I set the dark offsets for IM4 Trans QPD.  Attached screenshot has the QPD segment old offsets on the far left.  Center screens show current offsets.  Dataviewer plots on the right show data from last Sunday night with a locked IMC with the EOM power reading 5W (left dataviewer plot), and today with no light (right dataviewer plot).  Using the lock Sunday night, with the sum = 88.7, I calculated a differences of 6% in the pitch signal with the old offsets vs without offsets (offsets set to zero), and looking a back a year to O2, the difference in the pitch signal with and without the old offsets was 0.2%, low as expected.  Changes saved in SDF.