Guide to museum glazing: Museums demand the best

Museums and art galleries present challenges to glazing designs, given the tight environmental conditions that must be met. Consideration must be given to many parameters, often conflicting, including the quality and quantity of light, ultraviolet resistance, aesthetics, thermal performance, condensation resistance, strength, security performance, safety and cost. For a best-practice curatorial environment, temperature and humidity must be contained within a tight band of 70 degrees Fahrenheit and 50 percent relative humidity year-round. Artifacts and artwork must also be protected from UV and visible solar radiation at the same time as being displayed in “natural” light.

Designing for light control
A simple approach to lighting control is to omit glazing entirely and use only artificial light. While this approach gives good control over UV and light levels, the quality of light can be compromised. As a result, more museum and gallery curators adopt natural daylighting solutions that display, for instance, an artwork’s colors in the lighting environment experienced by the artist when creating the artwork.

It is now possible to simulate accurate daylighting performance with computer software. Such analyses must take into account more than just a sun-path study indicating shadows and direct sun patches but must include transmission and diffusion through glazing and also reflected light off internal surfaces. Once an initial daylighting study has been carried out, the location and size of windows and skylights, location of sunshades—fixed or operable—and types of glass can be determined.

The total amount of light entering a gallery space, called total lux hours, is a governing criterion for a curatorial space. This approach to light control allows short direct sun patches or hot spots to occur, provided that lower light levels exist at other times. A hot spot can have 1,000 times the intensity of background light, thereby creating a major advantage to a full-time diffuse light solution.

Architects for the High Museum of Art in Atlanta adopted an indirect light solution with a series of 3-foot diameter north-sloping skylights being shielded by ‘bishops’ hat’ sunshades (see story, p. 84). The Wexner Center in Columbus, Ohio, allowed hot spots within the gallery but not on any display walls. Diffusing and dark glass was selected to minimize the effects of the hot spots.

How to choose glazing
A color rendering index provides an average measure across the visible light spectrum of the difference in changes between the color bands. Colored glass, including normal clear glass, and colored coatings reduce the CRI. Neutral colors may reduce the total amount of light passing through but do not adversely affect the CRI. Low-iron glass provides the best CRI and now widely selected for use in art galleries. From a CRI perspective, low-emissivity coatings required for enhanced thermal performance become an unhelpful burden but an essential component of the glazing. The dark but neutrally tinted glass adopted for the Wexner gallery for solar control—and to match the original glass patterns—did reduce the CRI slightly, but with the use of low-iron glass for a sizeable percentage of the walls and skylight, a good quality of light was still achieved with good solar control.

Resistance to ultraviolet light is naturally provided by the polyvinyl butryal interlayers in laminated glass and regularly sold as such. This natural resistance remains true for a limited bandwidth of UV radiation. For curatorial environments, a broader band UV resistance will typically be required.

Designing for heat loss
At the macro level, the thermal performance of any glazing system must be sufficient to maintain internal gallery temperatures where artifacts and artwork will be displayed. At the micro level, condensation formation on internal surfaces must be predicted and ideally avoided for temperatures down to the low external design temperature.

For a cold condition design, “simple rectangular” museum and gallery space conditions can be predicted by conventional mechanical design software. For more complicated spaces, computational fluid dynamics analyses predict air flow and temperatures. Such CFD analyses provide much more information than required for glazing design. They can greatly assist with the mechanical design of a space and allow an optimal design for duct locations and input air temperatures and velocities. Related to glazing, a CFD analysis also helps determine the need for fin tube or trace heating at mid-height locations on tall glass walls.

Mechanical and glazing combos
Engineers working on the High Museum, the Wexner Center and the Cleveland Museum of Art extension all relied heavily upon CFD analyses to fine-tune the interaction of mechanical systems and glazing design.

In designing for macro-level performance, in either simple or CFD mechanical analyses, it is important to use the overall system U-value, not just the center-of-glass U-value. Frame effects reduce thermal performance because of direct thermal conduction paths from inside to outside through glass spacers, glazing seals and gaskets and framing. To optimize overall system thermal performance, frame and glass-edge performance should be matched to glazing performance. For example, there is little point in using a thermally broken framing system if single glazing will be used. Likewise, using a thermally unbroken frame or one with a very shallow thermal break with triple glazing becomes a waste of time.

As glazing tilts off the vertical, as with skylight applications, thermal performance decreases, up to 40 percent in some instances. This happens because a “short circuit” of the vertical air flow within an insulating glass unit can occur, creating a shorter air path between warm and cold inner and outer glass surfaces. This proves the case for all applications but becomes more critical for museums and galleries due to more demanding internal conditions and the unacceptability of dripping condensate.

Designing for condensation control
Condensation forms when the surface temperature of glass or framing falls below the dew point, the temperature where airborne moisture condenses. The warmer the air, the more moisture it can hold. Relative humidity, expressed as a percentage, measures the actual amount of airborne moisture, called vapor pressure, against the absolute amount of moisture the air can hold at a given temperature. The dew point temperature is reached when relative humidity reaches 100 percent. For a fixed vapor pressure, the relative humidity rises as the temperature falls. For a typical art galley environment of 70 F and 50 percent relative humidity, the dew point temperature is 50.5 F. That is, if any surface temperature falls below 50.5 F, then condensation will occur.  As a comparison, in a typical office environment of 70 F and 30 percent relative humidity, the dew point temperature is 37.2 F—much easier to design for.

Selecting an appropriate external design temperature is not straightforward. Engineers plan typical mechanical designs based upon the American Society of Heating, Refrigerating and Air Condition Engineers’ ASHRAE Heating Dry Bulb 99.6 percent temperature. This is a statistical value and the temperature will fall below this measure about 35 hours per year, on average. For a museum or art gallery, this tolerance may not be acceptable and a lower design temperature may be necessary, though this should be measured against the added cost of designing for a worst-case temperature.

Given the above, designing for condensation resistance gets down to micro-level design. We recommend carrying out solid thermal modeling to check the full internal surface profiles for areas with temperatures below the dew point.

Explore the following options to improve performance:
•   Low-e coatings. They provide approximately 50 percent better thermal performance than uncoated glass. Note the adverse effect that low-e coats may have upon CRI.
•   Triple glazing. It provides approximately 50 percent better thermal performance than double glazing.
•   Argon-filled insulating glass. It provides approximately 15 percent better thermal performance than unfilled triple glazing.
•   Warm-edge technology. Insulating glass spacers such as  stainless steel spacers have better performance than traditional aluminum spacers; a number of proprietary spacers, based upon polymers, perform even better.
•   Thermally broken framing. Performance becomes a function of thermal break depth, so a shallow thermal break does not perform as well as a deep spacer. Consider structural glazing as a poor man’s thermal break.
Active solutions to condensation control may typically be adopted in tandem with passive glazing designs. Active solutions include local heating of the glazing and framing through ducted air, fin tube or panel radiators or electric trace heating.

In some instances, condensation may be allowed to form and be controlled through an appropriately designed drainage system. Typically, invisible glazing at a slope greater than 20 degrees off horizontal—to prevent condensation dripping—is the only case where condensation formation would be considered.

Thermal tests of framing to predict condensation performance is possible following the American Architectural Manufacturers Association’s AAMA 1503-98 test method. This method, based upon standard sample sizes and test conditions, determines the condensation-resistance factor for frame and glass. CRF is a useful preliminary guide for frame and glazing selection, as it fairly rates relative performance between systems. However, it is not possible to derive direct condensation performance on all frame surfaces from the CRF, as it is based upon average temperature readings in a number of locations; so, make direct calculation as described.

Nonstandard testing of actual sample sizes, shapes and environmental conditions should not be undertaken by the unwary if reliable results are required. For the High Museum and Wexner Center, for instance, we conducted two full-scale mock-up tests using actual humidified internal air to visually inspect where condensation formed, in addition to relying upon thermocouple readings, to prove the design assumptions and verify the analyses. Close work with the researchers at the testing labs is required to run the tests in a reliable manner and to collect reliable data.

Other design issues
When designing for strength and deflection, museums or galleries do not have specific requirements. In designing for security, however, many museums and galleries rely on full-time guards monitoring closed-circuit television and movement alarms as their principal means of defense against theft. Laminated glass is almost universally used for UV protection, so there is at least some resistance to incidental instances of accidental glass breakage, break-ins and over-pressure.

High Museum, Atlanta
Many of these lessons can be seen in recent construction of the a main gallery for the Renzo Piano Building Workshop-designed expansion of the High Museum. More than1,000 3-inch-diameter skylights accent the upper galleries. These consist of a sloped glass layer 2 feet above the main roof line with a 5-foot-deep shaped tube below integrated into the ceiling and Bishop’s-hat sunshades on page 69. Piano’s arrangement was carefully analyzed from a lighting perspective to prevent any direct sunlight from entering the gallery spaces below. Low-iron glass was selected to minimize reduction in CRI. A low-e coating was used for increased thermal performance with a consequent but acceptable reduction in CRI. The complicated air space within the tube directly below the glazing was analyzed using CFD to predict actual air temperatures directly below the glass to be used in a condensation risk assessment. A mock-up of a skylight, including the tube below, was tested with a custom test box and procedure using actual humidified air to verify the design assumptions.

The Wexner Center, Columbus Ohio
Eisenman/Trott Architects’ Wexner gallery has recently undergone refurbishment that included replacing existing skylights and glazed walls. CFD modeling of the interlinked galleries was carried out to predict the air flows and temperature profiles. This analysis provided a realistic design air temperature to be used in the condensation analysis and enabled the omission of fin-tube heaters at mid-height levels on the glazed wall. Triple glazing with stainless steel spacers, argon gas and a low-e coating was used to minimize U-value. This was glazed into a thermally broken standard framing system. A large scale mock-up consisting of a 20-foot vertical section of wall and a 12-foot section of skylight was thermally tested, including full humidity control, to verify the design. From a lighting perspective, a full daylighting study was carried out on the gallery spaces to determine that hot spots did not occur on display walls and to establish lighting levels. A combination of body-tinted glass and opacifying and translucent interlayers was used to match the existing glass nomenclature and to provide the required lighting control.

Cleveland Museum of Art
The expansion to the Cleveland Museum of Art by Rafael Vinoly is in the design stage. It includes a number of cladding elements designed with condensation resistance in mind. A structural glass connector joins the existing gallery to the expansion. This was carefully analyzed using CFD to determine the maximum temperature of inlet air to provide the maximum amount of warming while still keeping the space comfortable and establishing that the local conditions within the connector did not extent into the gallery spaces by more than a small distance. Rather than using trace heating to control humidity formation, the glazing details were adapted locally to allow warmth to penetrate into the body of a few critical details. For example: The main new galleries have large glass walls. These were designed as cavity walls to meet the aesthetic requirement for frameless glazing with good condensation performance, so dryer and warmer air then allowed in a gallery space can be used. The main atrium space includes a potential major thermal bridge at the main roof support, however, three-dimensional solid thermal modeling demonstrated that this was not a problem area. In addition, back-of-house curatorial spaces have “conventional” triple glazed and thermally broken framed windows.

These museum projects and countless others prove fascinating to work on for every construction professional and provide visitors with lasting symbols of the fenestration arts.

Jack S. Blanton Museum of Art
www.blantonmuseum.org  
Congress Ave. & MLK Jr. Blvd.
Austin, Texas
Opening: February 2006
Architect: Kallmann McKinnell & Wood Architects, Boston
General contractor: SkanskaUSA, Parsippany, N.J.
Lighting design: Arup Associates, London
Contract glazier: Win-Con Enterprises, New Braunfels, Texas