Mark Barton, Fred Raab, Szymon Steplewski

While trying to test the L2 OSEMs on ETMY and ITMY I noticed that the power spectrum data in DTT was looking very strange for certain combinations of Low Freq, High Freq, and BW. Sometimes the power spectrum plot would show several enormous spikes repeated through the range, while other times nothing would show up. 

In Diagnostic Test Tools the user has an option to set the low frequency, high frequency, and BW. When I set these to low freq = 0 Hz, high freq = 10 Hz, and BW = 0.05 Hz, DTT returns some strange power spectra data, despite having normal looking time series data. Also using a BW >= 0.05 Hz (corresponding to <=20 seconds of time data) results in 16 seconds of time series data plotted in the DTT time series window, while BW <= 0.05 Hz brings in 32 seconds of data, which seems problematic.

Another issue that I am seeing comes from using 50% overlap in the measurement window. Setting the BW = 0.01 Hz  (corresponding to 100 seconds of time data) reads in 128 seconds of time series data with 64 seconds overlap, not the 100 seconds of data with 50 second overlap that I expect.

Although we came up with evidence of this bug, we don't really know what is causing it to happen for this combination of settings.  I have reported the bug at:
https://bugzilla.ligo-wa.caltech.edu/bugzilla/show_bug.cgi?id=389