPW fixture efficacy by the typical luminaire efficiency of 75 percent. If the outdoor lighting bill is enacted and the fixture efficiency stays the average, then the lamp/ballast efficacy would have to be 66.6 LPW to meet the required 50 LPW fixture efficacy. Again under that legislation when the minimum fixture efficacy increased to 70 LPW, the minimum lamp/ballast efficacy would 93.3 LPW assuming that the fixture efficiency stayed constant. 2.4.1.3 DOE Background Studies

In 2004, DOE published High-Intensity Discharge Lamps Analysis of Potential Energy Savings (Docket EE-DET-03-001). The scope of the document was HID lamps but focused on an in-depth analysis of MV lamps. While developing and after publishing the document, DOE received comments on the draft document as follows: In 2003, a representative of California Department of Transportation (Caltrans) stated, The affect on Caltrans if mercury vapor ceased to be an option would be minimal. At this time, we are installing no new MV street lighting fixtures, and have replaced most of those that were installed 30 to 40 years ago with HPS. 13 In 2003, the American Council for an Energy-Efficient Economy (ACEEE) stated, ACEEE agrees that it makes sense to establish a minimum efficacy standard that eliminates mercury vapor lamps for many, if not all, applicationsIt should be noted that some states are adopting even more stringent codes resulting in rapid adoption of pulsestart metal halide in new installations ACEEE believes that the Department should explore the potential for setting minimum efficacy requirements for HID that would effectively eliminate probe-start metal halide lamps greater than 150 W-the best high wattage probe-start lamps (i.e., 1000 W lamps) can match pulse-start efficacys and would not be affected. Pulse-start metal halide lamps offer immediate energy savings relative to probe-start lamps and offer the potential for even greater savings as electronic ballasts become more widely available. The technical, life cycle cost, and energy savings analyses should include the substitution of pulse-start metal halide for both mercury vapor and probe-start metal halide lamps to determine the cost and savings associated with a higher efficacy requirement. 14 In 2004, independent lighting designer Jim Benya of Benya Lighting Design stated: In my opinion, banning mercury vapor technology is warranted. 15

Fixture efficiency is the percentage of light emitted by the luminaire (fixture) divided by the amount of light generated by the lamp. e Downward efficiency is the percentage of light emitted by the luminaire between 0 and 90 . It is typically reported for area, parking, and roadway luminaires. Although additional light may be emitted above 90 , it provides no value to lighting the intended surfaces.

2-19

In 2005, NEMA stated that it would oppose any DOE rulemaking that imposed a lumens-per-watt standard on HID lamps such that a formal, regular reporting of measurements of metal halide and high pressure sodium lamp efficacies would be required to prove that no lamps with MV-type efficacies are being placed on the market. Such reporting would represent a gross administrative overkill to achieve the objective. 16 In 2005, ACEEE stated that it is disappointed that DOE is not addressing other savings opportunities from HID lamps, specifically the savings available from replacing probestart metal halide lamps with pulse-start metal halide technology. However, at this time several states are developing standards on these metal halide products, so we recommend that DOE let states take the lead on these products rather than doing further work in ths area at the present time. 17

In summary, DOE received no comments advocating for MV lamps and comments to set pulse-start MH as the efficacy standard. 2.4.1.4 State Regulator y Pr ogr ams

Effective January 1, 2008, MH luminaires with lamps in all operating positions ranging from 150 to 500 W cannot contain a probe-start ballast in Arizona, Oregon, or Washington. New York and Rhode Island have similar laws effective January 1, 2008. In New York and Rhode Island, only luminaires with lamps operating in a vertical position ( 15 degrees of vertical) shall not contain probe-start ballasts. 18 Under Title 20, California has an appliance efficiency program and requirements in place regarding metal halide luminaires (fixtures) since 2006 (Table 2.4.1). Table 2.4.1. California Metal Halide Fixture Requirements
General Title 20 Enactments Milestone Date Description 1/1/06 Luminaires with lamps oriented vertically with the base up at rated wattages 150500 W shall not contain a probe-start MH ballast. 1/1/08 Luminaires with lamps oriented vertically with the base down at rated wattages 150500 W shall not contain a probe-start MH ballast. Luminaires rated wattages 150500 W shall not contain a probe-start MH ballast. Luminaires rated wattages 150500 W shall not contain a metal halide ballast with a minimum ballast efficiency of 88 percent. Exceptions include: Luminaires with electronic ballasts that operate at 480 V. Luminaires rated at 150 W, rated for use in wet locations per NEC and contain a ballast rated to operate at ambient air temperature above 50 F. 1/1/10 Metal halide luminaires rated partially within the range of 150 to 5000 W shall not have probe-start ballasts and comply with Section 1605.3(n)(2)(A) as applicable.

2-20

Table 2.4.2. California (continued)

1605.3(n)(2)(A) Indoor metal halide luminaires manufactured on or after January 1, 2010 shall comply with at least one option of Section 1605.3(n)(2)(B). Indoor metal halide luminaires manufactured on or after January 1, 2015 shall comply with Section 1605.3(n)(2)(B)4 and shall also comply with at least one other option of Section 1605.3(n)(2)(B). Outdoor metal halide luminaire manufactured on or after January 1, 2010 may comply with Section 1605.3(n)(2)(B)3 and shall comply with at least on other option of Section 1605.3(n)(2)(B). A minimum ballast efficiency of 90 percent for 150250 W lamps or 92 percent for 251 500 W lamps. A minimum ballast efficiency of 88 percent and occupant sensors that are integral control as defined in Section 1602(n) of this Article, shipped with the factory default setting to reduce lamp power automatically through dimming by a minimum of 40 percent within 30 minutes or less after an area has been vacated. A minimum ballast efficiency of 88 percent and automatic daylight integral control as defined in Section 1602(n) of this Article, shipped with the factory default setting to reduce lamp power automatically through dimming by a minimum of 40 percent. A minimum ballast efficiency of 88 percent and a relamping rated wattage within only one of the four wattage bins specified. The luminaire shall be able to operate lamps within only one of the four wattage bins and shall not be rated for any lamp wattage outside of that wattage bin. The luminaire shall have a permanent, pre-printed factory-installed label that states the relamping rated wattage: 150160, 200215, 290335, or 336500. When operating 336500 W lamps, the luminaire shall be prepackaged and sold together with at least one lamp per socket having a minimum lamp efficacy of 80 lumens per watt, based on published mean lumens and rated lamp power (watts). The following metal halide lighting systems shall not have probe-start ballasts and are not required to meet the minimum ballast efficiency requirements: Luminaires that use regulated lag ballasts; Luminaires that use electronic ballasts which operate 480 V; or Luminaires that meet all three of the following requirements: Rated for use only with 150 W lamps, and Rated for use in wet locations per NEC, and Contain a ballast rated to operate at ambient air temperature above 50 F.

1605.3(n)(2)(B)

Exceptions to 1605.3(n)(2)

The effect of Californias Title 20 brings the state appliance requirements in line with the Federal requirements. The rules effective in 2010 push consumers towards electronic ballasts. The efficiencies in 1605.3(n)(2)(B) are the ballast efficiencies in EISA 2007 for nonpulse-start electronic ballasts. Dimming requirements (reduce power to 40 percent) per 1605.3(n)(2)(B) and can really only be achieved via electronic ballasts. It is possible with magnetic ballasts, but inefficient, more expensive, and harder to compete with electronic ballasts. Finally, Californias requirements of 80 mean LPW push consumers to better lamps. Better lamp lumen maintenance can be achieved with electronic ballasts than with magnetic ballasts, but also needs a better lamp to actually achieve the required system efficacy. Another example of an efficacy requirement indirectly affecting HID sources is the Minnesota Statute 216C.19 that states, [n]o new highway, street or parking lot lighting may be installed in violation of these rules. Existing lighting equipment, excluding roadway sign lighting, with lamps with initial efficiencies less than 70 lumens per watt must be replaced when worn out with light sources using lamps with initial efficiencies of at least 70 lumens per watt. Although the statute is technology neutral, it effectively excludes incandescent halogen lamps and virtually all CFL sources. This statute focuses on roadway lighting and affects applicable HID sources. MV lamps could not be

2-21

installed in any application under this statute. Several states and a number of municipalities passed dark-sky ordinances aimed at reducing light pollution. While most focus on eliminating light pollution by addressing luminaire type and shielding, some ordinances (e.g., Arizona) specifically address the light source. Often, these municipalities are trying to ban a certain portion of the visible light spectrum. Astronomical observatories and coastal areas can be negatively impacted by certain spectral content that affects nighttime sky observation and animal habitats. 19 However, regulatory authority is not consistent; some ordinances cover only the lighting owned by the regulatory entity (e.g., state or municipal), while others cover privately owned lighting as well. Examples of source-specific state requirements are as follows: Arizonas statute states, No new mercury vapor outdoor light fixtures shall be installed after the effective date of this section. No replacement equipment other than bulbs for mercury vapor lighting fixtures shall be sold in the state after January 1, 1991 and the use of mercury vapor light fixtures is prohibited after January 1, 2011. 20 The City of San Jos, California limits restricts source type because of the local observatory. The use of low-pressure (LPS) sodium lighting for outdoor, unroofed areas shall be required for all private development in the City of San Jos as a condition of approval on all Land Use Development Permits. 21

In addition to dark-sky ordinances, numerous state departments of transportation have specifications that focus on HID technology. In addition to their specifications, these departments and other municipalities have prequalified lighting-procurement programs that specifically govern HID sources. For example, the Kansas Department of Transportation has a list of prequalified lighting systems that specifically list MH and HPS technology. 22 Although this is not an outright regulatory program, it is an example of municipal action that directly affects HID technology. The Model-Lighting Ordinance (MLO) is a joint effort by IESNA and the International Dark-Sky Association (IDA) to provide a sample code that municipalities can adopt. The MLO pertains to outdoor lighting only. In contrast to the major energy codes, the MLO recommends restricting light output in terms of initial lumens for environmental zones. Where ASHRAE Standard 90.1 and California Title 24 Standard regulate installed power density for the different environmental zones, the MLO gives a total lumen value for the zones. At the time of this publication, the MLO had not been completed by the IESNA/IDA Joint Task Force. Therefore, no actual values were provided. California Title 24 requires spaces with two or more luminaires lighting building facades, parking lots, garages, non-sales canopies, and all outdoor sales areas to reduce the lighting power by 50 percent when the lighting is not needed. Compliance with this code can be achieved via continuous dimming and hi-lo ballasts (stepped dimming ballasts) as well as switching every other luminaire.

2-22

After an incident in a school in Oregon where teachers were injured by burns to their eyes (via excessive ultraviolet radiation exposure) by a damaged MH lamp, the State of Oregon mandated that all schools replace MH and MV lamps with self-extinguishing versions. 23 2.4.2 Non-Regulator y Pr ogr ams Affecting High-Intensity Dischar ge Lamps

DOEs review of non-regulatory programs promoting higher-efficiency HID sources found that there are four general types of programs that are either prescriptive or performancebased. A prescriptive program would provide an incentive for using an efficient technology (with the assumption that with the equipment installed, energy will be saved). A performance-based program would provide an incentive for saving energy with little restrictions on how those savings were achieved. The first type of program promotes energy efficiency through sponsored retrofits, either on a prescriptive (technology) basis or on a minimum efficacy (LPW threshold) basis. The second type of program encourages using the most efficient technology in new installations. The third type of program encourages higher-efficiency technology, with the objective of reducing peak-load demand. This type of program is more applicable to indoor HID luminaires, rather than outdoor lighting that typically operates during off-peak hours. Incentives in this category are based on kilowatts saved during peak periods without dictating the technology used to reduce the load. The fourth and final type of program focuses on reducing energy (kilowatt hours) through more efficient technology. Again, this does not prescribe specific technologies, but instead rewards energy savings. In Appendix A, DOE provides a summary of non-regulatory programs in the United States that focus on lighting and HID technology. Each program is described briefly, identifying the target technology, the incentives offered, and a website for more information. Although the list is not exhaustive, 23 states offer some type of program that specifically focuses on HID technology. Incentives exist for either installing new HID sources or replacing HID sources with non-HID sources. In general, the incentives treat MV and MH as the old inefficient sources to be replaced. The three most common incentives are: 1. PMH These incentives are for PMH fixtures or lamps/ballasts replacing incandescent, MV, MH, and HPS technology. The incentives can include either quartz or ceramic PMH lamps. The incentives range from $18 to $100. The most common PMH incentive is the 320W replacement for a 360W or 400W MH system. PMH being more efficacious and having better lumen maintenance than MH lamps means that, over the operating life, the lower wattage PMH system is equal or better than the standard higher wattage MH systems. 2. High-intensity fluorescent (HIF) These incentives are for multi-lamp T5 or T8 fluorescent low/high bay luminaires. Incentives range from $50 to $125 for replacing HID fixtures. The rationale for these incentives is that fluorescent lamps can be dimmed more easily than HID lamps. Multi-lamp fixtures can also be switched off in different pairings. These control features save energy. The high lumen maintenance (above 0.90) for fluorescent lamps, the high efficacy (above 90 LPW), and the long rated life when

2-23

operated for long periods (over 30,000 hours) makes HIF a good option compared to traditional HID fixtures. 3. Integrally-ballasted CMH lamp This is a PAR lamp with an integral electronic ballast. Other incentives exist for CMH lamps but those are similar to other PMH incentives. Incentives range from $12 to $25 for this type of lamp, which directly replaces halogen PAR 38 lamps. This type of CMH lamp yields at least a 50 percent energy savings compared to the halogen lamps. The CMH lamps typically have a rated life of 12,000 hours which is double the best rated life for halogen lamps. Programs or incentives for demand or energy reduction were not listed in Appendix A unless the program specifically mentioned HID technology. Other programs that addressed HID technology, but did not provide data (required user to contact agency), were also not listed in Appendix A.

2-24

CHAPTER 3.

TECHNOLOGY ASSESSMENT

High-intensity discharge (HID) lamps produce light by generating an electrical discharge between tungsten electrodes contained within an arc tube. By definition, HID lamps include mercury vapor (MV), metal halide (MH), and high-pressure sodium (HPS) lamps. HID lamps range from 20 to 2000 watts (W). The lamps are often subdivided by power into low (less than 150), medium (from 150 to 500), and high (greater than 500) wattages. Because the electrical discharge exhibits a negative resistance, ballasts regulate this electrical discharge and prevent it from drawing excessive current that would lead to premature lamp failure. Therefore, HID lamps always operate with a ballast, which is typically external to the lamp and located within the luminaire. There are limited exceptions to this configuration, specifically self-ballasted (integral ballast) MV lamps and the MH parabolic aluminized reflector (PAR) lamps. The following discussion examines and summarizes current specifications, standards, and guidelines governing the classification, testing, and evaluation of HID lamps. Because this analysis considers lamps only, issues concerning the effects of ballasts on HID lamp performance are outside the scope of this analysis. However, the availability, capability, and functional relationship of ballasts and lamps are important for understanding the technology and its market. Therefore, as appropriate, ballasts and their characteristics are addressed in this section. 3.1 SPECIFICATIONS, TEST STANDARDS, AND PROCEDURES

The U.S. Department of Energy (DOE) reviewed available evaluations of the electrical and photometric properties of HID lamps and ballasts. Table 3.1.1 presents a list of industry standards publications pertaining to HID lighting systems. Table 3.1.1 Documents Pertaining to High-Intensity Discharge Lamps
Source # ANSI* C78.1381 Title American National Standard for Electric Lamps70 W, M85 Double-Ended MH Lamps C78.380 2007 American National Standard for Electric Lamps-High-Intensity Discharge Lamps, Method of Designation C78.380a 2004 American National Standard for Electric Lamps-HID Lamps, Method of Designation (Amendment) C78.389 2004 American National Standard for Electric Lamps High Intensity Discharge R2009 Methods of Measuring Characteristics C78.40 1992 Specifications for Mercury Lamps R2003 C78.40a 1998 American National Standard for Specification for Mercury Lamps-Maximum Outline Drawing of Bulb BT56 C78.42 2007 American National Standard for Electric Lamps-HPS Lamps C78.43 2007 American National Standards for Electric Lamps-Single-Ended MH Lamps C78.44 2008 American National Standards for Electric Lamps-Double-Ended MH Lamps C78.45 2007 American National Standard for Electric Lamps-Self-Ballasted Mercury Lamps C78.62035 2004 American National Standard for Electric Lamps-Discharge Lamps (Excluding Fluorescent Lamps)-Safety Specifications Year 1998

3-1

Table 3.1.2 Documents (continued)

Procedures for HID Lamp Sample Preparation and the Toxicity Characteristics Leaching Procedure (TCLP) ANSI* C81.62 2007 American National Standard for Electric Lampholders (contd.) C82.4 2002 American National Standard for Ballasts for HID and Low-Pressure Sodium (LPS) Lamps (Multiple-Supply Type) C82.6 2005 American National Standard for Lamp Ballasts Ballasts for HID Lamps-Methods of Measurement C82.9 1996 American National Standard for HID and Low-Pressure Sodium (LPS) Lamps, Ballasts and TransformersDefinitions C82.14 2006 American National Standard for Lamp Ballasts-Low-Frequency Square Wave Electronic Ballasts-for MH Lamps LE 5B 1998 Procedure for Determining Luminaire Efficacy Ratings for HID Industrial Luminaires LSD 14 2002 Guidelines on the Application of Dimming to HID Lamps LSD 24 2003 Marking of Luminaire Codes on MH Lamps LSD 25 2008 Best Practices for MH Lighting Systems, Plus Questions and Answers about Lamp Ruptures in MH Lighting Systems LSD 31 2005 A Lighting Systems Division Information Bulletin: Changes to the 2005 NEC Will Impact Future MH Lamp Systems Options LSD 35 2006 ANSI Code Update to Include Letter C for Ceramic MH Lamps LSD 37 2008 Effects of EPAct 2005 on MV Lamp Ballasts and Lighting Systems: Frequently Asked Questions LM-31-95 1995 Photometric Testing of Roadway Luminaires Using Incandescent Filament and High Intensity Discharge Lamps IESNA LM-35-02 2002 Photometric Testing of Floodlights Using HID or Incandescent Filament Lamps LM-46-04 2004 IESNA Approved Method for Photometric Testing of Indoor Luminaires Using HID or Incandescent Filament Lamps LM-47-01 2001 IESNA Approved Method for Life-Testing of HID Lamps LM-51-00 2000 IESNA Approved Method for the Electrical and Photometric Measurements of High Intensity Discharge Lamps LM-61-06 2006 IESNA Approved Guide for Identifying Operating Factors Influencing Measured Vs. Predicted Performance for Installed Outdoor HID Luminaires LM-73-04 2004 IESNA Guide for Photometric Testing of Entertainment Lighting Luminaires Using Incandescent Filament Lamps or High Intensity Discharge Lamps * ANSI = American National Standards Institute IESNA=Illuminating Engineering Society of North America. Sources: National Electrical Manufacturers Association (NEMA) website, 2010; IESNA website, 2009. C78.LL 3 2003

As the Secretariat for ANSI, NEMA publishes standards for lamps (C78); bulbs (C79); lamp bases and holders (C81); ballasts (C82); for performing standardized tests to evaluate the energy efficiency of luminaires (LE); general information documents (LSD) as part of its lighting-standards program. 24 Table 3.1.1 presents the ANSI standards specifically related to HID technology, excluding the more generally-applicable standards for bulbs (C79) and lamp bases and holders (C81). ANSI also publishes a system of nomenclature for the designation of HID lamps in its C78 technical-support documents. The designation code, at a minimum, includes the class and electrical characteristics of the lamp. In addition, ANSI publishes a set of corresponding performance and physical characteristics specifications. The designations and specifications not

3-2

only provide a system of identification, but also ensure a level of product commonality and interchangeability among manufacturers of HID lamps and ballasts. IESNA is another organization promulgating standards for HID lamps and related technologies, including Lighting Measurement (LM) test procedure documents. IESNA LM-3502, 46-04, and 73-04 focus on testing luminaires using HID technology (as well as other light sources). IESNA LM-41-01 and LM-51-00 focus on life and electrical/photometric measurements of HID lamps, respectively. These publications build on ANSI standards to include procedures for both electrical and photometric measurements by defining testing conditions, lamp stabilization, and other critical aspects to ensure consistency in the testing of lamp performance. Publications by IESNA reflect the consensus of industry stakeholders, including NEMA members. 3.2 LAMP METRICS

Lamp characteristics and performance are evaluated on a number of factors, in addition to lumen output. The following discussion is a review of different metrics applicable to HID lamps. 3.2.1 Lumens

The lumen is the basic unit of light. The IESNA defines luminous flux as the time rate of flow of radiant energy, evaluated in terms of a standardized visual response:

= 683 Where: = luminous flux (in lumens, lm).

Eq. 3-1

The value of 683 LPW is a constant for the photopic luminous efficacy function. The equation essentially is the integral of the light sources spectral content expressed as spectral power distribution (SPD) over the visible range of the electromagnetic spectrum weighted by the photopic luminous efficacy function. In practical terms, this equation expresses integrating the energy emitted by the light source across the visible light portion of the spectrum (380 to 780 nanometers (nm)) and multiplying by the visual efficacy function (a function weighting the different portions of the visible light spectrum by the different photoreceptors in the eye) times the constant 683 lumens per watt. The units cancel out to just lumens. 3.2.2 Mean Lumens/Lamp Lumen Depr eciation/Lumen Maintenance

Lamp lumen depreciation (LLD) is also known as lumen maintenance. In use, a lamp loses light output over time. According to the IESNA, the LLD factor is the fraction of the initial lumens produced at a specific time during the life of the lamp. 25 The LLD value is used in lighting calculations to account for the reduction in output over the operating life of the lamp. Often the mean (also known as design) lumens are used for LLD. Mean lumens are published in

3-3

the manufacturers catalogs and are measured at a certain point in time, typically a specific percentage of the lamps rated lifetime. The IESNA does not define when to measure mean lumens; however, 70 percent of average rated life is the recommended criterion for lamp replacement for both group and spot relamping programs. This value is usually not used in design calculations because it is difficult to determine from manufacturers data. Often, designers must build margins into their calculations and use the mean/design lumen value presented in the catalogs. The basis for mean lumens can vary between manufacturers, as shown in DOEs survey of definitions from manufacturer catalogs: GE: Mean Lumens lamp light output at 40% of rated lamp life for (MH) lamps and 50% of rated life for mercury and HPS lamps. 26 OSRAM SYLVANIA: Mean lumens are measured on ANSI reference circuits at rated wattage (HQI, LUMALUX Super and SOx lamp ratings are based on input voltage) at 40% of average rated life except for those lamps with a + next to their life rating; these lamps are measured at 50% of average rated life. All measurements are based on ballast operation on systems with current crest factors of 1.8 or less. Higher current crest factors reduce values. In actual applications on constant wattage (CW) or constant-wattage autotransformer (CWA) ballasts, mean lumens may be higher than published ratings. 27 Philips: Approximate lumen output at 40% of lamp rated average life. 28 Ushio: Lumens, Mean: Average quantity of light output over the life of the lamp. High Pressure Sodium and incandescent lamps are measured for mean lumens at 50% of lamp life. Fluorescent and MH lamps are measured for mean lumens at 40% of rated lamp life. 29 Venture: A manufacturer of only MH lamps states, Mean Lumens: Light output at 40% of rated life. 30 Chapter 2, Market Assessment, discussed both draft legislation focusing on outdoor lighting and requirements in California already in effect that have requirements related to mean lumens. The absence of an agreed definition for mean lumens complicates any discussion of standards compliance or recommended relamping cycles, and hinders comparing similar products from different manufacturers. 3.2.3 Efficacy

Efficacy (luminous efficacy) is defined by the IESNA as quotient of the total emitted luminous flux and the total lamp power. Expressed mathematically: = luminous flux (lm) / lamp power (W) Where: Eq. 3-2

3-4

luminous efficacy (in lumens per watt, lm/W or LPW).

A 2001 paper, An Efficacy Analysis of HID Lamps examined the components of efficacy of HID lamps. 31 The paper was not entirely based on new research but long-standing lighting science. The paper focused on aspects of efficacy irrespective of chromaticity and color rendering. The paper only examined lamp efficacy and excluded lighting system (i.e., ballasts and fixtures). The paper converted the mathematical efficacy function into the three components, rewritten as: = vis max spectral spectral 683 LPW Where: vis = max spectral = spectral = LPW = visual efficiency, maximum spectral efficiency, spectral efficiency, and lumens per watt. Eq. 3-3

The value of 683 LPW is both a constant and the absolute photopic luminous efficacy value. Therefore, as the different values approach 1.0, the closer the efficacy is to 683 LPW. 3.2.3.1 Visual Efficiency

The visual efficiency of a lamp is the quotient of the power radiated in the visible spectrum (380 to 780 nm) and the total lamp power. Table 3.2.1 is from the 2001 Nelson et al. paper. Table 3.2.1. Visual Efficiencies and Efficacies of Various Sources
Lamp Type High-Pressure Sodium Pla W 150 400 1000 100 400 400 400 100 100 vis 0.22 0.31 0.33 0.20 0.15 0.35 0.24 0.35 0.38 LPW 90 124 135 51 55 110 80 98 95

White High-Pressure Sodium Mercury Vapor Quartz Metal HalideNaSc Pulse Start Quartz Metal HalideNaTl In Ceramic Metal Halide3000K Ceramic Metal Halide4000K

The data in the table reflect visual efficiencies ranging from 0.15 to 0.38. Newly developed (post-1990) HID lamps (pulse start ceramic and quartz MH) have visual efficiencies above 0.35. As lamp power increases so does visual efficiency as shown in the HPS data (150 W/0.22, 400/0.31, 1,000 W/0.33) data. Although related, changes in visual efficiency do not directly correlate to changes in efficacy. The white HPS lamp has a slightly lower efficacy than the mercury vapor and yet more visual efficiency, whose values are virtually the same for the 150 W HPS (0.22) and the white HPS (0.20). The efficacy of the 150 W lamp is almost twice that of the 100 W white HPS lamp.

3-5

Table 3.2.2 is excerpted from the IESNA Lighting Handbook table 7-49.2 Each of the HID lamps produce more IR than light. This distribution of energy directly affects the visual efficiency value of the lamp. If the HID lamps produced more energy in the visual light region of the electromagnetic spectrum, the visual efficiency values would be greater. Table 3.2.2. Sample Energy Output for High Intensity Discharge Lamps
Type of Energy MV 14.6 46.6 1.9 27.0 10.1 400 W % MH 20.6 31.9 2.7 31.1 13.7 HPS 25.5 37.2 0.2 22.2 14.9

Light Infrared Ultraviolet Conduction-Convection Ballast

Maximum Spectral Efficiency. The 2001 paper defined the maximum spectral efficiency a factor expressing the effect of the eye sensitivity curve for a given chromaticity. A given lamps location in color space factors into efficacy and that is the role of the maximum spectral efficiency. The 2001 paper provided the chromaticities maximum spectral efficiencies for the lamps that it provided the visual efficiency data for, which shown in Table 3.2.3. The paper found (and supported by the data republished) that the maximum spectral efficiency varied from 0.70 to 0.76 for HID lamps. Table 3.2.3. Chromaticities and Maximum Spectral Efficiencies of Various Sources
Lamp Type High-Pressure Sodium Pla W 150 400 1000 100 400 400 400 100 100 x 0.521 0.521 0.522 0.461 0.331 0.374 0.390 0.368 0.433 0.372 y 0.423 0.413 0.413 0.405 0.395 0.394 0.398 0.368 0.388 0.366 CCT 2100 2040 2040 2640 5680 4280 3890 4300 2950 4150 MPCD 11 -1 -1 -7 58 26 20 0 -19 -7 max spectral 0.75 0.72 0.72 0.75 0.73 0.75 0.76 0.71 0.73 0.70

White High-Pressure Sodium Mercury Vapor Quartz Metal HalideNaSc Pulse Start Quartz Metal HalideNaTl In Ceramic Metal Halide3000K Ceramic Metal Halide4000K High-Pressure Sodium

Spectral Efficiency. The spectral efficiency is a measure of how well a given spectrum is distributed relative to the optimally efficient two line spectrum for that chromaticity. There is a negative correlation with spectral efficiency. As spectral efficiency increases, color rendering index (CRI) decreases. This is demonstrated in the high spectral values for HPS and MV (0.82 and 0.77 respectively) and the very low CRI values of 11 (HPS) and 17 (MV). In contrast, the high CRI values of 93 and 89 from the HPS had lower spectral efficiency values (0.53 and 0.56, respectively; Table 3.2.4).

3-6

Table 3.2.4. Color Rendering Indexes and Spectral Efficiencies of High-Intensity Discharge Sources
Lamp Type High-Pressure Sodium Pla W 150 400 1000 100 400 400 400 100 100 CRI 11 21 17 84 17 36 65 62 89 93 max spectral 0.82 0.80 0.82 0.50 0.77 0.70 0.62 0.67 0.56 0.53

White High-Pressure Sodium Mercury Vapor Quartz Metal HalideNaSc Pulse Start Quartz Metal HalideNaTl In Ceramic Metal Halide3000K Ceramic Metal Halide4000K High-Pressure Sodium

Nelson et al. found in the 2001 paper that this negative relationship did exist and derived a function expressing the relationship of CRI and spectral efficiency for HID and LPS lamps. The function was a straight line with an R2 value of 0.95 and is: spectral = 0.8461 (0.0034 CRI).8 Potential Efficacy Limits. Nelson et al. hypothesized the possible efficacy limits for conventional HID lamps on the blackbody locus, based on their assumptions in the 2001 paper.8 The authors assumed a visual efficiency value of 0.40. Based on the lamps they surveyed, a value 0.38 was the highest for HID lamps. In the discussion section of their paper, this value was questioned. The authors defended their choice in that it represented a likely value for long life lamps with electrodes and were commercially available. The maximum spectral efficiency values were provided for the different chromaticities. Finally, a CRI of 80 was assumed and using the function provided the spectral efficiency values were derived. A CRI of 80 was chosen by the authors because end users will be more concerned with color quality in the future. A CRI of 80 seems reasonable when reviewing other DOE lamp rules. Table 3.2.5 is a review of the probable efficacy limits for HID lamps that lie on the blackbody locus (BBL) and have a CRI of 80. In review, the efficacy ranges from 100 to 120 LPW depending on the lamp chromaticity. Table 3.2.5. Probable Efficacy Limits by Color Correlated Temperature (CCT) for Conventional High-Intensity Discharge Lamps
CCT 2000 2800 3000 4000 5000 6000 vis max spectral 0.722 0.766 0.763 0.722 0.676 0.636 spectral 0.574 0.574 0.574 0.574 0.574 0.574 0.40 0.40 0.40 0.40 0.40 0.40 LPW 113 120 120 113 106 100

3.2.3.2

Moving Beyond the Limits

Nelson et al. hypothesized that an improvement in the power balance could increase the visual efficiency. Of the three values in the efficacy function, this value was below 0.4. Also, as stated in Table 3.2.2, only 15 to 26 percent of the energy being used by HID lamps is in the visual light portion of the spectrum. Nelson et al. hypothesized that the a source, such as the

3-7

sulfur lamp with a visual efficiency of 0.55 at 3000K on the BBL and a CRI of 80, could have an efficacy of 170 LPW. It should be stated that the sulfur lamp is no longer in production, but one reason that the lamp has a higher visual efficiency value is that it does not have electrodes. Nelson et al. hypothesized that spectral optimization would increase the spectral efficiency. As shown in Table 3.2.5, the spectral efficiency values are limited to 0.574 due to the CRI of 80. If the spectrum was more optimized and the CRI held constant the spectral efficiency values could be 0.80 resulting in a 170 LPW lamp. Nelson et al. point out that the measures could be used in conjunction yielding in a possible 230 LPW lamp. Finally, the authors note that even after more than 50 years of HID lamp development, the task of achieving a lamp with a high luminous efficacy (say, greater than 150 lm/w [LPW]), high CRI and a chromaticity near the black body line will not be an easy one.8 3.2.4 Color Render ing Index

The IESNA defines CRI as a measure of the degree of color shift objects undergo when illuminated by the light source as compared with those same objects when illuminated by a reference source of comparable color temperature. In CRI calculation, 14 samples are evaluated; however, the general color rendering index (Ra) is only an average of the first 8 samples. Samples R1R8 are not highly saturated and are pastel colors. R9 is an oft-cited color value because it is a deeply saturated red. The accurate rendering of this color is often desired in retail applications. Light emitted at the red end of the visual spectrum (above 700 nm) adds little to total lumen output, but can add important color qualities to the lamp. Incandescent and halogen lamps have R9 values of 100. The R9 value for HID lamps is typically much less than 100.8 3.2.5 Co rr elated Co lo r Temper atur e

The IESNA defines CCT as the absolute temperature of a blackbody whose chromaticity most nearly resembles that of the light source. Color temperature is a not a performance metric but rather characterizes the color appearance of a light source: Color temperature is a specification of chromaticity only. It does not represent the SPD of the light source.2 Chromaticity (expressed as chromaticity coordinates, e.g., x and y) is a representation of the light source color calculated with by the standard observer and spectroradiometric measurements. 3.2.6 Lamp Life

The rated lamp life is defined when 50 percent of a sample group of lamps have failed. It is expressed in hours, and varies with the method by which the ballast starts and operates the lamp. 3.3 MERCURY VAPOR (MV) LAMPS

Mercury vapor was the first commercialized HID light source, introduced in the 1930s, and remains fundamentally unchanged. Mercury vapor and fluorescent (FL) lamps are both MV discharge lamps, but FL lamps operate at low pressure and MV lamps operate at high pressure.

3-8

In the movement toward efficient lighting, one development was the invention of a self-ballasted MV lamp. This lamp originated as a longer-life substitute for the A-shaped general service incandescent lamp. By using a length of tungsten filament within the lamp structure to provide current regulation, these lamps function without ballasts. Although more efficient than incandescent lamps, these lamps are less efficacious than MV lamps with ballasts due to the resistive loss at the tungsten filament. Because of their low CRI (CRI of 15 is typical), MV lamps are used primarily in spaces that are not frequently occupied by people. A phosphor coating is added to clear MV bulbs to obtain better CRI, but the improvement is marginal. Outdoor security, street, and landscape lighting are some of the applications for MV lamps, but higher performing MH and HPS lamps are replacing these. MV systems continue to maintain market share in some niche applications (e.g., the security or barn light an arm-mounted luminaire used typically in residential applications) where long life and low initial cost are important. However, even in niche applications, MV is losing its advantage as other HID sources offer improved lifetime and initial cost. MV lamps are available in wattages from 50 to 1000. Mercury vapor lamps have applications aside from general illumination, including quality inspection, processing, scientific uses, reprographic (copy/duplication), fluorescent microscopy, and curing. The specific spectrum of MV lamps is beneficial to these applications. The selfballasted MV lamp has special niche because of its unique capability to directly replace highwattage incandescent lamps in standard mogul bases. Self-ballasted MV lamps are available in wattages between 100 and 400. Typical applications have included outdoor security lighting, area lighting, and some warehouse use. Although MV lamps offer modest energy savings over incandescent lamps, their more significant advantage over incandescent and other technologies is lamp lifetime, resulting in longer intervals between relamping. In recent years, the self-ballasted MV lamp has lost market share to integrally-ballast CFLs, integrally-ballasted MH lamps, and even LED replacement lamps. 3.3.1 Lamp Constr uction

In a MV lamp arc tube, an electric-current arc passes from the starting electrode to the main electrode. The initial arc strikes through the ionization of argon gas. Once the arc is established, its heat vaporizes the liquid mercury in the arc tube. The amount of mercury in the lamp essentially determines the final operating pressure, which is 200 to 400 kilopascals (kPa) (29 to 58 pounds per square inch) in the majority of lamps.2 The arc tube construction is fused silica (quartz) that can operate at up to 1000 C. An outer glass envelope (bulb) protects the arc tube and internal electrical connections from the environment. The bulb is usually in the shape of an ellipse and made of hard (borosilicate) glass, which absorbs most of the ultraviolet (UV) energy radiated by the discharge while allowing light to pass through. For lamps where UV emission is desired (e.g., germicidal lamps), the bulb is made of other materials that allow transmission of UV energy.2 In coated MV lamps, an internal coating of diffusing material coats the bulb to reduce source brightness of the lamp. This diffusing coating is usually a color-correcting phosphor that uses UV energy radiated by the arc tube to improve the lamps overall color-rendering properties, as in a fluorescent lamp. MV lamps are also available in a reflector lamp format.

3-9

There are wires within the outer bulb that are suitable for high temperatures to conduct electricity to the arc tube and structural components to support the arc tube. The atmosphere in the outer bulb is usually a low-pressure gas (usually nitrogen) or in rare cases, a vacuum.2 If the outer bulb breaks and the arc tube continue to operate, the lamp emits a significant amount of UV energy. Self-extinguishing lamps, designed to shut off the lamp should the outer bulb break, usually contain a tungsten filament in place of a portion of nickel wire that oxidizes quickly, opening the circuit. This separation renders the lamp inoperable. Figure 3.3.1 shows essential construction details typical of lamps with quartz inner-arc tubes within an outer envelope.2

Source: GE Lighting, High-Intensity Discharge Lamps Catalog, 2003. Reproduced with permission.

32

Figure 3.3.1. Mercury Vapor Lamp Construction 3.3.2 Lamp Per formance

The pressure at which MV lamps operate has a significant impact on their characteristic SPD. As pressure increases, the emitted radiation shifts towards longer wavelengths (from UV to visible light) and the spectrum lines spread into wider bands. Within the visible region, the mercury spectrum consists of five principal lines (404.7, 435.8, 546.1, 577, and 579 nm), which result in greenish-blue light at efficacies of 30 to 65 LPW, excluding ballast losses. a, 2 While the light source itself appears greenish-blue white, a deficiency of long-wavelength radiation distorts the perceived color of objects illuminated by that light source. Blue, green, and yellow are emphasized; orange and red appear brownish. MV lamps also emit radiation outside the visible region in the form of UV in the range of approximately 250 to 365 nm. These wavelengths do not contribute to seeing, but are useful for non-illumination purposes (e.g., reprographic or germicidal applications). Clear MV lamps generally have a CRI value of approximately 15 and are not desirable for use where people will occupy the space. Phosphor-coated MV lamps convert some of the
a

Ballast Loss Power drawn by the ballast in addition to the rated lamp power used to operate the lamp.

3-10

lamps UV radiation to visible light, as occurs with a FL lamp. The most widely used lamps of this type (designated DX) have a coating of vanadate phosphor that emits long-wavelength radiation (orange-red) where MV lamps are otherwise deficient, thus improving CRI.2 This phosphor may also blend with other phosphors to produce a cooler or warmer CCT between 3200 and 7000K. The phosphor coating can raise the CRI rating to 50. Figure 3.3.2 shows the SPDs of a clear lamp and coated lamp using these phosphors.
H39KB-175
0.80

Output (Watts per 5nm for 1000 lumens)

0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 200 300 400 500 600 700 800

Wavelength (nm)

(a)
H39KC-175/DX
Output (Watts per 5nm for 1000 lumens)
0.60

0.50

0.40

0.30

0.20

0.10

0.00 200 300 400 500 600 700 800

Wavelength (nm)

(b)
Source: Sylvania, 2003. Reproduced with permission. 33

Figure 3.3.2. Spectral Power Distribution of Mercury Vapor Lamps (a) Without Phosphors and (b) With Phosphors Efficacies of self-ballasted lamps are slightly better than those of the incandescent lamps they replace. Typically, efficacies range between 15 to 25 LPW. The spectrum consists of the line spectrum of mercury in addition to the blackbody radiation pattern from the incandescent component of the tungsten filament broadening the spectrum into long-wavelength range (red).

3-11

Although not significantly increasing total light output, the light in these wavelengths increases the CRI of the source. The CCT ranges from 3300 to 4000K, with a CRI of 50, depending on mercury pressure and phosphor coating used (if any). The light output from MV lamps decreases over the operating life of the lamp. Typically, lumen output depreciates about 25 percent from initial output level at 50 percent of its rated life. The depreciation in light output is a result of the bulb becoming darker because the cathode material evaporates through use and is deposited on the arc tube bulb wall. Typical rated life is 24,000 hours, but it is not uncommon for the lamp to continue operating long after its rated life. One of the major failure points for discharge lamps is that the cathode material eventually evaporates and fails to strike an arc. MV lamps are notorious for never burning out, but simply diminishing in output over time, because the evaporating cathode material has to diffuse through mercury vapor, a very heavy gas at high pressure, thereby slowing the evaporation process. The lamp power remains constant as light output declines, meaning that the efficacy of the light source steadily decreases over time. To achieve the necessary high vapor pressures within the arc tube, MV lamps require a significant warm-up time of approximately 5 to 7 minutes. Re-strike times are about as long, requiring 5 to 7 minutes for the vapor pressure to reach the level where the arc can restart. Warm-up time for self-ballasted lamps is approximately 3 minutes, with a re-strike time of 5 minutes. Table 3.3.1 summarizes the performance of common MV lamps. Table 3.3.1. Performance Summary of Mercury Vapor Lamps
Initial Efficacy Mean Efficacy @ 100 hrs @ 40-50% life LPW LPW 50 Minimum 20 3,200 1,050 6,000 22 13 Median 45 4,100 1,575 8,000 30 24 Maximum 50 5,700 1,900 24,000 40 33 75 Minimum 20 3,200 2,700 16,000 36 24 Median 45 4,100 2,700 18,000 36 30 Maximum 50 5,700 3,000 24,00 40 32 100 Minimum 15 3,200 1,100 6,000 11 8.3 Median 40 4,100 4,000 2,4000 40 26 Maximum 60 7,000 4,500 2,4000 45 37 175 Minimum 15 3,700 5,100 16,000 29 21 Median 40 4,300 7,800 24,000 45 39 Maximum 50 6,800 8,900 24,000 51 43 250 Minimum 15 3,200 3,200 12,000 13 11 Median 43 4,100 8,900 24,000 36 24 Maximum 50 6,700 13,700 24,000 55 45 400 Minimum 15 3,700 1,500 12,000 37.5 13.25 Median 43 4,000 21,000 24,000 53 36 Maximum 50 6,500 24,000 24,000 60 49 1000 Minimum 15 3,700 13,000 16,000 45 13 Median 40 4,100 45,200 24,000 58 45 Maximum 50 6,300 52,240 24,000 64 52 * A minimum of three different manufacturers had to produce the type and similar wattage to have that wattage class listed in this table. See Appendix D for a comprehensive list of lamp data from manufacturers catalogs. Correlated-color temperature is not a performance metric; the values are listed to show the range of available CCTs for the product. Lamp Power* W CRI CCT K Initial Lumens Life Hrs

3-12

 Initial efficacy is initial lumens divided by the lamp rated power (the power use of the ballast is not factored into this table). Mean efficacy is the mean lumen value provided by the manufacturer divided by the lamp rated power (the power use of the ballast is not factored into this table). Lumen depreciation still occurs after this point.

3.3.3

Electr ical Char acter istics

Because MV lamps have an additional electrode located at one end of the arc tube to assist in lamp starting, they do not require a separate starter circuit. These lamps typically require an open-circuit voltage of approximately twice the operational lamp voltage to initiate and sustain the arc discharge. MV lamps may operate with any of the commonly available MV and MH ballasts without an ignitor. All MH ballasts will operate MV lamps of the corresponding wattage, although most MV lamp ballasts will not operate MH lamps reliably. Ballasts for MV lamps tend to be small and inexpensive, and fall into four basic categories: (1) reactor (RX) ballasts; (2) high-reactance autotransformer (HX) ballasts; (3) CWA ballasts; and (4) CW isolated transformer ballasts. Ballast losses may reduce the system efficacy of MV sources (and other HID sources) by 5 to 15 percent. This results in an efficacy range for MV systems (lamp plus ballast) of approximately 26 to 55 LPW. Figure 3.3.3 shows the energy balance from a typical 400-W MV lamp. 34 Although this diagram dates from the mid-1970s and is for one lamp wattage, the diagram is representative of how the input power is used in this lamp type. The percentage values in parentheses are the portion of power for that input or output per rated lamp wattage. Typically, only 15 percent of the power used by the lamp is actually converted into visible radiation. The efficiency of the conversion of power into radiation increases with power. As shown in Figure 3.3.3, MV lamps produce equal or more energy in the IR region than in the visible light region.

Figure 3.3.3. Sample Power Flow of a Mercury Vapor Lamp

3-13

3.3.4

Other Issues

Although the Occupational Safety and Health Administration does not require lamp manufacturers to do so, all three major MV lamp manufacturers (General Electric, Philips, and OSRAM SYLVANIA) provide material safety data sheets (MSDS) for manufactured articles, such as HID lamps. The MSDS reports that MV lamps contain anywhere from 15 milligrams (mg) of mercury in 50 W lamps to 250 mg of mercury in the 1000 W examples. Although other toxic materials are in the lamps, these are present in trace amounts and are not classified as hazards. Lead is typically used in soldering the connection in the lamp. Although lead in the lamp and mercury in the arc tube pose little risk under normal use, a toxicity characteristic leaching procedure (TCLP) for lead and mercury, as prescribed by NEMA standards publication LL 3-1999, results in MV lamps being classified as hazardous waste. Therefore, the Universal Waste Rule published by the U.S. Environmental Protection Agency (EPA) on July 6, 1999 (64 FR 36466) prescribes specific disposal requirements for spent MV lamps. 3.4 HIGH-PRESSURE SODIUM LAMPS

HPS lamps were introduced in the 1960s, prior to the introduction of MH lamps. HPS lamps are currently the most efficacious HID light source ranging from 38 to 150 LPW, with typical efficacy being greater than 80 LPW. These values are initial lamp efficacy (rated lamp lumens/rated lamp power) and do not take in account any ballast losses. The 38 LPW value is from a directional (reflector-type) lamp. This is an atypical lamp and its low efficacy is not representative of HPS lamps in general. Typical HPS lamps have low CCT (e.g., 2100K) and low CRI (e.g., 22), but there are some enhanced-spectrum HPS lamps that have a higher CRI (and a higher CCT as a byproduct of the change in spectrum). HPS lamps are in use in public (e.g., outdoor roadway), commercial, and industrial sectors. Although HPS lamps are available in wattages from 35 up to 1000, the most common wattages in these sectors range from 50 to 500. Typical applications include vehicular traffic areas (e.g., roadways and parking lots) and pedestrian areas (e.g., public-access areas, subways, and parks). HPS lamps are also used in commercial and industrial applications where color quality may not be critical (e.g., storage and loading areas, and corridors). Many of these application overlap with MH, creating competition between the technologies for market share. 3.4.1 Lamp Constr uction

Figure 3.4.1 shows the basic construction of HPS lamps. Similar to MV lamps, they have an inner arc tube and an outer bulb. The arc tube is made of a ceramic material that contains electrodes, a sodium and mercury amalgam, and a small amount of xenon gas. No starter probe is present in the HPS arc tube, and a high-voltage, high-frequency pulse is used to ionize the starting gas. The high temperatures needed to vaporize sodium dictate the arc tube geometry and material. Further, the highly corrosive nature of sodium, especially at elevated temperature and pressure (700 C/27 kPa [3.9 pounds per square inch]), precludes the use of certain materials (e.g., quartz). An amalgam reservoir inside the arc tube helps to stabilize pressure, similar to

3-14

amalgam-based fluorescent lamps. This allows HPS lamps to operate in any physical orientation, simplifying stocking and installation. 35

Source: GE Lighting, High-Intensity Discharge Lamps Catalog, 2003. Reproduced with permission.9

Figure 3.4.1. High-Pressure Sodium Lamp Construction The lamps outer glass envelopes are typically elliptical in shape and are made of a hard glass that protects the arc tube from damage. Usually, an outer envelope contains a vacuum that reduces convection and heat losses from the arc tube to maintain high efficacy.12 In some HPS lamps, an additional coil is wrapped around the outside of the ceramic arc tube to assist in starting the lamp.12 This coil is similar in function to the starting probe in probestart MH and MV lamps. This type of HPS lamp can directly replace MV lamps and operate on the same ballast without an ignitor. However, HPS retrofit lamps produce significantly more light for the same wattage. Therefore, to realize any energy savings a lower-wattage HPS retrofit lamp should replace a higher-wattage MV lamp. 3.4.2 Lamp Per for mance

The high luminous efficacy of HPS lamps corresponds to the emittance of sodium in the yellow-orange portion of the visible spectrum, to which the human visual system responds most efficiently. The efficiency of HPS lamps is critically dependent on the specific vapor pressure of sodium in the arc tube. Any deviation from this optimum pressure results in reduced performance of the lamp. Notwithstanding, HPS lamps offer the highest efficacy of HID sources, achieving 64 LPW for 35-W sources and up to 133 LPW for 1000-W sources. HPS lamps of 100 watts or more have efficacies over 100 LPW. Rated lifetimes of 24,000 hours are typical, with some long-life lamps rated at 40,000 hours. These lamps have expected LLD value of 10 percent at mean rated life. A historical perspective of how these values have changed in the 40 years since HPS lamps entered the market can be found in another IESNA significant paper. A 1965 publication in Illuminating Engineering stated that with initial efficacy in excess of 100 LPW,

3-15

10,000 hour lumen maintenance has been measured on laboratory test lamps at over 90 percent. Median life at the present time is in the 3,000- to 6,000-hour range. From an engineering viewpoint it is reasonable to expect that this can be increased as techniques for constructing the lamps are improved. 36 From the data in Table 3.4.1, it appears that the comment was true in that life of HPS lamps have improved and efficacy somewhat improved. The historical perspective shows that in over 40 years, the efficacy of the lamp has increased 20 to 30 percent since the introduction of the lamp. Table 3.4.1. Performance Summary of High-Pressure Sodium Lamps
Initial Efficacy Mean Efficacy @ 100 hrs @ 40-50% life LPW LPW 35 Minimum 21 1,900 2,100 16,000 60 55 Median 22 1,900 2,250 16,000 64 58 Maximum 22 2,100 2,250 24,000 64 59 50 Minimum 17 1,900 2,300 6,000 46 40 Median 22 1,900 4,000 24,000 80 72 Maximum 85 2,700 4,000 30,000 80 72 70 Minimum 17 1,900 3,350 6,000 48 41 Median 22 1,900 6,300 24,000 90 76 Maximum 83 2,500 6,500 40,000 93 84 100 Minimum 17 1,900 4,900 10,000 49 42 Median 22 2,100 9,500 24,000 95 80 Maximum 85 2,700 10,500 40,000 105 95 150 Minimum 17 1,800 5,900 6,170 39 41 Median 22 2,100 15,000 13,500 100 90 Maximum 85 2,500 16,000 14,400 107 96 200 Minimum 21 2,000 21,400 24,000 107 90 Median 22 2,100 22,000 24,000 110 99 Maximum 25 2,100 22,000 40,000 110 99 250 Minimum 15 2,000 9,500 9,000 38 32 Median 22 2,100 27,000 24,000 112 99 Maximum 85 2,500 29,000 40,000 121 109 400 Minimum 22 2,000 18,000 9,000 45 38 Median 22 2,100 48,500 24,000 121 111 Maximum 85 2,500 54,000 40,000 135 122 1000 Minimum 21 2,000 125,000 15,000 125 112 Median 22 2,100 138,750 24,000 139 126 Maximum 25 2,100 146,000 40,000 146 136 * A minimum of three different manufacturers had to produce the type and similar wattage to have that wattage class listed in this table. See Appendix D for a comprehensive list of lamp data from manufacturers catalogs. Correlated-color temperature is not a performance metric; the values are listed to show the range of available CCTs for the product. Initial efficacy is initial lumens divided by the lamp rated power (the power use of the ballast is not factored into this table). Mean efficacy is the mean lumen value provided by the manufacturer divided by the lamp rated power (the power use of the ballast is not factored into this table). Manufacturers select the time at which they provide the mean lumen data, which vary between 40 and 50 percent of rated life. Lumen depreciation still occurs after this point. Also, not all manufacturers publish mean lumen values, which can be higher than the initial efficacy values because the mean lumens were not provided. This column is based on mean lumens published by manufacturers (a full list can be found in Appendix D). Lamp Power* W CRI CCT K Initial Lumens Life hrs

3-16

Figure 3.4.2 shows the SPD for a typical HPS lamp. During warm-up, the lamps color appearance shifts from red to blue, before stabilizing at the characteristic yellow-white.2 A typical CCT value is 2100K with a CRI of 22. Higher CCT values are possible at the cost of reduced efficacy and lifetime. The highest commercially available CCT is 2700K with a CRI of 85. 37 Applying diffuse coatings does not significantly change the SPD of HPS lamps because most of the radiation emitted by the lamp is in the visible portion of the spectrum.
LU400/ECO
Output (Watts per 5nm for 1000 lumens)

0.30 0.25 0.20 0.15 0.10 0.05 0.00 300 400 500 600 700 800 Wavelength (nm)

Source: Sylvania, 2003. Reproduced with permission.10 * Same distribution for bare and coated lamps.

Figure 3.4.2. Spectral Power Distribution* of High-Pressure Sodium Lamps 3.4.3 Electr ical Char acter istics

Given their arc tube geometry and lack of starting probes, HPS lamps require an ignitor to generate the high-voltage pulse necessary to initiate the electrical arc across the electrodes. Once the ignitor initiates the arc, the cold lamp requires approximately 34 minutes of warm-up time to attain full light output, with a re-strike time typically ranging 13 minutes. Figure 3.4.3 shows the energy balance from an example 400 W HPS lamp.11 Although this diagram dates from the mid-1970s and is for one lamp wattage, the diagram is representative of how the power is used. The percentage values in parentheses are the portion of power for that input or output per rated lamp wattage. Only 30 percent of the power used by the lamp is actually converted into visible radiation. The efficiency of the conversion of power into radiation increases with power. Very little lamp power (approximately 1 percent) is emitted in the UV range. Some power (20 percent) is emitted in the IR range; however, a larger portion of the power is emitted in the visible range. The values in Figure 3.4.3 are consistent in relationship to the energy output of an HPS lamp as listed in Table 3.2.2.

3-17

Figure 3.4.3. Sample Power Flow of a 400 W High-Pressure Sodium Lamp Unlike a typical MV or MH lamp, where voltage remains relatively constant throughout its lifetime, the arc-tube voltage increases significantly during the operating life of HPS lamps. Sodium is lost over the operating lifetime of the lamp, and the arc voltage rises as the ratio of sodium to mercury changes. Although the voltage rises over life, the technology needs to operate a constant wattage. Therefore, HPS ballasts must compensate for the change in operating lamp voltage to maintain an acceptable average wattage. The lamp/ballast combination constitutes a dynamic system in which the intrinsic voltage rise of an HPS lamp is either amplified or attenuated by the nature of the characteristic curve of the ballast on which the lamp is operated. 38 Figure 3.4.4 illustrates the boundary conditions that restrict the performance of the lamp and the ballast to certain acceptable limits. These boundary conditions, due to their shape on the graph, are referred to as the trapezoid-shaped boundaries (ANSI C78.42). HPS ballasts operate HPS lamps within this trapezoid for any input voltage within the designed range of the ballast.

3-18

Source: NEMA, 2001. Reprinted by permission from ANSI document C78.42-2001 High-Pressure Sodium Lamps. 39

Figure 3.4.4. Trapezoid Boundary for High-Pressure Sodium Ballast Design Toward the end of the lamps life, the lamp-operating voltage will rise to a level beyond the ballasts capability to sustain the arc. At this point, the lamp will start, warm up to full brightness, and extinguish. The repeating of this sequence is known as cycling and signals endof-life for HPS lamps. If the lamp is then not replaced within a reasonable amount of time, premature failure of the ballast can occur. In the 1990s, the industry introduced non-cycling HPS lamps. Manufacturers increased the xenon pressure and added a starting strip on the arc tube to assist starting. They also reduced the mercury dose and included lead-free construction. This type of lamp operates on standard HPS ballasts. Non-cycling lamps, with lowered additive dose construction, do not have a voltage rise at end-of-life. Because sodium loss cannot be eliminated completely, the lamp will gradually become fully unsaturated in the useful operating regime at some point in life. As the lamp then ages, its operating point will begin to shift downward and to the left on the watts-volts plot. 40 Other research in the 1990s found that a specific combination of mercury and sodium in the lamp to maintain mercury and sodium pressures lead to a lamp in which the voltage did not significantly increase over life. 41 Through the use of this specification relationship and changing the material of the emitter in the lamp, less sodium was lost and lumen maintenance was slightly improved. The simplest and most economical ballast for HPS is the RX ballast, which is commonly used with 35- to 150-W HPS lamps. HX ballasts are used where the input voltage does not generate sufficient voltage to operate the lamp. The CWA is the most popular ballast type for balancing cost and performance (similar to ballasts for MH and MV lamps). CW ballasts have some advantages over the previous three, but are physically larger, have higher losses, and are more expensive. Other ballasts may incorporate three coils in their design instead of the usual two or one; the third coil and the capacitor stabilize the lamp.

3-19

3.4.4

Other Issues

Typical HPS lamps contain mercury as an amalgam with sodium. Over the years, manufacturers lowered the dose of mercury in many of their lamps with common dosage for a 250 W lamp of approximately 15 mg. This is much less than that found in MH lamps because of the sodium-mercury amalgam used in HPS technology. In addition, some manufacturers are selling lamps free of lead solder. TCLP tests classify some HPS lamps as exempt and others as hazardous waste, subject to state and local disposal regulations. 3.5 METAL HALIDE LAMPS

Developed in the 1960s after HPS lamps, MH lamps offer an excellent combination of quality, performance, and flexibility for white light applications. The first lamps were probestart quartz (circa 1960), then pulse-start quartz (circa early 1990s), and then pulse-start ceramic (circa late 1990s). The sources provide significant amount of light for the package. Unlike MV and HPS lamps, MH lamps have many suitable uses in both interior and exterior applications. The lamps are available in a wide range of CCTs and CRIs ranging from 2,900 to 10,000K and 65 to 90, respectively. The small physical dimensions and high intensity of the arc make it an ideal point source around which to create application-specific light distributions through precise optical design in reflector lamps and fixtures. MH lamps are in use in public, commercial, industrial, and residential sectors. The lamps are even used in transportation activities like vehicle forward lighting and range in wattages from 20 to 2000. Low-wattage (below 150) lamps are used floodlighting and recessed lighting. Very low-wattage MH lamps (in the range of 20 to 60) can be found in exterior accent and landscape luminaires (both residential and commercial) and retail track lighting. Also, since around 2000, very low-wattage MH lamps can be found in automotive headlights. Medium-wattage MH lamps (150 to 500) typically are used in area (e.g., parking lot and pedestrian walkways), roadway, low-bay and high-bay applications. Finally, high-wattage MH lamps are used in some parking lots, search lights, and stadium/sports lighting. Some manufacturers produce MH retrofit lamps capable of directly replacing other lamps. Some lamps are designed to replace MV lamps of similar wattage, and operate on the same MV ballasts. However, like HPS retrofit lamps, MH retrofit lamps produce significantly more lumens for the same wattage. For example, replacing a 400 W MV lamp with a 400 W MH retrofit lamp generates about twice the light of the original MV lamp. MH lamp manufacturers have also developed self-ballasted, very low-wattage (under 40) PAR38 lamps intended as direct replacement for higher wattage halogen and incandescent reflector lamps. These PAR38 lamps are used in high-end retail applications and even grocery stores. 3.5.1 Constr uction

The standard MH lamp is similar to MV lamp technology; the lamp is essentially a MV lamp containing one or more MHs (usually iodides) which vaporize in the lamp arc. 42 The added compounds broaden the MV spectrum, producing better color rendering and lamp efficacy in MH lamps. The use of MHs presents two advantages. First, they are more volatile at typical lamp operating temperatures than pure metals. This allows metals with desirable emission properties to

3-20

be introduced into the arc at reasonable arc tube temperatures. Second, they do not readily react with a fused silica arc tube, instead remaining in the arc. When the lamp attains full operating temperature, MHs in the arc tube partially vaporize. As halide vapors approach the hightemperature central core of the discharge, they dissociate into halogen and metals, with the metals radiating their light spectrum. As halogen and metal atoms move near the cooler arc tube wall by diffusion and convection, they recombine, and the cycle repeats. 43 Figure 3.5.1 and Figure 3.5.2 compare and contrast power flow of two 400-W MH lamps. Although dated, the diagrams illustrate how different MHs used in MH lamps affect SPD and overall lumen output.11 Figure 3.5.1 is a tin iodide (SnI) lamp and produces 24,000 lumens, CCT of 4100K, and a CRI of 91. In contrast, another lamp at the same wattage and type (MH) has very different output. Figure 3.5.2 is a dysprosium iodide (DyI) lamp and produces 32,400 lumens, CCT of 5250K, and a CRI of 85.

Figure 3.5.1. Tin Halide

Figure 3.5.2. Dysprosium Iodide

MH lamp construction is similar to that of MV lamps. One significant difference is that arc tubes of MH lamps are usually smaller for equal wattages. The arc tube is typically made of either fused quartz or ceramic. MH arc tubes operate at a higher temperature and pressure than their MV counterparts. Quartz arc tubes typically use a white coating at their ends to increase temperature and vaporization of MHs by trapping thermal energy inside the arc tube. Replacing quartz in the arc tube with ceramic materials enables the lamp to operate at even higher temperatures and pressures, resulting in improved efficacy, CRI, and color stability. An elliptical-shaped outer bulb, usually made of borosilicate glass, contains the arc tube. The bulb protects and buffers the arc tube and internal electrical connections from the environment. The outer envelope contains low-pressure inert gas (i.e., nitrogen) or a vacuum, which not only helps minimize the oxidation of internal components, but also provides a margin of safety against threat of arc tube rupture (also known as non-passive failure). The outer envelope also provides additional thermal buffering for a more stable arc temperature. The glass itself absorbs the majority of UV energy.20

3-21

Manufacturers may coat the outer glass bulb with diffusing material. Usually, this is a phosphor coating that helps change emission spectra for small improvements in CRI. However, the primary purpose of the coating is to reduce source brightness of the lamp for glare control. The technology advantage of MH lamps is their capability to emit the desired spectra without the addition of a phosphor coating. Adding a diffuse coating to the lamp reduces the efficiency and effectiveness of luminaires with optics. Within the outer bulb are wires suitable for high temperatures to conduct electricity to the arc tube and structural components to support the arc tube. There might be other components, including resistors, diodes, and bimetallic switches used to help start the arc discharge as well as devices called getters to purify the atmosphere in the outer lamp. Most MH lamps use getters to control (reduce or eliminate) impurities that, if present in the outer jacket of MH lamps in sufficient concentrations, can compromise performance. Like other lamps in the HID group, MH lamps typically have screw bases (medium or mogul) made from brass, nickel, or special alloys to minimize corrosion. A characteristic of MH arc tubes is that they can take different forms and can be made from different materials. In a horizontally operated MH lamp, the arc discharge bows away from the axis of the arc tube due to convection currents. The MH pool (liquid) moves to the center of the arc tube. The bowed arc moves farther from the MHs than when the lamp is vertical, causing them to cool. This lowers the vapor pressure of MH chemicals and decreases the concentration of metals in the arc, resulting in loss of light. The bowed arc moves closer to the top of the arc tube wall, causing its temperature to increase. The higher heat loading on the arc-tube material can result in a decrease in life of approximately 25 percent. Therefore, manufacturers employ additional shapes to compensate for this sensitivity in operating position. High-output lamps designed to operate horizontally use a special base and socket, such as the ANSI EP39 Mog position-oriented mogul base, to help ensure proper arc tube orientation and position. Figure 3.5.3 shows a bowed arc-tube shape that follows the natural bowing of the horizontal arc.20

Note: POM means position oriented mogul. 20 Source: Sylvania, High Intensity Discharge Metalarc Lamps, 1998. Reproduced with permission.10

Figure 3.5.3. Formed Arc-Tube Configuration for Horizontal MH Lamps

3-22

Figure 3.5.4 shows an asymmetric arc tube with offset electrodes. In this design, the electrodes are lower in the arc tube body, and the arc discharge bows to the center of the arc tube.
Arc Tube Starter Electrode

Main Electrode
Source: Philips, 2003. Reproduced with permission.
44

Main Electrode

Figure 3.5.4. Offset Electrode Configuration for Horizontal MH Lamps Some lamps use a transparent sleeve called a shield or shroud, to surround the arc tube (Figure 3.5.5). The sleeves are useful as heat shields because they help the arc tube achieve a uniform temperature profile. Shrouds also prevent the outer jacket of the lamp from breaking in case of a violent arc-tube failure; however, they are not the only method of protection (e.g., some new ceramic tubes have a wrapped wire instead of a sleeve). HID lamps with protective shrouds carry the ANSI O-rating, signifying that they meet ANSI requirements for safe operation in an open-luminaire not equipped to protect against violent failure. To prevent user misapplication, the industry developed unique socket and base combinations in both medium and mogul bases for these open-luminaire lamps.

Source: Venture Lighting, 2003. Reproduced with permission. 45

Figure 3.5.5. Metal Halide Arc Tube with a Shroud Some applications, such as sports lighting, use MH lamps without outer jackets. Due to high levels of UV energy present in its emission spectra, luminaires that operate with these lamps (no outer jackets) require a glass cover to adsorb some of this harmful UV radiation.

3-23

3.5.1.1

Pr obe-Star t Quar tz Metal Halide Lamps

The first MH lamp was the probe-start lamp (Figure 3.5.6), where the arc tube is shaped into what is called the classic pinched body. These lamps require a starter electrode (or probe) in the arc tube. The lamp body must also contain a system that provides for cutting starting voltage after the lamp is started, typically with a bimetal switch. Continued application of high starting voltages can severely shorten lamp life. The location and type of bimetal switch can restrict the lamp operating position, as the bimetal must achieve a certain temperature to function.

Source: GE Lighting, High-Intensity Discharge Lamps Catalog, 2003. Reproduced with permission.9

Figure 3.5.6. Metal Halide Lamp Construction An evolutionary development of the probe-start lamp configuration is the formed-body arc tube. Manufacturers can mold arc tubes into unique shapes such as an ovoid (Figure 3.5.7), resulting in significant improvements in performance. The contour of the formed-body arc-tube wall follows the shape of the arc, thereby allowing for a more uniform temperature profile for the arc tube. Also, formed-body arc tubes have much smaller pinch-seal areas. This allows MH chemicals to heat up more rapidly than those in conventional pinched-body arc tubes, resulting in a shorter warm-up time than pinched-body arc tubes of the same wattage. Probe-start MH lamps typically employ pinch-seal designs (Figure 3.5.7) that have a starter electrode (or probe) to help initiate the breakdown of arc-tube gases.

3-24

(a) Ovoid

(b) Pinched-Body

Source: Advanced Lighting Technologies Inc., 2003. Reproduced with permission.22

Figure 3.5.7. Common Arc Tubes 3.5.1.2 Pulse-Star t Quar tz Metal Halide Lamps

Thirty years after probe-start MH lamps entered the market, pulse-start MH (PMH) lamps were introduced. These lamps are similar to probe-start, however, these lamps lack a starting probe in the arc tube. Pulse-start MH lamps (section 3.5.1.3) were the first to incorporate formed-body arc tubes, but this arc-tube shape is not exclusive to pulse-start configurations. Some manufacturers still use pinched-body arc tubes in their PMH lamps. 46 Pulse-start MH lamps have a higher starting-gas fill pressure to decrease the starting time and minimize the transport of electrode material to the wall. These improvements result in longer life (up to 50 percent longer) and improved lumen maintenance (up to 33 percent) compared with traditional probe-start MH lamps. 3.5.1.3 Pulse-Star t Cer amic Metal Halide Lamps

The most recent development in MH arc tube design incorporates the ceramic arc tube. Like formed-body arc tubes, this pulse-start technology lacks a starter probe. This arc tube represents the newest evolution in the development of MH lamps. The lamp performance and electrical-characteristic segments of this MH lamp chapter discuss the improvements in performance from these new shapes and materials.

3-25

3.5.2

Lamp Per for mance

Commercially available MH lamps have initial efficacies from 30 to 125 LPW (excluding ballast losses). The lowest efficacy is related to a type of directional lamp, and is not typical of the MH lamp family. Depending the on the type of MH lamp, the CRI ranges from 65 to 90, far superior to virtually every other HID lamp (MV and HPS). This unique capability results from manufacturers ability to tailor the spectral outputs of the lamps. Metal halides are binary compounds consisting of halogen and metal elements. Halogens are a group of non-metal including fluorine (F), chlorine (Cl), bromine (Br), iodine (I), and astatine (At). Iodine in the form of ions (also known as iodide) is the most common halogen used in MH lamps. Bromine is the other halogen used, but less often than iodine. Metal halides are additive compounds, so nearly any metal can be combined with halogens. However, certain metals are not used because of toxicity, instability, or other problems with the metals at the required vapor pressures. Excluding those metals, roughly 50 different metals can still be used in MH lamps. Using two metals at a time, the number of possible combinations (not counting variations in concentration) is 1225. The metals can be used three, four, or five at a time, giving 19,600, 230,300, or 2,118,760 combinations. 47 Although a significant number of possible combinations exist, metals are typically selected with strong emission lines in the desired portion of the spectrum. Typical metals used in MH lamps include indium (In), tin (Sn), and the rare earth (RE) elements which include scandium (Sc), yttrium (Y), and 15 lanthanoids: lanthanum (La), cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm), europium (Eu), gadodinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb), and lutetium (Lu). Other MHs that have been explored are alkaline earth haides, a series of metals that comprise Group 2 of the periodic table: beryllium (Be), magnesium (Mg), calcium (Ca), strontium (Sr), barium (Ba), and radium (Ra). A 1976 study examining improving MH lamps found that: In many cases, e.g., sodium and the rare earth halides, an increase in vapour pressure will result in a higher luminous efficacy. In other cases, e.g., the alkaline earth halides, one could enhance the emission at certain wavelengths and thus substantially improve the colour rendering properties of the lamp. 48 Metal halides in these lamps have characteristic emissions that are spectrally selective. Some metals emit visible radiation in narrow spectral bands, generating specific colors to the eye (i.e., Na for yellow-orange, Tm for green, and In for blue). A 1985 study examining the spectral properties of MH lamps with rare earth iodides found that the spectra of the metal can be selectively used: This results in the possibility of optimizing the spectra for distinct applications. In 400 W, lamps the highest efficacies obtained were reached with Ce-Na and Pr-Na mixtures, = 110 to 115 lm/W [LPW]. The best colour rendering properties can be obtained with Tm and Dy (Ra>80). 49

3-26

Other metals such as tin, when introduced as halides, radiate predominantly as molecules, providing continuous band spectra across the visible spectrum. Various combinations of these halides can create a wide range of CCT and CRI for these lamps. Phosphor coatings also change the spectral emission of these lamps. Because phosphors convert shorter-wavelength radiation to longer-wavelength radiation, these phosphors effectively lower the CCT of lamps by approximately 300K. Although a slight improvement in CRI may occur, the primary function of a phosphor coating of MH lamps is to create a more diffuse light source. These combinations of metals and their halides as well as phosphors (to a lesser extent) create an almost infinite palette of possibilities, giving MH lamps flexibility in light-output design unique to the HID sources. Figure 3.5.8 shows typical SPD plots for MH lamps with and without a phosphor coating.
MP100UMED
Output (Watts per 5nm for 1000 lumens)

0.30 0.25 0.20 0.15 0.10 0.05 0.00 300 400 500 Wavelength (nm) 600 700
TR20014366

800

(a)
MP100/C/U/MED 0.30 0.25 0.20 0.15 0.10 0.05 0.00 300 400 500 Wavelength (nm) 600 700
204443b

Output (Watts per 5nm for 1000 lumens)

800

(b)
Source: Sylvania, 2003. Reproduced with permission.10

Figure 3.5.8. Spectral Power Distribution of Metal Halide Lamps (a) Without Phosphors and (b) With Phosphors

3-27

Because the physical operating orientation of the lamp affects performance, MH lamps are typically life-rated and lumen-rated in the vertical operating position. For example, MH lamps classified as universal orientation (able to operate in any orientation) achieve their rated performance in the vertical position. 3.5.2.1 Pr obe-Star t Quar tz Metal Halide Lamps

When operated to specifications, probe-start lamps have a rated life from 10,000 to 20,000 hours with an average efficacy of 84 LPW. The actual efficacy varies by both the lamp wattage and type (omni-directional are more efficacious than directional lamps). The rated LLD is about 25 percent at 40 percent of rated life. Color quality is an important aspect of MH lamps. The average CRI for probe-start lamps is 76; actual values range as in Table 3.5.1). However, CRI only examines a sample of eight (R1R8) pastel color samples, atypical for some applications. Therefore in addition to the CRI values, many MH manufacturers are now reporting the R9 (deep red color) for specification in interior applications, especially retail. The R9 value of probe-start lamps is negative. 50 Therefore probestart lamps have limited use in interior applications and are not practical for retail installations. Over life, the color shift amongst the same batch of MH lamps can be from 300 to 800K. 51 Table 3.5.1. Performance Summary of Probe-Start Metal Halide Lamps
Lamp Power* W 50 Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum CRI CCT K 2,900 3,500 4,000 2,900 4,000 6,500 2,900 4,000 6,500 2,800 3,850 6,500 3,000 4,000 10,000 3,000 4,000 10,000 3,600 4,000 4,300 3,000 4,000 10,000 3,200 3,700 5,500 Initial Lumens 3,200 3,525 4,100 2,610 5,000 6,200 6,000 8,400 9,500 5,400 11,950 14,000 10,000 13,600 15,000 17,500 20,750 23,000 31,700 36,000 43,200 25,000 36,050 43,000 72,000 110,000 120,000 Life hrs 10,000 10,000 20,000 3,200 9,000 20,000 9,000 15,000 20,000 2,800 10,000 20,000 7,500 10,000 10,000 7,500 10,000 20,000 15,000 20,000 30,000 8,000 20,000 20,000 6,000 12,000 18,000 Initial Efficacy @ 100 hrs LPW 64 71 82 37 71 89 60 80 95 36 80 93 57 78 86 70 83 92 88 100 120 63 90 108 80 110 120 Mean Efficacy @ 40% life LPW 30 46 55 43 55 66 48 59 71 48 60 70 43 50 69 38 55 78 57 65 98 44 60 84 52 86 96

70

100

150

175

250

360

400

1000

70 85 92 70 92 96 65 92 96 65 95 96 55 65 75 62 65 90 60 65 80 60 65 90 65 65 72

3-28

Table 3.5.2. Performance (continued)

Minimum 60 3,400 125,000 3,000 83 67 Median 65 4,000 163,000 3,000 109 93 Maximum 92 5,500 170,000 6,000 113 102 * A minimum of three different manufacturers had to produce the type and similar wattage to have that wattage class listed in this table. See Appendix D for a comprehensive list of lamp data from manufacturers catalogs. Correlated-color temperature is not a performance metric; the values are listed to show the range of available CCTs for the product. Initial efficacy is initial lumens divided by the lamp rated power (the power use of the ballast is not factored into this table). Mean efficacy is the mean lumen value provided by the manufacturer divided by the lamp rated power (the power use of the ballast is not factored into this table). Manufacturers select the time at which they provide the mean lumen data, which vary between 40 and 50 percent of rated life. Lumen depreciation still occurs after this point. Also, not all manufacturers publish mean lumen values. The mean efficacy values can be higher than the initial efficacy values because the mean lumens were not provided, and this column is based on mean lumens. 1500

3.5.2.2

Pulse-Star t Quar tz Metal Halide Lamps

When operated to specifications, the rated life of pulse-start MH lamps is 15,000 to 30,000 hours with efficacies in excess of 100 LPW. The LLD is about 25 percent at 40 percent of rated life. Table 3.5.2 summarizes the performance of PMH sources. Table 3.5.3. Performance Summary of Quartz Pulse-Start Metal Halide Lamps
Lamp Power* W CRI CCT K Initial Lumens Life hrs Initial Efficacy @ 100 hrs LPW 58 66 82 64 74 106 58 85 95 73 83 100 74 93 100 66 90 104 70 94 106 88 Mean Efficacy @ 40 % life LPW 30 41 55 40 54 79 48 58 72 55 64 75 57 71 80 52 69 80 28 72 84 57

50

70

100

150

175

250

320

350

Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum Median Maximum Minimum

60 70 92 65 82 96 65 73 92 60 85 96 62 65 75 60 70 94 62 70 90 62

2,900 3,300 4,000 2,800 3,500 6,500 2,700 3,200 4,200 2,800 3,550 6,500 3,200 4,000 4,200 3,000 4,000 6,500 3,200 4,000 5,000 3,200

2,900 3,300 4,100 2,610 5,200 7,400 5,800 8,500 9,500 5,400 12,500 15,000 13,000 16,000 17,500 16,500 22,500 26,000 22,500 30,000 34,000 25,000

10,000 10,000 20,000 3,200 12,000 20,000 7,500 15,000 20,000 2,800 12,000 20,000 10,000 15,000 15,000 10,000 15,000 20,000 10,000 20,000 20,000 10,000

3-29

Table 3.5.4. Performance (continued)

Median 67 4,000 34,500 20,000 99 74 Maximum 90 5,000 51,628 30,000 120 98 400 Minimum 60 3,600 27,000 10,000 68 39 Median 65 4,000 39,850 20,000 100 75 Maximum 90 6,500 44,000 20,000 110 90 1000 Minimum 65 3,400 80,000 5,000 0 64 Median 65 3,900 110,000 12,000 110 88 Maximum 92 5,500 120,000 15,000 120 96 * A minimum of three different manufacturers had to produce the type and similar wattage to have that wattage class listed in this table. See Appendix D for a comprehensive list of lamp data from manufacturers catalogs. Correlated-color temperature is not a performance metric; the values are listed to show the range of available CCTs for the product. Initial efficacy is initial lumens divided by the lamp rated power (the power use of the ballast is not factored into this table). Mean efficacy is the mean lumen value provided by the manufacturer divided by the lamp rated power (the power use of the ballast is not factored into this table). Manufacturers select the time at which they provide the mean lumen data, which vary between 40 and 50 percent of rated life. Lumen depreciation still occurs after this point. Also, not all manufacturers publish mean lumen values. The mean efficacy values can be higher than the initial efficacy values because the mean lumens were not provided, and this column is based on mean lumens.

3.5.2.3

Pulse-Star t Cer amic Metal Halide Lamps

Advancements in arc tube materials led to the development of the ceramic arc tube. This material allows MH lamps to operate at even higher temperatures and vapor pressures. The technology is currently available only in wattages up to 400. Improvements in ceramic arc-tube design will soon make higher-power versions available. When operated to specifications, CMH lamps have a rated life from 6,000 to 20,000 hours with an efficacy of approximately 88 LPW. The LLD improves over quartz MH lamps to about 20 percent at 40 percent of rated life. In addition, CMH lamps have excellent CRI with an average CRI of 88. Typical R9 values are positive but very low. Recent advances in CMH lamp design allows for both high CRI (90), good R9 (over 40), greater efficacy and improved lumen maintenance. These new CMH lamps, currently produced by Philips, are medium wattage (315 and 210) and have an efficacy over 120 LPW. In contrast, other similar wattage CMH or PMH lamps are between 90 and 115 LPW. The lumen output on these new lamps is 90 percent at 40 percent of rated life; in contrast to typical 80 percent for CMH, 75 percent for PMH, and 70 percent for MH. The developers of this new CMH lamp attribute the performance enhancements to operating the lamp on an electronic ballast; utilizing new chemical additives (presumably different MHs); and optimized arc tube shape.27 Table 3.5.3 summarizes the performance of CMH lamps.

3-30

Table 3.5.5. Performance Summary of Pulse-Start Ceramic Metal Halide Lamps

Initial Efficacy Mean Efficacy @ 100 hrs @ 40-50% life LPW LPW 50 Minimum 81 2,900 1,400 6,000 72 49 Median 85 3,000 2,225 10,000 76 52 Maximum 92 4,200 3,500 15,000 82 57 70 Minimum 80 2,900 2,850 9,000 51 37 Median 88 3,000 5,800 12,000 81 61 Maximum 95 4,300 7,300 24,000 104 93 100 Minimum 81 3,000 5,700 10,000 57 40 Median 85 3,000 8,200 12,500 82 62 Maximum 93 4,100 9,500 24,000 95 72 150 Minimum 80 2,800 9,100 5,000 61 55 Median 90 3,000 12,800 10,000 85 63 Maximum 96 5,500 15,500 24,000 103 83 250 Minimum 85 3,000 20,750 12,000 83 66 Median 90 4,050 22,500 20,000 90 73 Maximum 94 4,200 25,000 24,000 115 103 400 Minimum 85 3,000 34,800 15,000 87 70 Median 90 4,000 38,000 20,000 95 78 Maximum 90 4,200 41,000 20,000 102 82 * A minimum of three different manufacturers had to produce the type and similar wattage to have that wattage class listed in this table. See Appendix D for a comprehensive list of lamp data from manufacturers catalogs. Correlated-color temperature is not a performance metric; the values are listed to show the range of available CCTs for the product. Initial efficacy is initial lumens divided by the lamp rated power (the power use of the ballast is not factored into this table). Mean efficacy is the mean lumen value provided by the manufacturer divided by the lamp rated power (the power use of the ballast is not factored into this table). Manufacturers select the time at which they provide the mean lumen data, which vary between 40 and 50 percent of rated life. Lumen depreciation still occurs after this point. Also, not all manufacturers publish mean lumen values. The mean efficacy values can be higher than the initial efficacy values because the mean lumens were not provided, and this column is based on mean lumens. Lamp Power* W CRI CCT K Initial Lumens Life hrs

3.5.3

Electr ical Char acter istics

The additional electrode located at one end of the arc tube in probe-start MH lamps allows operation without an ignitor circuit, making it possible for MH lamps to share the same ballast with MV lamps, within limits. However, MH ballasts typically have more rigorous operating requirements than MV ballasts. Thus, MV lamps can operate universally on similar wattage probe-start MH ballasts, but the reverse is true only with specially designed MH lamps. Lamp manufacturers took advantage of this and produced MH lamps that could operate on MV lamp ballasts. However, replacement raises issues of economy and utility, in addition to technological issues. The engineering analysis (chapter 4) of this document addresses these issues. Pulse-start MH lamps lack the secondary electrode found in probe-start MH lamps, resulting in different electrical requirements and performance. Using an ignitor, a high-voltage pulse (typically in excess of 3 kilovolts) is applied directly across the main electrodes to initiate the arc, resulting in faster starting and re-strike times than for probe-start MH lamps. The warmup and re-strike time are 2 minutes and 34 minutes, respectively. That is an improvement over

3-31

the typical probe-start MH warm-up time of 4 minutes and a restrike time of 1520 minutes; these results are also better than those of typical MV and HPS lamps. Common probe-start MH ballast systems for mid-wattage MH lamps include CWA and CW ballasts. Pulse-start MH lamps require ballasts similar to those for probe-start lamps, but also require an ignitor. Common pulse-start ballasts include CWA, HX, RX, regulated lag (magnetically regulated), and CW ballasts. Electronic ballasts also are available (in limited wattages) for these lamps, promising even better performance in the specific lamp/ballast combination (Appendix E). In addition, high-ignition voltages necessitate the use of high-voltage lamp sockets. 3.5.4 Other Issues

Mercury is a double-sided issue in the lighting industry. Although HID lamps are more efficacious than halogen/incandescent (mercury free) lamps; mercury is needed in most vapor discharge sources (both low and high pressure). However, the use of these sources displaces more mercury released in energy production, so it still is beneficial to use a discharge source over incandescent/halogen. The amount of mercury in an HID source varies by lamp type as well as by wattage. Table 3.5.4 is from a 2008 report about trends in mercury use in products. 52 In both 2001 and 2004, lamps represented about 8 percent of the total use of mercury in products. There was roughly a 6 percent reduction in overall mercury usage in lamps between 2001 and 2004. The total use of mercury in HID lamps increased by 15 percent, whereas the use by fluorescent lamps declined by 14 percent between 2001 and 2004. The reasons behind total mercury usage being relatively flat, while these two lamp types changed significantly is that fluorescent lamps have been designed to use less mercury and more HID lamps were specified between the two years. Table 3.5.5 breaks down the mercury sold by HID lamp type. The vast majority of mercury usage in HID lamps is in the MH family. Table 3.5.6. Total Mercury Sold in Lamps
All Lamps Total Mercury Weight lbs (kg) 2001 16,657 (7,555.48) 2,749 (1,246.93) 877 (397.80) 1,103 (500.31) 10 (4.54) 42 (19.05) 21,438 (9,724.11) 2004 14,372 (6,519) 3,156 (1,431.54) 1,479 (670.86) 1,070 (485.34) 17 (7.71) 24 (10.89) 20,118 (9,125.37)

Fluorescent High Intensity Discharge Compact Fluorescent Neon Short Arc Miscellaneous Total

Table 3.5.7. Total Mercury Sold by HID Lamp Type

Lamp Type 2001 Total Mercury Sold by Weight lbs (kg) 2,145 (972.95) N/A 401 (181.89) Portion of total HID % 78 0 15 2004 Total Mercury Sold by Weight lbs (kg) 2,426 (1100.42) 31 (14.06) 453 (205.47) Portion of total HID % 77 1 14

MH Ceramic MH HPS

3-32

Table 3.5.8. Total (continued)

MV Total 203 (92.08) 2,749 7 -213 (96.62) 3,156 7 --

Mercury is integral to the three lamp types, though more for some than others. HPS lamps have been designed with low mercury and even no mercury content (labeled as such in catalogs). The typical amount of mercury for HPS lamps from 35 to 1000 W is between 10 and 50 mg. By definition, MV lamps must have mercury in them. The typical mercury content in MV lamps from 50 to 1000 W is between 50 and 150 mg. Because MH technology is related to MV lamps, mercury plays a vital role in these lamps. The typical mercury content in MH lamps from 39 to 50 W is between 1 to 10 mg; 70 to 250 W is between 5 to 50 mg; 350 to 400 W is between 30 to 100 mg; and 400 to 2000 W is between 100 to 150 mg. In response to reducing mercury in products, mercury-free HID lamps have been investigated. A 2007 research paper considered replacing mercury with noble gases. Noble gases include helium (He), neon (Ne), argon (Ar), krypton (Kr), xenon (Xe), and radon (Rn). The study experimented with Ne, Ar, and Xe. The presented results emphasize Ar and Xe as proper Hg substitute candidates. 53 Lead is sometimes present in these lamps as solder for the coupling. A TCLP test on these lamps classifies them as hazardous waste for mercury and lead. Therefore, they are subject to the Universal Waste Rule (EPA, July 6, 1999). State and local regulations will vary on their disposal, but recycling is recommended. A small percentage of MH lamps contain the radioactive isotope, krypton 85. The MH lamps that contain this radioactive isotope are primarily used in special applications of theatrical and film luminaires. 3.6 HIGH-INTENSITY DISCHARGE LAMP BALLASTS

High-intensity discharge lamp ballasts apply sufficient voltage for lamp ignition, regulate the operating current, and essentially relight the lamp at each one-half cycle of the voltage. A 1983 paper focusing on the relationship between HPS lamp voltage rise and the ballast summarized the relationship: A lamp and ballast cannot be considered separate elements but each contributes to a dynamic system influencing each other in achieving total system performance.15 Although this quote was in relation to HPS lamps, it is applicable to HID lamps in general. For HPS and PMH lamps, an additional ignitor circuit generates the high-voltage starting pulse necessary to initiate the arc discharge. Typical input voltages for HID lighting systems are 120, 208, 240, 277, and 480 V. There are two types of HID ballasts: magnetic and electronic. The majority of ballasts sold for HID lamps are magnetic, also known as core-and-coil. The ballasts operate at 60 Hz. Within the magnetic HID ballast family, there are seven sub-types of ballasts: (1) high-reactance autotransformer; (2) reactor-normal power factor; (3) reactor-high power factor; (4) regulatedlag, also known as magnetically regulated; (5) CWA; (6) constant-wattage isolated; and (7) super constant-wattage autotransformer. CWA has become the most popular HID ballast for lamp

3-33

power levels above 150 W. 54 Ballast efficiency varies with rated lamp power, less ballast loses occur as the rated power increases. b The other major type of ballast is electronic; currently, electronic ballasts only exist for pulse-start lamps. Within the electronic HID ballast category, there are only two types: low frequency (less than or equal to 400 Hz) and high frequency (equal to or above 100 kHz). Most electronic HID ballasts are for MH lamps, with a small portion for HPS lamps, and no electronic ballasts for MV lamps. A 1990 paper comparing some of the first electronic HID ballast with conventional ballasts found that when electronic ballasts for high pressure lamps will be different than electronic ballasts for low pressure discharge lamps. For electronic ballasts, more sophisticated, forms of generators for lamp operation must be used than in the field of lowpressure discharge lamps. 55 Unlike electronic ballasts for FL lamps, HID electronic ballasts do not always make the lamp-ballast system more efficient. In 2006, the Lighting Research Center (LRC) compared 70 W MH lamps from various manufacturers operating on both magnetic and electronic ballasts by different manufacturers. The LRC study found that ballast efficiency was directly affected by the type of ballast: Because of design and manufacturing practicalities, the ballast efficiency of magnetic ballasts is approximately 10 percent lower than that of electronic ballasts. 56 There are some system (lamp plus ballast) efficiency gains from converting from magnetic to electronic HID ballasts, but there are other electrical issues associated with electronic HID ballasts that affect HID lamps. During the public meeting for the framework document for the MH lamp fixture energy conservation standard, both the disadvantages and advantages of electronic HID ballasts were discussed. A comment from the meeting sums up the issue of electronic HID ballasts: Some electronic ballasts do improve lamp maintenance and lamp life, others do not; as a mater of fact, some electronic ballasts can decrease lamp life. 57 One of the desirable features of HID lamps compared to other sources is the long-rated operating lifetimes of the lamps. Electronic ballasts can both start and operate a lamp more gently than a magnetic ballast, with positive benefits in the lamp performance. A study from the mid-1990s found electronic ballasts can extend lamp life once other electrical issues are addressed: [W]ith attention paid to the proper features of the electronic ballast, namely high extinguishing resistance, it can be concluded that lamp service life with electronic ballasts will be 1.5 to 2.0 times longer than with conventional ballasts. 58 One of the principal disadvantages of electronic ballasts is that the lamps operate at high frequency and this creates acoustic resonance. The LRC study comparing 70 W lamps operating on electronic ballasts describes the negative issue of acoustic resonance: At frequencies above 4 kHz, standing pressure sound waves (acoustic resonances) build inside the lamps arc tube, which can lead to visible distortions and instability of the electrical discharge, light flickering, and possible cracking of the tube.33 As stated earlier, lamp life can be extended via the use of an electronic ballast but also cut short if special precautions in either the ballast or lamp design are not utilized. Ballast design is not the purview of the document. Therefore, only aspects of the lamp that directly affect acoustic resonance are reviewed here, specifically arc tube size and geometry. Pressure waves are created in the lamp when the resonant frequency is close to the operating
b

Ballast Efficiency the ratio of lamp rated power/ballast input power (Pout/Pin).

3-34

frequency. These pressure waves move through the lamp distorting the arc tube path, causing flicker and, in extreme cases, arc contact with the tube wall. If the arc touches the tube wall, the arc tube can crack. A cracked arc tube seriously affects the lamp and can lead to non-passive failure (i.e., rupture) of the arc tube and bulb wall. Lamp Current Crest Factor (LCCF) c is an aspect of either magnetic or electronic ballasts that can affect both the HID lamp life and lumen maintenance. Many within the lighting industry consider lamp life and lumen deprecation the primary performance metrics of a lamp-ballast system.31 The lamp electrodes can be damaged by high LCCFs which directly affects lumen maintenance and lamp life. LCCF values can range from 1.0 to infinity; however, ANSI specifies maximum LCCF values. For example, ANSI C78.43-2007 covers single-ended MH lamps from 39 to 1650 watts and sets a maximum LCCF of 1.8 for these lamps. A 2008 paper comparing the effects on lamp life and lumen depreciation for two different types of ballasts for 400W MH lamps articulated the value of an LCCF limit: The rationale for placing an upper limit upon the LCCF is to prevent the lamp current from falling to an unsuitably low value for a significant period of time on a half-cycle basis.31 Lower LCCF values are better for the lamp electrode and thus the lamp life and lumen maintenance. The LRC study articulated the waveform and LCCF issues of magnetic and electronic ballasts in their comparison: An ideal sinusoidal waveform (and therefore, a magnetic ballast) cannot have an LCCF below 1.4. Most magnetic ballasts have LCCFs ranging from 1.5 to 1.7. The square current waveform of an electronic ballast allows the LCCF to approach 1.0.33 Appendix E provides summary descriptions of each ballast type.

Lamp Current Crest Factor (LCCF) the ratio of the peak lamp current to the root-mean-square (rms) value of the lamp current

3-35

CHAPTER 4.

ENGINEERING ANALYSIS

This chapter provides the technical support documentation for the engineering analysis of high-intensity discharge (HID) lamps. For purposes of this determination, the engineering analysis identifies representative equipment classes with baseline lamps and energy-efficient substitutes, and develops power rating estimates for the equipment analyzed. The outputs from the engineering analysis are critical inputs to subsequent cost-benefit calculations for individual consumers. In this determination, DOE developed end-user prices for analyzed lamp, lamp-andballast, and luminaire designs using a commercially-available pricing database. By combining the results of the engineering analysis and the equipment price determination, DOE derived the necessary inputs for use in the life cycle cost (LCC) analysis and national energy savings/net present value (NES/NPV) analysis. The LCC and NES/NPV analyses are discussed in chapters 5 and 6 of this determination, respectively. 4.1 ENGINEERING ANALYSIS FRAMEWORK

DOE based the engineering analysis on commercially-available equipment that incorporate the design options identified in the market assessment (chapter 2) and technology assessment (chapter 3). For this determination, the engineering analysis employed the following approach: 1. Representative Equipment Categories. DOE reviewed potentially covered lamps and selected certain broad categories as representative, based on their market share, wattage, and color characteristics. 2. Baseline Lamps. Within the broad equipment categories, DOE selected baseline lamps as representative equipment. Generally, a baseline lamp is one that represents the typical lamp sold. DOE selected specific characteristics such as certain color rendering index, operating lifetimes, and light outputs to characterize the most common lamps consumers purchase. DOE selected multiple baseline lamps to ensure consideration of different high-volume lamps and their associated consumer economics. 3. Substitute Lamps. DOE selected more efficacious substitute lamps for each baseline lamp in the representative equipment classes. In identifying more efficacious lamp designs, DOE recognizes that a lamps lumen package and performance characteristics are important design criteria for consumers. For example, if consumers do not have the option to purchase substitute lamps, lampand-ballast systems, or luminaires with similar lumen packages under an energy conservation standard, consumers would need to change the lighting design to maintain a similar light output. Therefore, DOE established energy-efficient designs for the LCC analysis so that potential light output is equal to or within 90 percent (or greater) of the mean light output of the baseline system.

4-1

The sections that follow discuss how DOE applied this methodology to create the engineering analysis. 4.1.1 Repr esentative Equipment Categor ies

For this determination, DOE developed representative equipment classes that represent the plurality of U.S. HID lamp shipments. The equipment classes are categorized broadly by lamp wattage and baseline lamp technology. HID lamp wattages were segregated into three categories: low (<150W), medium (150 500W), and high (>500W). Based on its market assessment, DOE selected medium wattage categories as most representative of lamp shipments, comprising an estimated 38 percent of U.S. HID lamp shipments. The equipment categories are further specified by lamp technologyin this case, mercury vapor (MV) and probe-start metal halide (MH) lampswhich broadly characterizes the color requirements of HID lamp applications. For the MV baseline equipment classes, DOE assumed that color appearance and color rendering are not critical (e.g., roadway lighting and industrial applications), which allows for energy-efficient substitute lamp types with poor or moderate color rendering (e.g., high-pressure sodium (HPS) and probe-start MH). DOE selected probe-start MH as the baseline lamp technology for equipment classes where color appearance and color rendering may be important (e.g., interior and exterior white light applications in commercial facilities). 4.1.2 Baseline Lamps

Within the broad wattage and lamp technology categories, DOE created multiple equipment classes to capture high-volume equipment and their associated economic impacts. The equipment classes are identified by a nominal initial light output and an associated baseline lamp (MV or probe-start MH), and are presented in section 2.1.3 below. In this determination, DOE based its equipment classes on lamps. This is because the potential energy conservation standards would apply only to lamps. However, DOE used a systems approach in analyzing the representative equipment classes because both lamps and ballasts determine a systems energy use and light output. Accordingly, DOE paired lamps with ballasts to develop representative lamp-and-ballast systems, which demonstrate the actual energy usage and light output of operating lamps. It should be noted that, for this analysis, DOE only considered magnetic ballasts. Although electronic ballasts exist for some HID lamps, these types of ballasts represent a very small segment of the market. 4.1.3 Substitute Lamps

For each representative equipment class, DOE considered the following when identifying acceptable substitute HID lamps: Light Output: Some substitutions are not suitable because of light output differences. Substitute lamps with significantly increased light output may not be appropriate, despite

4-2

their higher efficacy. Higher light output for a system in an existing luminaire may cause glare, which may be a safety hazard and/or create lighting quality issues. In this determination, DOE analyzed lamp-and-ballast systems that both save energy and maintain comparable light output. In particular, DOE considered lamp-and-ballast designs that emit mean lumens equal to, greater than, or no more than 90 percent less than the light output of the baseline system. DOE maintains light output across efficacy levels to ensure that products supply comparable service. In assessing light output, DOE makes a distinction between mean and initial lumens. DOE uses initial lumens in its test procedures to measure compliance with a standard. This is consistent with consensus industry standards on efficacy measurements. Therefore, DOE uses initial lumens to identify energy-efficient substitutes in this determination. However, the light output of a lamp decreases over time. To account for this real-world depreciation in lumens, DOE used mean lumens in determining equivalent substitutes in the LCC and NES/NPV analyses. Color Characteristics: Some substitutions are not suitable because of color (i.e., spectral) differences, which may exclude some technology options for specific applications. For example, an HPS system (with its characteristic amber color appearance and poor color rendering) would not be an appropriate substitute for a MH system in color-critical white light applications (e.g., an interior lobby). However, HPS can be a viable substitute for MH in some exterior applications (e.g., parking structures), where energy efficiency may be a higher priority than color quality. Fixture Constraints: Some substitutions are not suitable because of fixture constraints. A substitution of a different HID technology type may mean the use of a lamp with a different socket type (e.g., mogul, medium, pin) that does not exist in the luminaire. Replacement lamp options and their ballast may not fit in the existing luminaire. Some HID luminaires are not enclosed and cannot provide protection against the possibility of a lamp rupture for some technology replacements. Economics: Lighting economics could significantly affect the choice of a substitution. Lamps that provide acceptable light output and/or color quality can have a high first cost, making their LCC too high to be a reasonable substitute. Material and maintenance costs associated with shorter life and increased lamp replacements can also adversely affect the LCC of the substitute product. EQUIPMENT PRICE DETERMINATION

4.2

For substitutions requiring luminaire replacement, it is important to note that manufacturers typically package fixtures without a lamp, i.e., they package and sell only the fixture, which includes the housing, ballast, and other associated electrical and mechanical components (e.g., wires, sockets, mounting brackets, fasteners). Therefore, to calculate the total price of a luminaire for the LCC analysis, DOE determined the price of associated lamps separately and added this to the fixture price.

4-3

4.3

ENGINEERING ANALYSIS RESULTS

For purposes of this determination, DOE identified seven representative equipment classes of HID lamps that represent an estimated 35% of U.S. HID lamp shipments. The analyzed baseline lamps and their energy-efficient substitutes fall in two broad wattage categories (<150W and 150W-500W) and are classified by initial lumen output. The engineering analysis also captures associated lamp-and-ballast input power ratings, light output, lamp life and pricing information needed for the LCC and NES/NPV analyses. Table 4.3.1 summarizes the engineering data for all baseline and substitute equipment. Table 4.3.1. HID Lamps Determination Engineering Analysis Results
Wattage Category Baseline Lamp Type Equipment Class ID Approx. Light Output (mean lumens) 6,800 Efficiency Level Lamp Description System Input Power Rating (W) 202 185 122 283 185 185 454 285 185 208 185 122 291 209 185 400 367 295 456 367 295

Medium Wattage (150W500W)

MV

10,000

14,400

Probe-start MH

8,800

13,700

23,500

25,200

Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2 Baseline Substitute 1 Substitute 2

175W MV 150W PMH 100W HPS 250W MV 150W PMH 150W HPS 400W MV 250W PMH 150W HPS 175W MH 150W PMH 100W HPS 250W MH 175W PMH 150W HPS 360W MH 320W PMH 250W HPS 400W MH 320W PMH 250W HPS

4-4

CHAPTER 5.

LIFE-CYCLE COST ANALYSIS AND PAYBACK PERIOD ANALYSIS

5.1

Method and Data

To assess the life-cycle cost (LCC) to end users of designs that improve lamp efficacy, DOE conducted an analysis that compares the incremental first cost to the value of electricity savings. In particular, the LCC analysis compares the incremental first cost to the discounted value of electricity savings over the life of the lamp system. The simple payback period (PBP) analysis calculates the amount of time required for the electricity savings to equal the incremental first cost, assuming the electricity savings in future years is constant and equal to that in the first year. The analysis requires several inputs: 1. 2. 3. 4. 5. 6. 7. 8. 9. Typical annual operating hours; Typical price for the lamp and new fixture; Typical lamp lifetime; Typical ballast lifetime; Contractor mark-up; Sales tax; Typical maintenance costs; Electricity price; and Discount rate (to convert future operating cost savings into present value terms). DOE discusses these variables below. 5.1.1 Annual Oper ating Hour s

HID lamps were assumed to operate in both exterior and interior environments, with the following assumed operating schedules. Annual hours of use: 4,200 Exterior applications Annual hours of use: 5,840 Interior applications The exterior operating hours value was derived by assuming that the fixture is controlled by a photocell and operates roughly from dusk to dawn. This translates to roughly 4,200 hours annually 59. This is based on a typical operation of 11.5 hours per day. The interior operating hours value was derived from published research, as reported in three different documents focusing on high-bay luminaires. Reported operating periods ranged between 5,000 and 6,000 hours 60,61, with one document specifically citing 16 hours per day as typical operation 62. DOEs assumed operation of 16 hours per day (5,840 hours per year) is in line with surveyed research, as well as the 2004 draft report, which used a value of 5;080 hours per year.

5-1

5.1.2

Lamp Pr ices

Published prices for lighting equipment vary widely, depending on their source. DOE surveyed several typical pricing data sources in developing representative, conservative prices for HID lamps in this determination. First, DOE reviewed a commonly cited construction estimation reference book, RSMeans. Table 5.1.1 presents HID lamp prices from the 2008 version. Table 5.1.1. Lamp Prices from RS-Means
MV Lamps Lamp Power W --100 --250 400 1000 Price $ --$31.00 --$41.25 $33.00 $77.25 MH Lamps Lamp Power Price W $ --------175 $38.25 ----400 $41.00 1000 $99.00 HPS Lamps Lamp Power W 70 ----250 400 1000 Price $ $44.50 ----$50.50 $52.00 $143.00

DOE also reviewed the prices on municipal web sites. These are procurement contracts that the different municipalities negotiate with distributors. As illustrated in Table 5.1.2, these values can vary widely from state to state for the same lamp 63,64,65,66,67,68. Table 5.1.2. Municipal Lamp Prices
Lamp Type MV MV MH MH MH MH HPS HPS HPS PMH PMH PMH PMH Lamp Power W 250 400 175 250 360 400 100 150 250 150 175 250 320 Arkansas Price $ 44.15 45.69 31.68 35.87 39.11 88.52 ------45.93 39.95 66.52 68.79 West Virginia Price $ 18.89 19.54 15.31 15.20 13.95 32.56 10.17 --10.47 19.19 40.20 66.95 69.24 North Dakota Price $ --------------9.30 ----------Oklahoma Price $ ------------8.87 8.87 9.89 --------Ohio Price $ ------------11.65 11.65 ----------New Jersey Price $ ----------------7.49 ---------

Finally, DOE surveyed lamp prices on publicly-accessible websites for W.W. Grainger, Inc. (Grainger) and GoodMart.com 69,70 in March 2010. These online retailer price catalogues were selected because they offer a wide range of products (i.e., lamps, ballasts, and fixtures) for multiple types of HID lamps and wattages. Figure 5.1.1 presents the lamp type, wattage, and price from the data surveyed. This methodology of reviewing Grainger for prices is consistent with the pricing methodology used in previous rulemakings, including the Small Electric Motor Determination. 71 FR 38799, 38803 (July 10, 2006).
5-2

400 350 300

Comparison of Lamp Power (W) and Price ($)

Initial Price (U.S. $)

250 200 150 100 50 0 0 200 400 600 800 1000 1200 Lamp Rated Power (W)

MH MV HPS PMH

Figure 5.1.1. Lamp Wattage/Price Comparison DOE developed trendlines to interpolate the price for different wattages. The interpolated lines resulted in the following functions per lamp type where x is the lamp wattage and y is the lamp price: R2= 0.2472 Mercury Vapor: y = 4.1461x0.4271 0.1915 Probe-Start Metal Halide: y = 15.366x R2= 0.1059 Pulse-Start Metal Halide: y=5.4365x0.456 R2=0.5063 High-Pressure Sodium: y=3.6562x0.525 R2=0.4029 Based on its analysis, DOE chose to base HID lamp prices for the LCC-PBP analysis on the wattage-price functions developed from Grainger and Goodmart.com price data. This online retailer price data is publicly available (requiring no special log in to access the data) and offers product information that could be applied to the full range of HID lighting system technologies and components. Table 5.1.3 summarizes the assumed prices for the evaluated lamps. Table 5.1.3. Lamp Price derived from Online Distributor Web Sites Baseline Option Substitute Option
Lamp Type MV MV MV MH MH MH MH Lamp Power W 175 250 400 175 250 360 400 Lamp Price $ $37.64 $43.83 $53.58 $41.31 $44.24 $47.43 $48.40 Lamp Type HPS HPS HPS PMH PMH PMH PMH Lamp Power W 100 150 250 150 175 250 320 Lamp Price $ $41.02 $50.75 $66.37 $53.41 $57.30 $67.42 $75.45

5-3

5.1.3

Ballast Pr ices

As with HID lamps, DOE evaluated pricing for HID ballasts from typical, publiclyavailable sources. First, DOE reviewed a commonly cited construction estimation reference book, RS-Means. However, RS-Means only had prices for fluorescent ballasts, not HID ballasts. DOE also surveyed ballast prices on publicly-accessible websites for W.W. Grainger, Inc. and GoodMart.com in April 2010. Figure 5.1.2 provides the lamp type, wattage, and price from the data surveyed.

Comparison of Ballast Lamp Rated Power (W) and Price ($)

1000 900 800
Ballast Price (U.S. $)

700 600 500 400 300 200 100 0 0 500 1000 Lamp Rated Power (W) 1500 2000 MH HPS PMH

Figure 5.1.2. Ballast Prices Based on the data graphed, DOE used trendlines to interpolate the ballast price for different lamp wattages. The interpolated lines resulted in the following functions per lamp type where x is the lamp wattage and y is the fixture price: Probe-Start Metal Halide: y = 0.1582x+147.77 Pulse-Start Metal Halide: y =0.1775x+131.89 High-Pressure Sodium: y=0.4347x+151.61 R2= 0.3377 R2= 0.7368 R2=0.5627

As with HID lamps, DOE chose to base HID ballast prices on the wattage-price functions developed from Grainger and Goodmart.com price data. Although some MV ballasts exist, the long-term supply is restricted because these ballasts have not been manufactured or imported since December 31, 2007. DOE assumes that MV ballast shipments will decline to zero by the effective date of a possible energy conservation standard.

5-4

Table 5.1.4. Lamp Price derived from Online Distributor Web Sites
Lamp Type MV MV MV MH MH MH MH Baseline Option Lamp Ballast Price Wattage $ W 175 --250 --400 --175 175.46 250 187.32 360 204.72 400 211.05 Lamp Type Substitute Option Lamp Power Ballast Price W $ 100 150 250 150 175 250 320 195.08 216.82 260.29 158.52 162.95 176.27 188.69

HPS HPS HPS PMH PMH PMH PMH

5.1.4

Fixtur e Pr ices

As with HID lamps and ballasts, DOE evaluated pricing for HID fixtures (luminaires) from typical, publicly-available sources. First, DOE reviewed a commonly cited construction estimation reference book, RS-Means. Table 5.1.5 presents HID fixture prices from the 2008 version. Table 5.1.5. Prices from RS-Means for HID Fixtures
MH Fixtures Roadway / Area Floodlighting Power Price Power Price W $ W $ 175 325 400 465 400 405 1,000 525 1,000 555 HPS Fixtures Roadway / Area Floodlighting Power Price Power Price W $ W $ 70 224 400 485 400 365 1,000 550 1,000 625

DOE also surveyed fixture prices on publicly-accessible websites for W.W. Grainger, Inc. and GoodMart.com in March 2010. Figure 5.1.3 provides the lamp type, wattage, and price from the data surveyed.

5-5

700 600

Comparison of Interior Fixture Power and Price

Fixture Price (U.S. $)

500 400
MH

300 200 100 0 0 100 200 300 400 500 Fixture Power (W)

HPS

Figure 5.1.3. Lamp Wattage/Price Comparison Based on the data graphed, DOE used trendlines to interpolate the price for different fixture wattages. The interpolated lines resulted in the following functions per lamp type where x is the lamp wattage and y is the fixture price: Metal Halide: y = 0.5176x+180.75 R2= 0.0942 High-Pressure Sodium: y=0.0944x+204.18 R2=0.0271 As with HID lamps, DOE chose to base HID fixture prices on the wattage-price functions developed from Grainger and Goodmart.com price data. DOE added $100 to the cost (e.g., 180.75+100 = 280.75) for the exterior prices regardless of the source technology. DOE assumed that probe-start MH luminaires were 20 percent less expensive than pulse-start MH fixtures for both interior and exterior luminaires. Prices for comparable MV fixtures were generally not available; consequently, DOE assumed that the prices for MV fixtures were 67% (two-thirds) of the comparable pulse-start fixtures. Table 5.1.6 presents a summary of assumed HID fixture prices used in the LCC-PBP analysis. Table 5.1.6. Lamp Price derived from Online Distributor Web Sites
Lamp Type MV MV MV MH MH MH MH Lamp Wattage 175 250 400 175 250 360 400 Baseline Option Interior Fixture Price ------$217.09 $248.14 $293.69 $310.26 Exterior Fixture Price $248.81 $274.82 $326.84 $297.09 $328.14 $373.69 $390.26 Lamp Type HPS HPS HPS PMH PMH PMH PMH Substitute Option Lamp Interior Wattage Fixture Price 100 $213.62 150 $218.34 250 $227.78 150 $258.42 175 $271.36 250 $310.18 320 $346.41 Exterior Fixture Price $313.62 $318.34 $327.78 $358.42 $371.36 $410.18 $446.41

5-6

5.1.5

Lamp Lifetime

Lifetime for HID lamps is determined when 50 percent of a sample population has failed. Within the lighting industry, this metric is known as B50. After issuing the 2003 draft report, DOE received comments about the rated lamp lifetimes. Routine group relamping time for MV and HPS lamps is typically four years and MH lamps is two years. (Caltrans, No. 8 at p. 2; Allegheny, No. 12 at p.1) Allegheny further stated that the lamp life is generally the rated lamp life by the manufacturer. DOE surveyed the available lamp catalogs and found that for any given wattage, the rated life of the lamp can vary. Table 5.1.7 Table 5.1.13 list the minimum, median, and maximum rated lifetimes for the different lamps. DOE is using the median rated life as the basis for relamping schedules in the LCC-PBP modeling. Table 5.1.7. Lamp Life Data (175W MV Baseline)
Lamp Type Lamp Power W 175 150 100 Minimum Rated Life Hours 16,000 2,800 9,000 Median Rated Life Hours 24,000 12,000 24,000 Maximum Rated Life Hours 24,000 20,000 40,000

Baseline Substitute 1 Substitute 2

MV PMH HPS

Table 5.1.8. Lamp Life Data (250W MV Baseline)

Lamp Type Lamp Power W 250 150 150 Minimum Rated Life Hours 12,000 2,800 9,000 Median Rated Life Hours 24,000 12,000 24,000 Maximum Rated Life Hours 24,000 20,000 40,000

Baseline Design Option 1 Design Option 2

MV PMH HPS

Table 5.1.9. Lamp Life Data (400W MV Baseline)

Lamp Type Lamp Power W 400 250 150 Minimum Rated Life Hours 12,000 10,000 9,000 Median Rated Life Hours 24,000 15,000 24,000 Maximum Rated Life Hours 24,000 20,000 40,000

Baseline Design Option 1 Design Option 2

MV PMH HPS

Table 5.1.10. Lamp Life Data (175W Baseline)

Lamp Type Lamp Power W 175 150 100 Minimum Rated Life Hours 10,000 2,800 10,000 Median Rated Life Hours 10,000 12,000 24,000 Maximum Rated Life Hours 10,000 20,000 40,000

Baseline Design Option 1 Design Option 2

MH PMH HPS

5-7

Table 5.1.11. Lamp Life Data (250W MH Baseline)

Lamp Type Lamp Power W 250 175 150 Minimum Rated Life Hours 10,000 10,000 9,000 Median Rated Life Hours 10,000 15,000 24,000 Maximum Rated Life Hours 15,000 15,000 40,000

Baseline Design Option 1 Design Option 2

MH PMH HPS

Table 5.1.12. Lamp Life Data (360W Baseline)

Lamp Type Lamp Power W 360 320 250 Minimum Rated Life Hours 20,000 10,000 9,000 Median Rated Life Hours 20,000 20,000 24,000 Maximum Rated Life Hours 30,000 20,000 40,000

Baseline Design Option 1 Design Option 2

MH PMH HPS

Table 5.1.13 Lamp Life Data (400W Baseline)

Lamp Type Lamp Power W 400 320 250 Minimum Rated Life Hours 8,000 10,000 9,000 Median Rated Life Hours 20,000 20,000 24,000 Maximum Rated Life Hours 20,000 20,000 40,000

Baseline Design Option 1 Design Option 2

MH PMH HPS

5.1.6

Ballast Lifetime

In the technical support document for the 2003 draft report, DOE assumed a ballast lifetime of 50,000 hours 71. During the public meeting for the metal halide lamp fixture rulemaking framework, a manufacturer stated that NEMA, when in California for a meeting related to a similar California rule, agreed that the typical lifetimes for magnetic HID ballasts was 60,000 hours and electronic HID ballasts was between 30,000 and 50,000 hours 72. DOE reviewed Californias Database for Energy Efficient Resources website which lists the life of electronic ballasts at 70,000 hours and magnetic ballasts at 45,000 hours 73. The Environmental Protection Agency (EPA) developed a lighting maintenance checklist as part of the EPAs Green Lights program. Within the document, the EPA acknowledges that there is no reliable long-term test data and recognizes the manufacturers claims about lifetime. Additionally, the EPA states, that ballast failure rate can be expected to be small in the first 70% of average life and increase beyond that point. 74 Based on this research, DOE assumed that the average life of magnetic ballast was 50,000 hours. Electronic ballasts represent a very small segment of the HID market and thus were not considered in the LCC-PBP analysis. 5.1.7 Co ntr actor Mar k-up

DOE received comments after the 2003 draft report. The American Council for an Energy Efficient Economy commented that DOE should use 13% for contractor markup. (ACEEE, No. 11 at p 2) This value is consistent with the contractor markup used in previous rulemakings, including the Fluorescent Ballasts Energy Conservation Standards Final Rule 65
5-8

FR 56740 (September 19, 2000) as well as the current revision to that rule. 75 FR 14319 (March 24, 2010). 5.1.8 Sales Tax

DOE assumed a national average sales tax of 7% for the LCC-PBP analysis. This value is consistent with that presented in the Fluorescent Ballasts Energy Conservation Standards preliminary technical support document (TSD) 75. 75 FR 14319 (March 24, 2010). 5.1.9 Maintenance Costs

The costs associated with maintenance is a real cost in the life cycle analysis of a product. Typical applications of HID lamps often require lifts to access the fixture for lamp replacement. The actual cost of maintenance will vary if the fixture is located inside or outside; by geographic region; and by mounting height. The overall wattage of the lamp should not factor in the individual relamping costs; however, the higher mounting heights typically require higherwattage lamps with higher light output. To determine the maintenance costs, DOE surveyed a number of sources. During the January public meeting for the framework document for the Metal Halide Lamp Fixture energy conservation standard, a manufacturer commented: As far as maintenance costs, we'd urge you to consider that, particularly in exterior environments; maintenance costs can be quite significant. It's difficult to get a bucket truck to a site to work on pole-mounted luminaires for less than a thousand dollars just to get it there and the average that our experience shows needs to be attributed somewhere between $150 and $300 per luminaire per servicing. So that is higher than you would traditionally think it would be for interior and just wanted to make that comment 76. As part of DOEs solid-state lighting commercialization efforts, DOE conducts demonstrations of light-emitting diode (LED) technology in real-world applications through its GATEWAY program. A number of the GATEWAY demonstrations have focused on exterior lighting where the maintenance costs often help with the payback of the typically more expensive LED fixtures. The following are maintenance costs used in the GATEWAY demonstration program: A demonstration with the Federal Aviation Administration in Atlantic City, New Jersey assumed that annual maintenance costs for relamping was $39.24 for labor. These luminaires are mounted roughly 15 above finished grade (AFG) and thus do not require a truck. This annual relamping labor rate was estimated from a much larger general facilities maintenance contract 77. Assuming that the lamp will be replaced a 5.5 year schedule; the typical cost of relamping is then $215.82. A demonstration in Portland, Oregon replacing HPS fixtures with LED fixtures in a parking structure assumed an annual relamping cost of $10.5 including the lamp. The fixtures here were mounted just over 7 AFG and thus did not require a lift 78. Assuming that the lamp will be replaced a 2.8 year schedule; the typical cost of relamping is then $28.76.

5-9

 A demonstration with Minnesota Department of Transportation (Mn/DOT) where LED fixtures were installed on a bridge on an interstate found the typical cost for relamping was $137.25 per fixture. This was similar to average rate across all of Mn/DOTs territory of $130 - $150 per fixture 79. A demonstration with Pacific Gas & Electric that replaced HPS streetlights with LEDs assumed an annual maintenance cost of $24.44 80. Assuming that the lamp will be replaced a 5.5 year schedule; the typical cost of relamping is then $134.42. A demonstration with Pacific Gas & Electric that replaced probe-start metal halide fixtures in a parking lot in the Sacramento, California area had maintenance costs between $200 $215 per luminaire for lamp servicing 81. A demonstration in Portland, Oregon that replaced HPS street lights with LED street lights had the following annual maintenance costs of $32.40 (100W), $32.52 (150W), $32.76 (250W), and $32.88 (400W) 82. Assuming that the lamp will be replaced a 5.5 year schedule; the typical cost of relamping is then between: $178.20 and $180.84. A case study by Kwantlen Polytechnic University found that when replacing 16 incandescent fixtures with compact fluorescent lamps, a total of $3,500 in maintenance was deferred. This translates to a cost of $218.75 per luminaire for maintenance 83. Based on this research, DOE assumed the average ($225) of the range ($150 $300) provided during the metal halide lamp fixture public meeting for the maintenance cost per relamping for exterior applications in the LCC-PBP analysis. DOE could not find reliable cost data for interior relamping. Therefore, DOE assumed that the exterior maintenance is three times as expensive as interior relamping. This is consistent with the comment received during the metal halide lamp fixture public meeting that exterior relamping is more expensive than interior relamping. Therefore, DOE used $75 for interior relamping for the LCC-PBP analysis. 5.1.10 Electr icity Pr ice DOE determined the relevant price of electricity based on the type of entities that purchase and own considered HID lamps. DOE based commercial and industrial electricity prices on the average of the 2017 and 2030 forecasts from the Energy Information Administrations Annual Energy Outlook 2009 84 This approach is consistent with that presented in the Fluorescent Ballasts Energy Conservation Standards preliminary TSD 85. 75 FR 14319 (March 24, 2010). 5.1.11 Discount Rate The discount rate is the rate at which future expenditures are discounted to estimate their present value. For this determination, DOE used the industrial and commercial discount rates calculated in the Fluorescent Ballasts Energy Conservation Standards preliminary TSD, made available on March 24, 2010. 75 FR 14319 For commercial and industrial consumers, DOE estimated the cost of capital for commercial and industrial companies by examining both debt and equity capital, and developed an appropriately weighted average of the cost to the company

5-10

of equity and debt financing. The resulting average discounted industrial and commercial discount rates used in the LCC-PBP analysis are 7.6 percent and 7.0 percent, respectively. 5.2 RESULTS FOR MEDIUM-WATTAGE MERCURY VAPOR DESIGN OPTIONS

DOE did not present LCC-PBP analysis results for MV lamps in the notice of proposed determination for HID lamps. MV ballasts can no longer be imported or manufactured; consequently, DOE assumed that consumers would have to switch to another HID technology when existing MV ballasts failed, regardless of financial benefit or penalty. DOE also assumed that MV fixtures were not a viable option in a new construction or fixture scenario and does not present related LCC-PBP results here. All LCC-PBP analysis results for MV lamps are presented for reference only. 5.2.1 Event 1 Lamp-Only Replacement

Table 5.2.1-Table 5.2.3 below present the results for medium wattage (150-500W) MV lamps and higher-efficiency substitute HID lamps in a lamp-only replacement scenario. DOE considered MH lamps in exterior applications only. In this scenario, a failed baseline lamp is replaced either with an identical baseline lamp, or with a substitute lamp-and-ballast system. These analyses were based on representative, incremental lamp and fixture prices as well as maintenance costs. Given the assumed disappearance of MV products from the marketplace, DOE did not consider the negative LCC savings for MV substitutes in its proposed determination. Table 5.2.1. LCC-PBP Analysis for 175W MV Baseline
Baseline 175 W MV $ Ballast Price Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) -45.17 45.17 -294.90 -1,837.32 --Commercial/Exterior Substitute 1 Substitute 2 150 W PMH 100 W HPS $ $ 190.22 64.09 254.31 209.14 288.18 6.72 2,191.63 -354.31 31.12 234.10 49.23 283.33 238.16 263.26 31.64 2,059.27 -221.95 7.53

Table 5.2.2. LCC-PBP Analysis for 250W MV Baseline

Baseline 250 W MV $ Ballast Price Lamp Price -52.60 Commercial/Exterior Substitute 1 Substitute 2 150 W PMH 150 W HPS $ $ 190.22 64.09 260.18 60.91

5-11

Table 5.2.3. LCC-PBP (continued)

Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) 52.60 -326.94 -1,874.13 --254.31 201.71 288.18 38.76 2,193.63 -319.50 5.20 321.09 268.49 288.18 38.76 2,126.51 -252.38 6.93

Table 5.2.4. LCC-PBP Analysis for 400W MV Baseline

Baseline 400 W MV $ Ballast Price Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) -64.29 64.29 -394.58 -1,943.34 --Commercial/Exterior Substitute 1 Substitute 2 250 W PMH 150 W HPS $ $ 312.34 80.90 393.24 328.95 327.73 66.85 2,255.98 -312.64 4.92 260.18 60.91 321.09 256.80 288.18 106.40 2,126.51 -183.17 2.41

5.3 5.3.1

Results for Medium-Wattage pr obe-star t metal halide DESIGN Options Event 1 Lamp-Only Replacement

Table 5.3.1-Table 5.3.4 below present the results for medium wattage (150-500W) probestart MH lamps and higher-efficiency substitute HID lamps in a lamp-only replacement scenario. DOE considered MH lamps in both interior and exterior applications. In this scenario, a failed baseline lamp is replaced either with an identical baseline lamp, or with a substitute lamp-andballast system. These analyses were based on representative, incremental lamp and fixture prices as well as maintenance costs. A full rulemaking would yield more detailed results than the representative analyses conducted. Generally, the LCC of a high-efficiency lamp and ballast replacement is higher than the LCC of an inefficient lamp-only replacement. Table 5.3.1. LCC-PBP Analysis for 175W MH Baseline
Industrial/Interior Baseline Substitute 1 175 W MH 150 W PMH $ $ Ballast Price Lamp Price Total First Cost -49.58 49.58 190.22 64.09 254.31 Commercial/Exterior Baseline Substitute 2 175 W MH 100 W HPS $ $ -49.58 49.58 234.10 49.23 283.33

5-12

Table 5.3.2. LCC-PBP (continued)

Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) -149.23 -808.83 --204.73 141.02 8.21 1,056.34 -247.51 24.94 -297.28 -1,947.52 --233.75 263.26 34.02 2,059.27 -111.75 6.87

Table 5.3.3. LCC-PBP Analysis for 250W MH Baseline

Industrial/Interior Baseline Substitute 1 250 W MH 175 W PMH $ $ Ballast Price Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) -53.08 53.08 -178.85 -853.30 --195.54 68.76 264.30 211.22 149.59 29.26 994.23 -140.93 7.22 Commercial/Exterior Baseline Substitute 2 250 W MH 150 W HPS $ $ -53.08 53.08 -330.11 -1,983.51 --260.18 60.91 321.09 268.01 288.18 41.93 2,126.51 -143.00 6.39

Table 5.3.4. LCC-PBP Analysis for 360W MH Baseline

Industrial/Interior Baseline Substitute 1 360 W MH 320 W PMH $ $ Ballast Price Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) -56.92 56.92 -217.75 -788.24 --226.43 90.54 316.97 260.05 205.97 11.78 1,083.54 -295.30 22.08 Commercial/Exterior Baseline Substitute 2 360 W MH 250 W HPS $ $ -56.92 56.92 -373.22 -1,919.94 --211.52 79.64 291.16 234.24 331.69 41.53 2,146.17 -226.23 5.64

Table 5.3.5. LCC-PBP Analysis for 400W MH Baseline

Industrial/Interior Baseline Substitute 1 400 W MH 320 W PMH $ $ Ballast Price Lamp Price -58.08 226.43 90.54 Commercial/Exterior Baseline Substitute 2 400 W MH 250 W HPS $ $ -58.08 211.52 79.64

5-13

Table 5.3.6. LCC-PBP (continued)

Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) 58.08 -237.74 -810.40 --316.97 258.89 205.97 31.77 1,083.54 -273.14 8.15 58.08 -395.37 -1,937.06 --291.16 233.08 331.69 63.68 2,146.17 -209.11 3.66

5.3.2

Event 2 New Constr uction / Fixtur e Replacement

Table 5.3.5-Table 5.3.8 below present the results for medium wattage (150-500W) probestart MH lamps and higher-efficiency substitute HID lamps in a new construction or fixture replacement scenario. DOE considered MH lamps in both interior and exterior applications. In this scenario, a consumer selects either a baseline or substitute fixture and lamp. In the exterior lighting cases, the HPS substitutes have a lower LCC. These analyses were based on representative and incremental lamp and fixture prices as well as maintenance costs. A full rulemaking would yield more detailed results than the representative analyses conducted. Table 5.3.7. LCC-PBP Analysis for 175W MH Baseline
Industrial/Interior Baseline Substitute 1 175 W MH 150 W PMH $ $ Fixture Price (incl. ballast) Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) 260.51 49.58 310.09 -149.23 -1,069.34 --310.10 64.09 374.19 64.10 141.02 8.21 1,176.22 -106.89 7.81 Commercial/Exterior Baseline Substitute 2 175 W MH 100 W HPS $ $ 356.51 49.58 406.09 -297.28 -2,304.03 --376.34 49.23 425.57 19.73 263.26 34.02 2,201.51 102.52 0.58

Table 5.3.8. LCC-PBP Analysis for 250W MH Baseline

Industrial/Interior Baseline Substitute 1 250 W MH 175 W PMH $ $ Fixture Price (incl. ballast) Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) 297.77 53.08 350.85 -178.85 -1,151.07 325.63 68.76 394.39 43.54 149.59 29.26 1,124.32 Commercial/Exterior Baseline Substitute 2 250 W MH 150 W HPS $ $ 393.77 53.08 446.85 -330.11 -2,377.28 382.01 60.91 442.92 -3.93 288.18 41.93 2,248.34

5-14

Table 5.3.9. LCC-PBP (continued)

LCC Savings Payback Period (years) --26.75 1.49 --128.94 -0.09

Table 5.3.10. LCC-PBP Analysis for 360W MH Baseline

Industrial/Interior Baseline Substitute 1 360 W MH 320 W PMH $ $ Fixture Price (incl. ballast) Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) 352.43 56.92 409.35 -217.75 -1,140.67 --415.69 90.54 506.23 96.88 205.97 11.78 1,272.81 -132.14 8.22 Commercial/Exterior Baseline Substitute 2 360 W MH 250 W HPS $ $ 448.43 56.92 505.35 -373.22 -2,368.37 --393.34 79.64 472.98 -32.37 331.69 41.53 2,328.00 40.37 -0.78

Table 5.3.11. LCC-PBP Analysis for 400W MH Baseline

Industrial/Interior Baseline Substitute 1 400 W MH 320 W PMH $ $ Fixture Price (incl. ballast) Lamp Price Total First Cost Incremental First Cost Annual Operating Cost Annual Operating Cost Differential Life-Cycle Cost (7% Discount Rate) LCC Savings Payback Period (years) 372.31 58.08 430.39 -237.74 -1,182.71 --415.69 90.54 506.23 75.84 205.97 31.77 1,272.81 -90.10 2.39 Commercial/Exterior Baseline Substitute 2 400 W MH 250 W HPS $ $ 468.31 58.08 526.39 -395.37 -2,405.37 --393.34 79.64 472.98 -53.41 331.69 63.68 2,328.00 77.37 -0.84

5-15

CHAPTER 6.

NATIONAL ENERGY SAVINGS AND CONSUMER IMPACTS

For this determination analysis, the U.S. Department of Energy (DOE) estimated the following national impacts from representative energy-efficient substitutes for baseline HID lamps: (1) national energy savings (NES), (2) monetary value of the energy savings due to standards, (3) increased total installed costs of the considered products due to standards, and (4) the net present value (NPV) of energy savings (difference between value of energy savings and increased total installed costs). DOE prepared a spreadsheet model to forecast energy savings and national consumer economic costs and savings resulting from new standards. The next two sections present details of the methodology and results of the national impact estimates. 6.1 6.1.1 METHODOLOGY Shipment For ecasts

Shipment forecasts for HID lamps are key inputs to the NES and NPV calculations, and were developed by DOE using a three-step process. First, DOE used historical shipment data from NEMA to estimate the total historical shipments of each lamp type analyzed. Second, DOE calculated an installed stock for each lamp type based on the average service lifetime of each lamp type. Third, by applying estimated market growth rates and HID lamp shipment trends, DOE developed annual shipment projections. As discussed in chapter 2, NEMA provided two sets of HID lamp shipment data for the periods 1990 to 2002 and 2003 to 2008. Based on the NEMA shipments data, DOE estimated the average annual growth rate to be 1.5 percent until 2016. DOE assumed that mercury vapor (MV) lamps would decline to extinction and not exist for lighting applications by 2016. DOE also assumed that high-pressure sodium (HPS) lamps would represent one-third of all HID lamps, and that metal halide lamps (both probe and pulse-start) would constitute the remaining shipments. Historically, probe-start metal halide (MH) has dominated total metal halide lamp shipments, but DOE assumed that MH shipments will gradually decrease as luminaire requirements and other incentives promote the use of pulse-start MH (PMH) or non-HID technologies. DOE assumed that the overall number of HID lamps shipped would peak near 2020. The following are the factors in DOEs assumptions: DOEs metal halide lamp fixture rule could affect the price of fixtures. More efficient ballasts could require redesign of the fixtures as noted during the metal halide lamp fixture (MHLF) public meeting 86. The effective dates of future MHLF rules are 2015 and 2022. Many non-HID technologies are competing in the traditional HID market. For instance, the high-bay and low-bay market, traditionally dominated by HID technology, has lost significant ground to the fluorescent versions of these fixtures known as high-intensity fluorescent (HIF). During the metal halide lamp fixture public meeting, DOE heard that

6-1

since 2002 there has been a major shift away from metal halide high-bay fixtures to fluorescent high-bay fixtures 87. DOE believes that this trend will continue. Many energy codes (both state and the Federal ANSI/ASHRAE/IESNA Standard 90.1) have requirements in either parking lot or structures for increased use of controls. The controls turn off or reduce the lighting power in response to either daylight or low frequency of occupants. HID technology has dominated parking lot and structure applications, but the control requirements will require other technologies to be considered. DOE believes that induction, fluorescent, and solid-state lighting will fill the void left by HID from the increased lighting control requirements.

In summary, DOE assumes a growth of 1.5 percent from 2008 to 2016 in overall HID lamp shipments with MV trending to 0 percent by 2016, HPS consistently representing 33 percent of the total HID shipments, and metal halide (MH and PMH) constituting the remaining shipments. DOE assumes that overall HID lamps will peak and chose 2016 as the year of maximum shipments. After 2016, DOE assumed the HID lamp shipments would decline at a rate of 1.5 percent per year. Under these assumptions, the total HID shipments in 2024 will be roughly the same number of HID lamps shipped in 2008. 6.1.2 National Ener gy Savings Analysis

The inputs for determining NES are (1) annual energy consumption per unit; (2) shipments; (3) equipment stock; (4) national energy consumption (calculated from consumption per unit and equipment stock); and (5) site-to-source conversion factors. DOE calculated the national energy consumption by multiplying the number of units, or stock, of lamp-and-ballast systems (by vintage, which represents the age of the ballasts) by the unit energy consumption (also by vintage). Then, DOE calculated national annual energy savings from the difference between national energy consumption in the base case (without efficiency standards) and a higher-efficiency standards case. DOE estimated energy consumption and savings based on site energy, and converted the electricity consumption and savings to source energy. Cumulative energy savings are the sum of the annual NES, which DOE determined over the analysis period (2017 2046). Cumulative NES results are presented as undiscounted and based on a 7- and 3-percent real discount rate. 6.1.3 Net Pr esent Value Analysis

The inputs for determining NPV are (1) total annual installed cost, (2) total annual operating cost savings, (3) discount factor, (4) present value of costs, and (5) present value of savings. DOE calculated net savings each year as the difference between total operating cost savings and increases in total installed costs (including equipment price and installation cost). DOE calculated savings over the life of the equipment, accounting for differences in yearly energy rates. DOE calculated undiscounted NPV as the difference between the present value of operating cost savings and the present value of increased total installed costs, and used discount rates of 7 and 3 percent to discount future costs and savings to the present.

6-2

DOE calculated increases in total installed costs as the product of the difference in the total installed cost between the base case and standards case, with total annual shipments assumed as equal in both cases. Because purchase costs of the higher-efficiency equipment in the standards case are generally greater than the purchase costs of equipment in the base case, price increases appear as negative values in the NPV. DOE expressed operating cost savings as decreases in operating costs associated with the lower energy consumption of equipment in the standards case compared to the base efficiency case. Total operating cost savings are the product of savings per unit and the number of units of each vintage surviving in a particular year. 6.2 ESTIMATES OF POTENTIAL ENERGY AND CONSUMER IMPACTS

DOE estimated national energy savings and consumer effects of energy conservation standards for the considered HID lamps using its own engineering analysis data. Cumulative NES values are presented in quadrillion British thermal units (quads), and NPV values are presented in billion 2009$. 6.2.1 National Ener gy Savings

The results using DOEs analysis of design options indicate potential cumulative energy savings for medium-wattage HID lamps of 2.8 quads at a 7-percent discount rate over a 30-year analysis period (20172046). Table 6.2.1 presents the cumulative NES values for each pairing of baseline and substitute lamps, as well as the total cumulative NES. Table 6.2.1. Cumulative National Energy Savings for HID Lamps (20172046)
Baseline Lamp 175 W MV 175 W MV 250 W MV 250 W MV 400 W MV 400 W MV 175 W MH 175 W MH 250 W MH 250 W MH 360 W MH 360 W MH 400 W MH 400 W MH TOTAL Substitute Lamp 150 W PMH 100 W HPS 150 W PMH 150 W HPS 250 W PMH 150 W HPS 150 W PMH 100 W HPS 175 W PMH 150 W HPS 320 W PMH 250 W HPS 320 W PMH 250 W HPS Undiscounted -1.5 0.0 -1.5 0.0 -7.0 0.0 1.2 2.6 -1.9 3.7 0.5 5.1 1.2 5.8 8.2 NES Discount quads 3% -0.9 0.0 -0.9 0.0 -4.3 0.0 0.8 1.6 -1.2 2.3 0.3 3.1 0.8 3.6 5.1 7% -0.5 0.0 -0.5 0.0 -2.4 0.0 0.4 0.9 -0.7 1.3 0.2 1.8 0.4 2.0 2.8

6.2.2

Net Pr esent Value

DOE estimated the national economic effect on end users in terms of the NPV of cumulative benefits from 2017 to 2046. The results indicate a potential NPV of $15.0 billion (2009$) at a 3-percent discount rate and $3.5 billion at a 7-percent discount rate. Table 6.2.2 presents the cumulative NPV results for each pairing of baseline and substitute lamps, as well as the total cumulative NPV.
6-3

Table 6.2.2. Cumulative NPV Results for HID Lamps (20172046)

Baseline Lamp 175 W MV 175 W MV 250 W MV 250 W MV 400 W MV 400 W MV 175 W MH 175 W MH 250 W MH 250 W MH 360 W MH 360 W MH 400 W MH 400 W MH TOTAL Substitute Lamp 150 W PMH 100 W HPS 150 W PMH 150 W HPS 250 W PMH 150 W HPS 150 W PMH 100 W HPS 175 W PMH 150 W HPS 320 W PMH 250 W HPS 320 W PMH 250 W HPS Undiscounted -21.0 -0.5 -21.2 -0.5 -82.8 -0.5 8.6 29.0 -21.9 38.7 -1.0 47.4 4.1 53.6 31.9 NES Discount billion 2009$ 3% -13.0 -0.4 -13.2 -0.4 -51.3 -0.4 5.0 17.3 -14.2 23.3 -1.0 28.7 2.1 32.5 15.0 7% -7.4 -0.3 -7.5 -0.3 -29.1 -0.3 2.5 9.2 -8.6 12.5 -1.0 15.6 0.7 17.6 3.5

6-4

APPENDIX A. NON-REGULATORY INCENTIVE PROGRAMS A.1 CALIFORNIA INCENTIVE PROGRAMS

Alameda Power and Telecom offers rebates for its commercial customers to install energy-efficient lighting. The Alameda Municipal site claims rebates available for T8 fluorescent lamps and electronic ballasts, for compact fluorescent, high-pressure sodium (HPS) and metal halide lamps, for occupancy sensors, and other energy saving products. However, upon further investigation, Alameda only offers $100 per interior fluorescent high-bay fixture, also known as high-intensity fluorescent (HIF). More information: http://www.alamedapt.com/electricity/com_rebates.html. Anaheim Public Utilities not only offers rebates on certain lighting equipment, they actually give away equipment for free. As part of their Dusk-to-Dawn Lighting Program, residents can receive a 70W HPS or fluorescent (details not actually provided) fixture for free. In addition to the fixture, residents receive a photosensor so the fixture can operate the entire night. When APU is not giving away lighting equipment, they are performing lighting audits and making recommendations for improvements. At the end of the audits, APU provides specific rebates. More information: http://www.anaheim.net/article.asp?id=990. City of Palo Alto Utilities, through the Commercial Advantage Program, offers several incentives for their commercial customers to replace old equipment with new, more efficient equipment. CPAU offers rebates on complete new pulse-start metal halide (PMH) fixtures or retrofit kits replacing existing MV (MV), probe-start metal halide (MH) or HPS fixtures only. Replacements must be equipped with PMH lamps and either magnetic or electronic ballasts. New fixtures must replace, one for one, existing incandescent, MV, T12/High output Fluorescent, T12/Very High Output Fluorecent, MH, or HPS fixtures in interior installations. Rebates are available based on lamp wattages. Same incentives are applicable for exterior applications as well. The incentives are as follows: $10.00/Fixture for replacing a 175 watt lamp, up to 190 watt replacement fixture. $40.00/Fixture for replacing a 176 - 399 watt lamp, up to 275 watt replacement fixture. $45.00/Fixture for replacing a 400 watt lamp, up to 400 watt replacement fixture. $50.00/Fixture for replacing 400 watt or more lamp, up to 821-950 watt replacement fixture. $100.00/Fixture for replacing 400 watt or more lamp, up to 820 watt replacement fixture.

A-1

More information: http://www.cityofpaloalto.org/civica/filebank/blobdload.asp?BlobID=3976. The Modesto Irrigation Department (MID) 2009 Business Rebates offers rebates on complete new PMH fixtures or retrofit kits replacing existing MV, MH or HPS fixtures only. Replacements must be equipped with PMH lamps and either magnetic or electronic ballasts. New fixtures must replace, one for one, existing incandescent, MV, T12/High output Fluorescent, T12/Very High Output Fluorecent, MH, or HPS fixtures in interior installations. Rebates are available based on lamp wattages. Same incentives are applicable for exterior applications as well. The incentives are as follows: $7.50/Fixture for replacing a 175 watt lamp, up to 190 watt replacement fixture. $30.00/Fixture for replacing a 176 - 399 watt lamp, upto 275 watt replacement fixture. $35.00/Fixture for replacing a 400 watt lamp, up to 400 watt replacement fixture. $40.00/Fixture for replacing 400 watt or more lamp, up to 821-950 watt replacement fixture. $75.00/Fixture for replacing 400 watt or more lamp, up to 820 watt replacement fixture. Rebate of $9.50/ lamp is also offered for replacing existing reflector-type incandescent, PAR halogen, or PAR halogen IR lamps or fixtures with an Integrated Ballast ceramic metal halide (CMH) PAR lamps of 25 W.

More information: http://www.mid.org/services/rebates/rbts-com.htm. Roseville Electric - Commercial Energy Efficiency Rebate Program offers incentives for its commercial customers to increase the efficiency of their existing facilities. Rebates are offered upto $50/fixture for PMH. More information: http://www.roseville.ca.us/electric/business/rebates/default.asp. The Los Angeles Department of Water and Power (LADWP) manages the Commercial Lighting Efficiency Offer. This program seeks to promote energy-efficient lighting retrofits, including MH. High bay T8/T5 fixtures replacing 400W fixtures that result in a 45% energy savings are eligible for a $100 per fixture rebate. In applications replacing fixtures less than 400W, there is only a $50 rebate. In addition to the rebates for the equipment, occupancy sensors for fluorescent high bay fixtures have a $20 per sensor rebate. Installations where induction lamps replace fixtures with incandescent, MH, HPS, or MV lamps are also eligible for rebates $35 per 55W to 100W lamp and $50 for lamps greater than 100W. MH interiors or PMH in exterior applications replacing incandescent, HPS, or MV have rebates of $40 per fixture. Incentives for CMH are both lamp and fixture focused. $40 per CMH fixture 35W and $15 per CMH lamp that is less than 35W exists.

A-2

More information: http://www.ladwp.com/ladwp/cms/ladwp000572.jsp. Pacific Gas and Electric (PG& E) jointly manages a statewide program called Express Efficiency with Southern California Edison. The program is quite comprehensive, not only in lighting, but in many other areas. In the Display and Accent Lighting category, integrated ballast CMH PAR lamps have a $12.50 per lamp rebate and CMH adjustable accent lighting has a $45 per fixture rebate. For the PMH fixtures (exterior and interior), $10 to $100 rebates are offered per fixtures ranging in power from 1W to 950W. High bay fluorescent fixtures ranging from $35 to $125. More information: http://www.pge.com/biz/rebates/rebates_assistance/ (PG&E) or http://www.sce.com/RebatesandSavings/LargeBusiness/expressEfficiency.htm (SCE). Silicon Valley Power has a lighting rebate program. New high bay linear fluorescent fixtures under 250W have a $100 per fixture rebate if they are replacing incandescent, T12, or HID fixtures 400W and greater. Interior PMH fixtures ranging in power from 1W to 400W have rebates of $18 to $50 per fixture if they are replacing MV or incandescent technology. A lamp/ballast system efficacy of 45 LPW for sources under 100W and a 55 LPW for sources greater than 100W is required. There are no restrictions on ballast type. $100 to $36 per fixture is available for HID exterior fixtures ranging in power from 1W to 175W. No specific information is provided for the exterior HID exterior fixtures, but it should be assumed that the requirements are the same or similar to the interior PMH requirements. PMHde lamp/ballasts less than 350 replacing 400W or greater systems have a $45 per lamp rebate. CMH fixtures replacing incandescent or halogen fixtures have a $25 per fixture rebate. The lamp must be less than 75W and the mean lamp/ballast efficact must exceed 55 LPW. More information: http://www.siliconvalleypower.com/bus/?sub=rebatelight. A.2 COLORADO INCENTIVE PROGRAMS

Longmont Power and communication in collaboration with Platte River PowerAuthoritys offers the LightenUP program. Loveland Water & Power also collaborates with Platte River PowerAuthoritys to offer the same program. The program recommends replacing HPS, MV, and MH with fluorescent fixtures. Incentive amounts for retrofit of high performance T8 system is $500/kilowatt and of 8 ft. fluorescent retrofit or replacement is $600/kilowatt For more information, http://www.ci.longmont.co.us/lpc/bus/eep_homepage.htm Specifically for the LightenUP program, http://www.prpa.org/productservices/eeservicesbus/i/lightenup2009whatsnew.pdf Xcel EnergySM offers a variety of technology specific rebate incentives. The following rebates are offered: $ 85 - $210 / unit per fixture for replacing 250W 1000 W HID (HPS, MH, MV) with multi-lamp HIF.

A-3

$30 is available per fixture for replacing incandescent, HPS, or MV with 151W to 250W PMH. $45 per fixture for replacing incandescent, HPS, or MV with 251 W or more PMH. $25 per fixture for replacing incandescent, HPS, MV or MH with 175W or less PMH system. $50 per fixture for replacing incandescent, HPS, MV or MH with 176W to 319W PMH system. $45 per fixture for replacing incandescent, HPS, MV or MH with 320W to 749W PMH system. $140 per fixture for replacing incandescent, HPS, MV or MH with 750W or more PMH system. $55 per fixture for replacing incandescent, halogen or HID systems with 150W or less CMH fixtures. $60 per fixture for replacing incandescent, halogen or HID systems with 151W to 250W or less CMH fixtures. $75 per fixture for replacing incandescent, halogen or HID systems with 251W or more CMH fixtures. $25 per lumianire for replacing 75W or greater incandescent or halogens with 25W integrated CMH lamps.

More information: http://www.xcelenergy.com/Business/Programs_Resources/ConservationRebates_Incentives_Bu siness/Pages/Lighting_Efficiency.aspx A.3 CONNECTICUT INCENTIVE PROGRAMS

Connecticut Light and Power (part of Northeast Utilities) sponsors an Express Service Lighting Rebate program. The rebate works to promote energy-efficient lighting in retrofit applications. Incentive of $30 per fixture is offered for upgrading from MV, HPS, and MH to PMH. More information: http://www.cl-p.com/Business/SaveEnergy/BusinessRebates.aspx. Groton Utilities lighting efficiency program cash incentives for new fixture installation or retrofits of old lighting fixtures to energy efficient systems. Groton Utilities lighting efficiency program also provides rebates for installation of lighting occupancy controls and daylight harvesting and dimming controls. $20 rebate per fixture is offered for installing new PMH fixture.
A-4

More information: http://www.grotonutilities.com/elec_conserv_comm_light.asp. United Illuminating manages three programs promoting energy-efficiency lighting that applies to HID lighting. Energy Blueprint Program focuses on reducing energy consumption in a building. This program offers incentives on both a lighting power density basis, as well as, a technology basis. Outdoor Lighting, Lite the NiteSM provides commercial customers with exterior lighting design services, the ability to have equipment purchased, installed, and maintained for a monthly fee. The program offers a variety of fixture types with HPS and MH lamps available in a variety of wattages. Finally, United Illuminating also participates in the Express Service that Connecticut Light and Power also promotes. More information: http://www.uinet.com/uinet/connect/UINet/Top+Navigator/Your+Business/UI+Products+%26+ Services/. A.4 FLORIDA INCENTIVE PROGRAMS

Florida Power and Light (FP&L) manages a Business Lighting Program that works to reduce the utilitys peak demand from 3 pm to 6 pm during summertime weekdays. Installation of PMH and electronic ballasts in MH fixtures and CMH in hard wired fixtures qualify for lighting incentives. The incentive is an upfront discount for the work performed by an independent contractor. More information: http://www.fpl.com/business/savings/lighting.shtml. Progress Energy will pay between $1 and $5 per fixture replaced in a retrofit (for new construction, the space must exceed code by 15% to be eligible for incentives). CMH technology is eligible. More information: http://progressenergy.com/custservice/flabusiness/efficiency/efficiencyservices.asp. A.5 IDAHO INCENTIVE PROGRAMS

Avista Utilities Commercial Lighting Incentives offers incentives for HID interior and exterior fixtures. When replacing 400W MH high bay with multi-lamp fluorescent high bay, $125-$460 per fixture is available. Incandescent flood of 100 W or more replaced with CMH are eligible for $35 rebate per fixture. Avistas incentives are also available for new construction on a case-by-case basis. More information: https://www.avistautilities.com/business/rebates/washington_idaho/Documents/Lighting%20We b.pdf Idaho Power offers a series of incentives related to HID lighting. MH lighting is specifically identified. These incentives only apply to lighting upgrades for interior lighting (street, area, and parking lot lighting is excluded). Eligible metal halide fixtures include PMH or CMH are replacing higher wattage MH, MV, or other sources. The incentives are the following:

A-5

$18.00 for PMH fixtures between 30W to 70W when replacing a fixture drawing 20W more than the MH wattage. $22.00 for PMH fixtures between 70W to 150W when replacing a fixture drawing 25W more than the MH wattage. $26.00 for PMH fixtures between 150W to 250W when replacing a fixture drawing at least 40W more than the MH wattage. $55.00 for PMH fixtures between 250W to 360W when replacing a fixture drawing at least 80W more than the MH wattage. $75.00 for PMH fixtures between 360W to 500W when replacing a fixture drawing at least 120W more than the MH wattage. $105.00 for PMH fixtures for 500W and more when replacing a fixture drawing at least 200W more than the MH wattage.

More information: http://www.idahopower.com/EnergyEfficiency/Business/Programs/EasyUpgrades/default.cfm. A.6 INDIANA INCENTIVE PROGRAMS

Duke Energy offers $25 per fixture for replacing 400W MH with 320W PMH (if the fixture operates a minimum of 1800 hours per year). More information: http://www.duke-energy.com/indiana-business/energymanagement/energy-efficiency-incentives.asp. A.7 IOWA INCENTIVE PROGRAMS

Ames Electric Department manages the Commercial High efficiency Lighting Program offers only a handful of rebates for select technologies and power ranges. The incentives are available for replacing existing or for new equipment: $15 per fixture is available for MH fixtures with lamps rated under 250W $25 per fixture is available for PMH fixtures with lamps in the range of 250W350W $20 per fixture is available for HPS fixtures with lamps under 250W $25 per fixture is available for HPS fixtures with lamps in the range of 250W400W

More information: http://www.cityofames.org/SmartEnergy/

A-6

MidAmerican Energy Companys EnergyAdvantage offers rebates for some fixture technologies, but also for specific HID lamps. $ 25 per fixture is available for 320W or less and $50 per fixture for greater than 320W PMH fixtures. MidAmerican also provides a rebate of $3 per lamp for the 360W MH lamps. This lamp is for reduced wattage replacements. The incentives are available for replacing existing or for new equipment. More information: http://midamericanenergy.com/html/energy6b.asp. A.8 KANSAS INCENTIVE PROGRAMS

Kansas City Light & Power only focuses on PMH technology. $50 per fixture is avaible for 320W and 360W PMH replacing 400W MH and HPS fixtures. More information: http://www.kcplprograms.com/content/prescriptiverebates. A.9 KENTUCKY INCENTIVE PROGRAMS

Duke Energy offers $25 per fixture for replacing 400W MH with 320W PMH (if the fixture operates a minimum of 1800 hours per year). It also offers rebates of $30 - $75 for replacing HID lamps with T5 HO High Bay (2 -8) lamp. More information: http://www.duke-energy.com/kentucky-business/energymanagement/lighting-incentive.asp. A.10 MAINE INCENTIVE PROGRAMS

Efficiency Maine, a program of the Maine Public Utilities Comission, does not offer a rebate for HID fixtures. However, the program offers $ 65 (retrofit fixtures) and $35 (new construction) for High Intensity Fluorescent technology. No specific requirements were mentioned. More information: http://www.efficiencymaine.com/business_programs.htm. A.11 MASSACHUSETTS INCENTIVE PROGRAMS

National Grid offers incentives for new equipment as well as replacing existing equipment. The Energy Initiative for existing facilities offers rebates for both fixtures and lamp/ballast systems. For existing metal halide systems, National Grid offers $45 per lamp/ballast kit. No specific wattage requirement is listed, but a minimum savings of 13W is required. The incentive is only available for interior fixtures. Besides replacing the lamp/ballast system, National Grid offers $70 per fixture for completely replacing an existing MH fixture with a new PMH fixture. Again, the program does not list any specific wattage requirements for the existing or new fixture, but a saving of 20W is required. $100 per fixture is available for replacing a MH fixture with an HPS fixture. The program states that HPS is more energy efficient than MH. No specific fixture wattages are listed, but a minimum savings of 29 watts is required. No incentives specifically for HID fixtures are listed in National Grids new constrcuction program. However, the program offers $30 to $40 per fixture for HIF. The applications recommended for these fixtures are some of the typical locations for HID fixtures.

A-7

More information: https://www.nationalgridus.com/masselectric/business/energyeff/3_large.asp. Nstar manages a Construction Solution and Business Solutions program that aims to promote energy-efficient lighting in renovations and new construction projects. The Commercial Section of the program provides $70 for MH speciality fixture for a minimum of 55 wattage reduction per fixture. The Industrial Lighting section of solutions program offers the following incentives for HID or rival technology. $45 for a PMH kit with electronic ballast for a minimum of 37 wattage reduction per fixture. $60 for a PMH fixture with electronic ballast for a minimum of 37 wattage reduction per fixture.

More information: http://www.nstaronline.com/business/energy_efficiency/electric_programs/business_solutions.as p. Reading Municipal Lighting Department offers incentives to replace existing HID fixtures with new T5 or T8 HO fixtures. The program offers $100 per fixture and no other requirements are listed. For more information, http://www.rmld.com/from_priscilla/programs/light_rebate_08.pdf Western Massachusetts Electric (WMECO) offers rebates to commercial customers for lighting. New construction projects are not eligible for this program. $25 incentive is offered for replacing MH, MV or HPS fixtures with PMH fixtures. It offers rebate of $150 for replacing 250W MH to 54T5Ho (2/3 lamp) (117/176 W) and for replacing 400W MH to 54T5HO (4 lamp) (234 W). It offers rebate of $150 for replacing 250 W/400 W MH to high performance 4 4lamp/6- lamp T8 with electronic ballast and reflector. More information: http://www.wmeco.com/business/saveenergy/EnergyEfficiencyPrograms/Lighting.aspx. A.12 MINNESOTA INCENTIVE PROGRAMS

Alexandria Light and Power Commercial Energy Efficiency Rebate Program offers the Bright Energy Solution. The following incentives are offered: $75 incentive per fixture for replacing 100W or larger incandescent with 20W CMH. $75 incentive per fixture for replacing 150W or larger incandescent with 39W CMH. $75 incentive per fixture for replacing 500W or larger incandescent with 150W CMH.

A-8

$20 incentive per fixture for replacing 400W or larger HID or incandescent with 320W PMH fixture. $30 incentive per fixture for replacing 1000W or larger HID or incandescent with 750W PMH fixture.

There are 23 Utilities participate in this program. More information: http://www.brightenergysolutions.org/municipalities/?category=business&state=mn Anoka Municipal Utility has a Commercial Lighting Rebate program that includes both retrofit and new construction applications. Within their Lighting Rebate Guide, the utility recommends replacing MV with MH or HPS, MH or T12 fluorescent with high-bay multilamp CFL fixtures, MH or HPS with linear fluorescent, and MH or HPS with PMH. For retrofit applications, $33-$80 per fixture is available for replacing the existing 400W or MH fixture with a multi-lamp linear or CFL fixture. $25 to $80 per fixture is available for replacing standard MH or HPS with PMH. New construction has similar incentives. $15.40-$37.50 per fixture is available for using multi-lamp linear or CFL fixtures where HID high bay fixtures could also be used. $11 to $18.50 per fixture is available for using PMH rather than just installing standard MH or HPS. More information: http://www.ci.anoka.mn.us/index.asp?Type=B_BASIC&SEC={F366614A-5300-4465-BC0D279ADDD33B82} Austin Utilities offers incentives to its commercial and industrial customers to install energy-efficient equipment in their facilities. The incentives offered on HID lamps are as follows: $7.5 rebate per fixture for installing new PMH fixture of 32W, 50W, 70W and 100W. $6 rebate per fixture for installing new PMH of 150W and 175W $8 rebate per fixture for installing new PMH of 200Wand 250W. $12 rebate per fixture for installing new PMH fixture of 320W, 350W and 400W. $24 rebate per fixture for installing new PMH fixture of 400W with two lamps. $20 rebate per fixture for installing new PMH fixture of 575W. $15 rebate per fixture for installing new PMH fixture of 750W. $15, $30 and $45 rebate per fixture for installing new PMH of 875W with 1, 2 and 3 lamps respectively.

A-9

$8 rebate per fixture for installing 25W Integrated CMH. $20 rebate per fixture for installing 150W or less CMH. $10 rebate per fixture for installing 151W-250W CMH. $15 rebate per fixture for installing 251W or more CMH. $7.5 rebate per fixture for installing new HPS fixture of 35W, 50W, 70W, 100W and 150W. $15 rebate per fixture for installing new HPS fixture of 200W, 250W, 310W, 400W, 600W and 750W.

More information: http://www.austinutilities.com/pages/business_conserve.asp. City of North St. Paul (Rebates on hold till Jan 2010). HPS, PMH and CMH are eligible for incentives. The incentives are as follows: $28 and $45 for replacing incandescent, HPS or MV with MH with 151 to 250W and 251W or more repectively. $28 and $45 for replacing incandescent or MV with HPS with 151 to 250W and 251W or more repectively. $50 for replacing incandescent, halogen or HID with CMH fixtures of 150W or less. $45 for replacing incandescent, halogen or HID with CMH fixtures of 151 to 250W or less. $60 for replacing incandescent, halogen or HID with CMH fixtures of 251W or higher. $20 for replacing incandescent, halogen or HID with integrated CMH fixtures of 25W. $25 for replacing incandescent, HPS, MV or MH with PMH of 175W or less. $40 for replacing incandescent, HPS, MV or MH with PMH of 176 to 319W. $55 for replacing incandescent, HPS, MV or MH with PMH of 320 to 749W. $65 for replacing incandescent, HPS, MV or MH with PMH of 750W or higher. $50, $75, $100 and $125 for replacing HID lamps of 250W, 310- 400W, 750Wand 1000W with high bay fluorescent fixtures (T5HO) with electronic ballast having 2-3, 4-6, 8 and 10 lamps respectively.

A-10

$50, $75, $100 and $125 for replacing HID lamps of 250W, 310- 400W, 750Wand 1000W with high bay fluorescent fixtures (T8) with electronic ballast having 4, 6- 8, 12-16, and 18-20 lamps respectively.

More information: http://www.ci.north-saintpaul.mn.us/index.asp?Type=B_BASIC&SEC={F5C1E5FE-1ADD-49AF-ACD859C3FB4192FF} Connexus Energy, along with the Dakota Electric Association (East Central Energy, Elk River Municipal Utilities, Minnesota Valley Electric cooperative, Shakopee Public Utilities, Stearns Electric Association) offers rebates for both retrofit applications and new construction. Connexus recommends replacing T12 or metalide systems with MH and HPS fixtures. $25 per fixture rated 150W and under, $35 per fixture rated between 151W to 250W, and $55 per fixture rated over 251W. Incentives also exist for the installation of new PMH over older less efficient technology. Fixtures 175W are eligible for $35 per fixture, fixtures in the range of 176W to 319W are eligible for $55 per fixture, fixtures between 320W to 749W have a $70 per fixture rebate, and fixtures over 750W have a rebate of $80 per fixture.

CMH technology is also included in the retrofit program. $35 per fixture 150W $50 per fixture that draws power in the range of 151W to 250W $75 per fixture that draws more than 251W

Connexuss new construction program is similar. MH or HPS fixtures rated at less than 151W are eligible for a $6.60 per fixture rebate. PMH fixtures have the following incentives: $6.50 per fixture 175W $8.80 per fixture in the range of 176W to 319W $13.20 per fixture in the 320W to 749W range $19.80 per fixture rated over 749W

A-11

CMH technology rebates are as follows: $22 per fixture 150W, $11 per fixture that draws power in the range of 151W to 250W $16.50 per fixture that draws more than 250W

More information: http://www.connexusenergy.com/busrebates.htm http://www.dakotaelectric.com/cirebate.asp. https://www.eastcentralenergy.com/rebatesbusiness.aspx http://www.elkriverutilities.com/cipeleccomm.php http://www.mvec.net/business/grants_rebates.asp http://www.shakopeeutilities.com/commrebates.htm http://www.stearnselectric.org/grantsbus.htm

Otter Tail Power Company focuses on demand reduction with incentives based on a kW savings. Though, their program specifically addresses HID technology by including in their definition of eligible fixtures, efficient HID lighting. Rebates of 20/KWh saved is offered when MV lamps are replaced with HPV or high efficiency PMH. More information: http://www.otpco.com/SaveEnergyMoney/Rebates.asp. A.13 MISSOURI INCENTIVE PROGRAMS

Empire District Electric Company offers rebates to certain commercial and industrial customers for the installation of energy efficiency equipment. They provide a rebate of $50/fixture for installing a PMH. The lamp wattage must be either 320 or 360W as a replacement for 400W MH or HPS. $50 rebate on replacing high intensity HID with T5 or T5HO lamps with electronic ballast. More information: http://empire.programprocessing.com/content/Home. Kansas City Light & Power only focuses on PMH technology. $50 per fixture is avaible for 320W and 360W PMH replacing 400W MH and HPS fixtures. $50 rebate on replacing high intensity HID with T5 or T5HO lamps with electronic ballast. More information: http://www.kcplprograms.com/content/prescriptiverebates.

A-12

A.14

MONTANA INCENTIVE PROGRAMS

Flathead Electric Cooperative offers a variety of incentives for new and existing equioment in their Commercial Lighting Rebate Program. For new construction projects: $40 per CMH fixture ranging in wattage from 20W to 100W $50 per CMH fixture ranging in wattage from 101 to 400W $50 per HIF fixture

For retrofit applications: $80 per CMH fixture in wattage range of 99W or less $150 per CMH fixture for wattages 100W or more $150 per PMH fixture in wattage range of 200W to 399W $200 per PMH fixture in wattage range of 400W to 750W $80 to $140 per HIF fixtures ranging in wattage from 85W to 600W Finally, there are other incentives for replacing incandescent and MV sources with LEDs, fluorescent, or compact fluorescent.

More information: http://www.flatheadelectric.com/energy/Rebates.html. A.15 NEW HAMPSHIRE INCENTIVE PROGRAMS

National Grid offers incentives for new equipment as well as replacing existing equipment. For new construction, $15 per new PMH fixtures is available if lamps are greater than 150W with linear ballasts or 250W with CWA ballasts. HIF fixtures incentives include $30 for low-bay fixtures in the 125W to 219W range and $50 for high-bay fixtures greater than 219W. $90 for new low-wattage MH fixtures for retail or display lighting exists. Fixtures must be from 20W to 100W and may be track, recessed, or surface mounted and used for high quality display type lighting. $90 per HID fixture exists for HID controls. Either occupancy controlled hi-low systems or daylight dimming systems. For existing facilities, $40 per fixture is offered for new PMH lamp and ballast installed per the manufacturers specifications and applicable codes. $60 per fixture is offered for an entirely new PMH. $70 per fixture is offered for new HPS. They consider HPS more efficient than MH. More information: https://www.nationalgridus.com/granitestate/business/energyeff/5_light.asp

A-13

The New Hampshire Electric Cooperative has quite a comprehensive incentive program for lighting. $15 per new PMH fixtures is available if lamps are greater than 150W with linear ballasts or 250W with CWA ballasts. HIF fixtures incentives include $30 for low-bay fixtures in the 125W to 219W range and $50 for high-bay fixtures greater than 219W. $90 for new lowwattage MH fixtures for retail or display lighting exists. Fixtures must be from 20W to 100W and may be track, recessed, or surface mounted and used for high quality display type lighting. $90 per HID fixture exists for HID controls. Either occupancy controlled hi-low systems or daylight dimming systems. More information: http://www.nhec.com/business_energysolutions_newbusiness.php. The New Hampshire Electric Cooperative also offers incentives for its large businesses. HID interior fixtures are eligible for rebates. $40 per fixture is offered for new PMH lamp and ballast installed per the manufacturers specifications and applicable codes. $60 per fixture is offered for an entirely new PMH. $70 per fixture is offered for new HPS. They consider HPS more efficient than MH. More information: http://www.nhec.com/business_energysolutions_largebusiness.php. A.16 NEW JERSEY INCENTIVE PROGRAMS

New Jersey Clean Energy is a statewide program that promotes energy efficiency and renewable energy. The specific program with HID rebates is the New Jersey SmartStart Buildings and offers the following incentives: $25 per PMH fixture rated at 150W or more $75 per HID fixture controlled via Hi-Low or dimming ballasts $284 per T5 or T8 fixture replacing an HID, T12, or incandescent fixture rated at 1000 W $16 per T5 or T8 fixture for replacing HID lumainires rated between 75-99W $30 per T5 or T8 fixture for replacing HID fixtures rated between 100-175W $43 per T5 or T8 fixture for replacing HID fixtures rated between 175-249W $50 per T5 or T8 fixture for replacing HID fixtures rated between 250-399W $100 per T5 or T8 fixture for replacing HID fixtures rated between 400-499W

More information: http://www.njcleanenergy.com/

A-14

A.17

NEW YORK INCENTIVE PROGRAMS

The Long Island Power Authority manages a Commercial Construction Program to promote energy-efficient lighting in new construction projects. Applicable incentives are as follows: $45 per CMH fixture replacing one-for-one incandescent, halogen, MV, and HPS fixtures $35 per T5 or T8 high-bay fixture $40 per multi-lamp CFL high-bay fixture $20 per integrated ballast CMH PAR lamp $20 per PMH fixture with electronic ballast $75 per HID fixture controlled via Hi-Low occupancy sensor or daylight controls

More information: http://www.lipower.org/cei/commercial.pre.html. A.18 OHIO INCENTIVE PROGRAMS Dayton Power and Light offers the following incentives: $35 for PMH or CMH rated between 0-100 W $40 for PMH or CMH rated between 100-250W $50 for PMH or CMH rated above 250W $75 for MH track lighting

It offers rebates for replacing HID systems with the following T5 HO lamps: $30 per T5 HO High-Bay 2-lamp fixture replacing HID fixtures $40 per T5 HO High-Bay 3-lamp fixture replacing HID fixtures $50 per T5 HO High-Bay 4-lamp fixture replacing HID fixtures $60 per T5 HO High-Bay 6-lamp fixture replacing HID fixtures $75 per T5 HO High-Bay 8-lamp fixture replacing HID fixtures

More information: http://www.dpandl.com/bizprograms.php.

A-15

Duke Energy offers $25 per fixture for replacing 400W MH with 320W PMH (if the fixture operates a minimum of 1800 hours per year). More information: http://www.duke-energy.com/indiana-business/energymanagement/energy-efficiency-incentives.asp. A.19 OREGON INCENTIVE PROGRAMS

Columbia River Public Utility District sponsors a Business LightingRetrofit program. MV, HPS, and MH are listed as existing equipment with recommended replacement technology. Columbia River PUD recommends and offers rebates for replacing incandescent, T12 and MV with new fluorescent fixtures. In a separate section, rebates are offered for specific HID technology to replace incandescent and MV lamps. $80 per new CMH fixture ranging from 20W to 100W $150 per new CMH fixture ranging from 101W to 250W

For high bay or industrial applications, the program recommends replacing T12, MV, HPS, incandescent or MH with either high output fluorescent or PMH. The following incentives exist: $80 to $140 per fluorescent (T8, T5, or long biax CFL) fixture raninging in wattage from 85W to 600W $150 per PMH fixture in the 300W to 399W range $200 per PMH fixture in the 400W to 750W range

More information: http://www.crpud.net/business/lighting. Emerald Peoples Utility District previously sponsored a Lighting Retrofit Rebate Program. This program has changed to the Commercial Energy Program. EPUD will perform lighting audits of existing buildings. Recommnedations will be made for improvements. EPUD will estimate the possible rebates and perform economic analysis. If the project moves forward, EPUD can help with construction management assistance. EPUDs website about the progam explains past projects include fluorescent lighting retrofits in offices, gyms, and retail stores-MV replacements, to high efficiency metal halide, or HPS lights and dimming and lighting control upgrades. More information: http://www.epud.org/comnrg.aspx. Eugene Power offers incentives for energy-efficient lighting design. This is a two-tiered program. In the first tier, the lighting power density must be reduced by 10% but the design must still meet the lighting requirements of the program. The only HID technology allowed is PMH, all other are excluded. This tier has an incentive rate of 0.420. The second tier is similar, yet, the

A-16

lighting power savings must be a minimum of 20%. In this tier, no HID sources are allowed for interior applications. The incentive rate for this tier is 0.560. For replacing existing HID with fluorescent fixtures, the following rebates are offered: $50 for 120 or less wattage T5, T5HO, or T8 high-performance fixture replacing 175 to 250W HID $130 for 260 or less wattage T5, T5HO, or T8 high-performance fixture replacing 400W HID $90 for 310 or less wattage T5, T5HO, or T8 high-performance fixture replacing 400W HID $60 for 370 or less wattage T5, T5HO, or T8 high-performance fixture replacing 400W HID $25 per 25W integral ballast CMH PAR replacing 90W to 150W incandescent

More information: http://www.eweb.org/public/documents/energy/Commercial_Retrofit_Lighting_Catalog.pdf The Energy Trust of Oregon manages a statewide program called the Building Efficiency Program. This program aims to promote energy-efficient lighting upgrades that incorporate HPS or MH fixtures. The Busines Energy Solutions program has rebates for replacing MV and incandescent lighting with HPS or MH fixtures$25 for a HPS or MH fixture less than 175 watts, and $30 for a HPS or MH fixture greater than 175 watts. Energy Trust also has a set of incentives for New Construction: $35 per T8/T5 fixture for Hi Bay applications, no specific wattage requirements of the fixture, but lamping requirements; $75 per HID fixtures controlled in Hi-Lo applications; $100 per HID fixture used for daylight dimming; $50 per 150W PMH, $60 per PMH fixture in the 150W to 400W range, and $70 for PMH fixtures 400W and greater. More information: http://www.energytrust.org/business/index.html. Idaho Power offers a series of incentives related to HID lighting. Metal Halide lighting is specifically identified. These incentives only apply to lighting upgrades for interior lighting (street, area, and parking lot lighting is excluded). Eligible metal halide fixtures include PMH or CMH that are replacing higher wattage MH, MV, or other sources. The incentives are the following: $18.00 for PMH fixtures between 30W to 70W when replacing a fixture drawing 20W more than the MH wattage. $22.00 for PMH fixtures between 70W to 150W when replacing a fixture drawing 25W more than the MH wattage. $26.00 for PMH fixtures between 150W to 250W when replacing a fixture drawing at least 40W more than the MH wattage.

A-17

$55.00 for PMH fixtures between 250W to 360W when replacing a fixture drawing at least 80W more than the MH wattage. $75.00 for PMH fixtures between 360W to 500W when replacing a fixture drawing at least 120W more than the MH wattage. $105.00 for PMH fixtures for 500W and more when replacing a fixture drawing at least 200W more than the MH wattage.

More information: http://www.idahopower.com/easyupgrades/. Midstate Electric sponsors a Commercial & Industrial Rebate Programs. MV, HPS, and MH are listed as existing equipment with recommended replacement technology. Midstate recommends and offers rebates for replacing incandescent, T12 and MV with new fluorescent fixtures. In a separate section, rebates are offered for specific HID technology to replace incandescent and MV lamps. $50 per new CMH fixture ranging from 20W to 100W $80 per new CMH fixture ranging from 101W to 250W

For high bay or industrial applications, the program recommends replacing T12, MV, HPS, incandescent or MH with either high output fluorescent or PMH. The following incentives exist: $80 to $140 per fluorescent (T8, T5, or long biax CFL) fixture raninging in wattage from 85W to 600W $100 per PMH fixture in the 300W to 399W range $150 per PMH fixture in the 400W to 750W range

In addition to providing information about the rebates, Midstates website lists local business that have reduced their operating expenses by using Midstates lighting rebates. More information: http://www.midstateelectric.coop/ProductsAndServices/CommercialIndustrial/EnergyEfficiency Programs/default.aspx. A.20 TEXAS INCENTIVE PROGRAMS

Austin Energy sponsors a lighting rebate program, Commercial/Business Rebates and Incentives, with the objective of energy and demand through higher efficiency lighting systems. $300 per kW saved is eligible for replacing existing HID with T8 or T5 high-bay fixtures. More information: http://www.austinenergy.com/

A-18

A.21

VERMONT INCENTIVE PROGRAMS

Efficiency Vermont, a statewide non-profit organization funded out of an efficiency charge applied to the electricity bill, manages a program called Commercial Lighting Equipment. A $50 rebate per fixture is available when replacing 250W or greater MH or HPS fixtures with a new fixture with an electronic ballast. A caveat exists stating that additional funding is available if dimming controls are used. A $30 rebate per fixture is available for the installation of MH or HPS exterior fixtures that do not use more than 100W. CMH technology has a $75 per fixture rebate. The incentive is directed towards trackheads or recessed downlights (cans), but any technology that uses low-wattage CMH is applicable. In addition to subsidizing equipment, a rebate of $20 is offered for the 25W CMH PAR38 lamp with integrated ballast. More information: http://www.efficiencyvermont.org/pages/Business/HVAC/Lighting/LightingRebates/. A.22 WASHINGTON INCENTIVE PROGRAMS

Avista Utilities Commercial Lighting Incentives offers incentives for HID interior and exterior fixtures. When replacing 400W MH high bay with multi-lamp fluorescent high bay, $125 - $460 per fixture is available. . Incandescent flood of 100 W or more replaced with CMH are eligible for $ 35 rebate per fixture. . Avistas incentives are also available for new construction on a case-by-case basis. For replacing 400W HID fixtures with 4-lamp T5 high output fixture, 6 lamp T5 high output fixture, 6 lamp T8 fixture, 8 lamp T8 fixture and 200 Watt induction Fluorescent Fixture, rebates of $125, $90, $125, $110 and $220 are offered respectively. Rebate of $1000 is offered to replace 1000 Watt HID fixture with (2) 6 lamp T5 High output fixtures. More information: https://www.avistautilities.com/business/rebates/washington_idaho/Documents/Lighting%20We b.pdf Benton Public Utility District sponsors a Commercial Lighting & Special Projects program. MV, HPS, and MH are listed as existing equipment with recommended replacement technology. Benton PUD recommends and offers rebates for replacing incandescent, T12 and MV with new fluorescent fixtures $8 to $15 per fixture. In a separate section, rebates are offered for specific HID technology to replace incandescent and MV lamps. $50 per new CMH (from 20W to 100W) fixture and $80 per new CMH (from 101W to 250W) fixture for replacement of incandescent and MV sources.

For high bay or industrial applications, the program recommends replacing T12, MV, HPS, incandescent or MH with either high output fluorescent or PMH fixtures. Rebates are as follows:

A-19

$80 to $140 per fluorescent fixture raninging in wattage from 85W to 600W range $100 per PMH fixture in the 300W to 399W range $150 per PMH fixture in the 400W to 750W range

More information: http://www.bentonpud.org/save_energy_money/comm_lighting.html. Richland Energy Services sponsores Energy Efficient Commercial Lighting Program. Customers are eligible for: Incentives exist for replacing incandescent or MV sources as follows: $80 per new CMH fixture in the 20W to 100W range $150 per new CMH fixture in the 101W to 250W range

For high bay or industrial applications, the program recommends replacing T12, MV, HPS, incandescent or MH with either high output fluorescent or PMH fixtures. Rebates are as follows: $80 to $140 per fluorescent fixture raninging in wattage from 85W to 600W range $100 per PMH fixture in the 300W to 399W range $150 per PMH fixture in the 400W to 750W range

More information: http://www.ci.richland.wa.us/RICHLAND/Electric/docs/86/docs/EECLPRebates.pdf?CFID=2222963&CFTOKEN=84833416 Puget Sound Energy has rebates for lighting improvements for small business. The program has subsets for different technology conversions. The subset that applied to HID technology is Section II: Higher Wattage Incandescent & HID Conversions. Few parameters exist for the equipment with the rebates. $90 to $125 per fixture is available for higher efficiency HPS/MH fixtures. T8/T5 fixtures for fluorescent warehouse lighting are eligible for $110 to $225 per fixture. Finally, high wattage CFL wall packs, a traditional HID fixture, are listed with rebates in the range of $110 to $130 per fixture. More information: http://www.pse.com/solutions/forbusiness/pages/comRebates.aspx?tab=2&chapter=1 Seattle City Light manages a program called Energy Smart Services Financial Incentives that provides financial incentives to small businesses for replacing inefficient lighting with approved energy-efficient lighting equipment. This program offers incentives for specific technologies, but pays per kWh saved. MV is explicitly banned from being installed to receive rebates. In contrast, HPS and MH are specifically allowed for the incentive programs.

A-20

More information: http://www.seattle.gov/light/conserve/business/cv5_fi.htm. A.23 WISCONSIN INCENTIVE PROGRAMS

Baron Electric Cooperative offers $10 per MH and $10 per HPS fixture. Only retrofit applications are applicable. The only parameters are that the fixtures be hard-wired. More information: http://www.barronelectric.com/services/efficientlighting.html. Cedarburg Light & Water Utility offers some incentives. To be eligible, a representative from the organization must perform an energy audit. At the end of the audit, if potential savings exist, cash incentives might be provided to help reduce the cost of the equipment. Technologies listed specifically focus on more efficient HID technologies. More information: http://www.cedarburglightandwater.com/business_customers/participationrequest.asp?Category Number=1&SubcategoryNumber=2. Eau Claire Energy Cooperative offers $10 per MH and $10 per HPS fixture. Only retrofit applications are applicable. The only parameters are that the fixtures be hard-wired. More information: http://www.ecec.com/electric/rebates/rebateslighting.htm. Riverland Energy Cooperative offers $10 per MH and $10 per HPS fixture. Only retrofit applications are applicable. The only parameters are that the fixtures be hard-wired. More information: http://www.riverlandenergy.com/RIVERLAND%20HOMEPAGE/commercial/incentives.htm.

A-21

CHAPTER 7.

REFERENCES

. Updyke, C. Personal communication. National Electrical Manufacturers Association Lamp Section. E-mail to the U.S. Department of Energy. May 2, 2003. . . National Electrical Manufacturers Association. Personal communication. Meeting with the U.S. Department of Energy. January 21, 2003. U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. 2008 Buildings Energy Data Book. March 2009. <http://buildingsdatabook.eere.energy.gov/docs/DataBooks/2008_BEDB_Updated.pdf> U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Comments from the Public Meeting for Technical Support Document: Energy Efficiency Program for Commercial and Industrial Equipment: High-Intensity Discharge Lamps Analysis of Potential Energy Savings. December 2004. Transcript for Docket DOE EEDET-03-001, Comment 06. National Electrical Manufacturers Association. General Lighting High Intensity Discharge Lamps: Manufacturers. 2010. (Last accessed January 15, 2010.) <https://www.nema.org/prod/lighting/lamps/general/> National Electrical Manufacturers Association. High Intensity Discharge (Electronic Ballasts). 2010. (Last accessed January 15, 2010.) <http://www.nema.org/prod/lighting/ballasts/electronic/hid/> National Electrical Manufacturers Association. Electromagnetic Ballasts. 2010. (Last accessed January 15, 2010.) <https://www.nema.org/prod/lighting/ballasts/electromag> Vorsatz, D., L. Shown, J. Koomey, M. Moezzi, A. Denver, and B. Atkinson. Lighting Market Sourcebook for the U.S. 1997. Lawrence Berkeley National Laboratory: Berkeley, CA. Report No. LBNL-39102, UC-1600. Illuminating Engineering Society of North America. The IESNA Lighting Handbook: Reference and Application. M. S. Rea, Editor. 9th ed. 2002. New York, NY. . . Energy Policy Act of 2005, Public Law 109-58, 119 Statute 626. U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Building Energy Standards Program: Determination Regarding Energy Efficiency Improvements in the Energy Standard for Buildings, Except Low-Rise Residential Buildings, ASHRAE/ IESNA Standard 90.1-1999; Final Rule. Federal Register. July 15, 2002. vol. 67, no. 135: pp. 4646446488.

. .

10 11

7-22

12

McColgan, M. and J. Van Derlofske. Specifier Report: Parking Lot and Area Luminaires. Lighting Research Center. National Lighting Product Information Program. July 2004. vol. 9, no. 1. U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Comments from the Public Meeting for Technical Support Document: Energy Efficiency Program for Commercial and Industrial Equipment: High-Intensity Discharge Lamps Analysis of Potential Energy Savings. December 2004. Transcript for Docket DOE EEDET-03-001, Comment 08. Ibid, Comment 11. Ibid, Comment 13. Ibid, Comment 14. Ibid, Comment 15. New York StateDepartment of State: Division of Code Enforcement and Administration: Laws & Regulations. Title 19Appliance and Equipment Energy Efficiency Standards. Official Compilation of the Rules and Regulations of the State of New York. International Dark-Sky Association. IDA Suggested Lighting Zones for Codes and Ordinances. (Last accessed January 15, 2010.) <http://www.darksky.org/mc/page.do?sitePageId=90679> Arizona State Legislature Revised Statu[t]es. Title 49The Environment, Chapter 7, Article 1Light Pollution. 49-1104: Use of mercury vapor light fixtures. (Last accessed January 15, 2010). <http://www.phoenixdarkskies.org/arizonalightingordinances.html> City of San Jos, California. City Council Policy 4-3. Outdoor Lighting on Private Developments. Effective 3/1/83, revised 6/20/00. (Last accessed January 15, 2010). <www.sanjoseca.gov/planning/counter/policies/pol_lighting.pdf> Kansas Department of TransportationBureau of Materials and Research. List of Prequalified Roadway Lighting Systems & Equipment. PQL24.2Revised 05/01/2009. (Last accessed January 15, 2010.) <www.ksdot.org/burmatrres/pql/pql-24-02.pdf> (State of) Oregon Revised Statutes. Relating to school lighting. Amending ORS 332.207 and section 2, chapter 312, Oregon Laws 2007. 74th Oregon Legislative Assembly--2008 Special Session. (Last accessed January 15, 2010.) <http://www.leg.state.or.us/08ss1/measures/sb1000.dir/sb1067.intro.html>

13

14 15 16 17 18

. . . . .

19

20

21

22

23

7-23

24

. . . .

National Electrical Manufacturers Association. Standards and Publications. 2009. (Last accessed February 2, 2010.) <http://www.nema.org/stds> Illuminating Engineering Society of North America. The IESNA Lighting Handbook: Reference and Application. M. S. Rea, Editor. 9th ed. 2002. New York, NY. General Electric. GE Lamps & Ballasts 2010 Products Catalog. (Last accessed February 2, 2010.) <http://www.gecatalogs.com> OSRAM SYLVANIA. Lamp and Ballast Catalog. 2008. (Last accessed February 2, 2010.) <http://assets.sylvania.com/assets/documents/Complete-Catalog.b176dbb1-d6e0-40f0ab92-e768e58f5dc1.pdf> Philips Lighting Company. Lamp Specification and Application Guide: High Intensity Discharge Lighting. 2008. SAG100 2008-2009. (Last accessed February 2, 2010.) <http://www.lighting.philips.com/us_en/browseliterature/download/SAG100_2009_HID.pdf> USHIO America, Inc. General Lighting Product Catalog. 2006. (Last accessed February 2, 2010.) <http://www.ushio.com/products/catalog.htm#GE> Venture Lighting. North American HID Lighting Specification and Energy Saving Products Guide. 2008. Summer Edition. (Last accessed February 2, 2010.) <http://www.venturelighting.com/On-LineCatalog/VentureLightingSummer08.pdf> Nelson, G. J., R. G. Gibson, and A. D. Jackson. An Efficacy Analysis of HID Lamps. Journal of the Illuminating Engineering Society. Winter 2001. 30(1): pp. 6875. Howley, J. G. Personal communication. GE Lighting. E-mail permission to use graphic to DOE via C. Updyke, NEMA. May 2, 2003. Bleasby, P. Personal communication. OSRAM Sylvania. E-mail permission to use graphic to DOE via C. Updyke, NEMA. May 1, 2003. Jack, A. G. and M. Koedam. Energy Balances for Some High Pressure Gas Discharge Lamps. Journal of the Illuminating Engineering Society. July 1974. 3(4): pp. 323329. OSRAM SYLVANIA. Lumalux/Unalux High Pressure Sodium Lamps. 1998. Engineering Bulletin HID009. (Last accessed February 2, 2010.) <http://ecom.mysylvania.com/miniapps/FileNet2/EngineeringBulletins/ENGR_BLTN2.pdf> Louden, W. C. and K. Schmidt. High-Pressure Sodium Discharge Arc Lamps. Illuminating Engineering. 1965. 60(12): pp. 696702.

25

26

27

28

29

. .

30

31

. . . . .

32

33

34

35

36

7-24

37

Philips Lighting Company. High Intensity Discharge Lighting. Online publication SAG-100 2006. (Last accessed February 2, 2010.) <http://www.nam.lighting.philips.com/us/ecatalog/catalogs/2006_SAG100_HID.pdf> Ward, P. C. Estimation of HPS Lamp Voltage Rise as a Function of Ballast Characteristic Curve Parameters. Journal of the Illumination Engineering Society. October 1983: pp. 157161. American National Standards Institute (ANSI). High-Pressure Sodium Lamps. ANSI document C78.42-2001. Ingalls, P., R. Dolan, J. Plumb, H. Zhu, and E. Wyner. A Noncycling High Pressure Sodium Lamp. Journal of the Illuminating Engineering Society. Summer 1995. 24(2): pp. 1925. Plumb, J., G. Duggan, D. Goodman, R. Snellgrove, and E. Wyner. Emitter for UnsatruatedVapor High Pressure Sodium Lamps. Journal of the Illuminating Engineering Society. Summer 1995. 24(2): pp. 2632. Work, D. E. Chemistry of Metal Halide Lamps: A Review. Lighting Research and Technology. 1981. 13(3): pp. 143152. OSRAM SYLVANIA. High Intensity Discharge Metalarc Lamps. 1998. Engineering Bulletin HID-010-0995. (February 2, 2010.) <http://ecom.mysylvania.com/miniapps/FileNet2/EngineeringBulletins/ENGR_BLTN1.pdf> Work, D., C. Mannino, B. Ives, and M. Hewitt. Personal communication. Philips Lighting Company, Somerset, NJ. E-mail to E. Hong, Navigant Consulting, Inc. Final illustration and permission to use provided by e-mail from M. Hewitt of Philips to Navigant via C. Updyke, NEMA. August 4, 2003. Wypasek, D. Personal communication. Advanced Lighting Technologies, Inc. Permission to use graphic provided by e-mail to DOE via Craig Updyke of NEMA. May 16, 2003. National Lighting Product Information Program. Lighting Answers: Mid-Wattage Metal Halide Lamps 7(1). Troy, NY: Lighting Research Center. Waymouth, J.F. Electric Discharge Lamps. 1971. Cambridge, MA: The MIT Press. Lorenz, R. Improvement of Metal Halide Lamps by Complex Formation. Lighting Research & Technology. 1976. 8(3): pp. 136140. Tielemans, P., P. Drop, and R. Lorenz. Spectral Properties of Metal Halide Lamps with Rare Earth-Iodide or Sc-Iodide and Sodium Iodide. Lighting Research & Technology. 1985. 17(2): pp. 7983.

38

39

. .

40

41

42

. .

43

44

45

. . . . .

46

47 48

49

7-25

50

Tu, J., J. van der Eyden, T. De Baerdemaeker, J. Woods, K. Graham, A. Dunaevsky, and N. Caruso. New Advancements in Medium Wattage Ceramic Metal Halide Lamp. Leukos. January 2009. 5(3): pp. 152165. Deng L., C. ORourke, J. Fullam, and R Hosseinzadeh. Performance of Medium Wattage Pulse Start Metal Halide Lamps. In IESNA Annual Conference Proceedings, Technical Papers. 2004. New York, NY. Illuminating Engineering Society of North America. pp. 193202. Northeast Waste Management Officials Association (NEWMOA)/IMERC. Trends in Mercury Use in Products: Summary of the Interstate Mercury Education & Reduction Clearinghouse (IMERC) Mercury-Added Products Database. June 2008. Franke, S., H. Schoepp, O. Langenscheidt, R. Methling, H. Hess, H. Schneidenbach. Study of Noble Gases as Mercury Substitutes in High-Pressure Discharge Lamps. Leukos. January 2007. 3(3): pp. 217227. Flory IV, I. L. and S. Rahman. Metal-Halide Lamp Life and Lumen Depreciation: A Comparison of Two, 400 Watt Discharge Lamp Ballasting Systems. Leukos. January 2008. 4(3): pp. 173185. Rasch, E. and E. Statnic. Behavior of Metal Halide with Conventional and Electronic Ballasts. Journal of the Illuminating Engineering Society. Summer 1991. 20(3): pp. 8896. National Lighting Product Information Program (NLPIP). Low-wattage Metal Halide Lighting Systems. Specifier Reports. October 2006. 10(1). (Last accessed February 2, 2010.) <http://www.lrc.rpi.edu/nlpip/publicationDetails.asp?id=911&type=1> U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Comments from the Public Meeting for Energy Conservation Program: Energy Conservation Standards for Metal Halide Lamp Fixtures: Public Meeting and Availability of the Framework Document. Transcript for Docket EERE-2009-BT-STD-0018 Regulatory Identification Number (RIN) 1904-AC00. Fukumori, F., H. Nishimura, K. Uchihashi, and M. Fukuhara. A Study of HID Lamp Life when Operated by Electronic Ballasts. Journal of the Illuminating Engineering Society. Winter 1995. 24(1): pp. 4147. IESNA publication DG-13-98 Selection of Photocontrols for Outdoor Lighting Applications Esource resource guide High-Bay Lighting: HID versus HIF, published November 7, 2007 INFORM, Inc. published in December 2003 titled Purchasing for Pollution Prevention HighBay Lighting: Opportunities for Mercury Reduction and Energy Efficiency

51

52

53

54

55

. .

56

57

58

59 60 61

7-26

62 63

Esource document, New High-Intensity Fluorescent Lights Outshine Their HID Competitors Arkansas Department of Finance and Administration

(http://www.dfa.arkansas.gov/offices/procurement/contracts/Documents/sp070399ConsolidatedP ricing.pdf) Last accessed August 28, 2009

64

State of West Virginia Department of Administration Purchasing Division

(http://www.state.wv.us/admin/purchase/SWC/LIGHT07_01.pdf) Last accessed August 28, 2009

65

North Dakota State Government Office of Management and Budget State Procurement Online

(https://secure.apps.state.nd.us/csd/spo/services/bidder/listCurrentContracts.htm) Last accessed August 28, 2009

66

State of Oklahoma Lamp & Ballast Catalog (Contract #SW900025) Tier 1 Pricing

(http://www.dcs.state.ok.us/sw_contracts.nsf/6fe2a5d9256854f886256c63004e411e/94517660f0 46390c862575ba004ef018/$FILE/Oklahoma%20Lamp%20%20Ballast%20eCat.pdf) Last accessed August 28, 2009

67

Ohio Department of Administrative Services (http://procure.ohio.gov/proc/viewContractsAwards.asp?contractID=11568)-Last Accessed August 28, 2009 State of New Jersey Department of Treasury Division of Purchase and Property Index #:T0192, Solicitation #:39192

68

(http://www.state.nj.us/treasury/purchase/noa/contracts/t0192.shtml) Last accessed August 28, 2009

69

W.W. Grainger, Inc.

(www.grainger.com) Last accessed April 16, 2010

70

GoodMart

(http://www.goodmart.com/) Last accessed April 16, 2010

71

High-Intensity Discharge 2003 Draft Technical Support Document

(http://www1.eere.energy.gov/buildings/appliance_standards/commercial/pdfs/hid_draft_report.p df)

7-27

72

U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Comments from the Public Meeting for Energy Conservation Program: Energy Conservation Standards for Metal Halide Lamp Fixtures: Public Meeting and Availability of the Framework Document. Transcript for Docket EERE-2009-BT-STD-0018 Regulatory Identification Number (RIN) 1904-AC00. http://www.deeresources.com/ Lighting Maintenance. Environmental Protection Agency. EPA 430-B-95-009 January 1995 Fluorescent Lamp Ballasts Preliminary Technical Support Document

73 74 75

http://www1.eere.energy.gov/buildings/appliance_standards/residential/fluorescent_lamp_ballast s_ecs_prelim_tsd.html
76 77

Ibid 11 LED Walkway Lighting: Federal Aviation Administration (FAA) Technical Center PNNL17407 (http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/gateway_faa.pdf) Demonstration Assessment of Light-Emitting Diode (LED) Area Lights for a Commercial Garage PNNL-17995

78

(http://www1.eere.energy.gov/buildings/ssl/gatewaydemos_results.html)
79

Demonstration Assessment of Light-Emitting Diode (LED) Roadway Lighting PNNL-18687

(http://www1.eere.energy.gov/buildings/ssl/gatewaydemos_results.html)
80

LED Street Lighting Pacific Gas and Electric Company Application Assessment Report #0727 (http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/gateway_sf-streetlighting.pdf) Application Assessment of Bi-Level LED Parking Lot Lighting Pacific Gas and Electric Company Application Assessment Report #0815 (http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/gateway_raleys.pdf) Demonstration Assessment of Light-Emitting Diode (LED) Street Lighting PNNL-18947

81

82

(http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/gateway_lija-loop.pdf)
83

Sustainability Success Story Kwantlen Polytechnic University (http://www.kwantlen.ca/__shared/assets/0501_Success_Story__Incandescent_Exterior_Fixture_Conversion_-_Surey6592.pdf) Energy Information Administration, Annual Energy Outlook 2009, DOE/EIA-0383(2009), U.S. Department of Energy, Washington, D.C., February 2009

84

7-28

85

Fluorescent Lamp Ballasts Preliminary Technical Support Document

http://www1.eere.energy.gov/buildings/appliance_standards/residential/fluorescent_lamp_ballast s_ecs_prelim_tsd.html
86

U.S. Department of EnergyOffice of Energy Efficiency and Renewable Energy. Comments from the Public Meeting for Framework Analysis: Energy Efficiency Program for Commercial and Industrial Equipment: Metal Halide Lamp Fixture. January 2010. Transcript for Docket DOE EERE-2009-BT-STD-0018/RIN 1904-AC00 Comment 1.2.005

87

Ibid 1

7-29