The frequency independent squeezing spectrum became clean due to IRPS improvement and lens flip(elog2445). This means back scattered noise becomes negligible.

To check back scattered noise for frequency dependent squeezing with the same homodyne (BHD) situation. I aligned filter cavity injection and reflection (injection is not optimized). With vacuum going to filter cavity and back to homodyne, I took noise spectrum as attached figure.

Comparing with elog2053 and elog2054, we can see back scattered noise became worse. The back scattered noise was checked twice, which shows change during these two measurement. The difference is that I switch on the oplev laser inside PR/BS chamber. However, oplev laser has different wavelength with infrared, which should not be an issue.

To further examine this situation, we will optimize injection to filter cavity or try to tilt more the lens. Some more investigation about oplev laser is necessary as well.