Quantifying Daylight for a Big Box

The David L. Lawrence Convention Center was conceived as a daylit space from the beginning. Rafael Viñoly Architects created a grand scheme for the building, and Burt Hill Kosar Rittelmann developed the daylighting concept using basic rules of thumb. These preliminary concepts were tested and refined based on diffuse light analysis by Vladimir Bazjanac of Lawrence Berkeley National Laborat...

By Keith J. Yancey, AIA, P.E., LC, Senior Associate, Lam Partners, Inc., Cambridge, Mass. May 1, 2004

The David L. Lawrence Convention Center was conceived as a daylit space from the beginning. Rafael Viñoly Architects created a grand scheme for the building, and Burt Hill Kosar Rittelmann developed the daylighting concept using basic rules of thumb. These preliminary concepts were tested and refined based on diffuse light analysis by Vladimir Bazjanac of Lawrence Berkeley National Laboratory. Bythe time Lam Partners, Inc. got involved with the project, the overall building design was basically set. We were presented with computer-generated exhibit hall renderings that were beautiful, but technically inaccurate. Daylight factors from model testing were accurate to the fifth decimal place, but limited in scope. Our role was to understand the spirit of the architecture, review the performance of the proposed design and then weave it into a comprehensive luminous interior environment that would serve as a comfortable and useful space while satisfying the green aspects of the project.

People like daylight in buildings because it gives them a connection to the outside. Daylight is a high quality, high color-rendering light source, but it’s always changing—daily, seasonally and with the weather. Good interior lighting can be simply defined as lighting that enables people to see what they want or need to see and to be comfortable while doing it. That being said, daylighting design is inseparable from electric lighting design and must consider the following:

  • Brightness balance

  • Distribution of light in space and time

  • Appropriateness of illumination levels

  • Harvesting techniques

During the lighting design of the interior environment, we always considered the visual needs of its occupants as equal in importance to the programmatic requirements of the facility. To start the process, we felt that part of our task was to actually demonstrate the performance of the proposed scheme with both diffuse and sunny conditions. This was accomplished with a physical model, a sundial, a tilt table and a video camera.

A physical model was preferred because it was much faster to analyze many different scenarios, such as weather conditions, time of day and time of year, along with different architectural factors such as skylight configurations, shading devices and glazing types. The videos of the model testing proved an extremely effective way to convey the issues involved. For example, they immediately confirmed the need for some type of shading of the strip skylights if direct sunlight on the exhibit hall floor was to be avoided. Originally, we experimented with a fixed baffle system within the strip skylight wells. However, based on programmatic requirements for complete solar blackout, an adjustable shading system was preferred. Tests were run on various materials such as woven fabric that would allow a view of the sky while reducing sunlight transmission by about 85%. Direct sunlight patterns were still visible on the floor, but with a greatly reduced contrast. A translucent material was also tested that was very effective in diffusing and spreading direct sunlight throughout the hall, but views to the sky were obscured.

An active shading system gave us the opportunity to vary the conditions for admitting and altering the available sunlight. We therefore proposed a combination of three different types of shading modes: no shade, diffusing and blackout. In the diffusing mode the opaque shade portion could be overlapped incrementally to “throttle down” the amount of daylight in the space until full blackout was achieved—a solar dimmer of sorts.

The shading system was planned for flexibility in controlling natural light. Therefore, we decided that the strip skylight glass should be clear, without frit, with the highest visible transmission possible. This mode proved to be desirable for such events as car shows, flower shows and most any other exhibit that didn’t entail video projection or special lighting.

Beyond analyzing solar patterns within the exhibit hall, we also performed model tests to help predict the amount of illumination from both daylight and sunlight. We found that the measured amount of illumination far exceeded traditional levels found in the interior of a typical “black box” exhibit hall. Furthermore, we discovered that the south-facing curved diffusing glazing alone, (known fondly by the design team as the “banana wall”) was contributing over 250 fc average on a horizontal plane at 30 in. above the floor on a sunny day. One could argue the need for no strip skylights at all, but the analysis clearly showed that these elements were responsible for distributing the daylight more effectively to the north side of the exhibit hall, thus contributing to a reasonable brightness balance. The overall illuminance levels, although quite high, felt appropriate and could easily be tailored to desired levels with the shading devices, based on various activities.

Innovation or dumb luck?

We felt that one of the greatest innovations of the project was the integration of electric lighting, which completed the whole interior luminous environment. Although there are very few convention centers that employ daylighting as the primary source of illumination, we know of none this large that use fluorescent lighting as the primary electric illumination source. The original design concept of using fluorescent lighting came as a reaction to the building’s architecture. In our experience, we’ve found that lighting design is much easier in a well-designed building since the architecture itself suggests the solution; details tend to fall into place. Using T5 high-output fluorescent lamps allowed us to pack a lot of light into a small package. The paired fabric ducts, spaced at 60 ft. on center, served as large-scale luminaires in which we were able to integrate custom lighting hardware and conform to the basic “energy” of the building.

Our preliminary electric lighting calculations showed that we would receive about 50 fc average on the floor plane. The indirect component of the electric light that reflects off the ceiling not only contributes illuminance to the floor plane but also provides visual brightness at times of low daylight availability, such as dark overcast days or late winter afternoons. The illuminance and exitance calculations were modeled with a computer program known as Lightscape.

The multiple, continuous fluorescent rows were relatively easy to control by simple switching in four stages, yielding four levels of illuminance tailored to harvest as much daylight as possible. This could not have been done with HID sources such as metal-halide since the warm-up, or re-strike, time could take as long as 15 minutes. The fluorescent lighting was also used as emergency lighting because of its instant-on capability.

As a result, this type of lighting blended seamlessly with the daylighting design. When compared to metal-halide sources, the fluorescent scheme resulted in a more environmentally friendly solution based on lower overall mercury content, recycling programs for fluorescent tubes and better lamp life.

Keeping perspective

Today’s technology offers architects and engineers a wide range of tools to assist with daylighting analysis of large building projects. Patterns of sunlight and quantities of daylight can be predicted accurately during the design process. However, ultimate success still rests in the careful balance of art and science, employing the massive amounts of data wisely and creatively.