With Lucia.
We modified the LVDTALIGN matrix in order to reduce sensor diaginalization in the IP LVDTs.
The concept is as follows
- Actuated in L, and minimized the amplitude of the resonance peak in L, that appeared in T and Y transfer functions.
- Actuated in T, and minimized the amplitude of the resonance peak in T, that appeared in L and Y transfer functions.
- Actuated in Y, and minimized the amplitude of the resonance peak in Y, that appeared in L and T transfer functions.
The matrix LVDTALIGN was calculated with the Matlab script BS_IP_SENS_ALIGN.m, which is in /kagra/Dropbox/Subsystems/VIS/vis_commissioning/bs/IP_diag/BS_new/ . This script uses data exported from the DTT files at /kagra/Dropbox/Subsystems/VIS/vis_commissioning/bs/IP_diag/20221004/. The files with the name *02.xml and *03.xml refer to measurement before and after diagonalization.
The new value of the matrix is
0.7523 -0.2235 -0.2108
0.2654 0.9126 0.0014
0.6232 -0.2964 0.8169
We accepted the new values in the SDF while the suspension was in TWR_FLOAT (table file name "isolated").
The plots show the transfer functions after diagonalization. More quantitative information will follow in a future entry.