No matter the type of project, one thing is clear: lighting energy codes are not going away. Designers must learn to communicate with regulatory agencies as they seek a balance between efficiency and creativity.

By James Benya

» Lighting energy codes have been the topic of much discussion lately. In its June 2004 Exchange column, Architectural Lighting featured comments from the industry on energy codes. It seems many designers view these guidelines as an obstacle to design, since they regulate both the types of sources as well as the wattage, while others see them as a necessary step toward more sustainable design. But what very few people see, is the struggle for balance that modern energy codes try to achieve.

The energy crisis of the early 1970s caused massive changes in U.S. energy regulation, with a series of federal laws and executive orders since then requiring the implementation of energy codes. The Federal Energy Policy Act of 1992 made ASHRAE/IESNA 90.1-1989 the national reference standard, requiring states to adopt it or an alternative energy model at least as stringent. Periodically the Department of Energy is required to consider and adopt a more stringent standard, and since 2002, the national reference is ASHRAE/IESNA 90.1-1999.

LIGHTING CODE RESOURCES to view (but not copy or print) ASHRAE/IESNA 90.1-2001 for free. to download the California Energy Code (2001 and 2005 available) plus supporting documents. for the Seattle Energy Code.,_Commercial_111302012949_76rule.pdf for the Minnesota Energy Code. for the Oregon Energy Code. a well-updated reference for energy codes, URLs and the status of code development.

There are three significant codes in the United States today:

ASHRAE/IESNA Standard 90.1 is a large, daunting code that covers most aspects of building energy use. Compared to building envelope, HVAC, and other sections, lighting is comparatively simple, occupying only about 12 pages of 90.1's 600-page tome. The lighting section of 90.1 is straightforward, easy to understand and to implement. While Standard 90.1-1999 is the federal standard, 90.1-2001 and 90.1-2004 are more recent versions that include changes and updates. Many states use Standards 90.1-1999 or 90.1-2001; Standard 90.1-2004 is so new that few have probably even read it. Because ASHRAE and the IESNA are co-owners and developers of 90.1, it has considerable credibility among lighting designers.

The International Energy Conservation Code (IECC) is a simplified model roughly equivalent to 90.1. Many states do not want to enforce complex codes, and IECC offers a workable alternative. In terms of lighting, IECC is almost the same as 90.1. The IECC was developed largely by the New Buildings Institute and other contractors to the International Codes Commission.

California's Code of Regulations Title 24 Part Six ('Title 24') is the only major alternative to 90.1; it is equally large, and in some cases, more complex. For instance, in lighting, Title 24 has significantly more restrictions and requirements than 90.1. This complexity has both positive and negative aspects. Title 24 is also the only energy code to regulate lighting within living quarters. (Having been personally involved in developing this code since the 1980s, I admit to prejudice in claiming it is the best of the bunch.)

Owing to the size and complexity of 90.1, a number of 'state-specific' codes have been developed. After all, IECC did not appear on the scene until 2000, by which time states that wanted a simpler guideline had already created their own. For example, the Oregon Energy Code, which started out as a simplified code allowing 90.1 as an alternative, has now dropped the 90.1 relationship altogether.


Codes are comprised of a combination of several regulatory methods, including:

•Lists of exemptions and exceptions, such as surgical lighting or emergency exit signs.

•Rules, such as how much wattage must be counted for track and other loads.

•Mandatory measures, such as switching, circuiting and controls.

•Simple prescriptive requirements such as the unrestricted use of low-wattage, highly efficient and efficacious luminaires.

•Whole-building power allowance, which is determined by multiplying the building's total interior area by the permitted lighting power density.

•Area category power allowance, which calculates interior illumination of each area, separately.

•Space-by-space power allowance, which determines interior illumination for spaces, individually.

•Task-by-task power allowance, which determines allowed power for interior illumination by each task.

•Exterior lighting power allowance, where exterior illumination is determined individually for each exterior lighting task.

•Additional allowances, for spaces or tasks under certain conditions. Often these allowances, such as for display or ornamental lighting, are 'use it or lose it.' In other words, where general lighting power can usually be traded between spaces, use-it-or-lose-it allowances can only be used for the specific application and space. In California, the ability to use 'controls credits'-power reduction credits for the use of certain energy-saving control devices or systems-is also available.

Most models employ only some of these approaches, and the simpler the code, the fewer the options. For example, Title 24 incorporates all except the simple prescriptive method; Oregon only employs the whole-building and space-by-space methods, with no additional allowances.


Some might say any code restricting lighting creativity is bad, and to a certain extent, they are right. The issue is, not having a code is not an option. So the challenge is to develop guidelines that do not unreasonably restrict lighting power and applications.

Mandatory measures are very similar among most codes today. They tend to be principally concerned with control, ensuring for example that each space has at least one switch. The biggest issue with mandatory measures involves the extent of controls: In California, automatic shutoff, separate daylight zone switching, multi-level lighting controls, and in some cases, automatic daylight dimming are all required. For now, the mandatory measures are relatively 'common sense' and not much of an issue. But then again, there is the Seattle code that effectively requires automatic daylight dimming-even after limiting offices to 1.0 watts per square foot and schools to 1.2 watts per square foot in one of North America's most daylight-deprived areas. Cost effective? Not!

Power density values for the major codes-Standard 90.1 and Title 24-are developed using a process in which lumen-method models are analyzed for their power needs. For each space, a lumen-method model is created in which a task is assigned a pro-rated area, a specific CU (coefficient of utilization), a source LPW (lumens per watt), and an illuminance level. These are added to determine the average lighting power density for the space. If properly determined, the models include general lighting, task lighting, focal and display lighting, wallwashing, and decorative lighting, depending on the building or space type. For each lighting type, the model assumes the most reasonable high-efficiency lighting system. Models are vetted among professional lighting designers, IESNA committees, and the industry to ensure that the choices are reasonable. Lesser codes, such as state-specific guidelines, are generally not modeled as thoroughly nor are they vetted, but rather, they are an amalgamation of values and decisions by the code developers. In other words, while 90.1 and Title 24 have stood the test of public review, codes like those for Seattle and Oregon are too often full of politically correct requirements that often make good lighting design nearly impossible.

The real challenge for the authors of lighting regulations is that contractors, code officials, and just about everyone else want simple, headache-free codes. Code enforcement is a major issue, since inspectors don't have the interest or time in their already overloaded schedules. Furthermore, every member of the design and construction process sees energy codes as one more item that costs money-more design time, more paperwork, and more expensive materials and equipment. Hence, the demand for simpler codes.

For lighting designers, however, simplification is frequently problematic. If a code is too basic, either it will not effectively regulate, or it will unreasonably restrict lighting design. Nowhere is this more apparent than in specialty and high-end retail lighting. Within reason, most other building types can be designed with modern equipment, and achieve acceptable lighting levels while meeting code. (Lighting-intensive applications, such as lighting stores, nightclubs, theaters, and theme parks, are typically exempt.) The more demanding retail lighting environments, however, which concurrently require high display levels, elegant aesthetics, and excellent color, can reasonably require three to four times the lighting power levels needed by offices, schools and other common buildings. Moreover, it is difficult to differentiate between general retail, specialty retail (i.e. slightly upscale), high-end retail, and the exceptionally demanding jewelry and china sales environments. A designer taking full advantage of Standard 90.1, IECC, or Title 24 can properly justify and use 4.0 to 5.0 watts per square foot or more for high-end retail, and over 6.0 watts per square foot for jewelry stores. The secret to success is to utilize every allowed watt in the code, and to apply efficient equipment wherever possible. In general this means abundant use of high-efficacy lamps, but some halogen infrared display lighting and decorative incandescent lighting is still possible.

In states with their own codes, the reduced allowances for retail lighting will force designers to use more expensive, or very ordinary, lighting equipment. For example, in Minnesota, code-complying retailers must use either general lighting systems or expensive ceramic metal halide display technology. By code, the intermediate lighting options have been eliminated, and that is the problem. The broad acceptance of ceramic metal halide lighting in Europe-and for that matter in New York-is partly understandable when peak electric rates exceed 20¢ per kWh, and ceramic metal halide pays for itself in a few years. But at 8.5¢ per kWh, the US national average, ceramic metal halide is not cost justified, and compared to halogen, it can increase the cost of display lighting systems by 500 percent.

Then there are the Pacific Northwest criteria, which are some of the strictest in the country. Seattle's code allows a total retail lighting power of 3.0 watts per square foot, unless one is willing to spend heavily on ceramic metal halide and dimming ballasts for fluorescent lighting systems. But for retail scenarios, Oregon's code is one of the most difficult to work within, where retail stores are limited to 2.0 watts per square foot, with no alternative of any kind. (Jewelry stores are allowed 4.0 watts per square foot.) Oregon's code delivers a clear message to retail merchants: Don't build in Oregon unless you change your store dress and lighting to the basic level of a grocery store.

There are other building typologies that also suffer under energy codes. For instance, in Oregon, libraries are classified as schools. Try lighting a library at 1.0 watt per square foot! Or how about lighting a hotel ballroom at 2.0 watts per square foot? Bye-bye chandeliers.


Many energy advocates believe lighting energy codes can continue to be refined. For instance, in California's 2005 code development process, Pacific Gas & Electric (PG&E) pressed for retail lighting standards to be based on models using ceramic metal halide display lighting. This would have dropped the California watts-per-square-foot values closer to those in Minnesota or Seattle, essentially limiting jewelry store lighting to about 4.5 watts per square foot and high-end retail to about 3.5 watts per square foot. But the PG&E initiative failed because California requires its energy code provisions to be cost effective.

Should ceramic metal halide prices drop (and quality remain high) enough to make it a more broadly useful technology, then one more round of reductions may be possible. But remember, metal halide is not a dimmable source and can only be used in retail and other settings without frequent switching.

Beyond this point, it is probably a good idea for ASHRAE, the IESNA, the Department of Energy, and all others interested in lighting energy codes to back off for a while and fix what we have. Both Standard 90.1-2004 and Title 24-2005 are already based on 'super T8' and T5 technology. LED is not yet an energy-efficient lighting system and probably won't be for a decade or more. And dimmable fluorescent lighting is less energy efficient than non-dimmable, since one must dim fluorescent lights 25 percent just to get system power equal to the full output of the non-dimming system! There are simply no other pressing new technologies on the horizon that will make lighting sufficiently more efficient than the codes can be made any more restrictive.

Lighting professionals should learn to develop compliant designs and to reinforce reasonable codes and interpretations. By recognizing that good lighting design is conscious of energy use, we continue to make the architectural lighting design profession stronger. It takes genuine expertise to accomplish quality lighting within the boundaries of energy codes and other limitations.

For those of us in states with unreasonable codes, we need to let our state energy departments know that their actions are at least unfair, and typically, have major unintended and negative consequences. We need to change these regulations so that a good design in California is a good design in Oregon, rather than a code violation. A bad code means one of two things: either people won't build, or those that do, will find a way around the regulations.

James Benya is a professional lighting designer and principal of Benya Lighting Design, West Linn, Oregon. He serves on the editorial advisory board of A|L.