After tweaking the alignment into the pre-modecleaner, I measured the power noise spectra
for both photodiodes PDA and PDB when either was used as the in-the-loop sensor.

    The plot PDA.png is for when PDA was used as the in-the-loop sensor.  PDB.png is for when
PDB is used as the in-the-loop sensor.  When the ISS is off, in both cases the free-running
power noise measurements agree quite well.  However the in-the-loop and out-of-the-loop spectra
disagree quite badly when PDB is used as the in-the-loop sensor.  The in-the-loop noise suggests
there's enough gain in the loop in this case with the level being ~1.0E-8.  The out-of-the-loop
spectrum is more than a factor of 10 higher, which suggests there's a difference in the two optical
paths (most likely beam pointing).

    Not sure why at this stage.  I would probably use PDA as the photodiode to be used as the
loop sensor (for now).