We put offsets in the PRCL and MICH loops, and measured sensing matrices for each condition.
What we found was that PRCL offsets of order 1/20th a linewidth (calibration to be checked tomorrow) would give significant changes in the angles of the REFL signal sensing matrix elements. We broke MICH lock before we were able to put in a significant enough offset to see the demod phases change.
Because there are so many plots, I've put them together in a pdf. Each page has a set of radar plots for sensing matrix elements. On the bottom of each page I note what our MICH and PRCL offset values were, and where the data is saved (in the 40m scripts directory). To see the differences, make sure your pdf viewer is set to single-page, not scrolling.
One major thing that we noted was that putting in a PRCL offset also changed the MICH offset. When we increased the PRCL offset, we saw the AS port get brighter (but not as bright as when we were putting in large MICH offsets).
Tomorrow, I need to check the calibrations we were using, to see how many meters we were moving the optics. Also, Q, Gabriele and I need to meditate and do some modelling to figure out why the length offset could be affecting the degeneracy so strongly.