Summary:

• S-polarization leaking out of BS_AR is 20% size of P-polarization
• S-polarization is much noisier at low frequencies
• Clear indication of polarization rotation at the time of a lockloss (exact cause unknown).

Details:

Following the installation of the PBS on the HWS table to measure the polarization state of the 1064nm light leaking out of the BS_AR surface, I took a quick look at the data from the detectors. 

• DETECTOR1 = P-polarization
• DETECTOR2 = S-polarization

Looking at a 5 hour time series around a couple of lock acquisitions, we see 20-30mV on the S detector vs about 97mV on the P-detector. The X-arm transmission is shown as a guide for when the IFO is locked. Vertical axis is the voltage from the detectors.

Next, if we normalize the time series by the median values of each series, we can see much more noise on the S-detector (some of which is caused by the higher relative electronics noise) but significantly higher low-frequency variations during lock - for example, the amount of S-polarization drops by 30% in the second lock-stretch.

Lastly, the time just around the second lock-loss is shown in detail. The S-polarization oscillation is almost exactly out of phase with the oscillation in the arm and in the P-polarization (which is a strong indication for polarization rotation in the interferometer beam). I need to do some further analysis, but the fact that the SUM of the S+P still shows a strong oscillation indicatest that this is not polarization rotation in the HWS optics.