J. Kissel

At 2019-03-28 03:29:37 UTC, for about 5 minutes, I took a broad band PCAL injection to test the quality of the updates we did to the CAL-CS. This time can also be used to test GDS (but they've not yet updated anything from the new model parameters yet).

There's still some systematic error left over in the detection band -- a frequency dependent error, with excursions at the level of +/- 7%  and -10 deg.
We've still got work to do:
    - This data was taken at the very beginning of a lock stretch, unlike the reference data we took which was well into the thermalized realm.
    - We know that the relative delay between the paths is above half a 16 kHz clock cycle wrong
    - We know that the PUM and Sensing Function still has some level of systematic error, but they should all be impactful only below 20 Hz.

Also -- looking at the time dependent correction factors, the actuator kappa's are all over the place.

More tomorrow...


Broadband injection template is here:
/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs
    2019-03-28_H1_PCAL2DARMTF_BB_postCALCSupdate.xml