joan-rene.merou@LIGO.ORG - posted 17:51, Friday 11 October 2024 - last comment - 12:14, Wednesday 23 October 2024(80624)
Broadening of 60 Hz shoulders in H1 strain channel
After finding that the source of the broad 60 Hz shoulders in L1 appears to be a problem with line subtraction, as reported in LLO alog 773549, we have checked for possible related issues in H1. We have found that unlike in L1, the broad shoulders appear after October 1st instead of after September 17th as can be seen by checking at L1:GDS-CALIB_STRAIN vs L1:GDS-CALIB_STRAIN_NOLINES vs L1:GDS-CALIB_STRAIN_CLEAN at different points in time during these dates, where the shoulders appear in NOLINES and CLEAN. It would also appear like before September 17th the substraction was not ideal either, since there are still some small shoulders
September 17:September 26:
October 1st:
October 2nd:
October 11th:
Images attached to this report
Comments related to this report
the current gstlal-calibration was installed at both sites on September 3rd (not September 5th as I've been telling people). At LHO at 17:52 UTC and at LLO at 17:19 UTC.
Images attached to this comment
As in LLO, the shoulders seem to get worse getting closer to September 17th, but before then h(t) values for GDS_CALIB_STRAIN_NOLINES and GDS_CALIB_STRAIN_CLEAN shouldering 60 Hz still appear to be higher than GDS_CALIB_STRAIN, including before and after maintenance on September 3rd. Adding additional plots before and after September 3rd showing this behaviour. Example from June 10th:
Images attached to this comment
