Ok, it turns out these optics were purchased on purpose, as this elog shows. Jon considered building a mode-matching telescope with stock optics as an initial step before purchasing the custom optics (referred to as "design" optics in my elog).
I dialed in the new distances between the optics into the .kat file as described in this elog and pushed the changes to the repo. With the new distances, I got mode-matching of 87% for the full IFO and 89% for FPMI so there's probably no need to worry as the mode-matching with these optics was already designed.
I begin modeling the initial BHD setup using Finesse. I started with copying C1_w_BHD.kat from the 40m/bhd repo and making changes to reflect the current BHD setup:
1. OMCs were removed.
2. Only 1 PD per BHD port was left.
3. Transmission of PR2 was changed to 2.2%. The PRG was calculated to be ~15.5.
4. Actual RoCs of new optics were dialed in (Yesterday me and Paco went into the cleanroom to measure the RoCs and they seem to match the datasheets).
Here's a table comparing the old (design?) RoCs with the new RoCs:
The changes looked quite alarming, especially for LO4 and AS3, so I wrote a script to calculate the mode matching between the LO and AS beams called AS_LO_ModeMatching.ipynb and pushed it into the repo. In the notebook a bright AS beam is created by creating a small asymmetry between the arms of ~ 0.003 degrees (~10pm). Amplitude detectors were put at the input ports of the BHD BS to calculate the fields in the AS and LO beams. In particular TEM00, TEM02 and TEM20 were measured for each beam.
The calculation shows that with the old RoCs the mode matching between the LO and AS beams is 99% while for the new RoCs it is ~ 50%.