To conduct the test, a representative mockup of the specific curtain wall system must be prepared, measuring no larger than a single floor-to-floor height and 4.6 m (15 ft) wide. The mockup must contain wall components or details that are representative of the wall system and have been identified as possible condensation control concerns. All parts of the specimen must be full size, using the same materials, methods of construction, and anchorage as specified for the actual project. Portions of the interior wall representative of project conditions (e.g. drywall, insulation, etc.) should be included to at least 305 mm (12 in.) from the edges of the specimen to properly evaluate the perimeter conditions. Each operable element of the specimen must be securely closed and locked prior to testing. Any chamber fixture that is not included in the finished project (e.g. observation decks/floors and non-project support steel) and is located within 610 mm (24 in.) of a thermocouple location must be insulated. It may also be useful to include fenestration covering, such as blinds and curtains, if they could impact the eventual occurrence of condensation.
The mockup specimen is sealed into the opening of an insulated laboratory test chamber. The insulated outdoor side of the chamber is equipped with a means to lower its ambient temperature to the specified exterior winter design temperature and simulate exterior airflow. The latter is accomplished using wind generators to provide a flow of 5.5 ± 1.3 m/s (12.3 ± 3 mph), as calibrated using a wind speed measuring device (i.e. anemometer), applied so that the flow is generally parallel to the long face of the specimen test area being evaluated. The indoor side of the chamber is equipped with a means to both control temperature and limit RH to prevent the formation of condensation in the areas where surface temperature data is recorded. Indoor air circulation, used to minimize temperature stratification and fluctuation, is limited to 1.6 km/h (1 mph).
The interior surfaces of the test specimen are instrumented with a total of at least three sensors, such as Type T thermocouples, located near the top, middle, and bottom of the test area. Alternate temperature measuring equipment (e.g. non-contact pyrometers) may be used so long as the accuracy meets or exceeds that of ASTM C1363-19, Standard Test Method for Thermal Performance of Building Materials and Envelope Assemblies by Means of a Hot Box Apparatus. The thermocouples are taped tightly to the surface using aluminum foil tape and painted flat white or alternatively to meet the emissivity of the surface being analyzed. The exterior surface of the test sample may also be instrumented with thermocouples, but this is not a requirement for this method.
The test apparatus must control the average ambient temperature within ±2 C (±4 F) of the specified set points. Additional thermocouples are positioned at least 75 mm (3 in.) away from the wall specimen to measure interior and exterior ambient air temperatures. Individual ambient air thermocouples readings may vary more than the ±2 C range so long as the average of all thermocouples remains within these bounds.
The specimen is maintained at the specified interior and exterior conditions until the interior surface temperatures have reached steady-state conditions, the time for which varies with the material composition of the wall, such as masonry versus metal components. The interior and exterior air temperatures, surface temperatures, and RH are monitored and recorded for a period of not less than two hours after steady-state is achieved. All surface temperature data is compared to the specified dewpoint for the project, and all locations at which temperatures fall below the dewpoint are identified. All evidence of condensation is recorded, including quantity and location, and documented using photographs.
AAMA 501.9 enables building professionals to specify meaningful testing to ensure intended performance prior to occupancy. Since large curtain wall mockups require a lot of planning, coordination, and expense to build, it is often advantageous to add some of the other optional tests as well. After all, in the final analysis, testing is a lot less expensive than revisions after the fact.