jeffrey.kissel@LIGO.ORG - posted 12:59, Thursday 31 August 2023 - last comment - 14:40, Thursday 31 August 2023(72594)
Calibration Pushed / Updated for DARM2 Boost and TST Actuator Strength; DELTAL_EXT Systematic Error is Much Lower Now; Matches GDS Systematic Error
L. Dartez, J. Kissel More details to come, but as of 2023-08-31 19:10:00 UTC (12:10 PDT), we've updated several corners of the calibration for the first time since Jun 21 2023 (see LHO:70693) in order to: - Update the static model of the test mass actuation strength, to better match the current time-dependent-correction factor value (because it had gotten large enough that approximations used in all TDCF calculations would have started to breakdown) (LHO:72416) - Update the "DARM loop modeled transfer functions at calibration line frequencies" EPICs records in order to account for the new DARM2 FM8 boost (LHO:72562 and LHO:72569) - Update the sensing function (only a little bit) because we're now regularly operating with OM2 "hot" as of 2023-07-19 (LHO:72523) - start using the newly re-organized pydarm librarianship, including the use of new simulines-measured IFO sening and actuation function data. (aLOG pending) - fixed an unimpactful bug in the front-end computed version of the live measured response function systematic error, in which the local oscillator frequency for the demod that was demodulating the recently changed 102.13 to 104.23 Hz calibration had not been updated. (to be commented below) The exciting news is that, with all the metrics we have on hand, these calibration updates made everything better. - 1st attachment: at the boundary of the change, we see the "relative" time-dependent correction factors change rapidly from non-unity values to unity values (and the cavity pole doesn't change, as expected). - 2nd attachment: at the boundary of the change, we see the front-end computed live measured response function systematic error goes from large values to values close to unity magnitude and zero phase. We're still tracking down some bugs in the *modeled* systematic error budget that has been broken since yesterday, Aug 20 2023 19:50 UTC *and* we're not sure if the *GDS* processed live measured response function systematic error is running yet, but we'll keep you posted. The comments below will also contain some updated details on the process for this update.
Images attached to this report
Comments related to this report
attaching SDF tables for the cal update and for the H1:CAL-CS_TDEP_PCAL_LINE8_COMPARISON_OSC_FREQ change. All changes have been accepted and saved on OBSERVE and safe snap files.
I'm also including a screenshot of the H1CALCS filter updates (H1CALCS_DIFF.png).
Images attached to this comment
The interferometric-measurement-informed portion of this calibration push was informed by report 20230830T213653Z, whose measurement is from LHO:72573. parameter foton value physical units value ----------- ------------- ----------------------- 1/Hc [m/ct] 2.93957e-07 3.4019e+06 [ct/m] (* 2475726 [mA/ct] * 1e-12 [m/pm] = 8.422 [mA/pm]) f_CC [Hz] 438.694 L1/EX [N/ct] 7.53448e-08 1.60487 [N/A] L2/EX [N/ct] 6.24070e-10 0.03047 [N/A] L3/EX [N/ct] 1.02926e-12 2.71670e-11 [N/V^2] (with 3.3 [DAC V_bias] * 40 [ESD V_bias / DAC V_bias] = 132 [ESD V_bias])
I attach here a log of the process for updating the calibration. A lot of the work is much like it was in June -- see LHO:70735, but there are a few new bells and whistles that we used. Plus, there's a few extra steps at the end to validate that down-stream products look good -- namely, that the end-game plot -- the *measured* and *modeled* systematic error agree from https://ldas-jobs.ligo-wa.caltech.edu/~cal/. Indeed, in doing this, we found some bugs that we're still sorting out. I also note that Louis did a TON of work leading up to today, generating the last ~2 months of reports, re-organizing and re-creating them, defining epoch tags, etc. So steps (0) through (5) were taken care of before today, and we started around step (6). Steps (6)-(9) out of (11) -- using today's procedure's numbering -- worked really well, and went super smoothly. The procedure is getting quite good!
Non-image files attached to this comment
