There were a couple of broadband PCal injections done at LHO during O2, so far, to ascertain calibration in the full frequency band of ~10-200 Hz, as opposed to select frequencies used for sweeps. Attached are plots showing the comparison of PCal and various calibrated DARM data during those injection times.  The PCal, CAL-DELTAL-EXTERNAL and GDS data were obtained frames and DCS data (offline data that applies time varying corrections) were generated with kappa values applied by hand. Since calibration lines weren't available during the injections the kappas that track time varying corrections were obtained from data that followed the injection times and applied by hand for these comparisons. 

The first plot show the comparison for the injection done during Nov 30, 2016 (a-log 31994). In the plot, we see that the DCS data agrees well with PCal in the frequency band of ~10-200 Hz while the other two (CAL-DELTAL-EXTERNAL and GDS) have reasonable (?) agreement with PCal. The DCS data was generated using the following command,

GPS_START_TIME = 1164509182
GPS_END_TIME = 1164509732
FILTERS = 'H1DCS_1163173888.npz'
GST_DEBUG=3 gstlal_compute_strain --gps-start-time $(GPS_START_TIME) --gps-end-time $(GPS_END_TIME) --frame-cache H1_raw_frames.cache --data-source=frames  --filters-file $(FILTERS) --ifo=H1 --full-calibration --control-sample-rate 16384 --frame-duration=4 --frames-per-file=1 --compression-scheme=6 --compression-level=3 --frame-type H1_TEST --chan-prefix DCS- --chan-suffix _C01 --expected-fcc=341.0 --coherence-uncertainty-threshold=0.0001 --apply-kappatst --apply-kappapu --apply-kappac --expected-kappatst-real=1.009 --expected-kappapu-real=1.005 --expected-kappac=0.982

The kappa values used in the above command were taken from the DCS frames (just after 09:30:00 UTC). The code used to produce this plot is added to svn /ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/O2/H1/Scripts/PCAL/PcalBroadbandComparison20161130.m

The second plot show similar comparison for the injection done during Jan 04, 2017 (a-log 32942). Here we see that the DCS data agrees well with PCal while both CAL-DELTAL-EXTERNAL and GDS show significant differences. This confirms the trouble we were having with the GDS data during that week (see a-log 32973). This also confirms that the DCS data (availabel in CO1 frames) rectifies some of those problems.

GPS_START_TIME = 1167533440
GPS_END_TIME = 1167533990
FILTERS = 'H1DCS_1167436818.npz'

GST_DEBUG=3 gstlal_compute_strain --gps-start-time $(GPS_START_TIME) --gps-end-time $(GPS_END_TIME) --frame-cache H1_raw_frames.cache --data-source=frames  --filters-file $(FILTERS) --ifo=H1 --full-calibration --control-sample-rate 16384 --frame-duration=4 --frames-per-file=1 --compression-scheme=6 --compression-level=3 --frame-type H1_TEST --chan-prefix DCS- --chan-suffix _C01 --expected-fcc=341.0 --coherence-uncertainty-threshold=0.0001 --apply-kappatst --apply-kappapu --apply-kappac --expected-kappatst-real=1.0035 --expected-kappapu-real=1.00 --expected-kappac=1.0042

The kappa values used in the above command were taken from the DCS frames (just after 03:30:00 UTC). The code used to produce this plot is added to svn/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/O2/H1/Scripts/PCAL/PcalBroadbandComparison20170104.m