Abstract: This study predicts new hybrid lighting applications for LEDs. In hybrid lighting, LEDs provide a low-energy ?standby? light level while another, more powerful, efficient light source provides light for occupied periods. Lighting controls will allow the two light sources to work together through an appropriate control strategy, typically motion-sensing. There are no technical barriers preventing the use of low through high CRI LEDs for standby lighting in many interior and exterior applications today.

Abstract: This document is a report prepared in fulfillment of Task 1 of the Subcontract Statement of Work between Lawrence Berkeley National Laboratory and Broadata Communications, Inc. The statement of work that covers the requirements for this report is included in the following section.

Abstract: This report presents a framework for per-pixel analysis of the qualitative and quantitative aspects of luminous environments. Recognizing the need for better lighting analysis capabilities and appreciating the new measurement abilities developed within the LBNL Lighting Measurement and Simulation Toolbox, _Per-pixel Lighting Data Analysis_ project demonstrates several techniques for analyzing luminance distribution patterns, luminance ratios, adaptation luminance and glare assessment. The techniques are the syntheses of the current practices in lighting design and the unique practices that can be done with per-pixel data availability. Demonstrated analysis techniques are applicable to both computer-generated and digitally captured images (physically-based renderings and High Dynamic Range photographs).

Abstract: The goal of this project was to investigate broader building-level systems/strategies that enable further energy savings and control. This project investigated the potential savings offered by broader centralized control features and the potential advantages they may add to this system through such features as addressability and load shedding. This report documents the results of LBNL?s work in this area.

This report focuses on building-level systems and strategies and a multi-protocol gateway solution that is indifferent to the specific choice of lighting control/communications technique used to control the office lighting. The elegance of the IEEE 1451 intelligent gateway proposed in this report is that the overall building communications system should work regardless of whether the office lighting is controlled by DALI, UPB, IBECS, ZigBee or any other accepted communications protocol.

Abstract: The potential, limitations, and applicability of the High Dynamic Range (HDR) photography technique is evaluated as a luminance mapping tool. Multiple exposure photographs of static scenes are taken with a Nikon 5400 digital camera to capture the wide luminance variation within the scenes. The camera response function is computationally derived using the Photosphere software, and is used to fuse the multiple photographs into HDR images. The vignetting effect and point spread function of the camera and lens system is determined. Laboratory and field studies have shown that the pixel values in the HDR photographs can correspond to the physical quantity of luminance with reasonable precision and repeatability.

Abstract: In previous work, LBNL with Vistron Corp, developed an innovative lighting control system using a communications technology called Phase Cut Carrier (PCC). This report describes the performance of the desktop demonstration system that was developed to test this new controls concept. More detailed information on this project is given. This report is in fulfillment of deliverable #1 ?Report on Performance of Powerline-carrier Controlled Luminaire? from the FY2004 DOE Work Plan.

Abstract: One of the key deliverables for the DOE-funded controls research at LBNL for FY04 was the development of a prototype Personal Workspace Control system. The successful development of this system is a critical milestone for the LBNL Lighting Controls Research effort because this system demonstrates how IBECS can add value to today?s Task Ambient lighting systems. LBNL has argued that by providing both the occupant and the facilities manager with the ability to precisely control the operation of overhead lighting and all task lighting in a coordinated manner, that task ambient lighting can optimize energy performance and occupant comfort simultaneously [Reference Task Ambient Foundation Document]. The Personal Workspace Control system is the application of IBECS to this important lighting problem.

This report discusses the development of the Personal Workspace Control to date including descriptions of the different fixture types that have been converted to IBECS operation and a detailed description of the operation of PWC Scene Controller, which provides the end user with precise control of his task ambient lighting system.

Abstract: Despite a number of published studies on the effectiveness of lighting controls in buildings, only one US study examines the occupancy patterns of building occupants. Occupancy profiles allow one to determine, for example, the probability that an office is occupied for each hour of the workday. Occupancy profiles are useful for many purposes including: 1) predicting the effectiveness of occupancy sensors for reducing peak demand, 2) evaluating the impact of human activity on building lighting and other electric loads and 3) providing lighting equipment manufacturers with detailed lighting operation data to help evaluate the impact of advanced lighting controls on equipment life. In this paper, we examine the occupancy profiles for 35 single person offices at a large office building in San Francisco and analyze the data to obtain average occupancy as a function of time of day. In addition, we analyzed the data to identify how the use of occupancy sensors may affect switching cycles and lamp life.

Abstract: Integrated lighting controls can significantly improve building performance, increase energy efficiency, and enhance occupant comfort and satisfaction with the built environment. While previous research has shown that simple lighting controls such as occupancy sensors are effective at reducing the amount of electrical energy used for lighting in commercial buildings, advanced lighting control strategies have the potential to achieve even greater energy savings and offer many advantages over simple controls. But more advanced control strategies, such as daylighting or load shedding, which require a more systems-oriented approach, were less successful. Some of these difficulties are a result of the horizontal structure of the U.S. lighting controls market. While there are notable exceptions, the market is comprised largely of manufacturers of components (ballasts, switches and controls devices) rather than systems. Lighting controls components often do not work well together when specified as systems, especially in dimming applications where wiring is more elaborate. Lighting control equipment for implementing more complex strategies such as daylighting has proven difficult to commission in the field, leading to poor operation and user complaints. Finally, the lack of agreement on communications protocols was identified in as another market barrier.

Abstract: This report presents the results of laboratory measurements performed on The Watt Stopper?s LS-201 photosensor at the Lawrence Berkeley National Laboratory in January, 2003. The purpose of these measurements was to characterize the spatial and spectral response function of the LS-201 photosensor. This work was performed in fulfillment of a Project 3.3 Deliverable for the California Energy Commission?s Lighting Research Program.

The report has three sections. The first section describes the measurements performed to determine the spectral response function (sensitivity to light of different wavelengths) of the photosensor. The second section presents the results of our measurements of the sensitivity of the photosensor to light from different directions ? the spatial response function. The third section presents our conclusions and preliminary suggestions for product improvement consistent with project goals and objectives.

Abstract: Solid state lighting in the form of Light Emitting Diodes (LEDs) is bring new sources with different operating characteristics to the market. With the control in dimension, optics, intensity and color, these sources have the potential to transform the way we use light. This paper will review the recent improvements in performance that have been achieved by these devices, focusing on those product attributes identified as being critical to end users. The paper will conclude with a consideration of applications capitalizing on the LEDs" unique operating and physical properties.

Abstract: This paper discusses the successful development and testing of the first ballast/IBECS network interface that will allow commercially-available controllable ballasts to be operated from the Internet via IBECS (Integrated Building Environmental Communications System). The interface, which is expected to cost original equipment manufacturers (OEMs) only about $1-2/unit, has been hardened so that it is impervious to electronic noise generated by most 0-10 VDC controllable ballasts.

Abstract: Dramatic improvement in the efficiency of inorganic and organic light emitting diodes (LEDs and OLEDs) within the last decade has made these devices viable future energy efficient replacements for current light sources. However, both technologies must overcome major technical barriers, requiring significant advances in material science, before this goal can be achieved. Attention will be given to each technology associated with the following major areas of material research: 1) material synthesis, 2) process development, 3) device and defect physics, and 4) packaging.

The discussion on material synthesis will emphasize the need for further development of component materials, including substrates and electrodes, necessary for improving device performance. The process technology associated with the LEDs and OLEDs is very different, but in both cases it is one factor limiting device performance. Improvements in process control and methodology are expected to lead to additional benefits of higher yield, greater reliability and lower costs. Since reliability and performance are critical to these devices, an understanding of the basic physics of the devices and device failure mechanisms is necessary to effectively improve the product. The discussion will highlight some of the more basic material science problems remaining to be solved. In addition, consideration will be given to packaging technology and the need for the development of novel materials and geometries to increase the efficiencies and reliability of the devices. The discussion will emphasize the performance criteria necessary to meet lighting applications, in order to illustrate the gap between current status and market expectations for future product.

Abstract: We present analyses of two years" lighting data in open office areas, controlled by occupant sensors, time scheduling, or wall switches alone. We compare the energy savings using a before-after analysis of time scheduling and a conservative moving baseline analysis of both occupant sensing and time scheduling. We found that both techniques saved energy effectively when no occupants were present compared with wall switches alone. Time scheduling saved from 0.7 to 6.6% or an average of about 5%. Occupant sensors in similar areas saved from 9.0 to 14.6%, with an average of about 10%. False triggering of occupant sensors (by passersby) that would have caused energy waste was avoided by the presence of wall switches that positively turned lights off. Variations in occupant schedules and habits affected overall lighting energy use and the appropriateness of different control types. Little savings were found during the normal 8am to 5pm workday from either technology in large offices, but significant savings occurred after hours and on weekends. Our findings contrast with previous results for private offices in which only a single occupant is present, strengthening the evidence that different types of office space can be controlled appropriately with different types of control systems. However, small percentage savings in open areas result in larger actual savings due to the large number of fixtures controlled.

Abstract: Dimming controls for electric lighting have been one of the mainstays of the effort to use daylighting to reduce annual lighting energy consumption. The coincidence of daylighting with electric utility peak demand makes daylighting controls an effective strategy for reducing commercial building peak electric loads. During times of energy shortage, there is a greatly increased need to reduce electricity use during peak periods, both to ease the burden on electricity providers and to control the operating costs of buildings. The paper presents a typical commercial building electric demand profile during summer, and shows how daylighting-linked lighting controls and load shedding techniques can reduce lighting at precisely those times when electricity is most expensive. We look at the importance of dimming for increasing the reliability of the electricity grid in California and other states, as well as examine the potential cost-effectiveness of widespread use of daylighting to save energy and reduce monthly electricity bills.

Abstract: This report describes the work performed to design, develop, and demonstrate an IBECS network/ballast interface that is useful for economically dimming controllable ballasts in commercial buildings. The first section of the report provides the general background of the IBECS (Integrated Building Environmental Communications System) research and development work as well as the context for the development of the network/ballast interface. The research and development effort that went into producing the first proof-of-concept circuit and the physical prototype of that concept is detailed in the second section. In the third section of the report, we describe the lessons learned from the first demonstration of the network/ballast interface at an office at LBNL. The fourth section describes how electrical noise interference encountered with the first generation of interface led to design changes for a refined prototype that hardened the interface from electrical noise generated by the ballast. The final section of the report discusses the performance of refined prototype after we replaced the proof-of-concept prototype with the refined prototypes in the demonstration office at LBNL.

Abstract: This report describes the work performed to develop and test a new switching system and communications network that is useful for economically switching lighting circuits in existing commercial buildings. The first section of the report provides the general background of the IBECS (Integrated Building Environmental Communications System) research and development work as well as the context for the development of the new switching system. The research and development effort that went into producing the first proof-of-concept (the IBECS Addressable Power Switch or APS) and the physical prototype of that concept is detailed in the second section. In the third section of the report, we detail the refined Powerline Carrier Based IBECS Title 24 Wall Switch system that evolved from the APS prototype. The refined system provided a path for installing IBECS switching technology in existing buildings that may not be already wired for light level switching control. The final section of the report describes the performance of the IBECS Title 24 Switch system as applied to a small demonstration in two offices at LBNL"s Building 90. We learned that the new Powerline Carrier control systems (A-10 technology) that have evolved from the early X-10 systems have solved most of the noise problems that dogged the successful application of X-10 technologies in commercial buildings. We found that the new A-10 powerline carrier control technology can be reliable and effective for switching lighting circuits even in electrically noisy office environments like LBNL. Thus we successfully completed the task objectives by designing, building and demonstrating a new switching system that can provide multiple levels of light which can be triggered either from specially designed wall switches or from a digital communications network. By applying commercially available powerline carrier based technologies that communicate over the in-place lighting wiring system, this type of control can be economically installed even in existing buildings that were not wired for dual-level lighting.

Abstract: In a major test of different lighting control technologies in a typical office building, we present analyses of seven months" results from five control scenarios in private offices. We compare the energy savings and effectiveness of various combinations of occupant detection, daylight dimming, and switching techniques. Comparing measured energy use with occupant sensors against baseline energy use calculated using wall switch operation only, we found that occupant sensors saved 20-26% lighting energy compared to manual switching alone. In offices where light sensor controls were installed and properly commissioned, additional savings up to 27% for a total of 46% were obtained over a seven-month period, even in an area with unusually high minimum lighting requirements. Dimming the lighting system to desired task levels (task tuning) also resulted in significant (23% additional, 43% total) energy savings in overlit areas. On the base case floor, where only bi-level switches were installed, we found significant usage of only one switch resulting in an additional 23% savings over single-level switches-an unexpected result withimplications for building code requirements. We found that the energy savings due to occupant sensing vs. dimming depended on the behavior of occupants. In offices whose occupants tended to stay at their desks all day, dimming controls saved more energy, and vice versa. The lighting requirements of occupants appear to depend on their type of work.

Abstract: The paper presents the technical design concept for a low-cost building communications network that will allow most building equipment loads, including individual light fixtures, operable window blinds, motors and environmental sensors, to be controlled and monitored from an existing Ethernet. IBECS (Integrated Building Environmental Communications System) is a practical networking system that will provide building managers with an unprecedented degree of control over most building electrical loads, allowing them to implement load shedding, real time pricing and aggregated building load strategies with confidence. At the same time, IBECS will let individual occupants exert appropriate control over their environmental conditions (light and, ultimately, heat), initially using a virtual control panel served from their web browser, and eventually through voice command. Recent developments in hardware and software have enabled this new networking concept. New manufacturing techniques can produce semiconductor devices that incorporate a microprocessor, unique IP address, controller, and simple LAN communications for $0.25/chip. The private sector is already hugely invested in software that makes appliances, peripherals and telecommunications devices addressable via the Internet (e.g., Sun Microsystem"s Jini technology - a software infrastructure for providing services on a network). Because it directly exploits recent developments in hardware and software that are far beyond the development capabilities of the building industry, IBECS could provide the infrastructure that, for the first time, truly integrates the operation of building equipment to improve both human and building performance.

Abstract: The United States Postal Service (USPS) has made numerous efforts to improve the lighting quality and efficiency in their facilities. These efforts have included both traditional retrofits such as the transition to T8 lamps/electronic ballasts and more experimental approaches such as light pipes and sulfur lamps. However, these efforts have focused primarily on their industrial and plant facilities and have had little impact on their small and medium sized facilities, which comprise roughly 90% of their total building stock. These efforts have also neglected the affinity between task and ambient lighting functions.

The objective of this project was to develop and demonstrate an integrated lighting system that saves energy while improving the lighting distribution and quality in small and medium sized USPS facilities. Work included the evolution of a novel task lighting fixture designed explicitly to improve the light distribution within the carrier case letter sorting station. The new task light system was developed to work in combination with a high efficiency, low-glare ambient lighting system mounted on the ceiling. The use of high-performance task lighting allowed the ambient lighting component to be reduced, thereby limiting the amount of glare produced and reducing the amount of energy consumed.

Abstract: A new era of technology is emerging in lighting. It is being propelled by the dramatic improvements in performance of solid state light sources. These sources offer an entirely new array of design aspects not achievable with current light sources. At the same time, their performance characteristics continue to improve and are expected to eclipse those of the most common light sources within the near future.

High efficiency is one of these performance attributes motivating the Department of Energy (DOE) to work with the manufacturers of this new technology to create a program plan sufficiently comprehensive to support an industry-driven Solid State Lighting Initiative before Congress. The purpose of the initiative is to educate Congress about the potential of this technology to reduce the electric lighting load within the United States and, consequently, to realize the associated environmental benefits. The initiative will solicit congressional support to accelerate the development of solid state technology through investment in the research and development necessary to overcome the technical barriers that currently limit the products to niche markets.

Abstract: A study was conducted by LBNL/DOE in order to better understand the lighting usage and energy consumption patterns in typical hotel guestrooms. This involved a multi-phase research, development and demonstration program. This program started with the identification and characterization of common lighting technologies in hotel guestrooms and ended in the measurement and monitoring of newly developed and existing technologies at a hotel test site. Emphasis was placed on determining where lighting energy was being used by guests and identifying the savings potentials that energy efficient technologies could present. Initial data indicates that compact fluorescent lamps (CFLs) and occupancy sensors offer significant energy savings potentials in hotel guestrooms.

Abstract: Specific recommendations are made to improve the lighting quality and energy efficiency of the lighting system at the Social Security Administration Frank Hagel Building in Richmond, CA. The main recommendation is to replace the recessed fluorescent lighting system in the general office area with indirect lighting. Indirect lighting will improve lighting quality, will provide an energy efficient solution and will be about the same cost as the direct lighting system originally proposed.

Abstract: A fundamental premise of lighting is that light has value. This is easy to show when the choice is between no light and some light. When the choice is between some light and more light, it is much more difficult to demonstrate. Because of this difficulty, lighting recommendations have been based on consensus judgments of the value of light. There is no formal connection between these judgments and the basic premise upon which they are founded so they are not guaranteed optimal.

In tlis paper, we report on a successful attempt to formally analyze a practical problem: an evaluation of the relative economics of lumen maintenance control system vs the standard no-controls system. We digress briefly to describe a lumen maintenance control system.

Abstract: One of the most difficult barriers in high-lumen, high wattage fiber optic lighting is coupling to light guides efficiently without allowing the high lamp temperature to affect the structural stability of the light guide. Lawrence Berkeley National Laboratory"s Lighting Systems Research Group is currently developing a novel fiber optic coupler that will employ a variety of high-wattage electrodeless lamps. The goal of the project is to use an electrodeless lamp in a coupler whose design is based entirely on the principle of total internal reflection, as demonstrated in Figure 1.

In the coupling system showing in Figure 1, an electrodeless lamp built in the quartz coupler guides the light from the lamp to the acrylic fiber optic light guides. A major concern with this design is the structural stability of the material surrounding the lamp, since the extreme heat of lamp could cause most plastic or glass materials to melt or crack. The typical operating temperatures of electrodeless lamps range from 800-900 deg. C. Many lighting and glass experts recommended quartz as the material of choice fro the coupler because of its resilient thermal properties and ability to transmit visible light efficiently.

Abstract: Preliminary results from a large-scale testbed of advanced lighting control technologies at the Phillip Burton Federal Building at 450 Golden Gate Ave. in San Francisco are presented. The first year objective of this project is to determine the sustainable energy savings and cost-effectiveness of different lighting control technologies compared to a portion of the building where only minimal controls are installed. The paper presents the analyzed results from six months of tests focused on accurately characterizing the energy savings potential of one type of daylight-linked lighting controls compared to the lighting in similar open-planned areas without dimming controls. After analyzing a half year"s data, we determined that the annual energy savings for this type of daylight-linked controls was 41% and 30% for the outer rows of lights on the South and North sides of the building, respectively. The annual energy savings dropped to 22% and 16% for the second row of lights for the South and North, respectively, and was negligible for the third rows of lights.

Abstract: This paper describes the history, technology development, technology transfer and application of the energy efficient compact fluorescent torchiere. A review of the essential efforts that went into the development of the first commercially available CFL torchiere technologies is described. Also included is a review of the performance issues related to lumen matching capabilities. Furthermore, the paper overviews the critical steps and successes that occurred as this technology made the transition from laboratory to marketplace.

The energy efficient torchiere promises to have one of the single largest energy saving potentials of any DSM program developed to date. This project represents unique spectrum of industry-laboratory collaborations and addresses an important national energy and safety problem.

Abstract: The report presents a long-range plan for a broad-based, coordinated research, development and market transformation program for reducing the lighting energy intensities in commercial and residential buildings in California without compromising lighting quality. An effective program to advance lighting energy efficiency in California must be based on an understanding that lighting is a mature field and the lighting industry has developed many specialized products that meet a wide variety of light needs for different building types. Above all else, the lighting field is diverse and there are applications for a wide range of lighting products, systems, and strategies. Given the range of existing lighting solutions, an effective energy efficient lighting research portfolio must be broad-based and diverse to match the diversity of the lighting market itself. The belief that there is one solution - a magic bullet, such as a better lamp, for example - that will propel lighting efficiency across all uses to new heights is, in our opinion, an illusion. A multi-path program is the only effective means to raising lighting efficiency across all lighting applications in all building types.

This report presents a list of 27 lighting technologies and concepts (key activities) that could form the basis of a coordinated research and market transformation plan for significantly reducing lighting energy intensities in California buildings. The total 27 key activities into seven broad classes as follows:

Light sources

Ballasts

Luminaires

Lighting Controls

Lighting Systems in Buildings

Human Factors

Education

Each of the above technology classes is discussed in terms of background, key activities, and the energy savings potential for the state. The report concludes that there are many possibilities for targeted research, development, and market transformation activities across all sectors of the building lighting industry. A concerted investment by the state to foster efficiency improvements in lighting systems in commercial and residential buildings would have a major positive impact on energy use and environmental quality in California.

Abstract: Since the summer of 1995, several organizations have been in pursuit of what many consider the Holy Grail of lighting technology - a low-cost, drop-in, energy-efficient replacement for the incandescent lamp. This paper summarizes the international experience in attempting to catalyze the commercialization of a mass-market replacement product that could have a major impact on residential lighting energy consumption in U.S. and EU homes.

The technology procurement effort was originally spearheaded by the U.S. Federal Government through a loose collaboration between the Department of Defense (DOD), the Environmental Protection Agency (EPA) and the Department of Energy (DOE). The DOD agreed to serve as the anchor buyer for a low-cost, drop-in replacement product for standard-sized light bulbs that provide at least 30% energy savings compared to traditional incandescent lamps. In parallel to the U.S. effort, the International Energy Agency launched a cooperative technology procurement effort by assembling large buyers" groups in Finland, the Netherlands, Sweden, and the United Kingdom to pull a similar efficient lighting product into the European market. The lukewarm response from lamp manufacturers to these two technology procurement efforts illustrates the challenges of transforming residential lighting from incandescent to efficient lighting.

Abstract: California?s Title 24 Energy Efficiency Building Standard requires multiple lighting level control in all individual offices. Usually, this requirement is fulfilled using bi-level switching. With bi-level switching, each office occupant is provided with two wall switches near the doorway to control their lights. In a typical installation, one switch would control 1/3 of the fluorescent lamps in the ceiling lighting system, while the other switch would control the remaining 2/3 of the lamps. This allows four possible light levels: OFF, 1/3, 2/3 and FULL lighting.

Abstract: We present a tone reproduction operator that preserves visibility in high dynamic range scenes. Our method introduces a new histogram adjustment technique, based on the population of local adaptation luminances in a scene. To match subjective viewing experience, the method incorporates models for human contrast sensitivity, glare, spatial acuity and color sensitivity. We compare our results to previous work and present examples of our techniques applied to lighting simulation and electronic photography.

Abstract: A series of novel energy efficient torchiere systems have been developed using compact fluorescent lamps (CFLs). These systems were studied photometrically and compared with the performance of traditional commercially available tungsten halogen sources. Gonio-photometric data and power assessments indicate that significant lighting energy savings can be obtained by utilizing CFL sources instead of standard tungsten halogen sources. This energy savings is jointly due to the higher source efficacy of the CFLs and the surprisingly poor performance of the imported 300 Watt halogen lamps. Experimental data shows that a 50 to 60 Watt CFL will effectively lumen match a variety of 300 Watt tungsten halogen sources with 5 to 10 times the efficacy. CFL torchieres have additional benefits of higher power quality and cooler lamp operating temperature, making them safer fixtures.

Abstract: This paper focuses on performance variations associated with lamp geometry and distribution in portable table luminaires. If correctly retrofit with compact fluorescent lamps (CFLs), these high use fixtures produce significant energy savings, but if misused, these products could instead generate consumer dissatisfaction with CFLs. It is our assertion that the lumen distribution of the light source within the luminaires plays a critical role in total light output, fixture efficiency and efficacy, and, perhaps most importantly, perceived brightness. We studied nearly 30 different integral (screw-based) CFLs available on the market today in search of a lamp, or group of lamps, which work best in portable table luminaires. Our findings conclusively indicate that horizontally oriented CFLs outperform all other types of CFLs in nearly every aspect.

Abstract: High-lumen light sources represent unique challenges and opportunities for the design of practical and efficient interior lighting systems. High-output sources require a means of large-scale distribution and avoidance of high-luminance glare while providing efficient delivery. An indirect lighting system has been developed for use with a 1000 Watt sulfur lamp that efficiently utilizes the high-output source to provide quality interior lighting. This paper briefly describes the design and initial testing of this new system.

Abstract: Extensive development of designs and prototyping of energy efficient torchiere systems using compact fluorescent lamps (CFLs) has lead directly to the production of CFL torchieres by a major US manufacturer. This paper compares the electrical and photometric characteristics of one of the new CFL torchieres to standard tungsten halogen torchieres (halogen uplighters). Power assessments and gonio-photometric data indicate that the new CFL torchiere provides significant energy savings over the standard tungsten halogen torchiere while producing more luminous flux. The energy savings is jointly due to the high source efficacy of the CFLs and the poor performance of the imported halogen lamps. The paper also presents results from a test site in the student dormitories at Stanford University where a torchiere lamp swap program was initiated in which students voluntarily traded their halogen torchieres for CFL torchieres. Out of the 500 torchieres involved in the lamp swap, a random sample of nearly 100 halogen lamps (seasoned in the field and considered to represent a typical population) were collected and photometrically and electrically characterized in the laboratory. These laboratory results indicate that the CFL torchieres use 65 watts to produce 25 percent more light than the 300 W tungsten halogen torchieres they are designed to replace. Additionally, the CFL torchieres have the benefit of a cooler lamp operating temperature, making them safer luminaires.

Abstract: Lighting controls have the potential to capture significant energy savings in commercial buildings if properly specified, installed, commissioned and maintained. Proper commissioning is often absent in lighting projects and the lack of commissioning can significantly reduce a project"s energy savings potential. The paper focuses on the importance of good commissioning practice for obtaining satisfactory performance from lighting control systems and discusses the difficulties of commissioning today"s systems. Some practical suggestions for commissioning lighting controls is offered as well as advice to specifiers and early adopters to aid them in selecting controls that are most appropriate to their requirements.

Abstract: The extraordinary market growth of the high power halogen torchiere (halogen uplighter) presents significant global energy savings opportunities for energy efficient alternatives. Extensive development of prototype designs of energy efficient torchiere systems using compact fluorescent lamps (CFLs) has lead directly to the production and commercialization of CFL torchieres. This paper analyzes the current global market for torchieres and compares the electrical and photometric characteristics of one of the new CFL torchieres to standard tungsten halogen torchieres. Power assessments and photometric data indicate that the new CFL torchiere provides significant energy savings over the standard tungsten halogen torchiere while producing more luminous flux. The energy savings is jointly due to the high source efficacy of the CFLs and the poor performance of many cheaply made halogen lamps. Laboratory and in-situ experiments indicate that the CFL torchieres use 65 Watts to produce 25 percent more light than the 300 W tungsten halogen torchieres they are designed to replace. Additionally, the CFL torchieres have the benefit of a cooler lamp operating temperature, making them safer luminaires (Brooks, 1997; New York Times, 1997). This safety benefit, coupled with the potential for significant reductions in global greenhouse gas emissions, has prompted the insurance industry to form a unique alliance with energy conservation groups to promote energy efficient torchieres.

Abstract: Previously we have determined, with a full field of view, the relative contributions of scotopic and photopic luminance to pupil size at light levels typical of building interiors. Those studies were carried out in a white room with uniform reflectance, and with the viewed surfaces having an approximately uniform luminance distribution. To enhance the usefulness of the past results to lighting practice, we have constructed a simulated office where the viewed walls can have one of four very different colors, with quite different luminance distributions. This allows examination of interaction effects between wall spectral reflectivity and light source spectral distribution.

In the present study pupil sizes were obtained while subjects were viewing a very small screen television. A remote pupillometer was used, allowing subjects to sit in a comfortable chair without the inconvenience of chin rests or head gear. Seventeen subjects between the ages of 27 and 47 years were studied using illumination provided by conventional lamps, either a WW or a daylight fluorescent lamp. Pupil size variation was predicted by the value of the scotopic vertical illuminance at the eye. Even though the WW lamps are 50% more efficacious than daylight lamps in terms of photopic lumens per watt, daylight lamps can be as much as twice as efficacious in eliciting pupil size.

Abstract: This report provides background data upon which residential lighting fixture energy conservation programs can be built. The current stock of residential lighting is described by usage level, lamp wattage, fixture type, and location within the house. Data are discussed that indicate that 25% of residential fixtures are responsible for 80% of residential lighting energy use, and that justify targeting these fixtures as candidates for retrofit with energy-efficient fixtures. Fixtures determined to have the highest energy use are hardwired ceiling fixtures in kitchens, living/family rooms, dining rooms, and outdoors. An assessment of the market for residential fixtures shows that nearly half of new residential fixtures are imported, 61% of new fixtures sold are hardwired, and about half of all new fixtures sold are for ceiling installation.

Abstract: With U.S. sales approaching 50 million lamps per year, compact fluorescent systems are seeing increased application in the lighting of commercial interiors primarily as a conservation measure and an approach to reduce maintenance costs because they are longer-lived than incandescent sources. Lamp and fixture manufacturers, utilities, government agencies, and environmental groups have created programs designed to accelerate the adoption of compact fluorescent lamp (CFL) technology for commercial applications. However, efforts to accelerate the penetration of CFLs in the residential market have been narrowly limited to utility rebate programs for screw-based compact fluorescent technology. Concerns relating to performance, consumer acceptance and economics, suggest that screw-based (as opposed to pinbased) technologies are at best an impermanent and short-term approach to the ultimate, widespread residential application of the compact fluorescent lamp. The proper implementation of CFLs deserves attention not only because the technology saves significant amounts of energy, but also because CFLs have become a pervasive symbol for energy-efficiency.

This paper reviews some background on residential CFL applications, proposes a national program to accelerate the adoption of efficient CFL technologies in the home, and introduces strategies to bring about their success. This program is based on the premise that dedicated fixtures designed for the explicit operation of the CFL offer the most effective and permanent solution for efficient residential lighting.

Using pin-based lamps in dedicated fixtures allows for easy re-lamping with maximum economic benefits. It also ensures and maintains the full realization of the energy saving potential and lighting quality when displacing incandescent loads. Equally important, dedicated fixtures can be designed to optimize the performance of the CFL in comparison to fixtures intended for screw-based technologies. With these prospective benefits in mind a national program aimed at accelerating the use of CFL systems in residential applications should focus its efforts to promote dedicated compact fluorescent fixtures. Thus, the program should be a long-term effort addressing multiple issues in different market segmentsswith the cooperation of various partners, including manufacturers, utilities, designers and consumer groups.

Dedicated CFL fixtures can find great success in the residential sector in much the same way as pinbased long fluorescent tubes displaced incandescents in the commercial sector in the 1940s, and pin-based halogen lamps swept into popularity in homes in the 1980s.

Abstract: This paper describes the results of a photometric and energy analysis that was conducted on a new light guide and sulfur lamp system recently installed at the U.S. Department of Energy"s Forrestal Building. This novel system couples two high lumen output, high efficiency sulfur lamps to a single 73 m (240 ft.) hollow light guide lined with a reflective prismatic film. The system lights a large roadway and plaza area that lies beneath a section of the building. It has been designed to completely replace the grid of 280 mercury vapor lamps formerly used to light the space. This paper details the results of a field study that characterizes the significant energy savings and increased illumination levels that have been achieved. Comparisons to modeled HID lighting scenarios are also included.

Abstract: We have previously demonstrated that brightness perception in full field is influenced by scotopic luminance, even at light levels in the photopic range. We asked here if glare discomfort is distinguishable when comparing a scotopically enhanced source with a scotopically deficient source at the same photopic luminance. Similarly to our previous study on discomfort glare, we used both objective and subjective techniques to assess glare response.

Discomfort glare responses were assessed for 12 subjects who viewed two broad-band glare sources (illurninants) of size 1.22 x 0.91 m (4 x 3 ft) with maximum photopic luminances of 3,700 cd/m2. The two sources were approximately matched for photopic luminance but due to their spectra they had markedly different scotopic luminances. Both sources were presented separately at three different photopic luminance levels. These six glare conditions were each presented five times, for four second periods, in a randomized sequence. Electromyographic (EMG) responses from the facial orbicularis oculi muscles were subjected to Fourier analysis and integration of the power spectrum provided a measure of EMG activity. Our objective index of the response to glare was the ratio between EMG samples taken before and during the presentation of the source. For the subjective method, discomfort severity was indicated by subjects marking a Visual Analog Scale (VAS) punctuated with four descriptors - perceptible, annoying, disturbing and intolerable.

Both objective and subjective indices systematically increased with increasing luminance. However, the objective measure showed a significantly higher value for the scotopically deficient source at all luminances, while the subjective response was higher only at the greater luminances.

We conclude that discomfort glare is related to both the photopic luminance of the source, and its spectral composition with the absence of long wave length energy in the spectrum associated with lower levels of discomfort.

Abstract: The present study extends our prior visual performance studies to a complex resolution task which is representative of tasks in typical workplace environments: word reading presented at a fixed high contrast (black print on white background), but with varying sized letters.

We examined the effect of pupil size on the letter size-acuity function using accuracy of word recognition as the endpoint. Word reading acuity has been extensively used in vision research as a measure of visual performance and has been shown to correlate well with face recognition and other complex recognition tasks. In this study, the task was shielded from the surround lighting, allowing the luminances of the surround and task to be controlled independently. Two pupil size conditions were compared, where pupil size is controlled by high or low luminance levels of a single surround illurninant. We chose to use a single illuminant to control pupil size to avoid changes in induced color which occur when pupil size is changed by varying the surround spectrum.

The results here for nine subjects, ages 23 to 59, years replicate and extend our prior visual acuity studies using Landolt C tasks, and show again that smaller pupils improve visualsperformance even though task retinal illuminance is substantially reduced. We also found that improvement in visual performance with smaller pupils occurs despite an increased disability glare under the high luminance surround condition. Our results are directly applicable to self-illuminated tasks (e.g., computer terminals) operating with black print on a white background.

Abstract: When pupil size is changed by varying the surround spectrum, there is a perceived color shift of the task towards the complementary hue of the surround. This occurs even though none of the surround light falls on the task, and the task illumination is unchanged. This induced color effect is a neural process. To investigate whether such a mechanism is an alternative explanation of our results on the effects on visual performance of spectrally controlled pupil sizes, we studied visual performance both with and without mydriasis (pharmacologically dilated and fixed pupil). If the induced color hypothesis is valid, then it should occur with both fixed and light-responsive pupils. In addition, we studied whether the pupil size effect on visual performance can occur in accurately refracted subjects, or if it is enhanced by the addition of a small amount of optical blur ( 0.50 DS). We studied 12 subjects, 21 to 35 years of age, correctly refracted and with added blur, under each of two conditions: normal pupils and mydriasis. We compared Landolt C recognition, with a fixed task luminance but variable contrast, for two different surround spectra, each at 50 cd/m2. The two different surround spectra controlled subject"s pupil size. For normal pupils, performance was better with smaller pupils, and the improvement in performance due to switching to a scotopically enhanced surround was greater with added blur, even though blursreduced overall performance. Under mydriasis, change of the surround spectrum had no effect on performance, whether there was blur or not. However, the added blur reduced performance under the mydriasis condition, showing our measures are sensitive to these parameters. These experiments rule out the induced color hypothesis and demonstrate the benefits of smaller pupils on Landolt C contrast sensitivity even when subjects are correctly refracted. Further, the results indicate that the measured improvement due to pupil size change is greater when there is imperfect refraction (blur).

Abstract: The residential lighting sector represents a significant opportunity for energy conservation due to the almost exclusive use of inefficient incandescent sources. Compact fluorescent lamps (CFLs) have the potential to transform this market by using one fourth as much power as an incandescent to provide the same amount of light. While technical advances such as triphosphors and electronic ballasts have addressed issues of color rendition, flicker, and hum, CFLs still face significant market barriers, particularly their perceived brightness level in traditional fixture applications. When operated in fixtures originally designed for A-lamps, CFLs with equal total lumen packages can appear dimmer due to differences in their light distributions. One such fixture, the common table lamp, is typically operated for more than 3 hours a day, and thus represents a significant opportunity for energy savings. LBL conducted a series of goniophotometric candela distribution studies of table lamps with the initial objective of matching with CFLs the light distribution of the consumer accepted A-lamp. While goniometric testing was done on numerous CFL and incandescent sources, this paper focuses on three typical sources which have very different distributions. Our photometric studies indicate that horizontally oriented CFLs may produce a more desirable distribution than either A-lamps or vertically oriented CFLs by minimizing shade losses and thus maximizing the amount of useful light leaving the fixture. Optimizing fixture geometry and lamp position can significantly increase the efficiency of these CFL fixtures. Ongoing research with the fixture industry seeks to identify and develop efficient source/fixture configurations.

Abstract: Multi-layer polarizing panels are a technology that has been proposed for improving lighting quality in interior environments. Different authors have examined a variety of different conditions, have used different performance models in their evaluation of the potential of this technology, and have come to markedly different conclusions. This paper discusses how variations in conditions and performance models affect the evaluation of the performance potential of polarized lighting systems. The importance of the light source-task-eye geometry, which determines the benefit that can be achieved with polarizing luminaires, is discussed in detail.

Abstract: This paper describes the results of a photometric and energy analysis that was conducted on a new light guide and sulfur lamp system recently installed at both the U.S. Department of Energy"s Forrestal building and the Smithsonian Institution"s National Air and Space Museum. This system couples high lumen output, high efficiency sulfur lamps to hollow light guides lines with a reflective prismatic film. At the Forrestal building the system lights a large roadway and plaza area that lies beneath a section of the building. It has been designed to completely replace the grid of 280 mercury vapor lamps formerly used to illuminate the space. At the National Air and Space Museum a similar system illuminates Gallery 114, which houses the large rocket displays from the U.S. Space program. This paper outlines the unique operational and design characteristics of this hightly efficient distribution system and details the results of field studies that characterize the significant energy savings and increased illumination levels that have been achieved. The projected savings in maintenance costs, due to longer lamp life and a reduction of the total number of lamps, is also presented.

Abstract: The VL concept is based on measurements of accuracy thresholds where response times are essentially unlimited. Rea has produced a model based on reaction times, which include response times. Analysis of phenomena such as the Pulfrich illusion suggest that there is a delay time between signal and recognition that depends upon the adaptation luminance. This suggests an analysis of reaction times in terms of a simple sum of a signal strength (VL) term plus response term(s). A comparison of Rea"s reaction time and this additive VL model showed that the models are similar, but that the VL model fits the data better and has a more consistent theoretical foundation.

The luminance-dependent response term appears to be needed to fit performance on the numerical verification task developed by Rea, but not the word chart reading task developed by Bailey. The above additive model suggests that the difference may depend on whether the task is processed serially and includes the delay time, or is processed in parallel and does not.

Abstract: This paper explores numerical techniques for comparing real and synthetic luminance images. We introduce components of a perceptually based metric using ideas from the image compression literature. We apply a series of metrics to a set of real and synthetic images, and discuss their performance. FinaIly, we conclude with suggestions for future work in formulating image metrics and incorporating them into new image synthesis methods.

Abstract: Global illumination researchers tend to think in terms of mesh density and sampling frequency, and their software reflects this in its user interface. Advanced rendering systems are rife with long command lines and parameters for tuning the sample densities, thresholds and other algorithm-specific variables, and the novice user is quickly lost in a sea of possibilities. This paper details a successful effort of making one such global illumination system usable by people who understand their problems, even if they do not understand the methods needed to solve them, through an assisted oracle approach. A single program is introduced to map a small set of intuitive control variables to the rendering commands and parameter settings needed to produce the desired output in a reasonable time. This new executive program then serves as the basis for a graphical user interface that is both friendly in its appearance and reliable in its performance. Finally, we conclude with some future directions for improving this interface.

Abstract: For several generations lighting practitioners have had suspicions that calibrated light meters and photometers do not accurately reflect their perceptions of lit environments even for light which is whiteish in color. The troublesome perceptions arise when comparing lightings of different spectral quality and where the metered photometrics are equal but contrarily where brightness sensation and visual sensitivity are perceived to be different. An observer can readily experience these perceptions of different lightings by comparing the same environment under equally lit photopic conditions by low color temperature lamps (3000 K) versus high color temperature lamps (6000 K) or incandescent lighting versus natural lighting or compare high pressure sodium lighting with metal halide lighting. Vision scientists have tried to resolve this problem by introducing small corrections into the calibration function which are not included in its classic determination based on flicker photometry. These refinements have not resolved the differences and the problem continues to plague lighting practice.

Abstract: Most discharge lamps, including fluorescent lamps, metal halide lamps, and high pressure sodium lamps, contain Mercury, a toxic chemical. Lighting professionals need to be able to respond to questions about the direct hazards of Mercury from accidentally breaking lamps, and the potential environmental hazards of lamp operation and disposal.

We calculated the exposures that could occur from an accidental breakage of lamps. Acute poisoning appears almost impossible. Under some circumstances a sealed environment, such as a space station, could be contaminated enough to make it unhealthy for long-term occupation.

Mercury becomes a potential environmental hazard after it becomes methylated. Mercury is methylated in aquatic environments, where it may accumulate in fish, eventually rendering them toxic to people and other animals. Lighting causes Mercury to enter the environment directly from lamp disposal, and indirectly from power plant emissions. The environmental tradeoffs between incandescent and discharge lamps depend upon the amounts released by these two sources, their local concentrations, and their probabilities of being methylated. Indirect environmental effects of lighting also include the release of other heavy metals (Cadmium, Lead and Arsenic), and other air pollutants and carbon dioxide that are emitted by fossil fuel power plants. For a given light output, the level of power plant emissions depends upon the efficacy of the light source, and is thus much larger for incandescent lamps than for fluorescent or discharge lamps. As disposal and control technologies change the relative direct and indirect emissions from discharge and incandescent lamps will change.

Abstract: Global illumination methods allow us to calculate physical values such as radiance and luminance at each pixel, but they do not tell us how to display these values. If we attempt to fully reproduce the world luminance levels on a computer monitor, slide or print, the results are usually disappointing, for a couple of reasons. First, contemporary display media fall short of the real world"s dynamic range by a few orders of magnitude. It isn"t possible to reproduce the luminances in bright and dark regions accurately. Second, the adaptation luminance, or light level to which the viewer is accustomed, is almost always different under the real world and computer display viewing conditions, and this affects visual response dramatically.

Abstract: Although it is relatively easy to perceive and report the sensation of discomfort caused by the presence of an offending light source of high luminance, no one has yet found a reliable objective correlate correlate of discomfort glare.

In order to find an objectively measured correlate of discomfort glare, we have examined electrical activity associated with the two major facial muscles that surround the eye, viz. the orbicularis oculi and the corrugator supercilii. We have made electromyographic (EMG) recordings using small silver/silver chloride electrodes applied to the skin above the muscles and measured electrical potentials while lighting glare conditions have been changed. Intensities were varied over a range of glare luminance determined by a separate procedure according to subjective ratings. For this subjective method, subjects indicated the level of discomfort by marking a Visual Analogue Scale (VAS) punctuated with four descriptions ofsdiscomfort level: perceptible, annoying, disturbing, and intolerable. We have determined that the VAS is much more reliable with much less variability than thespreviously used border Between Comfort and Discomfort (BCD) method.

For 19 subjects, discomfort glare was assessed under three conditions: 2 deg. diameter glare source with low room illumination, 2 deg. glare source with medium room illumination, and I deg. glare source with medium room illumination. The glare source was a projector beam, 11 deg. to the right of a fixation target on a video monitor. Six different glare luminance levels were presented for 2 second periods. Each glare level was presented six times in a randomized order giving 36 presentations.

EMG responses were subjected to Fourier analysis and the power frequency spectrum was determined with appropriate digital filtering used to eliminate power line artifacts. Blinking causes an artifact whose power spectrum is markedly different and can be determined independently of the glare source. The integrated power spectrum of the EMG activity during exposure to the glare source was compared to the same integral prior to exposure to obtain a quantitative measure of glare induced activity. For each of the 19 subjects and a variety of glare conditions, the objective measure and the VAS have been plotted as a function of glare luminance. For individual subjects we found increasing objective measure and increasing subjective discomfort with increased glare luminance. We conclude that the EMG technique is a vlaid objective means of assessing discomfort glare.

Abstract: Many residential applications of the integral ballast, screw base compact fluorescent lamp (CFL) require a base-down orientation, as exemplified by table and floor lamps. Positioning CFLs base-down results in a 10-25% loss in lumen output, due to excess mercury vapor pressure in the lamp. These losses may limit consumers" acceptance of these new CFLs as a viable retrofit strategy. To mitigate output losses, researchers are developing thermal bridge systems that cool the small tubulation of the lamp, thereby optimizing the mercury vapor pressure. Thermal bridge systems consist of small copper heat sinks that fit entirely within the integral ballast compartment. Experimental data indicates most of the thermally based losses are recoverable, allowing for a lamp system that produces approximately the same lumen output regardless of orientation. As base-down positioning has wide residential application, the improved lumen output of these lamps would have a significant impact on residential retrofits.

Abstract: Monte Carlo techniques for image synthesis are simple and powerful, but they are prone to noise from inadequate sampling. This paper describes a class of non-linear filters that remove sampling noise in synthetic images without removing salient features. This is achieved by spreading real input sample values into the output image via variable-width filter kernels, rather than gathering samples into each output pixel via a constant-width kernel. The technique is non-linear because kernel widths are based on sample magnitudes, and this local redistribution of values cannot generally be mapped to a linear function. Nevertheless, the technique preserves energy because the kernels are normalized, and all input samples have the same average influence on the output. To demonstrate its effectiveness, the new filtering method is applied to two rendering techniques. The first is a Monte Carlo path tracing technique with the conflicting goals of keeping pixel variance below a specified limit and finishing in a finite amount of time; this application shows how the filter may be used to clean up areas where it is not practical to sample adequately. The second is a hybrid deterministic and Monte Carlo ray-tracing program; this application shows how the filter can be effective even when the pixel variance is not known.

Abstract: This paper describes a physically-based rendering system tailored to the demands of lighting design and architecture. The simulation uses a light-backwards ray-tracing method with extensions to efficiently solve the rendering equation under most conditions. This includes specular, diffuse and directional diffuse reflection and transmission in any combination to any level in any environment, including complicated, curved geometries. The simulation blends deterministic and stochastic ray-tracing techniques to achieve the best balance between speed and accuracy in its local and global illumination methods. Some of the more interesting techniques are outlined, with references to more detailed descriptions elsewhere. Finally, examples are given of successful applications of this free software by others.

Abstract: At light levels typical of building interiors it is the optical quality of the eye rather than retinal illumination that limits visual acuity and contrast sensitivity. Because aberrations generally occur in the eyes of the majority of the population and because smaller pupils reduce the defocusing effects of aberrations, pupil size becomes the major factor that light spectrum can control to influence visual performance. Our studies demonstrate that for conditions of full field of view and light levels typical of interior spaces, pupil size as measured by objective infrared pupilometry, is predominantly controlled by scotopic spectrum. We also demonstrate in a series of 10 separate studies with different subjects that pupil size is the ultimate factor in limiting acuity and contrast sensitivity at typical interior light levels. A large number of subjects are employed in these studies which take place under conditions of binocular viewing in a 2 m cubic room lit by a variety of fluorescent lamps.

Abstract: Conventional photometry is based on the psychophysical response of observers with the field of view confined to 2 deg. That region of the human retina contains only cone photoreceptors and does not contain rod photoreceptors. However, the majority of the retina is dominated by rods. Perhaps because the measurement of rod spectral response requires extremely low light levels in order to remain below cone threshold, it has been erroneously assumed that rods are inactive at normal interior light levels. Thus, any effects of rod receptors have not been included in the measurement of light quantity that is applied to lighting practice. Our studies demonstrate that rods are indeed active at typical interior light levels and that under conditions of full field of view they dominate the spectral determinants of pupil size, as well as, contributing significantly to brightness perception. Rod intrusion is well known to vision scientists, but nevertheless many useful past studies of lighting effects have fallen by the wayside because the obvious invoking of rod contributions have not been part of interior lighting science. We present here evidence that indicates that the time has come for the C.I.E. to recognize that lighting practice requires an enhanced photometry which incorporates realistic viewing conditions and the resultant scotopic sensitivity.

Abstract: This paper describes a novel light distribution concept that uses a centralized distribution system, which is optically connected (with light guides) to a series of integrated work stations. The system uses a single HID source with a series of light guides over each task plane. This centralized lighting/work station maintains illuminance and contrast independent of occlusion effects from furniture and body shadow, while increasing the efficiency of light delivery to the task plane. Using a single HID to illuminate multiple task planes can reduce the number of lamps required leading to reduced mercury recovery costs associated with fluorescent lighting systems as well as reduced energy consumption. This concept also has the potential to reduce the costs of both materials and maintenance.

Abstract: Subjects judged the orientation of a 2 min. gap Landolt C located at a distance of 2.4 m. The stimuli were presented in central vision on a CRT, at low to medium contrast. The effects of varying the spectrum and luminance of surround lighting were assessed on both pupil size (measured using infrared pupillometry during task performance) and task accuracy. The task display was protected from the surround lighting, so that its luminance and contrast could be varied independently of the changes in the surround lighting. Indirect surround illumination was provided by either two illuminants of very different scotopic spectral content but with the same photopic luminance (Experiments 1 and 3), or by using the same illuminant at two different luminance levels (Experiment 2). In Experiment 3, the effect of changing surround spectrum was coinpared to the effect of varying task background luminance between 12 cd/m2 and 73 cd/m2. In all experiments, scotopically enhanced surround lighting produced pupil areas which were reduced by almost 50% in comparison with surround lighting with relatively less scotopic luminance. Concomitantly there was improvement in Landolt C task performance with the scotopically enhanced surround lighting at all contrast and luminance levels. In these experiments, smaller pupil sizes were associated with significantly better visual-task performance in spite of lower task retinal illuminance when compared to the condition with larger pupils. These results suggest that changes in surround spectrum can compensate for the effect on task performance of a reduction in task luminance and supports the hypothesis that lighting energy savings could accrue in the workplace by shifting lamp spectra to obtain greater scotopic efficacy.

Abstract: In a previous study with young adults ages 20 to 40 years, we demonstrated that low contrast Landolt C recognition was better with smaller pupils than with larger pupils even though task retinal illuminance was higher for larger pupils. Pupil size in these studies was controlled by the spectrum of the far surround illumination which was prevented from affecting task luminance.

The present study used the same procedures as our previous study with 7 healthy elderly adult subjects between the ages of 61 and 66. Because senile miosis is a characteristic of the aging eye, spectral changes in the surround might be expected to produce relatively smaller changes in pupil size and hence performance. Surprisingly, although the scotopically enhanced surround lighting produced relatively smaller pupil size changes than in young adults, the performance enhancements were comparable to those of young adults.

As in the previous study the task was recognition of the orientation of the gap in the C that was presented on a CRT with contrasts varying from 18% to 80%. Two surround illuminants were compared, both provided a luminance of 53 cd/m2 on the front wall at visual angles larger than 30. Subjects had at least 20/20 vision and perform the task with their spectacles if normally used.

Abstract: Recessed downlights are one of the most popular fixtures used for lighting commercial buildings. The vast majority of these fixture use incandescent lamps" in sizes ranging from 50 to 150W. Because cost and energy efficiency are a growing concern in the lighting design process, fixtures using compact fluorescent lamps" (CFLs) have made significant inroads into the incandescent market. While providing high efficiency, these fixtures also reduce operating cost. CFLs have an efficacy about four times that of incandescent lamps; they also last about 10 times longer, thus providing maintenance savings.

Abstract: This paper investigates the application of nonimaging concentrators to the design of reflectors for luminaires. An interpretation of the concentration ratio -- a statement of the conservation of flux -- relative to the properties of a source and reflector is given. The result is used to develop practical compound parabolic (CP) reflector geometries that accommodate modern lamps. For the cylindrical and compact fluorescent lamps, we use the concentration ratio to show how the size and output performance of the CP reflector can be improved relative to the luminous and geometric properties of the lamp. The paper concludes by considering the addition of a nonimaging louver as a potentially significant designsstep once the reflector has been suitably designed relative to the lamp. It is noted that accurate data on the luminous emitting properties of lamps is a prerequisite to the development of optimum compact reflector designs.

Abstract: An experimental protocol and apparatus was developed to assess the relative differences in dirt depreciation between vented and unvented compact fluorescent recessed downlights under simulated conditions. A simulated plenum/ceiling chamber is designed to expose both vented and unvented fixtures simultaneously to a controlled dust environment over an extended period of time. Experimental data shows that the unvented fixture depreciated faster over time due to dust exposure than the vented fixture.

Abstract: In most compact fluorescent recessed downlights, hot air stratifies within the fixture, causing the lamps to overheat and lose up to 25% of light output and efficacy. Thermal management techniques, including passive venting of the fixture, are being developed to mitigate these losses in fixture efficiency. This paper demonstrates a sequence of venting configurations and techniques, from an unmodified compact fluorescent downlight as a control fixture, through a series of venting strategies, to the development of a highly efficientsdownlight that incorporates a new angular concept in convective venting. With this new venting design, an increase in light output of nearly 25% can be attained without the optical losses generally associated with some venting geometries.

Abstract: A Variational theory of the radiant emittance of the Mercury-Argon discharge is developed and applied to an investigation of the effects of alterations of the isotopic composition of the Mercury. The theory includes the effects of transport of resonance radiation, elastic and inelastic collisions, resonant exchange of excitation, diffusion of the Mercury atoms, the isotope- and hyperfine-shifted structure of the resonance line, and the surfaces of the discharge on the emittance. Two sensitive parameters - the branching ratio for radiative decay of the resonance state and the rate constant for resonant exchange of excitation - are fitted to data on natural Mercury. The remaining insensitive parameters are given values based on microscopic estimates. The resulting theory reproduces the experimentally observed results for Mercury - 196 enhanced mixtures. We predict small effects for mixtures with othersisotopes enhanced. The crucial role played by resonant exchange of excitation is emphasized.

Abstract: A new device for measuring the spatial reflectance distributions of surfaces is introduced, along with a new mathematical model of anisotropic reflectance. The reflectance model is both simple and accurate, permitting very efficient reflectance data reduction and reproduction. The validity of the model is substantiated with comparisons to complete measurements of surface reflectance functions gathered with a new reflectometry device. This device uses imaging technology to capture the entire hemisphere of reflected directions instantly, which greatly accelerates the reflectance data gathering process and makes it possible to measure dozens of surfaces in the time that it used to take to do one. Example measurements and simulations are shown, and a table of fitted parameters for several surfaces is presented.

Abstract: The speed of reading unrelated words as a function of luminance, size, and contrast, was measured with an eye movement monitor for fifteen young adults. Subjects read up to 5,000 words in a test session, with the exact number depending upon their acuity. The size of the smallest legible print at a given luminance and contrast for these subjects was found to fit well to the Blackwell-Taylor detection threshold data above about 1 minute of arc. At lower sizes inclusion of a resolution size term provided an excellent fit.

Reading speed was fit to a number of visual performance models. It was found that for most subjects that a ratio of the print size to an estimate of the threshold print size (a VLsize) gave the best fits to the data. The threshold size was computed with a fit to the Blackwell-Taylor detection threshold data, modified to include a resolution size term as above. For the sole remaining subject a slightly better fit was obtained with a VLcontrast model, where again the thresholds were modified by a limiting size term. The implication of these results for visual performance modeling is discussed.

The reading speed for all subjects varied rapidly with size near the acuity limit, but became almost independent of visibility parameters as long as size is two times the acuity limit. These results show that size is a powerful determinant of reading speed, and suggest that minification of about 1/2 power could be used as a field test for adequate visibility.

Abstract: The Illuminating Engineering Society of North America develops light level recommendations for tasks where visual performance is important. The 1959 and 1972 recommendations for illumination levels were based on the principle of delivering a fixed level of performance as predicted by the visual performance models of the time. This same principle is being considered for future revisions to the recommendations. There is currently no explicit method for determining whether a given fixed performance level is in any sense optimal or best.

Visual performance increases with lighting levels, but so do economic and environmental costs. These costs lessen the economic benefits of the improved visual performance. A formal method for including these factors in light level recommendations is to restate the problem in terms of net benefits (benefits minus costs). The resulting equations have well defined optima versus light level, and thus give an explicit estimate of what the best lighting levels are in terms of current visual performance models, and current economic conditions.

A simple net-benefit procedure is described, and sample calculations are shown for two current visual performance models. Fixed performance levels do not provide econornically optimal recommendations with either model. There are also differences between models, but they are less significant than the large differences between the principles of fixed performance levels and economic optimization.

Abstract: A new method for improving the accuracy of a diffuse interreflection calculation is introduced in a ray tracing context. The information from a hemispherical sampling of the luminous environment is interpreted in a new way to predict the change in irradiance as a function of position and surface orientation. The additional computation involved is modest and the benefit is substantial. An improved interpolation of irradiance resulting from the gradient calculation produces smoother, more accurate renderings. This result is achieved through better utilization of ray samples rather than additional samples or alternate sampling strategies. Thus, the technique is applicable to a variety of global illumination algorithms that use hemicubes or Monte Carlo sampling techniques.

Abstract: A potentially energy saving heat sink device for use in common fluorescent luminaires is presented. This heat sink, or Spot Cooler, increases luminaire efficiency by eliminating losses of light output and system efficacy that normally occur when lamps are heated beyond optimal temperatures in the thermally constrictive lamp compartment typical of most recessed luminaires. Unlike cooling devices suggested in the past, the Spot Cooler presented is inexpensive and completely passive, yet maintains an optimum performance level by recovering light losses that can vary from 10 to 25%. While the prototypes presented were designed for and tested with F40T12 type lamps utilized in standard recessed troffers, the design could be modified for any cylindrical fluorescent lamp type utilized in many standard fixture geometries.

Abstract: An advanced electronically ballasted lighting control system was installed in a portion of an office building to measure the energy and demand savings.

The lighting control system used an integrated lighting control scenario that included daylight following,slumen depreciation correction, and scheduling. The system reduced lighting energy on weekdays by 62%sand 51% in the north and south daylit zones, respectively, compared to a reference zone that did not havescontrols. During the summer, over 75% energy savings were achieved on weekdays in the north daylit zone.sEven in the south interior zone, which benefited little from daylight, correction strategies (scheduling andslumen depreciation) and adjustment of the aisleway lights to a low level resulted in energy use of only halfsthat of the reference zone. Although, in general, the savings varied over the year due to changing daylightsconditions, the energy reduction achieved with controls could be fit using a simple analytical models(Figure ES-1).

Significant savings also occurred during core operating hours (6 A.M. to 6 P.M.) when it is more expensivesto supply and use energy. Compared to the usage in the reference zone, energy reductions of 49%, 44%, and 62% were measured in the south daylit, south interior, and north daylit zones, respectively, during coresoperating hours throughout the year.

Lighting energy usage on weekends decreased dramatically in the zones with controls, with the usage insthe north daylit zone only 10% that of the reference zone.

A simple survey developed to assess occupant response to the lighting control system showed that thesoccupants were satisfied with the light levels provided.

During the load-shedding test, the lighting control system was used to lower light levels by about 15% forsshort intervals in the afternoons in simulated response to a peak demand situation. The dimming controlsswere effective in implementing load-shedding control; occupant response to the strategy was neutral.

A novel two-part control photosensor was effective in providing appropriate control; it allowed daylightingsand lumen maintenance strategies to be implemented simultaneously with the same hardware. With thissphotocontrol system, design light levels were maintained at the workplane regardless of the daylightscontribution or the age of the lighting system.

Abstract: We present a simple technique for improving the efficiency of ray tracing in scenes with a large number of light sources. The sources are sorted according to their potential contribution, and only those sources whose shadows are above a specified threshold are tested. The remainder are added into the result in proportion to a statistical estimate of their visibility. The algorithm requires very little storage, and produces no objectionable artifacts.

Abstract: A spreadsheet program for determining system efficacy, power input and light output of common 4 ft fluorescent lighting systems under realistic operating conditions is described. The program uses accepted IES engineering principles to precisely account for ballast factor, existing thermal conditions and maintenance practices. The spreadsheet, which includes a data base of lamp and ballast performance data, can be used to calculate the cost-effectiveness of many common lighting retrofits.

Abstract: Many compact fluorescent fixture systems present a highly constricted thermal environment to the lamp. This results in elevated lamp wall temperatures causing reduced light output and efficiency of the fixture system. Dissipating heat from the interior of the fixture to the surrounding plenum can increase the fixture"s light output and efficacy by more than 15%, allowing the light output and efficacy to reach 98%-99% of maximum. One way of achieving this thermal dissipation is through the convective venting of the fixture; this inhibits thermal stratification within the fixture housing and permits upward convective flows to cool the lamp. In an enclosed fixture, minimum lamp wall temperature (MLWT) can exceed 50?C; the proper venting configuration can reduce this temperature to 40-45?C.

Abstract: Recent experiments at Lawrence Berkeley Laboratory (LBL) have demonstrated that rod receptors, which are widely thought to be important only for night vision, also contribute actively to vision processes at typical office light levels. At these light levels the studies found that pupil size and brightness perception are strongly affected by rod activity. These results suggest that light sources with scotopically richer spectral content need less photopic luminance to enable a given level of visual performance, visual clarity, and brightness perception. Such phenomena can explain the confusing results of many earlier visual performance studies where performance and visual clarity differences obtained under different lamps could not be explained on the basis of photopic luminance A re-analysis of these past studies, together with an examination of currently available lamps and phosphors, suggests that there is a substantial opportunity to increase lighting energy efficiency in a highly cost-effective manner solely by considering lamp spectrum.

Abstract: Measured characteristics of compact fluorescent lamps (CFL) and incandescent lamps with diode type devices show low power factor (47% to 67%) and/or high total harmonic distortion (>100% THD). There is little effect on a building power quality if the CFL comprise less than 10% of the buildings load. Even if the load due to CFL is as high as 26% of the buildings total, the voltage distortion is less than 5%. Circuits along with estimated costs are presented that correct for power factor and suppress the harmonics.

Abstract: The performance of several typical fluorescent lighting retrofits are examined using analysis methods of varying sophistication. Estimates of energy and lighting performance based on the simple non-application specific data generally available tend to overestimate the energy savings obtained with the various retrofits by 10-30%. Adding a simple correction to account for cooling benefits exacerbates the error unless heating penalties are also considered. An analysis method that takes into account the thermal application factor of the lighting system shows that the error is typically due to systematically overestimating the energy usage of the original lighting system. If thermal application factor is considered, then detailed HVAC calculations do not significantly improve the energy-savings estimate.

Abstract: A limited set of illuminants and phosphor colors are commonly used in video display terminal (VDT) working environments. This study attempts to identify any combinations of such conditions that influence performance on a visually demanding counting task. Experiments were performed to test whether the phosphor color, ambient lighting spectrum, or temporal frequency characteristics of the display and ambient lighting could alter performance. Under conditions where potentially contaminating variables such as reflectance level and screen glare patterns were equalized, no significant differences in performance were produced by the particular sets of stimulus conditions tested. The results suggest that displays and illuminant types that are in common use allow substantially equivalent visual performance.

Abstract: Part of a lighting designer"s craftsis to imagine the appearance of ansilluminated area before it is built.sExperienced designers, however, cansbe surprised once construction begins.sNew uses of construction materials,snew techniques and experimentationsoften lead to profitable discoveries asswell as occasional mistakes. Designsexploration is to be encouraged, asslong as it does not conflict with a client"sspreference. Unfortunately, a clientsmay insist on an unwise choice,sresulting in confrontation and potentialsdissatisfaction. For these reasonssand others, designers and their clientssfrequently prefer to previewsmany decisions before they are made.

Computers offer the ability to predictsvariables in design including appearance, thereby making designersand client education more economical.sBy simulating the behavior ofslight, computers can predict luminancesand illuminance levels, andsmake synthetic images of a space beforesit is built. In lighting design, ascommon practice is to begin with astarget illuminance distribution, thensuse manufacturer"s tables and simplescalculations to arrive at an initialslighting arrangement. For criticalsareas, flexibility is built into the systemsso that the light distribution cansbe tuned during installation. This approachsworks well in most applications,sbut when daylighting is a keysfeature, or aesthetics are stressedsover illumination levels, it becomessmuch more difficult for the designersto be convinced, and even harder tosconvince the client of a quality design.sIn this situation computer prediction of appearance would be invaluable.

Abstract: In a previous paper we analyzed Rea"s data from asnumerical verification (NV) experiment. We used asmodel based on discrete eye fixations, with fixationsduration, composed of visual and nonvisual components.sThe present paper investigates one of thesimplications of this model, which involved an experimentson the effect of instructions on performance onsthe NV task. On any task requiring discrete eye fixations,sthe number of fixations needed for the task andsthe minimum time required per fixation puts a lowersbound on the time to complete the task. Thesminimum time per fixation has been reported to beson the order of 200-300 ms for simple tasks? Thestwo columns of numbers in the NV task are separatedsby about 7 degrees, which makes it difficult to clearlyssee them both in one fixation? On the other hand, assingle five-digit number is only about 1.4 degreessacross, so it should be visible in one fixation. Thissleads to an estimate of a minimum of 40 fixations toscomplete the task, and thus an estimate of 8-12 s as asminimum time per chart.

Abstract: This paper describes the thermal and light output performance variations forssome representative compact fluorescent fixtures. Experimental data isspresented showing the changes in relative light output and temperature forsboth screw-in and dedicated compact fluorescent fixtures systems. Initialsexperimental data shows that substantial losses in fixture efficiency can occursin compact fluorescent fixtures due to elevated temperature conditions insidesthe lamp compartment. These elevated temperatures affect the minimumslamp wall temperature of the compact fluorescent lamp and thescorresponding light output and efficacy characteristics of the lamp ballastssystem. This paper also presents experimental data on various methods tosreduce compartment and lamp wall temperatures in order to increase fixturesefficiency.

Abstract: Based upon the years of development andsmanufacture as well as a documented three andsone-half year life test the reliability of the electronicshigh frequency ballast concept has beensestablished. There are several characteristics thatscan be used to assess a ballasts performance withsrespect to cost effectiveness, lamp life and powersquality. The higher cost of the electronic ballast issdue to the demand exceeding the supply. Withsmore competition in an expanding market the pricesof ballasts will be significantly reduced.

Abstract: This paper describes the development of thermally efficient compact fluorescent futures. Experimental data shows that decreases in fixture effiaency can approach 20% due to elevated temperature conditions inside the lamp compartment. These elevated temperatures increase the minimum lamp wall temperature of the compact fluorescent lamp and reduce the light output and efficacy of the lamp ballast system.

A series of prototype fixtures are described that employ convective venting and heat sinking to reduce elevated lamp temperatures. These cooling strategies can produce 20% increases in light output and efficacy.

Abstract: Compact fluorescent lamps that can be inserted into conventional light fixtures are rapidly gaining acceptance as both a viable retrofit and new design approach to reducing lighting loads. Ideally, the compact fluorescent lamp should have the same light output as the incandescent lamp it replaces, but overheating inside typically small enclosed fixtures can reduce lumen output and hence lighting fixture efficiency by 15 to 20 percent. Fortunately, simple fixture modifications can erase this efficiency penalty, so that the full efficiency benefit of replacing incandescent lamps with fluorescent lamps can be realized. The paper describes such modifications and presents experimental data documenting the potential efficiency enhancement associated with thermal control systems.

Abstract: Two approaches to the design of lighting systems have developed over the years, embodied in the respective roles of the lighting designer and the illuminating engineer. The illuminating engineer relies on lookup tables and calculations to make lighting decisions, emphasizing the importance of light level over most other considerations. The lighting designer, in contrast, is guided by experience and aesthetic sense to illuminate environments in a way pleasing to the eye. For the illuminating engneer, the current revolution in computer technology has brought accurate illuminance predictions for simple spaces, making selection more reliable and, some would argue, easier. However, the computer revolution has done little to assist the lighting designer, whose evaluation more often depends on visual qualities than on numerical quantities.

The emerging field of visualization in computer science combines calculations with computer graphics to bring another dimension of understanding to scientists, engineers, and designers. This new development holds particular promise for lighting systems design, which is both a numerical and a visual endeavor Rather than being restricted as an engineer to a lighting decision based on a table of illuminance values, or as a designer relying only on experience, one could refer to a computer simulation that predicts quantity and displays quality.

Abstract: The ability of daylight-following lighting systems to provide a minimum specified lightslevel at the task surface is influenced by 1) the control algorithm used, 2) the spatialsresponse of the ceiling-mounted control photosensor and 3) the location of the photosensorsrelative to task and window. Best performance was obtained with a closed-loopsproportional control system controlled by a photosensor, with a large field of view butsshielded from direct light from the window. A minimum specified illuminance level couldsbe maintained at specific points on the task surface regardless of daylight condition or roomsgeometry provided that the system gain was properly calibrated to account for the localsluminous environment.

Open-loop proportional control also performed adequately but offered less precise controlsthan closed-loop systems due to the necessity of using a photosensor that was not shieldedsfrom direct window light. Integral-reset systems that were tested performed poorly, butsperformance could be improved slightly by completely shielding the photocell from directswindow light.

Abstract: The contrast multiplier, or VL technique, was originally derived to describe the influence of contrast and illuminance upon the accuracy of detection or resolution. The CIE 19/2 model extended the technique to predict the performancekisibility relationship but it did not define performance beyond stating that it was some mix of speed and accuracy. The report was vigorously criticized for this, and other reasons, and the model has seen little acceptance. The reaction to the CIE model has been so adverse that Rea recently developed a performance/visibility model that avoids even introducing the VL concept. Rea actually fit only speed as a function of visibility, but he notes that the form of the accuracy data is not substantively different. The disadvantage of Rea"s approach is that it provides no explanation for why the VL technique provided good fits to simple accuracy data.

Abstract: This paper describes the energy conservation potentialsassociated with thermally efficient fluorescent fixtures.sVarious approaches to improving fixture efficiency throughsthe optimization of lamp wall temperature are discussed.sThe energy saving potential using thermally efficientsfixtures is estimated for a sample lighting layout for ansoffice space. This analysis indicates that both power densitysand the number of fixtures can be reduced with thermallysefficient fixtures. Conservation potential on a nationalslevel is also estimated, using market profiles obtained fromsfixture manufacturers in conjunction with measuredsfixture performance data.

Abstract: This paper describes some of the currentsresearch in the area of lamp wall temperature control forsthe purpose of enhancing light output under elevatedstemperatures. More specifically a thermally conductivesheat pipe has been employed to maintain optimumslamp wall temperature under varied ambientstemperatures. Advantages of a heat pipe over previoussmethods of enhancing light output include no externalspower necessary and flexibility in orientation within thesfixture. The experiments described include thescalibration of a heat pipe against a known thermal load,sthe application of a heat pipe to a lamp in a test chambersand the placement of a heat pipe within a fixture.sTheory and experimental data indicate that optimumslamp operating conditions can be maintainedsthroughout an ambient temperature range whichstypically reduces light output by 20%.

Abstract: The study examines how luminaire type and spacing affect task contrast and visibility in unobstructed office spaces. The Lumen-Micro program was used to calculate visibility levels in a model open-office space as a function of luminaire candlepower distribution, fixture spacing, and illuminance level. Three representative luminaires were parametrically examined: a lensed troffer, a high-performance parabolic, and a high-performance uplight. Small differences in average task visibility levels were observed for the different luminaires and fixture spacings examined, but these differences were slight compared with variability within an installation. The study indicates that if one does not know the location of the task a priori, then it is not possible to achieve consistently high visibility levels throughout a space without resorting to lighting solutions that are intrinsically inefficient.

Abstract: In 1982, John Ott summarized his views on adverseseffects of fluorescent lighting on human health in Light,sRadiation, and You. One claim in this popular book wassthat fluorescent lighting could influence human musclesstrength, as measured by so-called kinesiologystesting.

Kinesiology testing claimed by some to aid insdiagnosing a variety of illnesses, involves a subjectivesevaluation of muscle tone in a contracting musclesIn the method used by Ott, the subject stands with ansarm held forward, parallel with the floor, and with thespalm down. The tester then pushes downward at theswrist while the subject resists this motion. Ott hasspublicly demonstrated that when a subject looks at ascool-white fluorescent light he is noticeably weakersthan when the light is turned off or when a full spectrumsfluorescent tube is viewed. Using the sameskinesiology testing he also concluded that electric-field shielding (by wire mesh) was needed on fluorescent light fixtures to prevent muscle weakness.

Abstract: An efficient ray tracing method is presented for calculatingsinterreflections between surfaces with both diffuse and specular components.sA Monte Carlo technique computes the indirect contributionssto illuminance at locations chosen by the rendering process.sThe indirect illuminance values are averaged over surfaces and usedsin place of a constant ambient term. Illuminance calculations aresmade only for those areas participating in the selected view, and thesresults are stored so that subsequent views can reuse commonsvalues. The density of the calculation is adjusted to maintain asconstant accuracy, permitting less populated portions of the scene tosbe computed quickly. Successive reflections use proportionally fewerssamples, which speeds the process and provides a natural limit tosrecursion. The technique can also model diffuse transmission andsillumination from large area sources, such as the sky.

Abstract: Time-averaged human electroretinogram (ERG) responsesswere determined for several workplace visualsstimuli which are temporally modulated at rates exceedingsthe perceptual critical fusion frequency (CFF).sA clearly identifiable synchronous response was insevidence for a video display terminal (VDT) stimulussoperating with a refresh rate as high as 76 Hz. A directlysviewed fluorescent luminaire with controllable drivingsfrequency elicited a synchronous response at rates asshigh as 145 Hz. In addition, an intense stimulus createdsby modulating the light from a slide projector producedsresponses at least as high as 162 Hz. The implicationssof these high-frequency responses as representing aspotential basis for visual symptoms are discussed.

Abstract: Linear interpolation is used to estimate missing values fromscandlepower, reflectance and other data tables. We have developed an errorsestimating program, and have validated it against known functions. Sampleserror estimates are presented. The procedures should be useful to both thesdevelopers and users of data tables.

Abstract: The ability of a photo-electrically controlled lighting system to maintain a constant total lightslevel on a task surface by responding to changing daylight levels is affected by the controlsalgorithm used to relate the photosensor signal to the supplied electric light level and by thesplacement and geometry of the photosensor. We describe the major components of astypical control system, discuss the operation of three different control algorithms, andsderive expressions for each algorithm that express the total illuminance at the task as asfunction of the control photosensor signal. Using a specially designed scale model, wesmeasured the relationship between the signal generated by various ceiling-mounted controlsphotosensors and workplane illuminance for two room geometries under real skysconditions. The measured data were used to determine the performance of systemssobeying the three control algorithms under varying daylight conditions. Control systemssemploying the commonly-used integral reset algorithm supplied less electric light thansrequired, failing to satisfy the control objective regardless of the control photosensor used.sSystems employing an alternative, closed-loop proportional control algorithm achieved thescontrol objective under virtually all tested conditions when operated by a ceiling-mountedsphotosensor shielded from direct window light.

Abstract: This study has measured the performance of energy-saving 34-watt F40, T-12, rapid-start, lite white fluorescent lamps being operated by solid-state ballasts and lightingscontrol equipment. The performances of these lamp systems are compared with thosesof 40-watt F40, T-12 rapid-start cool white fluorescent lamp systems studied in the prior phase of this project.

With the 34-watt F40 lamps and various solid-state ballasts, system efficacy ranged froms67 to 84 lumens per watt and ballast factor from 0.756 to 0.908. Average system efficacysusing the 34-watt lamps exceeded that of systems using 40-watt lamps and the samessolid-state ballasts by only 1 percent even though the 34-watt lamps is about 6 percentsmore efficacious than the 40-watt lamp. This apparent discrepancy is due to increasedsballast losses when operating the 34-watt lamps. However, the system efficacy of the 34-swatt lamps used with a solid-state-ballast exceeded that of a 34-watt, two-lamp systemsusing the standard core-coil ballast by as much as 29 percent.

A T-8 fluorescent lamp system with a smaller lamp diameter was also included in thesstudy. Operating this lamp with a solid-state ballast produced a high system efficacy ofs90 lumens per watt, a 39 percent improvement over the efficacy of a 40-watt F40 systemsusing the standard core-coil ballast.

The use of static controllers with 34-watt F40 lamps can result in excessive flickering (46 percent) and the generation of a second harmonic as high as 96 percent of thesfundamental frequency.

The dynamic controllers, when used to dim the 34-watt lamps generally cannot besdimmed as low as the 40-watt lamp system without flickering.

In general, the 34-watt energy-saving lamps are appropriate as a retrofit to reducesillumination levels. However, for new construction, the 40-watt F40 argon filled lampsscost less, perform better, and provide a more reliable system.

Abstract: This paper discusses fluorescent lamps, ballasts, compact fluorescent lampssand lighting controls. The introduction of these new energy efficientsproducts has provided new choices to obtain optimum results in meetingsthe illumination requirements of spaces. However, the lack ofsunderstanding their fundamental characteristics and mutual interactionsshave led to serious design errors. We provide some of the essentialsinformation one must know about these new lighting systems, in order tosevaluate proposed lighting designs and equipment selection.

Abstract: The perception of room brightness over photopic luminances ranging froms30 cd/m2 to 67 cd/m2 was judged by 12 subjects in an almost uniformly whitesexperimental chamber. Two different illuminants were compared which hadsdifferent spectral compositions, but were color matched. Brightness judgementsswere often opposite to large differences in photopic luminance. These results aresinconsistent with models of brightness perception that depend solely on conesreceptors. At the luminance levels considered here subjective evaluation of lightsintensity depends upon both photopic and scotopic spectral contributions. Thesesresults imply that aspects of the visual system operate mesopically under mostsinterior lighting conditions.

Abstract: The electric and photometric characteristicsshave been compared for 40 watt (40W) and 34 watts(34W), F40 T-12 fluorescent lamps by operating them swith electronic ballasts, static controls and dynamicscontrols of different designs. The energy savingsskrypton filled 34W lamp system is, at best, slightly moresefficacious than the standard 40W argon filled lampssystem by virtue of the use of lite white phosphor. Thes34W system has limitations that include higher startingsvoltages, reduced temperature range of operation, smallersdimming range and poorer color rendering than thesstandard cool white 40W lamp system.

Abstract: A review of techniques to control the lamp wall temperature of fluorescent lampssin luminaires is presented. Past results show large increases in efficacy and lightsoutput can be obtained (to 25%) if the lamps can be operated at their optimum lampswall temperature. It may be judicious to review their cost effectiveness in view ofsthe increased energy cost and advances in the devices technology.

Abstract: There are many opportunities to improve the performance of existing lightingssystems in which the power and energy use can be reduced while maintaining orsimproving the lighting quality. Too often, lighting retrofits do not meet theirsexpectations because the decision malting criteria were base solely upon economics.sFurthermore, some economic analysis are wanting because of the lack ofscomprehension of the performance of commonly used lighting equipment or aresbased upon the exaggerated claims of the manufacturers of energy saving and energysefficient lighting components. That is, some equipment is energy saving due toslowering light levels, however, there may be an improvement in efficiency due tosthermal effects and not due to action of the equipment. This is detrimental insanalysis when comparing different retrofit strategies when these factors are notsunderstood. Often, we are convinced that occupants will not notice the change inslight level, however, even if the change does not invoke a response it is possible thatsproductivity may be impacted, or the salability of the space may suffer. This reportswill attempt to address the above issues to assist the decision makers in reaching assound decision with a lighting system that meets their needs. The considerations andsinformation that follows will provide a guideline for evaluating and comparingsdifferent retrofit strategies.

We will confine ourselves to the consideration of retrofitting fluorescent systems inscommercial or industrial spaces. Data presented represents performance datasmeasured in our laboratory.

Abstract: Using infrared pupillometry the pupil sizes of 24 healthy young adult subjectsswere compared under indirect illumination from four different light sourcessoperating at either the 60 Hertz line frequency or at a high frequency ofsapproximately 30 Kilohertz. Several different luminance levels were used in thesfrequency comparison ranging from 5 to 160 candelas per square meter (cd/m2).sUnder conditions where pupil size differences have been observed due to bothsspectral composition of the lamps and luminance changes, no differences weresfound between high and low frequency operation within the accuracy of thesesexperiments.

Abstract: An advanced ray-tracing technique for computing luminances insunempty spaces is presented. The method is a hybrid of finite elementsanalysis and Monte Carlo techniques. Interreflection between specularssurfaces of complex geometries can be computed to any desired accuracy.sApplications to synthetic imaging for lighting design are highlighted.

Abstract: A cost-benefit analysis based on visual-performance models is derived by consideringsthe current consensus lighting recommendations, RQQ #6, as implicit estimates ofscost-effective lighting. There is only partial consistency between models and thesesrecommendations. The consistency analysis provides a quantitative basis for thesesrecommendations, delineates their limits of applicability, and suggests changes insthem for handling low reflectance tasks, aged workers, high electrical costs, orsconditions where high visual speed and accuracy are very valuable.

Abstract: The first phase of this program has developed a lamp-ballast system that improves the efficacy of the currently used fluorescent lamp-ballasts by 27 percent, from 49 lm/W to 62 lm/W. Equally important has been the development of a ballast that improves the power factor from 60 percent to over 90 percent and the reduction of all the line harmonics to below 3 percent. The specifications for the above system have been prepared and will be evaluated on board a ship in the near future.

In the course of the above Phase I study, further improvements in the system efficacy were measured with the T-8 (1-inch diameter) lamp. This report describes the development and performance of the T-8 fluorescent lamp-ballast system. Similar to the Phase I effort, the cooperation of the lamp and ballast companies (GTE, North American Philips Lighting, and IOTA Engineering) were required for the successful result. The above companies submitted prototype lamps and ballasts to the Lawrence Berkeley Laboratory Lighting Group to evaluate. The lamp designs were modified as required and the optimum system selected. Two-lamp designs and one ballast were chosen and tested to determine if they met the Navy specifications.

The second section describes the lamp-ballast specifications. The measured electrical input-output performance of the lamp-ballast systems are listed in the third section. This section includes the measured thermal performance from an ambient of 10 degrees C to 60 degrees C. The measurements of the electromagnetic conducted and radiated intensities and the line current harmonics are presented in the forth section. In the fifth section the quality assurance, the life and lumen depreciation of the system are presented. The results and status of the T-8 project is discussed. The final section presents the conclusions of this development.

Abstract: A common lighting design problem issevaluating distributions of power levels, orsilluminances, luminances, or some other lightingsparameter. It is a problem for IES committees,sas well as the practioners. We discuss thesadvantages and disadvantages of threestechniques that are being used.

Abstract: By controlling the lamp wall temperature of fluorescent lamps with a Peltier device,sthe optical efficiency of luminaires can be measured directly. Measurements of thesfixture efficiency for a luminaire can vary by over 8% depending upon the lamp-ballastssystem used in the measurement. The authors suggest that optical efficiency may besan improved metric for evaluating fixture designs.

Abstract: Long-life replacement lamps for the incandescent lamp have been evaluated with regard to their cost effectiveness. The replacements include the use of energy buttons that extend lamp life as well as an adaptive fluorescent circline lamp that will fit into existing incandescent lamp sockets. The initial, operating, and replacement costs for one million lumen-hours are determined for each lamp system. We find the most important lighting cost component is the operating cost. Using lamps that are less efficient or devices that cause lamps to operate less efficiently are not cost-effective. The adaptive fluorescent circline lamp, even at an initial cost of $15.00, is the most cost effective source of illumination compared to the incandescent lamp and lamp systems examined.

Abstract: Increased utilization of daylighting in commercial buildings is one method of reducing the requirement forsdaytime lighting energy and for moderating peak demand. For daylight to efficiently supply some portion ofsthe design light level at the task, the electric lighting system should be photo-electrically controlled so that itsresponds (dims) in proportion to the amount of available daylight entering the building space. The locationsand spatial response of the photosensor that controls the electric lighting system must be chosen so that thesphotosensor"s output is approximately proportional to the illumination at the task surface. Furthermore, thessystem"s control algorithm, which relates the photosensor signal to the output of the electric lights, should besselected to properly account for the location of the control photosensor relative to the task and the sources ofsillumination within the controlled space. If the above considerations are not properly accounted for, then thesillumination at the task will deviate significantly from the design level (Rubinstein 1984) and the occupantssmay respond negatively, especially if the control system supplies less than the design light level.

The paper describes experimental work done to analyze how the control algorithm and the photosensor"sslocation and spatial response affect the ability of a daylight-following lighting system to maintain a constantslight level at the task by responding to changes in daylight levels.

Abstract: This paper describes a new technique for optimizing fluorescent lamp performance underselevated temperature conditions. This approach uses a thermo-electric Peltier device tosproduce a localized cold spot temperature of approximately 40?C, allowing the lamps tosmaintain maximum light output and efficacy independent of prevailing ambient temperaturessinside a luminaire.

Experimental data shows that a 20% increase in light output and a 10% increase insefficacy over typical lamp performance in a warm fixture environment can be obtained usingsthis device. Only 0.25 watts must be supplied to the Peltier device to produce these results.

Abstract: The transverse profile of the monochromatic radiance of an optically thick resonance linesfrom a cylindrical discharge is inverted exactly to give the radial distribution of radiatingsatoms. In contrast to the Abel transform, this result is valid for all optical depths.

Abstract: Eight healthy young adult subjects produced significantly larger steady state pupilsareas, as measured by infra-red pupillometry, when exposed to indirect lightingsfrom high frequency high pressure sodium lamps compared to photopicallysmatched levels of indirect incandescent lamps at three levels of luminance: 30, 60sand 90 candelas per meter squared (cd/m2). Three additional intensities weresstudied, which were not matched photopically between lamps. Analysis of all datasshowed that a scotopic spectral distribution accounted for pupil size better thanseither a photopic spectrum or an Alpern-Campbell pupillary response spectrum.sBecause pupil size can affect visual functioning, these results suggest that control ofspupil size should be considered in lighting design and that the scotopic spectralsoutput from lamps should be important in determining the effectiveness of aslighting environment.

Abstract: This paper describes the effects of fixture type and lamp compartment air extract characteristics on lamp/ballast performance.

A luminaire/plenum/HVAC simulator was used to measure minimum lamp wall temperature inside four fixture types while varying lamp-compartment extract conditions. Experimental data show that the lumen output of the lamp/ballast system varies by 20% and system efficacy by 10%, depending on the type of fixture and lamp compartment extract technique employed.

Abstract: Current efforts at the Lawrence Berkeley Laboratory to improve the operationalsefficiency of fluorescent lamps are based upon novel techniques intended to reducesinternal resonance radiative transport losses. Such losses have been associated withsentrapment of resonance radiation in the optically thick positive column of thesmercury vapor discharge. Two promising techniques for entrapment reduction are providedsrespectively to selective mercury isotope enrichment and by the broadening of the hyperfinesstructure (hfs) of the resonance line lambda2537 via the Zeeman effect. These techniquesshave, thus far, led to fluorescent lamp UV production efficiency enhancements, respectivelysof about 6.8% and 7%.

Additional efforts to improve HID sources are in progress. Interesting metallicshalides have been combined with electrodeless arc technology to produce a laboratorysdischarge system operating with an efficacy exceeding 170 lumens per watt, and a colorsrendering index (CFI) of 52.

Abstract: A computational method was developed to examinesthe effect of lamp, ballast, and fixture selection on the quality andsquantity of illumination, energy consumption, and life-cycle cost ofslighting systems. Applying this analysis to lighting layouts usingsdifferent lamp/ballast/fixture combinations suggested thatscombinations with higher lumen outputs reduced the uniformity ofsthe illuminance distribution at the workplane but did not reducesvisibility levels. The use of higher lumen output lamp/ballast/fixturessystems and higher efficiency components tended to reduce life-cyclescosts as long as the premium cost of the components was notstoo high.

Abstract: This paper discusses the unique features of a multistory office building thatssuccessfully permit most of the space to be daylighted and the electric lights to besdimmed by a cost-effective centralized system. This system includes: the use of lightsshelves, sloped ceilings, and proper building orientation and symmetry, and suppliessonly the ambient illumination. Measurements of the daylight illumination levelssand the performance of the lighting control system indicate that daylighting cansprovide over 70% of the required ambient illumination through the year. Based onsthe installed cost of the lighting control system, its payback period is 2.21 years.

Abstract: This user"s manual explains how to work with the CONTROLITE lighting control system analysis program on your IBM Personal Computer. It contains:

- A listing of what you need to use the CONTROLITE system

- Instructions on how to get started

- Details of all features of the CONTROLITE system

- Example problems to help you learn how to use the program

The CONTROLITE manual assumes that you have read and are familiar with the IBM Personal Computer Guide to Operations, Section 2 in particular.

Abstract: High intensity discharge lamps excited by solenoidal electric fields (SEF/HID) were examinedsfor their ability to give high brightness, high efficacy and good color. Frequencysof operation was 13.56 MHz (ISM Band) and power to the lamp plasma ranged fromsabout 400 to 1000 W. Radio frequency transformers with air cores and with air core complementedsby ferrite material in the magnetic path were used to provide the voltage forsexcitation. Electrical properties of the matching network and the lamp plasma were measuredsor calculated and total light from the lamp was measured by an integrating sphere.sEfficacies calculated from measurement were found to agree well with the positive columnsefficacies of conventional HID lamps containing only mercury, and with additives of sodium,sthallium, and scandium iodide. Recommendations for future work are given.

Abstract: This paper describes a new apparatus designed tosmodel the physical parameters that affect fluorescentslamp performance under realistic operating conditions.sThese parameters include fixture type, mountingsconfiguration, HVAC integration, and room airstemperature, which directly determine the minimumslamp wall temperature (MLWT) and, therefore, thesresulting light output of the lamp/ballast system.sThis apparatus is used principally to measure MLWTsunder operating conditions, which enables us to identifysthe effects the major parameters have onslamp/ballast system performance. Initial parametricsresults illustrate the use of this apparatus to providesrepresentative MLWTs for a range of applicationsconditions.

Abstract: The aim of the U.S. Department of Energy (DOE), Lawrence Berkeley Laboratory (LBL) lighting program is to reduce the nation"s lighting energy use by 50% through the introduction of more efficient and cost-effective systems and strategies. We work toward this goal in close partnership with the lighting industry to enhance its long-term viability.

To achieve this goal, the lighting program undertakes research in lamp science and technology, application of lighting to buildings, and the impacts of lighting on human health and productivity. The technical methodologies employed by LBL are described herein.

Abstract: Analysis of roadway tasks in terms of the CIE 19/2 visibility parameters has beensrestricted to calculation of VL at a fixed distance because there has been no way to adjust forsthe change of apparent size with distance. We use a rough ft of the absolute threshold contrastsdata as a function of luminance and size to allow the computation of VL as a function ofsdistance. Sample VL versus distance and VL versus time calculations are presented. Aspreliminary estimate of detection probability as a function of time is compared to Gallagher"ssmeasured results to see how current detection models compare to real results.

Abstract: Automatic lighting controls can significantly reduce operating costs in commercial buildings. However, the design of these dynamically operated lighting systems has not kept paceswith advancements in equipment. This paper focuses on determining how many lights should besgrouped into sectors and controlled as a unit in order to minimize the life-cycle cost of ansautomatically controlled lighting system. The analysis is based on occupant behavior patterns, the cost of the lighting controls, the energy costs, and a decision criterion based on life-cycle cost.

Occupant switching patterns in the morning and afternoon were experimentally determined at a lighting control demonstration at the World Trade Center (WTC) building in New YorksCity. The experimental procedure used to estimate the occupancy probability distributions atsthe WTC site is described in section 2. Section 3 develops a general analytical model for determiningshow large to make independently controlled lighting sectors in order to most cost-effectivelysuse the scheduling control strategy. In Section 4, the model is applied to the occupantsbehavior patterns measured at the WTC site. Section 5 discusses the significance of the results;sthe final section summarizes the study.

Abstract: Standard 40-watt, F-40, rapid-start, fluorescentslamps were operated with solid-state ballasts followingsthe standard life-testing cycle of 3 hours on ands20 minutes off for more than 20,000 hours at highsfrequency. Lamp operating characteristics (starting voltage, filament voltage, arc current, and current-crest factor) were studied as factors affecting lampslife. Measurements show that fluorescent lamps can attain rated life at high frequency using solid-state ballasts. When lamps are operated in the dimmedsnode, full filament power is required to sustain lampslife. The rate of lamp lumen depreciation is dependentson the lamp loading and not the operating frequency.

Abstract: Two techniques are used to measure the effect ofsremoving the filament power from a two-lamp, F-40,srapid-start fluorescent lamp system. The changes aresmeasured for a standard CBM core-coil ballast and forsa solid-state ballast operating the lamp at high frequency.sThere is a 4 to 6% increase in system efficacyswhen the filament power is removed. Removal ofsfilament power also decreases filament temperaturesfrom 1000 degrees C below 700 degrees C in lamps operated at 60sHz, and from above 600 degrees C to 300 degrees C in lamps operatedsat high frequency. The study shows that the arcscurrent and anode fall also determine filament temperature.

Abstract: This study discusses parameters for selecting the most suitable auxiliary lightingsdevice to- operate and control gas-discharge lamps. The devices tested in this studysinclude solid-state, fluorescent, and high-intensity discharge (HID) ballasts;scurrent limiters; and dynamic lighting controls. They have been evaluated whensoperating the standard, 40-W, F-40, T-12, rapid-start, cool-white fluorescent lamps.

Solid-state ballast performance varied widely, from 68 t o 79 lumens per watt (lm/W)sin efficacy, and from 0.83 to 0.98 in ballast factor. System efficacy was measured at up to 26% higher than standard core-coil ballast efficacy.

Current limiters used with standard core-coil ballasts reduce light output and inputspower by 30 to 50% and may be suitable as retrofit devices to reduce light insoverilluminated spaces. In operation a t a constant lamp wall temperature, thesesdevices either maintain or reduce system efficacy.

Dynamic lighting controls (which vary the light output) condition the power to standardscore-coil fluorescent ballasts, and low-voltage-type controls alter the oscillating circuit of solid-state ballasts. Power conditioning systems are generally designed to operate many ballasts and most can dim lamps to 50% of full light output. The solid-state ballasts control individual lamps and can dim them to 10% of full light output.

Abstract: This report summarizes results of fluorescent lamp studiesscarried out during the time period Jan. 1, 1984 tosDec. 31, 1984 by GTE Products corporation for the DOE undersa subcontract from Lawrence Berkeley Laboratories. Thesstudies are divided into the following four areas: MagneticsField Effects, Lamp Isotope Experiments, Modeling of IsotopesEffects, and 196Hg Isotope Separation in a Flow Reactor.

In the previous annual report we mentioned preliminarysresults utilizing transverse magnetic fields withsfluorescent lamps. Here effects of both axial andstransverse magnetic fields are studied in more detail andsunder controlled lamp operating conditions. Effects ofschanges in the Hg isotopic distribution are also studiedshere but specifically for uncoated fluorescent lamps inswhich emitted U.V. radiation is measured. Monte-Carloscalculations were carried out to model the 253.7nm radiationstransport process through a F40T/12 lamp. Solutions of thesHolstein-Biberman equation were utilized to match thesmeasured U.V. hfs and determine the near resonant energystransfer cross section. Finally, isotope separation processsdevelopment continued with the goal of obtaining asreproducible high enrichment process. Substantial progressswas also made in developing diagnostics for the enrichmentsprocess.

Abstract: This article discusses the major parameters that affect fluorescent lamp performancesunder fixture conditions. These parameters include fixture type, and HVACsintegration, which directly determine the minimum lamp wall temperatures(MLWT), and therefore, the resulting light output of the lamp/ballast system.sExperimental data is presented showing that the lumen output of the lamp/ballastssystem can vary by as much as 20% and that the system efficacy can vary by 10%sdepending upon the type of fixture and HVAC system employed.

Abstract: This paper describes an experimental integrating chamber for measuring the photometric and electrical performance of fluorescent light sources over a wide range of bulb-wall temperatures. The system was used to measure the relative light output and system efficacy of various solid-state ballasted fluorescent lighting systems for minimum bulb-wall temperatures between 20 degrees and 65 degrees C. For lamp-wall temperatures between 30 and 60 degrees C, the relative efficacies of the solid-state ballasted systems were found to be less sensitive to changes in bulb-wall temperatures than standard or efficient (low-loss) core-coil ballasted systems.

Abstract: We reanalyze the Smith-Rea check value verification experiment. This experiment has been discussed i n a number of articles, and is one of the 20 experiments used to support the CIE 19/2 model. A preliminary data sheet from Smith and Rea listed an incorrect score function and contained a large number of arithmetic errors in converting raw times to scores. Correction of these errors changes the CIE fit. We argue that the W123 parameter of this fit is not related to the critical visual processes as claimed.

We use the corrected data to examine basic trends. Subjects achieved their maximum scores for a large fraction of runs under all visibility conditions. There was no statistically significant difference in scores for tests from 100 to 5000 lux. Furthermore, illumination level was less important to performance than the other variables studied: subject, practice, and check set (legibility and contrast).The RQQ #6 recommended illumination levels f o r such tasks range from 200 to 750 lux, indicating that recommended levels may overstate the need for illumination.

There was a distinct practice effect, and this effect is correlated to visibility. The practice effect was largest where there was least visibility. The same set of checks was used in each run. It is not clear how much of the practice effect is due to this experimental artifact and how much can be generalized. The long-term magnitude of the visibility performance trend is rendered extremely uncertain by uncertainty over the source of the practice effect. There is no question that there is at least a short-term visibility performance trend.

The CIE regression is re-examined to see how efficient is its empirical description of the visibility/performance relationship. This analysis tests the hypothesis that even though the CIE model may not be theoretically correct, it may still be a good approximation. The fourpoint fit used in CIE 19/2 had only one degree of freedom and would be rejected if it was linear. Using less, or unaveraged data, we found that although the CIE fit s explain a statistically significant amount of variance, they were less efficient than a simple ln(VL) fit.

It has been suggested that since VL is based on threshold contrasts, it is not an appropriate measure for supra-threshold real-world tasks. We performed a rank-order test of an alternative visibility measure, conspicuity, against performance, but found no correlation. As a hypothesis we suggest that visual performance is inherently bounded by threshold visibilities. There are several mechanisms that would lower nominally suprathreshold visibilities towards threshold levels in a visual performance experiment. The mechanisms are sufficiently different that there should be no unique visibility/performance relationship. Instead we argue that the relationship will depend on the type of the experiment (and hence the mechanism) and the details of the scoring function.

Abstract: Lighting control equipment is reviewed with respect to its role in thesbuilding application. The elements of a control system include theslight controller, sensor, and communicator. Each type of equipment hassa best use that depends on the specific control strategies and application.sThe applications discussed are retrofit, renovation, and new construction.sThe simplest controls are best used in retrofits wheresinstallation costs must be minimized. In new construction the mostssophisticated controls can be cost-effective.

Abstract: A novel mercury-argon discharge tube is described.sIt incorporates two identical condensation pump cold spots thatsare employed to determine both the mercury sorption rate and thescoefficient of free diffusion for mercury in argon. Equilibrationsperiods for this system are found to depend in part on the temperaturesdifference between the two cold-spot surfaces, and on thesseparation of these surfaces. Sorption of mercury at the dischargestube"s surface, and the subsequent cleanup of the mercury charge, issdetermined by the time rate of change of the equilibration period.

Data accumulated during these tests are incorporated into thesdesign scheme of discharge tubes intended for Zeeman-effect andsisotope blending studies described elsewhere.

Abstract: This paper discusses the basic parameters of a fluorescent lighting system thatsaffect the illumination level. The parameters include the thermal performance ofsthe fixture and the ballast factor, voltage regulation, and thermal regulation ofsthe ballast/lamp system. Fixtures determine the minimum lamp-wall temperature ofsthe lamps and are described as hot or cold. That is, the lamp-wall temperatures cansvary from 39 to 61 degrees C. In general, cool fixtures tend to provide higher light levelssand are more efficacious for a given ballast/lamp system.

Solid-state fluorescent ballast/lamp systems have been measured and show a variationsin light output from 6170 to 3780 lumens for the two-lamp, F-40, T-12, rapid-startslamps. Lighting designers must obtain this information in order to accuratelyspredict illumination levels in a space.

Abstract: The role of technology in the development of lighting control systems is examined. Prior to 1973, control systems were primarily functional. They were used to switch lamps on and off, or to create a mood. Nearly all dimming was done with incandescent lamps.

Since 1973, the rising cost of electrical energy has made the operational cost of systems a design factor. Lighting controls are used to decrease power, reduce the time of use, and lower peak power demands. New lighting control equipment is based on existing technologies that have been modified to meet the needs of lighting systems. The technologies came primarily from the field of electronics.

These developments have made it technically feasible to automatically control the intensity and distribution of lighting in an area illuminated by a single fixture. These technologies form the basis for meeting future goals of lighting designs to alter illumination levels over time.

Abstract: Solid-state ballasts with 20-30 kHzsfundamental frequencies were used to operate fluorescentslamps in office, screen room, and open-fieldssites. The test equipment and layout are describedsand the radiated and conducted electromagnetic interference (EMI) data are given for six commerciallysavailable ballasts from different manufacturers. EMIsdata from each location are presented that indicatesthe suitability of using an office site for future EMIsmeasurements. Broadband EMI data are compared withsdata from a spectrum analyzer, which justifies the usesof an RMS voltmeter in these measurements. A reductionsin radiated EMI of 36 dB is achieved by replacing asstrip fixture with an EMI-hardened fixture that has aswire grid embedded in its lens. Results of increasingslamp/antenna distance are given; we verify that, forsdistances greater than 1 meter, the radiated EMIsamplitude drops off as a function of l/d3.

Abstract: This paper describes the components of asdaylight-linked lighting system and presents threessimple control algorithms that can be incorporatedsinto a control system to achieve the design objectivesof constant task illuminance.

Abstract: There have been many new lighting products introduced since the 1973 energy crises. The products that entered the market most rapidly were based on existing technologies and relied primarily on reducing light levels to reduce electrical power use. These systems worked for retrofitting existing spaces that were over-illuminated. In the mid-70s, new products employing new technologies were developed that increased the efficacy of producing and distributing visible light. These included the high-frequency operation of fluorescent lamps, replacement of incandescent light sources, and effective management of lighting. These technologies will be described and their present status, with regard to their current use, will be discussed. Their impacts on energy use and on the quantity and quality of illumination will be presented. The results will provide evidence that proposed lighting power densities of 1.5 watts per square foot for lighting can be realized.

Abstract: Designers who wish to compare and select lighting systems and to designslighting layouts that meet both illuminance and energy code requirementssrequire accurate data documenting the combined performance capabilitiessof fluorescent lamp, ballast, and fixture systems operating under realisticsapplication conditions. These application conditions may differssignificantly from the standard ANSI conditions under which routine photometricstests are performed, introducing temperature-dependent errorssinto the lighting design calculation procedure.

To meet the need for accurate performance data, we plan to develop andsdemonstrate an experimental method for determining the performance ofslamp, ballast, and luminaire combinations operating under a broad rangesof realistic conditions.

Abstract: Experimental results from a major demonstration of an advanced lighting energy management system at the World Trade Center are presented. The energy-saving benefits of automatically scheduling the operation of the lighting system to conform to occupancy patterns are examined. The energy saved by scheduling was measured by comparing lighting energy consumption without scheduling to consumption with scheduling. The benefits of a variety of switching scheduling are compared and the relationship between energy savings and sector size discussed. Using a loose automatic schedule with 1000 ft2 zones, lighting energy consumption was reduced by 30% relative to baseline consumption. With a tighter schedule, energy consumption was reduced 36-37%. Based on a simple economic analysis, scheduling is shown to be a cost-effective strategy for reducing energy consumption in buildings.

Abstract: A complete analysis of the cost-effectiveness of daylighting strategies should include the impact of daylighting on peak electrical demand as well as on energy consumption. We utilized an hour-by-hour building energy analysis program to study the thermal and daylighting impacts of fenestration on peak demand. Fenestration properties and lighting system characteristics were varied parametrically for office buildings in Madison WI and Lake Charles LA. Peak electrical demand was disaggregated by component and by zone, monthly patterns of peak demand were examined, and impacts of fenestration performance on chiller size were studied. The results suggest that for daylighted office buildings, the peak electrical demand results from a complex trade-off between cooling load due to fenestration parameters, lighting load reductions due to glazing and lighting system characteristics. Lowest peak demands generally occur with small to moderate size apertures. With daylighting, peak electrical demand is reduced by 10 to 20% for the building configuration studied (37% perimeter zone, 63% core zone). This work indicates that solar gain through fenestration must be effectively controlled in order to realize the potential of daylighting to significantly reduce peak electrical demand.

Abstract: In this paper nine combinations of fluorescent F40 ballast/lamp systems are compared on the basis of initial and operating costs for a new lighting layout using two- and four-lamp fixtures. The factors that influence the input power and light output are studied. Manufacturer"s tolerance (lamp and ballast), minimum lamp wall temperature, fixture efficacy, ballast and lamp efficacy, and the cost of energy are considered. The ballast/lamp system that is most efficient and provides the greatest light output has the lowest initial cost in a four-lamp fixture. The same system has the lowest operating cost in a two-lamp fixture.

Abstract: The CIE 1912 model of performance cannot be used to predict performance or productivity. We present counter-examples to the link assumed in CIE 19/2 between performance and productivity. Statistical arguments show that the fitting parameters are not physically determined as was thought and that the curve fitting in the CIE 1912 report does not constitute a validation.

Some critics of CIE 1912 have suggested that RQQ #6 be used in its place for lighting calculations. RQQ #6 is simply a consensus of present practice. The more robust visibility trends presented in CIE 19/2 are inconsistent with the recommendations in RQQ #6. Careful use of the material in CIE 19/2 could lead to better recommendations than are exemplified by RQQ #6.

Abstract: Energy-efficient light bulbs a r e being developed to replace t h e incandescent lamp where they can satisfy the design criteria and be used in sockets that have long hours of annual use. The four technologies discussed here include the compact fluorescent lamp, coated-filament lamp, electrodeless fluorescent lamp, and compact high-intensity discharge lamp. The systems demonstrate efficacy improvements of two to four times that of their incandescent counterparts. These new lamps have required considerable advances in lamp technology. They offer the potential for achieving efficacies close to 80 lumens per watt.

These new lamps will reduce the energy used annually by incandescent lamps (190 BkWh) by more than 50% in the 1990s, at which times they will be commonly employed.

Abstract: An operator of a visual display terminal typically is confronted with two flickering light sources: VDT phosphor with its inherent refresh rate, and the ambient fluorescent illumination reflected from the screen. Due to differences in frequency, the two sources of flicker can produce low-frequency beats in the net light intensity presented to the VDT operator. These low frequencies, on the order of a few hertz, can lead to adaptation of visual sensory mechanisms.

We have constructed an experiment in which a stable, high-amplitude beat is produced, with frequency chosen to enhance the possibility of detecting a visual system response. By adjusting the environmental lighting to produce a sinusoidal flicker at 69 Hz, and employing a VDT with a 61-Hz refresh rate, we produced an 8-Hz beat frequency in the light emanating from the VDT screen.

By measuring temporal contrast sensitivity at 8 Hz as a test of sensory visual system response, we found that two of three subjects showed a reduced sensitivity when exposed to this stimulus, as compared with three sets of control conditions in which the 8-Hz beat was not present. We discuss possible relations between this sensory adaptation and the factors of subjective visual fatigue and reduced levels of task performance. Further work is underway to examine whether adaptation occurs under conventional office lighting and typical mains frequency variations.

Abstract: As the cost of energy continues to increase, sophisticated lighting control systems designed to minimize energy consumption are proving more cost-effective. If a designer is to evaluate the performance of lighting control systems for their energy-saving capabilities, an unbiased mathematical model which compares different control systems on an equal basis is an important tool. The purpose of the computer program described in this report is to provide that model.

In theory, a lighting control system which continuously monitors the available artificial and natural light within a space and supplies only the power required to provide light to meet some specified criteria illuminance level, and no more, should be able to optimize the energy-saving capabilities of that lighting control system. Because of the potential for constantly varying power consumption due to this Equi-Illumination Dimming (EID) system, the computer is employed to help in modeling the performance of these types of systems.

Abstract: The problem of determining the appropriate light levels for visual tasks is a cost-benefit problem. Existing light level recommendations seriously underweight the importance of economic factors. Furthermore, the relative importance of the visibility factors in determining the optical light levels appears inconsistent with the importance of these factors in determining visibility and visual performance.

We show that calculations based on acuities give a lower limit of 100-200 lux for cost-effective light levels for office tasks. Upper limits are calculated from correlations of task performance to visibility levels. Visibility levels become progressively insensitive to luminance as luminance increases. Average power densities above 100 watts/m2 are cost-effective only when visibility is very low. However, there is a 3-to-10 times larger increase in benefits from improving contrast or contrast sensitivity than from using more than 10 watts/m2. Contrast or contrast sensitivity can be improved by using forms with larger print, using xerographic copy instead of carbon or mimeo, making sure office workers have the right eyeglasses, or even by transferring workers with visual problems to less visually demanding tasks. Once these changes are made it is no longer cost-effective to use more than 10 watts/m2. This conclusion raises serious questions about recommendations that lead to greater than about 10 watts/m2 of installed lighting for general office work.

Abstract: Lighting is one of the largest energy loads in a large commercial building. Lighting typically accounts for 35-50% of the electrical consumption which, in turn, dominates the total energy costs in a building. Since Edison"s day, there has been a 100-fold increase in the efficacy of lighting sources. Relatively little progress, however, has been made in reducing consumption through effective lighting management - using the optimal amount of light, where needed, and when needed.

Commercial lighting control is an area where the potential for major energy saving exists. A number of new products have begun to emerge which focus on lighting control. To identify promising technologies and to expedite their adoption by building owners, the Department of Energy funded a program by Lawrence Berkeley Laboratory to test new, commercially available lighting controls in an actual office environment. The tests were designed to demonstrate the following:

(1) which control strategies have the greatest impact and why

(2) economic trade-offs between control cost and savings potential

(3) acceptability of the controls to occupants

(4) control reliability.

Abstract: The energy-saving benefits of two lighting control strategies - scheduling and daylighting ? were investigated at demonstration sites in two large commercial buildings. A continuously-dimmable lighting control system was installed a t the Pacific, Gas & Electric Co. building in San Francisco and an on/off switching system was installed at the World Trade Center building in New York City. By automatically scheduling the operation of the lighting systems to conform with occupancy patterns, lighting energy consumption was reduced 10 to 40%. Several scheduling techniques were investigated and the influence of switching zone size on energy savings was examined. Using photo-electrically controlled lighting systems which switch or dim lighting in accordance with available daylight, the energy consumed for lighting in daylit areas was reduced 25 to 35%. The influence of clear and overcast conditions on the energy savings associated with daylight-linked controls is discussed.

Abstract: On March 18 and 19, 1982, a conference was held at Lawrence Berkeley Laboratory to discuss and identify any existing or potential problems of electromagnetic disturbances caused by light sources and lighting systems. Concern over possible problems has arisen as new lighting systems are developed to operate at high frequencies. These systems include solid-state ballasts, powerline communication, electrodeless fluorescent lamps, and phase-control systems.

The conference drew more than eighty participants. It consisted of one day of invited presentations and one day of breakout sessions, which met informally to discuss and share information about selected topics. The results of each group discussion were presented to the entire conference by the session chairmen.

This proceeding reviews the activities of the Lighting-Electromagnetic Compatibility Conference. It includes several of the papers that were presented, a summary of each breakout session in the format chosen by the various chairmen, a summary of the conclusions, and an outline of the future effort the lighting community must make in this area.

Abstract: In visual performance experiments, accuracy is as much a function of the relative worth of speed and accuracy as it is of visibility. In a task that consists of subtasks, such as comparing two lists, it may be necessary to explicitly model the effect of each subtask"s visibility on speed and accuracy. Finally, changes in print may be significant in that they can lead to changes in visual performance without a corresponding change in visibility.

The current Commission Internationale de L"Elairage (CIE) model for visual performance, CIE 19/2, does not consider the above factors. Although we believe that the CIE model is not useful as an application model, we do feel that the general features of the visibility/visual performance relationship are clear and are important to lighting design. We close with a brief discussion of the failure of the RQQ #6 lighting recommendations to explicitly consider cost-effectiveness.

Abstract: Lamp and ballast manufacturers have shown increasing interest in having a high-frequency testing procedure to determine the peak and rms starting voltages for fluorescent lamps. This, paper discusses a starting voltage procedure and typical test results for a single-lamp circuit. These peak/rms relationships are shown as a function of frequency, cathode voltage, and ambient temperature for three types of F40 lamps (the standard F40, the F40 35-watt and the F40 35-watt without a conductive coating).

Abstract: General office lighting visibility specifications were analyzed with respect to optimization of cost effectiveness. Cost effectiveness is a function of visual performance, not visibility per se. It was found that the present procedures which utilize ESI are not readily adaptable to cost optimization calculations. An alternative procedure using log VL is presented which appears suitable for this use. A sample calculation of net benefits versus light level is presented. For the same calculation the net benefits from visibility saturate at lower levels of visibility than are normally prescribed for office environments.

Abstract: Long-life replacement lamps for the incandescent lamp have been evaluated with regard to their cost effectiveness. The replacements include the use of energy buttons that extend lamp life as well as an adaptive fluorescent circline lamp that will fit into existing incandescent lamp sockets. The initial, operating, and replacement costs for one million lumen hours are determined for each lamp system. We find the most important component lighting cost is the operating cost. Using lamps that are less efficient or devices that cause lamps to operate less efficiently are not cost-effective. The adaptive fluorescent circline lamp, even at an initial unit cost of $20.00, is the most cost-effective source of illumination compared to the incandescent lamp and lamp systems examined.

Abstract: This study describes the impact of lighting management systems that dynamically control lights in accordance with the needs of occupants. Various control strategies are described: scheduling, tuning, lumen depreciation, and daylighting. From initial experimental results, the energy savings provided by each of the above strategies are estimated to be 26, 12, 14, and 15%, respectively.

Based upon a cost of $0.05-0.10 per kWh for electric energy and a 2-, 3-, or 4-yr payback, target costs for a simple and a sophisticated lighting management system are found to be $0.14 and 1.89 per ft2 respectively, for a cost-effective investment.

A growth model, based upon an extrapolation of the increase in building stock since 1975, indicates that the commercial and industrial (C & I) building stock will grow from 40 x 109 ft2 in 1980 to about 67 x 109 ft2 by the year 2000. Even with the use of more efficient lighting components, the energy required for this additional C & I stock will be 307 x 109 kWh compared to the 230 x 109 kWh used today. Adopting controls would reduce this requirement to 243 x109 kWh, an increase of only 13 x 109 kWh above current use.

The specified information is used to analyze the economic impacts that using these systems will have on the lighting industry, end users, utility companies, and the nation"s economy. A $1-4 x 109 annual lighting control industry can be generated, creating many jobs. The estimated return on investment (ROI) for controls for end users would be between 19 and 38%. Utilities will be able to make smaller additions to capacity and invest less captial at 7-10% ROI. Finally, the annual energy savings, up to $3.4 x 109 for end users and about $5 x 109 for utilities, representing unneeded generating capacity, will be available to capitalize other areas of our economy.

Abstract: A solid state ballast has been designed for the efficient operation of a 10 circline fluorescent lamp. The circuit can be manufactured using power hybrid technology. Eight discrete component versions of the ballasts have been delivered to LBL for testing. The results show the solid state fluorescent ballast system is more efficient than the core-coil ballasted systems on the market.