I updated Simulink noise budget model to incorporate balanced BS/PRM actuator.
Updates made since last alog (#10441) are;
- ISCINF, (M2|M1)_LOCK filters are added (see alog #10559)
- PRM suspension model correction factor is added (see alog #10674)
- PRMI sensing matrix measured is added (see alog #10725)
PRMI sensing matrix I used is;
PRCL MICH
REFLAIR 45 I 4.8e5 1.5e4
REFLAIR 45 Q 3.9e3 -2.6e4 W/m
Some plots from NB model are attached. Data I used is taken from Mar 12, 12:43:00 UTC for 2048 seconds, when PRMI is sideband locked with REFLAIR45.
OLTF_MICH_1078663396.png: MICH OLTF. The magnitude from the model is higher than the measured by ~40 %. This means that there are some errors in the sensing matrix measurement. We have sensor/actuator calibration error of ~10%. Also, there was a power drop in the cavity build up (POP18) by ~10%. However, these errors don't explain the disagreement between the model and measurement. We need more investigaton on this. (There was is a suspision about the factor of sqrt(2) in BS, but I've checked that)
OLTF_PRCL_1078663396.png: PRCL OLTF. The model and measurement agrees within 10%. Also, phase looks reasonable (compare the OLTF before output matrix diagonalization in alog #10441; we had a dip).
PRCLtoMICH_1078663396.png, MICHtoPRCL_1078663396.png: Measured coupling between MICH loop and PRCL loop (TF from H1:LSC-MICH_EXC to H1:LSC-PRCL_IN1 or vice versa). Compared with the same measurement done before output matrix diagonalization, coupling is at least factor of 2 less (see alog #10481). Model explains some features qualitatively.
NB_MICH_1078663396.png, NB_PRCL_1078663396.png: Noise budget for MICH and PRCL loop. Note that seismic noise is not real (copied from LLO model).
