louis.dartez@LIGO.ORG - posted 11:34, Monday 07 August 2023 - last comment - 12:34, Monday 07 August 2023(72030)
mismatch in ETMX TST foton filter between CAL-CS and pyDARM/GDS
pyDARM and the GDS pipeline (which produces H1:GDS-CALIB_STRAIN) relies on exports of the CAL-CS Foton filterbanks installed on the front end. The filters installed on the front end and those used by the calibration pipeline need to be in sync to properly calibrate the IFO.
I'm attaching a comparison of the inverse sensing and ETMX actuation Foton tfs as installed in the CAL-CS path (plots). The inverse sensing, ETMX UIM, and ETMX PUM tf exports match well between what is currently installed on the front end and what pyDARM thinks is installed. That's good. However, the ETMX TST stage has a significant deviation above 10Hz. This can be seen on the last page of the attached PDF, which shows both tfs overlaid on the same bode plot (left column) and their residual, also in the form of a bode plot (right column).
This discrepancy is not ideal and should be fixed by re-exporting the TST foton export for pyDARM to use it and be included in the next calibration export at LHO as per LHO:69563.
T2000022 is a good resource with instructions on how to export the appropriate TFs from the CAL-CS foton banks.
Non-image files attached to this report
Comments related to this report
Jeff informed me that this discrepancy is likely due to a 3.2kHz pole that needs to be compensated for, similar to LHO:33927. There's been some back and forth as to whether the "correct" thing to do is to include the compensation in the front end vs the GDS pipeline. There's confusion as to whether the pole should be compensated for in the front end (it currently isn't) or in the GDS pipeline. It's not clear yet if this is properly included in the GDS pipeline.