Reports until 10:48, Thursday 02 January 2025
H1 SQZ
sheila.dwyer@LIGO.ORG - posted 10:48, Thursday 02 January 2025 - last comment - 11:35, Friday 03 January 2025(82082)
running psams script

Ryan Short, Sheila

We've followed Camilla's suggestion to run the PSAMs scan script with a smaller step size.  814908

In userapps/sqz/h1/scripts/SCAN_PSAMS.py I've edited lines 267 and 269 so that     PSAM_ZM4_LIST = [5.6, 5.7, 5.8, 5.9, 6, 6.1, 6.2, 6.3, 6.4]     PSAMS_ZM5_LIST = [-0.7, -0.6, -0.5, -0.4, -0.3, -0.2, -0.1]  .  We will let this run until noon, and see how many points it gets. 

Comments related to this report
sheila.dwyer@LIGO.ORG - 11:51, Thursday 02 January 2025 (82086)

The script ran for a while then had an error, so we got some of this data (see attachment).  The error is probably due to the way I edited the lists.

Traceback (most recent call last):
  File "/opt/rtcds/userapps/trunk/sqz/h1/scripts/SCAN_PSAMS.py", line 415, in <module>
    PSAMS_scan(freq, cycle, amp_SQZANG, dT_ZM4, dT_ZM5, amp_sus)
  File "/opt/rtcds/userapps/trunk/sqz/h1/scripts/SCAN_PSAMS.py", line 361, in PSAMS_scan
    _best_sqz, sqz_angle = _sweep(ii,jj,fig,axs,dirname,zm4,zm5,'ZM4')
  File "/opt/rtcds/userapps/trunk/sqz/h1/scripts/SCAN_PSAMS.py", line 287, in _sweep
    best_sqz, min_sqz, sqz_angle, OMC_Q, ZM4_SV, ZM5_SV = sweep_SQZANG(freq, amp_SQZANG, cycle, axs[ii,jj],
IndexError: index -3 is out of bounds for axis 1 with size 2
ICE default IO error handler doing an exit(), pid = 70032, errno = 32

 

Images attached to this comment
camilla.compton@LIGO.ORG - 11:35, Friday 03 January 2025 (82100)

The plots created are attached and in /sqz/h1/scripts/PSAMS_data/PSAMS_SCAN/250102103351/ as SQZ_DB_fit.png and heatmap.png. No clear patterns seen when looking at plots of range/BLRMs data. From the heatmap,  ZM4 at 5.2V or 6.0V with ZM5 -0.6V is best. Current is 6.0V/-0.4V so the only suggested change would be to decrease ZM5 from -0.4V to -0.6V.

Comparing our SQZ_BD_fit to LLO's (LLO#72749) our BLRMS seem to oscillate when LLO's don't. Maybe we need to adjust our BLRMs filtering to improve our SQZ DB fits.

Images attached to this comment