Methodology for the Evaluation of Natural Ventilation in ... - Cham

smoke pencils were used for localized airflow visualization and to determine the direction of theairflow at inlets and outlets. It was found, however, that the smoke they gave off dissipatedwhen there were large amounts of air flowing at a particular location in the model and due toturbulent flow within the model. A fogging machine was used when the smoke given off by thepencils did not perform satisfactorily. The fogging machine was found to produce a neutrallybuoyant, white fog that could be used to trace the flow within the model. The fogging machineuses heat to generate the fog, so a long flexible tube was used from the outlet of the foggingmachine to the outlet nozzle that introduced the fog into the space to ensure that the fog was atthe ambient temperature when it entered the model. This technique minimized the potential forthermal effects that potentially could be introduced in experiments by the use of fog as avisualization tool.6.4.2 MeasurementsAir velocity measurements were taken at both the inlet and outlet openings for each experimentin order to calculate the heat loss due to airflow, or advection. The heat loss through conductioncould be determined from the total heat input into the model by the heaters and the heat loss dueto air flowing through the model. Air velocity measurements were taken at each window andstack that was open during a particular experiment to ensure uniformity in airflow across a givenfaçade. The measurements were taken at the center of each window, at the exterior face of thewindow opening on each façade. The average of the measurements for a particular set ofwindows or stacks in each experiment was used in calculating the airflow balance and heat lossdue to advection. In cases where there was only a single opening, only one measurement wastaken. Mass flow balances were conducted to ensure that there was not significant air leakage.The power output of the heaters was measured to make certain that the internal load input wasconstant and known. Using a clamp-on ammeter and a multi-meter, both the current and voltagefor the heaters were measured, and the power output of the heaters calculated. The single-phaseheaters were wired in parallel. The heaters are rated at 300 Watts; however, through themeasurements described above, it was determined that the heaters were only providing 88percent of their rated output.The accurate measurements of ambient conditions were important both to determine if anyvertical temperature stratification existed, and to identify the best parameters to use as inputs tothe CFD model. The supply air temperature was kept constant, set to 13°C, but the ambienttemperatures varied from a value of 16 to 18°C due to the number of heaters used in a givenexperiment for the buoyancy-driven ventilation cases. With the wind-driven experiments, theambient temperature reached 24°C. The increase in ambient temperature of the test chamber wasless the fewer heaters used in a given experiment. This not only provided insight into the impacton ambient conditions on the model, but also contributed to the understanding of heat loss due totemperature difference. For both cases, the ambient temperature was carefully monitored toensure that it did not vary over the course of the experiments. The measurement locations for theambient temperature recordings occurred at 0.5m intervals from the floor to the ceiling of the testchamber in the corner facing the supply diffuser.105