Yuhang, Michael

We continued having SHG lock issues when investigating the stability of the green generation/stabilization path. We checked LO, it is at the correct frequency and level (5 dBm). The signal from the photodetector at the SHG mixer also looks fine on the spectrum analyzer (88 MHz peak going up and down with automated cavity scan). However, the output of the mixer PD x LO is quite bad.  It was eventually found that direct reflection from BAB is coming back the the SHG reflection PD that we use to lock.

Originally, locking of the SHG was done via transmission through the SHG modulated at 15.2 MHz. This is within the linewidth of the SHG, so the PDH sidebands have strong transmission. However, it also was causing phase noise in squeezing due to being inside the linewidth of other green optics, so we decided to replace the 15.2 MHz modulation with 88.3 MHz. Since we expected the PDH sideband transmission through the SHG to be much weaker (and it is), we placed a PD in reflection of the SHG to lock. However, the reflection signal is transmitted through the same BS that splits the main laser to BAB.

Since I can't upload images right now I will just draw text art:

                               ^

                               I
                               I
                               V
                              To and from BAB

                              \
<------- To SHG----- \ ---------> To SHG refl PD
                                \
                               ^
                               I
                               I
                             From main laser

In fact, the BAB is injected at the OPO HR surface, so it is also reflected quite strongly, and the BS is 80:20 R:T, so there would be quite a lot of BAB reaching the SHG refl PD.

I tried to switch the error signal acquisition PD to the SHG transmission one - I took the cable connected to SHG refl PD (the one with the DC block) and connected it directly to SHG tra. I could see a nice cavity mode sweep on the oscilloscope, about 1.4 V or so, however, I could not see any mixer signal, nor could I see any 88 MHz sideband on the spectrum analyzer. There was a 0.2 ND filter going to the PD which I removed, but no difference. I tried to lock with very low positive, or negative, threshold on the servo but still nothing. So we have to reconsider how we lock SHG, otherwise we cannot have BAB and green at the same time, which is necessary to optimize the OPO for green and infrared.

I returned the ND filter put the DC block cable back to SHG refl.