WMOD00284 - Laserfiche WebLink

Home Browse Search

each of the eight SF6 flights and these can be found ln the individual discussions of each test case. The primary data reduction effort was directed toward the raw SF6 data. In particular, software was written to remove baseline drift from the analyzer output and also to filter out short term noise. Typically the output signal would drop with increasing altitude and remain fairly constant while flying level, Therefore a program was written to adjust the SF6 data to a baseline which corresponded to a zero reading, and to remove erroneous noise generated by small changes in cabin pressure or by rapid acceleration. Once a baseline had been established a correction equation was applied to the raw data to convert analyzer output (volts) to concentration in parts per trillion (ppt). These correction equations were obtained from the pre- and postflight calibrations (Table 4-5). For this report delta-h (height of SF6 plume above release elevation) values for individual plumes were hand extracted from the data. An example of the baseline correction is shown in Figure 4.5. Note that the baseline is smooth after the correction and that the y-axis is portrayed in units of concentration instead of voltage. When rUnnlng the raw SF6 data through the software a minimum concentration value was used to search for the occurrence of a SF6 plume. Typically a value of 10 ppt was chosen since the noise of the analyzer ran around 5 ppt. Therefore plume widths were calculated with the baseline set at 10 ppt. Plume widths were calculated from the times the 10 ppt limit was exceeded and the true air speed of the aircraft. Corrections were made to these calculations when the aircraft transect 4-14 I I f I 'I I , I , I J I f ,I I I I I ,