Issues in Window Selection: Energy-Related
Heat Transfer Mechanisms | Measuring Properties | Overview of Energy Use | Codes and Standards
Overview of Energy Use in Commercial Buildings
The Concept of Zones
All commercial buildings are divided into zones which represent areas of the building served by discreet portions of the heating, cooling and ventilating system. There are also lighting control zones which may not correspond to mechanical system zone. In a sense, a mechanical system zone operates like a separate building, receiving heating, cooling, and ventilation from either its own packaged unit or a central system as needed. The reason for dividing a building into zones is that different spaces have different requirements and require separate control. For example, a highly-ventilated auditorium and a storage room with almost no ventilation would require separate zones. Separate zones are also needed for a north-facing space which may require heat in winter at the same time a south-facing space requires none because it is heated by the sun. Similarly, an interior zone with no windows may require cooling at the same time a perimeter zone with windows requires heating.
Some buildings like a school with long narrow wings of classrooms, may have mostly perimeter zones, while others like a massive office block may only have a small percentage of spaces on the perimeter. The use of a court or atrium in the center of a building creates interior perimeter zones that have some of the characteristics of both. If the atrium is large and well daylighted, these interior perimeter zones can behave like exterior perimeter zones with respect to light and view. If the atrium is fully conditioned however, there may be no heat loss or gain or air movement as there would be through an exterior wall. On the other hand, an atrium may be unconditioned or a semi-conditioned buffer space that behaves more like the outdoors with moderate temperature fluctuations and perhaps fresh air available.
Energy Loads in Commercial Buildings
Lighting, heating, cooling, fans, pumps, and any equipment or furnishings plugged into electrical circuits are the main uses of energy in commercial buildings. Electric lights, machines (i.e. computers, copiers, etc.), and people generate heat referred to as internal loads or internal heat gain. In an internal zone of the building there may be no other loads besides electric lighting, plug loads, and internal gains. Usually, these zones require cooling even in colder climates. Since ventilation with fresh air is also required in spaces with people, the outside air must be heated and cooled as well. Depending on the climate, during certain periods the outside air is the right temperature to provide cooling in an overheated space.
In a perimeter zone, heat losses and gains due to transmission through the building envelope (roof and walls) are also part of the heating and cooling loads. If a window or skylight is present, these losses and gains can be even more significant. Besides having relatively high rates of heat transmission, windows permit solar radiation to enter directly, resulting in major heat gain potential.
In some commercial buildings, the internal heat gains are greater than losses and gains through the envelope. This is referred to as an internal load-dominated building. This will naturally occur in massive buildings with a high ratio of interior to perimeter zones. When the losses and gains through the building envelope outweigh the internal gains, it is referred to as a climate-dominated or skin-dominated building.
The Effect of Windows on Heating and Cooling
At the scale of the perimeter zone, the role the window can be significant. Windows are the modulator of heat, light and air. The overall energy use patterns of the internal zones do not directly affect window design and selection. Windows have an important influence on the energy use and people in the perimeter zone, even if it is a small percentage of the total building floor area. If enclosed rooms such as private offices are on the building perimeter, then these automatically define the perimeter zone by their dimension, usually 10 to 20 feet. In larger spaces, such as an open office area, the depth of the perimeter zone can vary. The heating and cooling effects may only influence the first 10 feet near the windows, but the daylight may penetrate up to 25 feet or more if properly designed. The desire of occupants for daylight, view, and fresh air is leading to buildings that are thinner in profile with more perimeter and less interior zones.
As shown in Figures 2-19 and 2-20, a typical perimeter office has either heat gains or heat losses through the roof, walls, and windows. The impact of window choice on annual energy use is illustrated in Figure 2-21 for a 15-foot-deep south-facing perimeter zone in Chicago. Heating energy use diminishes considerably in Chicago as the window U-factor improves from 1.25 (Window A-clear single glazing) to 0.14 (Window I-quadruple glazed units). High performance windows also reduce cooling energy use. In Chicago, lower electric use for cooling corresponds to windows with lower SHGC (both lighting and other equipment use the same amount electricity in all cases). Although the effect is not show here, operating windows may reduce mechanical cooling costs by providing natural ventilation during certain periods of the year.
The Effect of Windows on Lighting Energy Use
In addition to transmitting heat gains and losses, windows and skylights also transmit light. Typically, this natural light is a desirable amenity but the electric lights continue to burn resulting in no energy savings. In recent years, methods and technologies to use this daylight to reduce electric lighting have emerged. Daylight is brought into the building by sidelighting with windows or toplighting with skylights, light monitors, or clerestory windows. The examples throughout this book focus on sidelighting but the general concepts and performance issues apply to toplighting as well.
The first requirement for an integrated daylight/electric light system is that lights are controlled in a way that allows for the energy reduction to occur. For example, lights near windows must be switched off separately from the rest. In addition, individual fluorescent tubes within light fixtures may be switched separately allowing for a range of light levels instead of only 100 percent on or off. Dimmable light fixtures also permit electric light levels to be reduced. To take advantage of the natural light from a window, either people or automatic controls must switch off the electric lights. Occupant switching can be effective but requires active participation and usually will not be done optimally to reduce energy use. If the daylighting is plentiful and uniformly distributed, there is a greater chance that people will switch off the lights. Portions of the electric lighting can also be switched off or dimmed automatically in response to a photosensor. This type of system is designed to operate optimally without depending on occupant participation. However, these systems are more expensive than simple switching and represent emerging technology where their installation and operation must be carefully monitored to ensure the projected savings.
Figure 5-36 illustrates the impact of using a daylight control system in a south-facing perimeter zone in Chicago. Compared to the cases where there are no daylight controls, considerable reductions in electric lighting energy occur for most window types (except Window D which has a very low VT). It can also be noted that without as much heat gain from the electric lights, cooling energy use is slightly smaller and heating energy use slightly greater in many cases.
The perimeter zone used in these examples is a 10-foot-wide by 15-foot-deep enclosed office with a 9-foot-high ceiling. A 6 x 6 window (36 square feet) is located on the exterior wall. There are no special techniques such as light shelves used to project daylight deeper and more evenly into the space.
While all the electrical energy use comparisons in Figure 5-36 show significant savings by using lighting controls, they assume conditions with reliable dimmable ballasts controlled by daylight sensors. In reality, there can be a range of performance depending on the design and operation of the system. Even if today's dimmable ballasts and light sensors may not provide optimal performance or be cost effective, they are likely to be in the future. Since the window system has a relatively long life, it should be designed based on the assumption that daylight control systems will be installed in the future.
This interaction between the building envelope and lighting system is one of the key synergistic opportunities in developing high-performance buildings. The design and selection of windows is a pivotal aspect of this integrated design approach.
The Effect of Windows on Mechanical System Design
Traditionally, windows have affected mechanical system design in two ways--they make perimeter heating necessary and they increase size of the mechanical equipment. Areas near conventional windows are usually places of the greatest temperature variation and discomfort in a building. While a majority of the heating and cooling in commercial buildings is delivered through forced air HVAC systems, additional radiant perimeter heating is often required near windows. High performance windows with low U-factors reduce the heat loss through the windows significantly and glass temperatures remain relatively high. At heat losses below 450 Btu per hour per linear foot of wall, baseboard heat is not required. At 250 Btu/hour per lineal foot of wall, slot diffusers at the windows are not necessary either (Grimm and Rosaler, 1998). In the examples shown in Figure 2-26, the triple and quadruple glazed windows (H and I) would clearly not require perimeter heating in Chicago.
The second impact of high performance windows is the reduction of the mechanical system size due to lower peak heating and cooling loads. This has a cascading effect on reducing initial costs. Reduced peak loads means smaller chillers and boilers, smaller ducts, and smaller fans. In addition, since electric peak loads usually occur on summer days when demand charges are highest, windows that reduce peak loads can result in energy cost savings as well. Reduced peak loads also lessen the need for additional power plant capacity.
Reducing the SHGC of the window reduces the peak demand. If daylighting controls are used in the perimeter zones, peak loads are reduced even further.
This interrelationship between windows, mechanical system, and lighting system again illustrates the potential benefits of the integrated systems approach to design. Even though higher performance windows and daylight controls may cost more initially, they may be offset by reduced costs for many components of the mechanical system and the elimination of perimeter heating. In addition, operating costs will be less and people are likely to be more comfortable and productive.