Reports until 16:34, Wednesday 11 January 2023
H1 ISC (TCS)
camilla.compton@LIGO.ORG - posted 16:34, Wednesday 11 January 2023 - last comment - 17:02, Friday 13 January 2023(66755)
ITMX YAW beam spot raster tonight from 9pm at different CO2 powers.

Georgia, Camilla

After CO2 stepper tests in alog 66732, we saw ADS YAW3 (PRM pointing onto ITMX) moves a lot. Are the changes in buildups due to worse TCS settigns or worse beam spots?  In partuclar, higher CO2X and lower CO2Y decreased the 201Hz frequney noise line and difference in arm circilating powers but buildup got worse. 

9pm -11pm Run ITMX YAW beam spot raster if the IFO is in NLN using ligo/gitcommon/labutils/beam_spot_raster/a2l_gain_stepper.py.

11:30pm-12:30am will gradually adjust CO2 settings to higher X, lower Y. So CO2X from 2.2 to 2.8W, CO2Y from 1.1 to 0.5W.

1:30am - 3:30am Repeat ITMX YAW beam spot raster

4am-6am will gradually adjust CO2X settings to pre-Dec 2022 settings: from 2.8 to 1.5W, CO2Y from 0.5 to 0W.

7am - 9am Repeat ITMX YAW beam spot raster

Comments related to this report
camilla.compton@LIGO.ORG - 09:22, Thursday 12 January 2023 (66768)
Link

The first test with nominal 4W CO2X and 1.7W CO2Y test ran, see attached. Both RF18 and RF90 buildups change togehter.

The middle test with higher CO2X and lower CO2Y did not run as IFO lost lock.

The last test with pre-December CO2 settings of 2.7W CO2X and 0.0W CO2Y ran, see attachted.. The ADS changed a lot for the high 2.1 Y2L gain.

Currently have the CO2s moving back to thier nominal. They will get there at 10:30am.

Images attached to this comment
georgia.mansell@LIGO.ORG - 12:37, Thursday 12 January 2023 (66773)
Link

We were curious about what happened towards the end of the second test, where the alignment got a kick (including in PIT, see ADS PIT3), the arm powers and buildups went up while the range went down. Craig postulated that this was in response to a big glitch just after -1 hour mark in Camilla's second attachment.

I made a plots of DARM during the two CO2 tests, comparing times before the Y2L gain steps and when the range and buildups had the most change (just after the max Y2L step).

  • During the first test (blue and grey traces on the top plot) we probably lost the squeezer (shot noise got worse) but otherwise DARM seems the same.
  • With the old CO2 settings (yellow and red traces) the low freqeuncy frequency noise coupling (25 Hz line) is worse (so our nominal CO2 settings are better for frequency noise), and the MICH coupling (40.7 Hz line) increases
  • The red trace is with the old CO2 settings and the IY Y2L gain increased, and while the IFO alignment is going a little crazy, so maybe it's not surprisng that the results are confusing.
    • The ADS lines are higher (to be expected since they were still converging).
    • There's a line around 22Hz that is much worse - idk what it is
    • The freqeuncy noise and PRCL couplings (25 Hz and 35.3 Hz) further increase, and there is additional noise around 100+ Hz that is coherent with PRCL and frequency noise (bottom plot).

Note the legend wrongly says IY but it should say IX!!!

Images attached to this comment
camilla.compton@LIGO.ORG - 17:02, Friday 13 January 2023 (66807)
Link

AWG_LINES injected during first test are attached.

201Hz and 4500Hz frequency lines increases during the increase in ITMX_Y2L.

  • awgs['frequency_25.2Hz_201Hz_4500Hz']
  • awgs['prcl_35.3Hz']; awgs['srcl_38.1Hz']; awgs['mich_40.7Hz']  

Using Dan Browns's code: python /ligo/gitcommon/labutils/demod_lines/demod_lines.py

Plot for whole second lock shows 201Hz line >10% higher for the old lower CO2 settings!

Images attached to this comment
Non-image files attached to this comment