Last week I accidentally changed the polarization of the beam to the 35.5 MHz EOM. So I optimize it again to minimize an RFAM effect.

I used the signal of the transmitted beam behind RCAV on the PD for beat signal. Since the cavity pole is around 37 MHz, I should be able to see the

signal at 35.5 MHz easily. I connected the RF signal from the PD to a spectrum analyzer and adjusted the 1/2 wave plate to  minimize the peak at 35.5 MHz.

However I also notice two peaks at 35.29 and 35.71 MHz (35.5 +/- 0.21 MHz) which are approximately the same size as the 35.5 MHz peak.

I'm not sure where they come from.