The daqd was restarted around 12:21 localtime.

F. Mera, M. Pirello

Per WP11175 & WP11172

Replaced Aging Kepcos with upgraded supplies in H1-VDC-C2 CER Mezanine Power Rack.  Racks affected:

H1-ISC-C3 +/-18V, +/-24V (all RF chassis)

H1-ISC-C2 +/-18V also H1-ISC-C1 uses this +/-18V, now documented. (Signals Chassis)

H1-ISC-C2 Slow Controls +24V  (this is all beckhoff and acromag in ISC-C2 sans Corner 6 and Corner 2)

H1-SQZ-C1 Slow Controls & OMC IO Chassis +24V (Entire squeezer rack in CER)

We also found that Slow Controls Corner 6 & Corner 2 are powered by the IO Chassis ASC, LSC +24V supply, so these two did not go down during the swap.

Keywords: aging ageing

We ran the functionality test on the main turbopump at EY. The scroll pump is started to take pressure down to low 10^-02 Torr, at which time the turbo pump is started, the system reaches low 10^-08 Torr after a few minutes, then the turbo pump system is left ON for about 1 hour, after the hour the system goes through a shut down sequence.
No issues were encountered while performing the functionality test on this station.

EY Turbo:

Bearing Life:100%

Scroll Pump Hours: 198.8

Turbopump hours: 1266

Closes WP 11179   FAMIS 24935

The 12:59 UTC lockloss this morning was caused by SWAP_BACK_ETMX state of SUS_CHARGE, when IFO L3 DARM control was being swapped from ITMX to ETMX. The in-lock charge injections themselves started at 5.45am alog 69412 and completed without issue.

Although I shortened some ramp times in alog 69234, this was only the ramp times in RUN_ESD_EXC guardians, for the injections and changing the bias values. That worked fine, we lost lock after these states had completed.

There was a glitch in ETMX int he last state of ESD_EXC_ETMX when the L2L filters were changed even though the L2L gain was zero, see attached.  This seems strange. After this point, the ADS monitors show an excursion (12:59 UTC lockloss) which could have caused the lockloss. ( <- Elenna clarfifed that the ADS control signals don't show anything when the lines are off...d'oh. )

Tue May 09 10:05:05 2023 INFO: Fill completed in 5min 5secs

Travis confirmed a good fill curbside.

I ran the OPLEV charge measurement for ETMX this morning. I was able to get 4 measurements in, during the 5th one an IPC error occured and we lost connection briefly.

The charge is above +\- 50[V] on all quads and has a small upward trend (~ 5-10 [V]) over the past month on all quad DOFs except for UR in both P&Y which saw a small decrease.

The gds broadcasters where updated to send a new broadcast list.

 * 90 fast channels were added for detchar.
 * 1 channel was added for calibration.
 * 8 channels were added at the request of the DMT maintainers.

I've attached 2 files.

 * H1_missing.txt - the channels that were added.
 * gds_broadcast_h1.txt - the full broadcast list for H1.

TITLE: 05/09 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Preventative maintenance
CURRENT ENVIRONMENT:
   SEI_ENV state: CALM
   Wind: 4mph Gusts, 3mph 5min avg
   Primary useism: 0.01 μm/s
   Secondary useism: 0.12 μm/s
QUICK SUMMARY: IFO was relocking when I arrived, making it to engaging ASC on its own so far. Already some maintenance activities going - EY wind fence, LVEA flooring, EY turbo pump.

diaggui updated to 3.1.5~dev13 to fix a crash when repeatedly running a failed test.

Workstations were updated and rebooted.

I tried my hand at another calibration sweep.
Took IFO to NLC_CAL_MEAS at 5:50 UTC
All the measurmesnts completed succefully.
I forgot to take a screen shot of the CALIBRATION_MONITOR while the Calibration was running so I grabbed a timemachine screenshot of both the OPS_OVERVIEW Screen and the calibration monitor screen while the calibration was running. I hope thats helpful.

Report lives here:
/ligo/groups/cal/H1/reports/20230509T070754Z/H1_calibration_report_20230509T070754Z.pdf and is attached.

TITLE: 05/09 Eve Shift: 23:00-07:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Calibration
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 3mph Gusts, 2mph 5min avg
    Primary useism: 0.01 μm/s
    Secondary useism: 0.11 μm/s
QUICK SUMMARY:
Nothing notable happened since my last A log except:
Took IFO to NLN_CAL_MEAS at 5:50 ish and started a Calibration measurement.
Rick stopped by to make some PCAL measurements and we did some analysis in the control room.
Calibration Measurments finished, IFO ISC_LOCK_STATE taken back to NOMINAL_LOW_NOISE.  7:36 UTC
IFO INTENT BIT: OBSERVING.  7:36 UTC
 

Lockloss around 23:33 Potentially caused by Calibration measurements being driven too hard
https://ldas-jobs.ligo-wa.caltech.edu/~lockloss/index.cgi?event=1367623789

Once unlocked I reloaded ALS_Diff to load in TJ's Changes from earlier today.
Guardian gave us an ALS_DIFF Error after reload of ALS_DIFF:
Camilla and I checked the log and subsequently line 240

We made some changes to ALS_DIFF after ALS_DIFF to see if it we could fix it.

Index: ALS_DIFF.py
===================================================================
--- ALS_DIFF.py    (revision 25588)
+++ ALS_DIFF.py    (working copy)
@@ -237,11 +237,11 @@
             return False
         elif self.freq_set:
             log('Getting 5sec of "dark" time')
-            self.avg = timeout.call_with_timeout(cdsutils.avg, 5, 'H1:LSC-TR_Y_NORM_INMON')
+            self.avg = timeout(cdsutils.avg, 5, 'H1:LSC-TR_Y_NORM_INMON')
             if abs(self.avg) > self.acceptable_dark:
                 log('Need to reset TR Y QPD B offset')
                 # I think just taking the average here should be good enough
-                dark_avg = timeout.call_with_timeout(cdsutils.avg, 5, 'H1:LSC-TR_Y_QPD_B_SUM_INMON')
+                dark_avg = timeout(cdsutils.avg, 5, 'H1:LSC-TR_Y_QPD_B_SUM_INMON')
                 ezca['LSC-TR_Y_QPD_B_SUM_OFFSET'] = -dark_avg.round(3)
                 return True
             else:

We didn't get much further than the error  before we got a lock loss that did not take the IFO to down. It took us back down to AQUIRE_DRMI_1F but stayed locked for a few moments then eventually got a lock loss that took us to down.
Lockloss https://ldas-jobs.ligo-wa.caltech.edu/~lockloss/index.cgi?event=1367626591

Accepted some SDF changes for H1:ASC-DSOFT_Y_GAIN & H1:ASC-CSOFT_P_GAIN.  Screenshot included.

PI Mode 24 Rung up and Vicky and I manually stepped through some phase shifts to damp it down.
It then continued to ring it's self up for the next 40 minutes.

Daniel, Naoki, Vicky

Today we used the HAM7 in-vac picomotors on D:M1 and D:M2 to re-center the squeeze beam on the in-vacum RLF QPDs, which are used for both filter cavity ASC + LSC. Before pico-ing, the FC retro-reflected beam was quite mis-centered on RLF_QPD_B, and we had to walk the beam considerably in pitch to re-center on the QPD's, see the before vs after QPD centerings. Now that we are servo-ing the beam going to the filter cavity (with FC beam spot control LHO:69374), hopefully now we will have a stable FC injection axis, and a stable retro-path through HAM7.

Jennie, Jeff, Louis

Executive Summary: After discussion with Jeff who did some measurements using swept sine, we have decided that the best offset is -290.

Before starting measurements today and looking at our measurements from last week more closely, I suspected that there was probably a small spring (as opposed to the previous anti-spring) in the SRC at the detuning of -265 which we have had since Friday.

To get clearer measurements, I have increased the DARM injection amplitude by 2 (to 10e-2 in noise elliptical filter).

As the measurements went on I realised we were still on the anti-spring side of the resonance when at an offset of -265, the feature at 10Hz that looked like spring resonance went away when we increased the measurement SNR by increasing the DARM injection amplitude.

Attached to this post is a plot of the magnitude and phase of the sensing function responses measured today and on Friday (note that the ones from Friday will be noisier due to the lower injection amplitude used for the DARM OLG measurement).

The second plot is a smaller selection of offsets that seem to be around the turing point between spring and anti-spring.

For reference, additional spring detunings were carried out while tuning the DARM offset and the squeezing, and while thermalising.

As before, the plot contains the measurements I took and the folders in which the data and figures are saved.

The injection code is darm_noise_injection_caller.py, pcal_noise_injection_caller.py. The processing is done by darm_sensing_function_processor.py, and plotting files on the same axis to compare sensing functions is done by darm_sensing_comparison.py.

Note that darm_sensing_function_processor.py has been updated to refer to the new pydarm file with the most up-to-date calibration parameters which I copied over today into /ligo/gitcommon/noise_recorder/data/darm_noise/pydarm_H1.ini

The files attached are saved in /ligo/gitcommon/noise_recorder/figures/darm_noise/darm_sensing_comparison_plot