TOday I did another round of noise budget measurements at 30Watts input power with the IMC DOF4P alinged to minize the jitter peak at 260Hz.  I don't have a plot ready but got measurements for the standard things but not frequency noise.  I thought that the noise from 20-40 Hz was mostly explained, and that we could gain some range by improving there, but there must be an additional noise source at those frqeuencies.  

• The difference between our range and Livingston's best mostly comes from between 20-40Hz.  The first attachement shows the range integrand difference between the two reference curves, this was made using code from Grant meadors several years ago. (The absoulte range number might not agree with other calculations, but it gives us the right idea).   The second attachment shows the cumulative range integrand (we get most of our range below 100Hz). The first set of noise measurements I did indicated that SRCL, CHARD P and CHARDY were all within a factor of 2-3 of DARM at these frequencies, so I decided to try working on them. 

• SRCL FF-  I changed back to Jenne's JFF filter, which does better at low frequencies than the split feedforward can.  This is in the guardian now with a gain of -0.7.  The third attached screenshot shows the improvement at low frequencies when there is a SRCL excitation on, and a small penalty at around 100 Hz.I also made a PRCL excitation before and after retuning this (4th attachment). 
• CHARD both P+Y were adding a similar level of noise at 20-30 Hz.  (Stefan just made the cut offs more agressive last week, (31298)).  I lowered the gain of CHARD Y by 3db (not it is -0.1) and added and even more agresive cut off. I also lowered the gain of CHARD Y by about 3dB (from -0.14 to -0.1) and made the ELP10 cut off more agressive.  

So now SRCL and both CHARD control noises should be at least a factor of 5 or more below DARM around 30Hz, but the DARM noise hasn't improved, so there must be another noise source there.  I checked MICH ASC (P is not a problem except maybe at 18Hz, yaw could be limiting us at 20Hz but not 30Hz), and PRC2 ASC (the loop from REFL WFS to PR3) which is not a problem.