While commissioners were MIA this morning I took the time to try sensor correction at the beam splitter again. Previously, I had found that it pretty much just made stuff worse, at least when I tried HEPI Z. Unfortunately, that still seems to be the case. This time I tried HEPI Z, and ISI X & Y, and in every case it just made things slightly worse. My attached plot shows a comparison between the ISI X for the BS (red) and ETMX (blue). Sensor correction is off  for the dashed traces, on for the solid. ETMX shows our expected improvements, but the BS gives no joy. Also interesting that the BS doesn't seem to be performing as well as the ETMX generally, but the mechanical and controls configurations are quite different. I've talked to Hugh, and the input and output matrices have errors on the BS (not on ETMX), and the STS2 the BS uses is a little wonky, it reads ~2000 cts, while the other cornerstation seismometers are in the hundreds. The STS spectra all  look similar in the .1hz and above region, but they disagree a bit below, see my second plot. When I get a chance, I'll try using a different seismometer, but I'm pretty sure I've gotten positive results at ITMX using the BS seismometer.