I created a new version of the HTTS TF plotting script, plotHTTS_ddtfs_M1.m, to correct for the fact that the wrong dewhitening filter has been installed at both sites, and that the LHO ISC HAM-A coil drivers have only recently been changed per E1201027. Setting meas.badFilt = '40:1.6' (rather than ' ' or anything else) enables the correction for the bad filter, and meas.E1201027=true enables the new gains.

The actual hardware whitening filter is 0.4:10 per D1002818/D080276. The dewhitening filter installed in Slot 1 of the the OSEMINF blocks claims to be 10:0.4 but has actually been 40:1.6.

After checking with Stuart to ensure that there was no LLO data that hadn't been committed, I reran plotHTTS_ddtfs_M1.m on all the data sets listed in measList in plotallhtts_tfs_M1.m, so as to regenerate the summary .mat files and plots at ^/trunk/HTTS/ifo/optic/SAGM1/Results . 

I used meas.badFilt = '40:1.6' in all cases and meas.E1201027=true for all except the measurements of H1:RM1 and H1:RM2 in 11/2013, which were before the LHO resistor change. I lost a day to head-scratching before I realized that the L1 resistors had been changed shortly after the ECR was approved in 12/2012, whereas the LHO ones were changed almost exactly a year later (12/2/13, LHO alog 8792).

As might be hoped, the new plots were free of the hump caused by the misalignment of poles and zeroes between hardware and software. The new files have been committed to the SVN.

The two attached comparison plots show  L1:RM1, L1RM2, H1:RM1 and H2:RM2 before (2013-11-25) and after (2014-02-12) the fixes. Note that the H1 data in the second plot is from later (12/20/13), and fixes a second filter issue that has not been identified but only affects the 11/24/13 data).

I prepared a new filter file for LHO with the correct dewhitening filters, /opt/rtcds/userapps/release/sus/h1/filterfiles/H1SUSHTTS_dewhitefix.txt , and will install it at the earliest convenient opportunity.