I measured the performance of the new sensor correction and blending filters. Analysis will be reported later.
I wrote the blending filters lp_part1 and lp_part2 for LVDTL, and hp_part1 and hp_part2 for ACCL. Each blending filter had to be split in two multiplicative factors because their orders were very high.
The aim was to measure the ASDs of three different quantities when the suspension was swinging free. Namely, we wanted the readout of
- The corrected LVDT, blended with the geophones.
- The corrected LVDT only.
- The uncorrected LVDT.
In order to measure these quantities simultaneously, I did the following:
- Set the suspensionsion in SAFE state.
- Turned off the old blending filters (LP80_new and HP80_new) and turned on the new ones.
- Set the state to TWR_FLOAT, and then I paused Guardian.
- Turned on the sensor correction for LVDTL.
- Modified the matrices LVDT2EUL and LVDTALIGN in order to get the readout in L in the channel for T (see klog 22376).
- Ran the measurement with diaggui.
- After the measurement, I reverted the changes to the matrices and made sure the suspension went to LOCK_ACQUISITION state successfully.
- I letf the new filters on and modified the SDF in SAFE state.
In the diaggui file, the relevant channels are as follows:
- Corrected LVDT without blending: K1:VIS-BS_IP_BLEND_LVDTL_IN1.
- Corrected LVDT blended with the geophone: K1:VIS-BS_IP_IDAMP_L_IN1.
- Uncorrected LVDT: K1:VIS-BS_IP_BLEND_LVDTT_IN1 (as explained above, this is a T channel, arranged to provide information about the L channel).
- Seismic motion along L, pre-filtered and in units of micrometers: K1:VIS-BS_IP_SENSCORR_L_IN1.
The location andname of the diaggui file is
- /kagra/Dropbox/Subsystems/VIS/vis_commissioning/bs/data/
- bs_ip_twr_float_04.xml
The new filters are written the file attached.