Flow Response - Volume
This experiment tried to determine how our PV system was affecting the air in the room around the zone of interest. Ultimately, we wanted to be able to use the results to map temperatures onto the volume of air and compare it to the CFD predictions. It had been expected that the temperature would die off as the distance from the diffuser head increased. For the full flow volume results please download the excel document linked here.
The first conclusion we came to was that high fan speeds cause a greater level of disturbance in the volume than lower speeds, illustrated by the following two graphs:
Unfortunately, due to the high levels of measurement noise, we were unable to show conclusively that the temperature does die off as expected. The two graphs above agree with the hypothesis, but many also oppose it, e.g. below:
In reality, what we found was that the experimental setup was not sensitive enough to measure anything in detail. High levels of measurement noise, errors in calibration and inaccurate sensors all contributed to both systematic error and random error playing an overwhelming role in the resulting data. The systematic error can be seen in effect in the following two graphs:
The shape of the temperature profiles are nearly identical, indicating that the sensors are behaving as desired, but the overall the shape doesn't make sense: the peaks and troughs of the line don't correlate to any specific area of the grid. This would appear to be primarily due to errors in the calibration of the sensors. A number of reccommendations have been made in order to improve this experiment in the future which are discussed here.