Before using the random tape in test operations, it is essential to verify
the spectrum shape of the synthetic random voltage over the 17.5 minutes of
recording. This is especially important when recording with older random
equalization systems where the level in individual filter bandwidths can drift
off the set point.
4-5.3.1 It is
recommended that the analysis be made with a real time analyzer using narrow
band filters of 10 Hz bandwidth or less. The sampling parameters should be
adjusted to yield a minimum statistical accuracy of 100 degrees of freedom. A
block diagram of the setup for spectrum analysis is shown in Figure
4-5.3.2 As a minimum, four samples of the recorded random
spectrum should be analyzed:
a. After 1 minute of elapsed time
b. After 5 minutes of elapsed time
c. After 10 minutes of elapsed time
d. After 15 minutes of elapsed time.
4-5.3.3 Rewind the
tape to the counter reading corresponding to 1 minute of elapsed time in the
recorded random spectrum (see Figure 3-3).
4-5.3.4 Activate the playback mode of the tape and adjust
the output control level to a convenient setting for the analyzer to be
4-5.3.5 Start the analyzer sampling of the playback data.
When the analyzer has sampled sufficient data for the required averaging
confidence, the tape deck should be stopped.
4-5.3.6 The random
spectrum computed by the analyzer should then be compared with the required
synthetic random spectrum (shown graphically in Figure
4-7). A convenient way of doing this is to plot the analyzer output
directly on the graph paper containing the graphic outline of the required
spectrum. It should be noted that the absolute values of the recorded voltages
are not significant so the "Y" axis pen position can be adjusted to
superimpose these plots. (provided the span adjustment is not changed -10
dB/inch in the sample shown in Figure
4-5.3.7 A ±1.0 dB tolerance should be applied in comparing
the recorded spectrum with the required spectrum. Exceedance in excess of ±1.0
dB will require the re-equalization of the random spectrum and re-recording of
4-5.3.8 The tape should then be advanced to the counter
reading corresponding to 5 minutes of elapsed time in the random recording and
the analysis repeated as described in proceeding paragraphs. Repeat with tape
at elapsed time of 10 and 15 minutes.
4-5.3.9 Following the successful verification of the
recorded random spectrum, the recording tabs on the tape cassette should be
moved or broken off (per manufacturer's directions) to prevent accidental
erasure of the data.
If the reverse side of the tape is to be used do not position
the "record" defeat tabs until both sides have been