ASHRAE Journal - June 2009 - 42

During the underfloor testing of each floor, the following AHUs could be tested individually and successfully. These unprocess was followed. (The data below was used for the second derfloor Categories 1 and 2 leakage tests were made soon after a floor was substantially completed, and its floors were installed/ floor to show the math for one typical AHU zone.) The second floor was sealed all around with an underfloor sealed and air-handling systems were able to be run. Automatic area of approximately 42,440 ft2 (3943 m2). Design airflow for control systems were used to monitor all of the control points the AHU was approximately 20,000 cfm (9437 L/s), or 0.47 and handheld test instruments were used to calibrate and verify their readings during the testing. cfm/ft2 (2.39 L/s·m2) After the underfloor was successfully tested as totally sealed Floor gaps were measured at an average of 0.008 in. (0.2 mm) gap, and at 0.1 in. w.g. (24.8 Pa) static (Photo 2). According to on an entire floor, the next phase, including the installation of manufacturer-provided data, they are designed to leak 0.35 cfm/ all the floor diffusers, carpets and grilles, could be started and ft2 (0.165 L/s·m2), which would result in 14,854 cfm (7009 L/s). completed to prepare for the next series of tests, precommisTwo test points of supply airflow were measured by the TAB sioning and functional performance testing of the HVAC&R contractor, one at VFD speed at 100% (average pressure at systems. 0.075 in. w.g. [18.6 Pa]), resulting in 17,286 cfm (8157 L/s) of total leakage (Category 1 + Category 2); and again at 90% VFD Precommissioning & Functional Testing The two sets of commissioning events of the UFAD systems speed (average pressure at 0.06 in. w.g. [14.9 Pa]), resulting in for each floor was performed on the sealed floor, then its air~16,000 cfm (7550 L/s) of total leakage. The AHU could not produce the full 0.10 in. w.g. (24.8 Pa) handling units, floor boxes, fan-powered VAV boxes and reheat coils, open VAV floor boxes, and static pressure under the floor. related control systems at the same Extrapolating (using fan laws) time. The operations of the outside out to the cfm at 0.10 in. w.g. (24.8 and exhaust/relief air delivery Pa) static pressure, the total leakage units and controls were tested and would be 19,699 cfm and 19,280 verified. Precommissioning took cfm (9296 and 9098 L/s) when an average of four days for each extrapolated from the leakage rates major subsystem. Functional testat fan speeds of 100% and 90%, ing took five to six days for each respectively. major subsystem. Coordination Subtracting the estimated floor was scheduled with all witnessing plate leakage through the panel parties to ensure full participation. seams of 14,854 cfm (7009 L/s) Ongoing problems and discrepan(from manufacturer published Photo 2: Measuring floor tile gaps. cies were recorded in the official data), leaves a balance of 4,426 and 4,845 cfm (2088 and 2286 L/s) leakage from places other than issues log and handled during the course of construction and final occupancy turnover. The commissioning team for this the floor tiles specifically (Category 1 leakage). The specifications allow for 0.1 cfm/ft2 (5 L/s·m2) of Cat- project operated as a cohesive unit and helped each other to egory 1 leakage, which is approximately 4,244 cfm (2002 L/s). determine, understand, and solve problems and challenges The Category 1 leakage exceeds the specifications by 182 cfm during the commissioning process. and 601 cfm (85 L/s and 283 L/s) and, therefore, had to be reAHU Commissioning inspected, and the underfloor retested. The AHUs were commissioned by testing and verifying the Upon visual inspection under and around the floors, more potential leakage spots were found. The commissioning team acceptable performance of all of the dampers, valves, controls sealed them, and re-tested, resulting in passing the performance devices, sensors, actuators, fans, motors, and safety devices. All occupied, unoccupied, cool-down, and off modes were requirements of the project specifications. Many attempts were made to test the floor leakage rates dur- tested. Power outages, generator/UPS takeover, and fire-alarm ing the construction process for Floors 1, 2, 3 and 4. During the shutdowns of HVAC&R systems were simulated as a part of our process, many unsealed orifices were discovered. Several leak tests. All temperatures, humidities, pressures, carbon dioxide detection methods were considered, such as smoke machines, levels, and flows were calibrated and tested per the specifications. Each of the floor boxes were observed operating during all etc. Because of cost and scheduling, orifices under the floor were manually sealed. On average, it took two to three runs modes and during minimum and maximum flow rates (40% to at each floor to obtain an acceptable leakage rate called for 100%). The perimeter fan-powered VAV boxes with hot water reby the designer and required by specification. It was further heat were tested during heating and cooling operations. The flow complicated because there were two AHUs per floor, operated output of the floor diffuser plates were tested with and without and controlled separately, with the underfloor return plenum their diffuser covers on (the final test having the cover on), and common to both AHUs. Leakage from open orifices located there was a significant difference of 50% lower flow rates noted on the complete floor had to be isolated and sealed before both through the floor diffusers when their outer grilles were installed. 42 ASHRAE Journal ashrae.org June 2009

ASHRAE Journal - June 2009

Table of Contents for the Digital Edition of ASHRAE Journal - June 2009