Solar design for windows

by Catherine Howlett | April 1, 2013 3:50 pm

[1]

by Karol Kazmierczak, ASHRAE, NCARB, LEED AP
Since glazing is the most advanced and expensive part of many façades, it warrants a good design that goes more than skin-deep.

Glass can be engineered to provide natural light, limit occupant discomfort, and make energy use more efficient, while maintaining the appearance desired by architects. The ‘coolness factor’—balancing the transmission of heat and light—remains the most important and, ironically, least-known performance characteristic of architectural glass.

In an era of widespread curtain walls and sloped glazing merging into vertical planes, the definition of what constitutes a ‘window’ can be thought of mainly in the context of code requirements. Unfortunately, it can seem like the International Building Code (IBC) treats the window as too obvious to specifically define. To make matters more complicated, the code introduces the noun “glazing,” which, depending on context, could be interpreted as a synonym of ‘glass’ or ‘window,’ but not always.

One of the frequently quoted industry standards, American Architectural Manufacturers Association/Window and Door Manufacturers Association/Canadian Standards Association (AAMA/WDMA/CSA) 101, Standard/Specification for Windows, Doors, and Unit Skylights, defines a window as:

an opening constructed in a wall or roof and functioning to admit light or air to an enclosure, usually framed and spanned with glass mounted to permit opening and closing.

IBC seems to agree with the assessment that a window’s primary function is to provide natural light, and establishes a bottom threshold:

Section 1205.2–Natural light. The minimum net glazed area shall not be less than 8 percent of the floor area of the room served.

This command would have been difficult to meet several hundred years ago, when glass was prohibitively expensive, production limits were severe, and real estate tax was charged per window in some places. (This is why some Parisian housing was purposefully built without windows.) Now, the requirement seems simple, but it is still frequently misconstrued. Strictly read (and against the apparent code intent), it allows opaque, glazed spandrel areas to be counted against the eight percent floor area demand. This can be problematic.

A good example would be a building designed with a visible light transmittance (VLT) in the single digits (e.g. five percent) for vision glazing in windows sized to comply with the minimum eight percent window-to-floor ratio mandated by code. This is akin to wearing two pairs of sunglasses. Such a design may be understandably objectionable to building occupants, regardless of the demonstrated building code compliance.

The aforementioned eight to 10 percent code requirement was developed with hindsight in days when dark-tinted glass was only used in vitrages, and the glass installed in windows was invariably clear, with the visible transmittance roughly 80 percent. Therefore, substituting five-percent VLT glass would logically require a net-glazed area approximately 16 times larger, making the interpolated window-to-floor ratio close to 130 percent. This is seldom feasible in typical construction.

Codes, terminology, and coolness
As its name suggests, VLT is the percent of total visible light (wavelength ranging from 380 to 780 nanometers [nm]) coming through the glass system—it is from zero to 100 percent, the lower the number, the less visible light. This is the only glass characteristic that can be verified by the traditional way architects choose the material: by comparing samples.

A better way can be found by trying to achieve the Indoor Environmental Quality (EQ) Credit 8.1, Daylight & Views, under the U.S. Green Building Council’s (USGBC’s) Leadership in Energy and Environmental Design (LEED) program. This entails a minimum two percent daylight factor (DF) in 75 percent of all occupied space for critical visual tasks.

[2]

The daylight factor is a measure of expressing the daylight availability in a room. It is a percentage ratio of the instantaneous illumination level at a reference point inside a room to what is occurring simultaneously outside in an unobstructed position, calculated under an overcast sky (excluding any direct sunlight). Using the same computer model, it is easy to analyze glare, as well as natural and artificial illumination as an additional benefit. Ultimately, an architect could specify VLT based on the DF analysis, and then correlate it with other necessary physical glass characteristics, which is explained later in this article.

Of course, when not properly specified, glass that brings in copious light can also bring in copious heat. This can mean an increase in the building’s energy consumption, as air-conditioning systems kick in to prevent occupant discomfort.

Another of the glass benchmarks allowing comparison of different glass types, solar heat gain coefficient (SHGC) illustrates glass transmittance of solar radiation, as opposed to the U-factor, which is for thermal transmittance. According to the European Standard (EN) 410, Glass in Building: Determination of Luminous and Solar Characteristics of Glazing, cited by the National Fenestration Rating Council (NFRC) standards, SHGC is the glass transmittance in range from 300 to 2500 nm. This range embraces ultraviolet (UV) rays, visible (VIS) range, and near-infrared (NIR) range, which is also called the ‘solar’ or ‘short infrared’ range.

The architectural glazing industry spends a lot of effort in increasing the ‘coolness’ of glass, illustrated by the proportion of the SHGC to the VLT. This is the most important characteristic of modern architectural glass, and part of the historical quest to bring natural light deeper into buildings while avoiding occupant discomfort.

[3]

With respect to glass coolness, the low-emissivity (low-e) coating is one of the most important developments. Most typically placed on the #2 glass surface (i.e. the interior surface of an exterior ply of glass), it reduces the portion of heat absorbed and radiated by the inner ply of glass. Part of the transmitted shortwave infrared radiation is absorbed by the glass itself and re-radiated as heat. (Therefore, generally speaking, tinted glass should be positioned closer to the exterior because it absorbs more radiation than clear glass.)

A useful tool for visualization of the infrared glass performance involves placing different glass samples between an infrared lamp and NIR sensor. Seemingly identical glass would exhibit different solar performance. A similar comparison could be done outdoors, using the sun in lieu of an infrared lamp, and an inexpensive pyrometer instead of the NIR sensor. These elementary demonstrations enable the project team to visualize the performance described by otherwise meaningless acronyms and numbered labels.

Dealing with the heat
This author was recently approached by an architect who specified and designed a large window replacement project in his own condominium building complex in Long Island, New York. Ironically, he found the new fenestration to be unbearably cold, drafty, and covered with water condensation. Some suggest the secret of success is to know how to pass the blame—therefore, he asked how to prove the contractor provided inferior windows. This author suggested relevant field testing that could be used to verify whether the specs were met.

In this author’s experience, most field testing is performed solely to confirm inferior specifications were followed. A lot of money and aggravation would be spared if owners verified the adequacy of the architectural specifications before embarking on expensive testing. Even more could be spared if the verification happened before the construction.

[4]

As experienced by the architect in question (and his family and clients), the heat transmittance of different glass types is easily noticed. There are receptors in our skin that detect the rate of heat loss, making it simple to tell a good window from a bad window. For benchmark purposes, this is described as the U-factor; it illustrates thermal transmittance in range from 2500 to 50,000 nm (the ‘long’ infrared range).

As perimeters (i.e. framing and glazing spacers) are typical bottlenecks of the U-factor, it pays to design them correctly. This author often analyzes a glazing design in the finite element analysis (FEA) computerized testing, and typically suggests simple and inexpensive improvements such as stainless steel spacer substitution, if glazing failed the virtual testing.

To put the whole thermal transmittance (i.e. U-factor) discussion into perspective, it is important in locations where the ambient temperature differences are large, like Scandinavia. In places like the United States, which mostly lies south of Europe, designers should be primarily concerned with the solar heat gain. (As a bonus, the economics are there—in most cases, the better SHGC is almost free, while the better U-factor sometimes comes with an unreasonably long return on investment [ROI].)

Another obvious factor is the draftiness (air infiltration)—its reduction can be challenging in the sliding fenestration popular in the United States, requiring core solutions such as continuous gasketing, stiff frames, and mitered framing and sash corners.

Understanding climate differences is critical—a better window in the North is not necessarily the same in the South, and may require adjustments in the mechanical system. This author was recently approached in Miami, Florida, by a mechanical engineer whose son contracted chronic respiratory illness after old projecting windows in his house were replaced with new, hurricane-proof fenestration. These assemblies are stiff and, in turn, much more airtight, causing a significant drop of the cooling load in hot and humid climates. Therefore, the air-conditioner, whose evaporator acts as a dehumidifier, works less, contributing to the inferior interior air quality (IAQ). This is why retrofitting airtight windows may require adding ventilation and dehumidification.

[5]

Northern Europeans, with their heavy and tight masonry buildings relying on gravity ventilation, learned it the hard way—they install window frame venting slots that automatically open when relative humidity (RH) is too high inside.

The energy conservation codes and standards generally tie the U-factor in with the glazing ratio for prescriptive compliance method. Figure 1—reproduced from American Society of Heating, Refrigerating, and Air-conditioning Engineers (ASHRAE) 90.1, Energy Standard for Buildings Except Low-rise Residential Buildings—ends at the 50 percent window-to-wall ratio (WWR) and five percent skylight-to-roof ratio. (As this is only a prescriptive path, designers can move away from these charts [e.g. using either a trade-off or energy cost budget options]. However, ‘off-the-chart’ projects are sometimes found uncomfortable by occupants, and not code-compliant when carefully examined.) The reason is simple: thermal performance of glazing is either far worse than that of opaque assemblies or prohibitively expensive (with negligible ROI).

The human body generally cannot directly and instantaneously detect solar radiation other than the VIS range, but NIR radiation is absorbed by objects (including skin) and re-radiated in the thermal range, which is felt as heat. The NIR radiation stretches from 780 to 2500 nm. This phenomenon contributes to the greenhouse effect because glass lets the NIR through, but stops the thermal range.

[6]

Unless one’s intention is to experiment with passive solar design, the heat gain via windows is generally undesirable, particularly in warmer climates. This author knows of a famous, ostensibly green, LEED Gold-certified, fully glazed office tower built in the South with the glass VLT specified lower than the SHGC. This means the glass coolness factor was less than 1, with approximately 50 percent window-to-floor ratio. Both SHGC and VLT are in their mid-30s. The ultimate result is the occupants are hot, and the cooling bills high.

This author is not aware of any daylighting studies performed on this tower. However, under the common-sense assumption the 15 percent VLT would be just enough (given the high window-to-floor ratio and good window access), then the SHGC could theoretically be four to five times lower should the more spectrally selective glass (at almost exactly the same initial cost) have been specified. Since the building elevations are fully glazed, the solar heat gain correlates well with the cooling energy costs. Essentially, the energy bills could probably be two times lower and occupants twice as happy.

This case illustrates the difference between the two alternative ways of approaching the solar heat gain challenge—either meeting the minimum threshold set for the cooling sizing, or the threshold set by glass production limitations with respect to natural lighting. In many cases, due to architectural over-glazing and improved glass technologies, the latter would result in a better glass, translating into a lower energy consumption and higher occupant comfort in cooling-dominated places like the United States. This is why this author believes determining the SHGC should begin with daylighting studies.

[7]

Sticking to vision
With glass, the most important light obstruction is dirt, and deferred maintenance has become the norm thanks to the current economy. This author has seen opportunities for doubling the light transmittance, had only the glass been regularly washed in the most severe cases. A ‘self-cleaning’ glass has been developed with a titanium hydrophilic surface. Unfortunately, this material is occasionally specified for flat skylights or under large overhangs where it serves no useful purpose because it requires sufficient rainwater flow to clean itself. The best way to ensure clean glass is to design for permanent window-washing access.

Another important way to improve visibility is to reduce reflections. To this end, anti-reflective glass can be particularly useful for showroom and retail storefronts, where it can better expose the merchandise on display. This way, one can only gain around three or four percent, but significantly improve the subjective perception of transparency. High glass reflectivity, on the other hand, may be desired in some applications where privacy is needed—a common example is a multi-family high-rise building.

A third way to increase transparency involves eliminating iron from the glass melt. Iron makes glass slightly greenish, which is more easily detectable in thicker materials. The two to four percent gain is justified in some thick, high-end applications, requiring numerous plies of laminated glass to achieve the necessary structural redundancy. Such a low-iron glass may be advisable for high-tech additions to historic landmark buildings.

The thicker the glass, the less light goes through. The necessary glass thickness can be estimated using ASTM E1300, Standard Practice for Determining Load Resistance of Glass in Buildings. The result should be the glass thickness of each glass ply. Then, using the free Windows and Optics software developed by Lawrence Berkeley National Laboratory (LBNL) or similar programs, a designer can select the sufficient VLT for daylighting combined with the lowest possible SHGC before ordering samples for comparison purposes.

[8]

Window design basics
Designing the most appropriate window can begin as soon as floor plans are developed. One should pick the worst-case layout and run the DF analysis to optimize the size and VLT of the glazing and shading. Based on this VLT, and using the chart in Figure 2, a designer needs to select the feasible SHGC range. This should be verified against energy code requirements, which in the case of many large buildings is met by the energy cost budget path. Therefore, a window’s total SHGC should be coordinated with the mechanical Basis of Design (BOD), then verified with the mechanical engineer, and re-calculated to receive the nominal glass SHGC by removing projections, shading, framing, and screens. (Calculating shading offered by miscellaneous architectural assemblies can be challenging; this author advises architects to compare shading options side-by-side with energy-modeling software.)

Additional challenges come into play when assumed and calculated physical characteristics must be captured in the construction documentation for the benefit of future construction. These conditions, and others affecting engineering estimates, should be communicated with all parties to avoid misunderstandings. Ultimately, the resulting numbers illustrating the glass performance should be reflected in MasterFormat’s 08 80 00–Glazing specification section or on its referenced glass schedule.

[9]

The typical large project can use 20 or more types of glass for optimal performance. Therefore, at this point, it would be better to leave the further refinement of the glass specification to the delegated design teams. Unfortunately, some architects pick the glass based on how it looks, leaving the design entirely up to the contractors, and often ending up unpleasantly surprised with the results.

The use of building information modeling (BIM) and various engineering simulations are sometimes criticized by those who note the discrepancies between the virtual and real world. Suffice it to say, software is only as good as the data fed to it. Such information includes physical characteristics of materials and spaces, including visible transmittance and reflectance of surfaces, which are often misunderstood by architects.

This author was once asked to calculate the daylight factor separately for mirror-like east and west quarters of a building, but the project’s architect did not respond to questions relating to the building surroundings. She also admitted ignorance with respect to material properties, such as surface reflectivity, which is important for such analysis. Therefore, predictably, east and west DF studies yielded identical results (at a double price), and neither result of the analysis was expected to emulate the real world.

Site conditions are not always fully taken into account. Walls and roofs of adjacent buildings and water surfaces sometimes send a strong reflection in the direction of a building, which should be addressed in the glazing design.

Other items to consider
Several factors besides glass performance affect a window’s VLT and SHGC.

Shading
Many aspects, such as external and internal shading, louvers, framing-to-glazing ratio, and opening projections all have an impact on physical glazing characteristics. In some cases, printed advertising media placed on a building can lower interior daylight factors significantly; in other instances, poorly designed sunshades can not only add to cooling energy expenditures, but also increase the need for electrical lighting when occupants keep shades down and lights on.

HVAC and interior design coordination
This author often finds a room’s windows are frequently not addressed by the HVAC system. In northern climates, this leads to damage of moisture-sensitive materials by water condensing on windows. In the South, lack of air-conditioning in part of the room adjacent to hot windows renders this area unable to meet ASHRAE 55, Thermal Environmental Conditions for Human Occupancy, and, therefore, the criteria of a conditioned space.

For example, this author was recently in a large media conference room in a new building in the Bahamas. The half of the room adjacent to the south-west elevation featured large strip windows and was not air-conditioned, with diffusers located close to the entrance doors and the return grilles.
A speaker was placed in a literal ‘hot spot’ because the podium was located in front of the windows.

The excessive window-to-floor ratio is sometimes caused by lack of coordination between building shell and core designs. Tenants often design and build the spaces long after the shell is constructed. This includes air-conditioning and internal shading.

Internal shades
Most sunshades are ‘roller’ furnishings coming down from a window head. Therefore, any mid-position leaves a horizontal strip of light above the sill (where glare hurts eyes the most and lights the room the least). Were the shades operated upside-down (or translucent strips introduced in a double configuration), a bright strip could be left at the top, providing some useful daylighting, saving artificial lighting, and preventing glare.

Cheap fixes
When a problem arises, there can be temptation to employ a ‘Band-Aid’ approach. However, in some cases, this may hurt more than help. This author is wary of interior-applied items that increase the glass temperature, such as the common aftermarket tint films or translucent insulation panels.

When incorrectly specified, they may induce glass breakage because the edge of the glass stays cool, chilled by its framing. It may be akin to pouring hot water on the center of a glass dish freshly removed from a fridge. A good example is glass cracking induced by users placing an aluminum kitchen foil behind a window glass to limit the sun heat transmission, frequently observed by this author in South Florida. For this reason, such an approach often voids the glass warranty; therefore, the glass fabricator should be consulted first.

 

Karol Kazmierczak, ASHRAE, NCARB, LEED AP, is the senior building science architect and president at Building Enclosure Consulting LLC. He has 17 years of experience in envelope design, engineering, consulting, and inspection. Kazmierczak can be contacted via e-mail at info@b-e-c.info[10].

Source URL: https://www.constructionspecifier.com/solar-design-for-windows/

---