Sheila, Jenne, Georgia

We took some time coming back from Tuesday maintence to work on our DRMI ASC, continuing the work we started on Friday 49586  We have been able to run all the DRMI ASC loops and offload them before doing the CARM offset reduction. 

• We moved the picomotor 1 for the POP QPDs, not well enough to be centered on the diode but well enough that we have a reasonable signal.  When DRMI was aligned so that the build ups were reasonably good, Jenne adjusted the POP A offsets so that we can use POP A as an error signal for PRM alignment in DRMI. 
• We increased the gain for SRC2 in the guardian, as we did manually on Friday. 
• We also phased AS72 so that we could use ASA72Q as a signal for the SRM control.  However, we found that we weren't able to increase the full IFO power with this phasing, which is similar to what Georgia reports here 49393  Instead, we were able to engage the SRC1 loop with the same offset that Georgia used for full lock.  
• We sometime between O2 and O3 we started using AS45Q for MICH control when we first lock DRMI, then switching over to 36 after all the ASC was on and the mode hopping offset was off.  This greatly helped with the reliablity of the DRMI ASC, so we want to keep this stratey.  However, we had some difficulty transitioning the BS control to 36.  We decicded to just phase ASA 36 to 0 the Q signal in each quadrant with DRMI locked, which has allowed us to transition. 
• Georgia found and fixed a bug in the state generator for offloading ASC that was causing the offloading to misalign things.  

Chris S., Gerardo M., Kyle R.

We used the HILTI core drill to core (4) ea. 1 1/8" x 6" holes in the LVEA floor at the Y-mid.  I then installed 2 of 4 threaded inserts but had a problem with the installation tool.  I'll fix the tool and finish installing the threaded inserts next Tuesday. 

These new threaded inserts are for anchoring the new turbo station which is scheduled for installation on 6/25. 

J. Kissel

After restarting the front wall Main DELTAL EXTERNAL ASD figure of merit, I saw that PCALX did not show the expected high-frequency roaming line. A quick look at the PCALX overview, and it was easy to see that the optical follower servo (OFS, the intensity stabilization and linearization servo for PCAL) was railed. The main time-series trend on the overview, of the OFS PD -- H1:CAL-PCALX_OFS_PD_OUTMON -- was steady at -7.7 V, with no sign of the expected ~kHz oscillation. 

I trended back to find out when the OFS had railed, and it appears coincident with the Beckhoff restarts this morning at 15:06:46 UTC (08:06:46 PDT). The OFS PD went to zero then, and then a few minutes later (at 15:13:44 UTC, 08:13:44 PDT), after bouncing around between OFF and RAILED, it settled on RAILED.

I fixed this by turning OFF, then turning ON the OFS loop, i.e. toggling the orange H1:CAL-PCALX_OPTICALFOLLOWERSERVOENABLE "loop enable" switch next to the diagram of the loop -- also on the PCALX overview. I did so at 23:11:55 UTC (16:11:55 UTC), before we resumed any observation ready segments, so this has no permanent impact on that quality of data.

We should consider adding "check on the PCALs!!" to the recovery checklist after any maintenance day in which a Beckhoff restart has taken place.
We should also add "Restart the wall FOMs!" to the recovery checklist after any maintenance day in which a DAQ restart has taken place.
Such activities should be considered akin to a physical VEA sweep done at the close maintenance.

TITLE: 06/04 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
INCOMING OPERATOR: Cheryl
SHIFT SUMMARY:

Maintenance day went well. The remainder of the day was used by commissioners to work on adding and optimizing DRMI ASC loops.
LOG:

1515:01 GRBV alert

15:04 Lockloss - Sheila induced

15:16 Sheila out to optics lab or squeezer bay to look for beam profiler

15:17 Sensor correction turned off at end stations.

15:18 Chris taking pest control into LVEA

15:21 Kyile out to MY - WP 8226

15:24 Fil out to pull cable from HAM5/6 to CER - climbing on chambers

15:25 Niko and Driptha out to EY for PCal work

15:26 Fire System maintenance on site

15:34 Dave B turning off Fire Pump Cell Phone notices until noon

15:36 EY to LASER HAZARD

15:46 Chris and pest guys back

15:47 Richard out to LVEA HAM5/6 area

15:52 IMC_LOCK guardian is now OFFLINE and unmanaged at Dr Kissel's request.

15:55 Richard back

15:57 Dick out to LVEA and CER for measurements

15:58 Karen leaving EY

15:58 Vanessa out to LVEA

16:03 Karen out to MY

16:09 Amber called about tour coming to CR around 10:00ish

16:10 Christina out to LVEA

16:23 Hugh heading out to EY to tend wind Fence

16:30 Karen leaving MY

16:30 Coffey Refigeration coming through gate - Jeff Bartlett

16:31 Norco on site - dewar 71

16:40 Rick and Laurence out to EY

16:50 Betsy and Sheila out to LVEA

17:09 Norco on site(second load) _ dewar 77

17:12 Vanessa back and heading out to EX

17:32 Marc out

17:35 Richard out to roof

17:36 Pest control done

17:39 Gerardo out to LVEA

18:00 Tour into control room

18:10 Initial Alignment re-locking ALS

18:10 Kyle back from MY

18:13 Sensor correction re-engaged

18:17 Niko out to optics lab

18:22 Input Align - following ALS attention

18:51 Begin locking

18:55 Daniel and Patrick out to LVEA

20:17 Pep out to CER to move a magnetometer

20:22 Pep back

I'm not sure if this looks any better, but all the slow controls displays are now in one block, including buttons to open the SDF screens. Also the SDF diffs are now marked with a RED circle since guardian DIAG_SDF is monitoring these systems. Snippet is attached.

Dave, Hugh:

The h1hpipumpctrlsdf_monitor.req file I put together this morning was not complete. Hugh and I have extended it to include all the detailed PID settings. I then restarted h1hpipumpctrlsdf and ACCEPT+MONITOR'ed all the new channels into h1hpipumpctrlsdf_OBSERVE.snap.

Attached screen shows the complete set of channels.

WP 8227

Daniel, Filiberto, Patrick, Richard

The code in PLC1 on h1ecatc1 was changed to use the Vantage Pro 2 library that I created prior. The weather station console was connected to port 2 (RS232 for channel 1) of an EP6002-0002 serial interface module. This module was named Corner MSR Weather and added to the end of the MSR Auxiliary device chain. A screenshot of the COM settings is attached.

FRS 9765

New Davis weather station installed in CS. Unit running on Beckhoff code. Old unit was removed.

F. Clara, R. McCarthy, P. Thomas

WP8231 Remove h1calcs from OBSERVE SDF snap exception

Jeff, TJ, Ed, Dave

Actually the scope of this WP was extended to removing all SDF safe->OBSERVE transitions exceptions. The current exceptions were for h1calcs (it is unknown why this was) and most, but not all, slow controls systems (again, it is not known why).

h1calcs: Jeff has made both safe and OBSERVE symlinks to the h1calcs_safe.snap file. I removed this model from the exception list.

slow controls: I created a checker script (check_beckhoff_sdf) which 1) informs of any systems with different safe and observe references and 2) systems which do not start up in safe.

1) only h1hpipumpctrl had different safe and observe snaps, and its safe snap was incorrect. I made the two point to h1hpipumpctrlsdf_OBSERVE.snap and updated this file with the new Beckhoff setpoint channels for the EX system (CS and EY are still using Ben's Purple Box).

I restarted all the slow controls SDF systems on h1ecatmon0 to resync to the new start state and the changes Daniel made this morning.

The changes were made to the file /opt/rtcds/userapps/release/isc/h1/guardian/switch_SDF_source_files.py

• -exclude_dict = {1024: 'sysecatc1plc1',
• -                1026: 'sysecatc1plc3',
• -                1027: 'sysecatx1plc1',
• -                1029: 'sysecatx1plc3',
• -                1030: 'sysecaty1plc1',
• -                1032: 'sysecaty1plc3',
• -                1033: 'hpipumpctrl',
• -                1034: 'pslopc',
• -                117:  'calcs',
• -            }
• -#                117:  'calcs',
• +exclude_dict = {}

Ed reloaded ISC_LOCK to get the new code.

WP8232 Guardian SQZ node channels in DAQ-DMT broadcaster

Dave:

I removed the not-needed GRD-SQZ channels and add the one which was needed to the broadcaster list:

-[H1:GRD-SQZ_BEATNOTE_OK]
-[H1:GRD-SQZ_SHG_OK]
+[H1:GRD-SQZ_MANAGER_OK]

This change went in when the DAQ was restarted.

Beckhoff Slow Controls Work

Daniel, Patrick, Dave:

New Beckhoff code resulted in six new INI files (C1PLC[1,4], X1PLC[1,2], Y1PLC[1,2]), the same list of autoburt.req and three new SDF monitor lists (C1PLC4, X1PLC2 and Y1PLC2).

The DAQ restart and SDF monitor restarts installed these.

DAQ Restart

Dave:

The DAQ was restarted for the above changes. To prevent out-of-band data showing up in EDC channels, I used my new DAQ restart procedure:

1 - generate the new H1EDC.ini file

2 - check the DAQ has no duplicate channels

3a - restart the DAQ at the appropriate time

3b - a few seconds later, after h1dc0 daqd has stopped, restart h1edc on h1susauxh34

This appears to have succeeded.

   Both TCS chiller water levels are at the full mark. I did not add any water to either.  

J. Kissel, P. Covas

Part of the focus of Pep's current LSC fellowship was to continue work mitigating sharp features in DARM to help out the CW group's search sensitivity. Recently, he's suspected that features in the ~15-50 Hz region of DARM (and confirmed in PRCL and SRCL as well -- see LHO aLOG 48947) are corner station triple suspensions, predicted to have their highest frequency vertical (a.k.a "bounce") and roll modes (V3 and R3) in this region. As such, we've remeasured all of these suspensions' V3 and R3 mode frequencies, using this morning's maintenance period to complete the inventory -- see LHO aLOG 49643.

Armed with this inventory, we then perused all triple suspension notches for these mode frequencies. Below are the list of changes we've made based on our findings, and the expected features that the changes are expected to improve -- ranked by the hopeful amount of improvement.

SR3
    - [Attachment 1]Changed/fixed the frequencies of FM9 "SB27.7" and FM10 "SB45.3" elliptic bandstop filters in the M1 DAMP banks for every top mass DOF (L, T, V, R, P, Y).
     >> We expect this will reduce the 27.71 Hz feature in DARM.

PRM
    - Turned ON FM8 "SB27.5" and FM9 "SB40.9" elliptic band stop filters that were already there in the M1 DAMP banks for every top mass DOF (L, T, V, R, P, Y).
     >> We expect this will reduce 27.59 Hz and 40.87 Hz features in DARM.

BS
    - [Attachment 3] Added previously non-existent FM8 "SB17.79" and FM9 "SB26.06" elliptic band-stop filters in the M1 DAMP banks for every top mass DOF (L, T, V, R, P, Y).
    - [Attachment 2] Updated the frequencies of the FM5 "NotBR_BS" elliptic band-stop filter collection in the optical lever OLDAMP banks for P and Y.
     >> We expect these actions will help reduce the 17.79 and 26.06 Hz features in DARM.
    - [Attachment 4] Turned OFF FM2 "not19.1" and FM3 "not23.1" elliptic bandstop filters in the M3 ISCINF P and Y banks. These, we believe, are unneccessary -- intended to be temporary -- notch filters single-line oscillator excitations for exploratory ASC loop tuning. These were accpeted inthe SDF by Jeff Kissel back in Oct 2018 (see LHO aLOG 44277) during the height of commissioning, and were likely just accidentally left ON after a night of loop tuning.
    >> We expect this will claw back a bit of phase for the MICH ASC loops, but otherwise have little to no effect (other than cleaning up an unneeded mess.)
    - Changed the names of FM5 "NotBR_BS" and FM7 "NotBR_Quad" from their original ambiguous names "BounceRoll" and "EvanBR"
    >> We expect no change  in DARM from this change, just cleaning up the filter banks.

PR3
    - Moved FM2 "SB46.09" elliptic band stop filter in to FM10 (overwriting the unused, incorrect frequency filter that was there before) in M1 DAMP bank, in order to 
     >> We expect no change in DARM from this change, just making PR3's filter location consistent with every other SUS.

For all of the above changes, the new settings have been saved into both the safe.snap and OBSERVE.snap such that they stick.

MC2
    - [Attachment 5] We *looked* at the various implementations of notch filters for the highest bounce and roll modes (V3 and R3) for MC2 (and comparing against MC1 and MC3), but decided that the notches were well centered enough, and because we don't see any of MC2's resonances in DARM, we elected not to make any changes.

All other corner station suspensions have been confirmed to have correct notches for their newly re-measured highest bounce and roll (V3 and R3) modes. 

WP 8229
ECR E1900035
E1600386

Following Beckhoff terminals were added to SQZ Controls Chassis 3:

1. EL9410 QTY 2
2. EL3104
3. EL4132
4. EL1124
5. EL2124

New DB25 cable was pulled from SQZ-C1 rack in the CER to SQZ-R1 rack in the LVEA. Part of the AM modulated AOM amplifier installation, alog 48353.

F. Clara, R. McCarthy, M. Pirello, D. Sigg

Operators have been saying that they have difficulty adjusting the ALS polarizations lately.  I tried to adjust the Y fiber polarizaton today and saw the stange behavoir in the attached screenshot.  There are certain steps that are not like the others, and seem to make a large jump in the polarization. 

Pep Covas, Jeff Kissel

This is an update on past's week alog regarding bounce/roll measurements:

Due to maintenance work requiring the fire pumps to be operated, I've bypassed these cell phone alarms.

Bypass will expire:

Tue Jun  4 12:36:36 PDT 2019

For channel(s):

    H0:FMC-CS_FIRE_PUMP_1

    H0:FMC-CS_FIRE_PUMP_2

I looked at 30 days of H1 range and the BLRMS channels, and found a clear increase in the frequency range 100-450Hz (H1:OAF-RANGE_BAND_5), and 400-420Hz (H1:OAF-RANGE_BAND_10).   I found 6 channels that have a clear decrease, and two channels that are either flat or jump around, making their contribution unclear with this first look.

PLOT 1: Increasing BLRMS:

• H1:OAF-RANGE_RLP_5_OUTPUT, 100-450Hz
• H1:OAF-RANGE_RLP_10_OUTPUT, 400-420Hz* (both RLP_9 and RLP_10 show 400-420Hz on the main medm)

PLOT 2: decreasing BLRMS:

• H1:OAF-RANGE_RLP_1_OUTPUT, 10-20Hz
• H1:OAF-RANGE_RLP_2_OUTPUT, 20-29Hz
• H1:OAF-RANGE_RLP_3_OUTPUT, 38-60Hz
• H1:OAF-RANGE_RLP_7_OUTPUT, 1K-2KHz
• H1:OAF-RANGE_RLP_9_OUTPUT, 400-420Hz* (both RLP_9 and RLP_10 show 400-420Hz on the main medm)

PLOT 3: flat/variable BLRMS:

• H1:OAF-RANGE_RLP_4_OUTPUT, 60-100Hz
• H1:OAF-RANGE_RLP_6_OUTPUT, 530-980Hz

M. Pirello, D. Gustafson

We installed new v3 Baluns on the following signals in the CER:

ISC-C3(37-1) 40Mhz TCS AOM Return exchanged v1-117 with v3-S002

ISC-C3(33-4) 158.8Mhz Fiber Beat Note Return exchanged v1-193 with v3-S003

ISC-C3(33-5) 79.4Mhz SQZ VCO Return exchanged v1-151 with v3-S004

ISC-C3(33-6) 203.125MHz SQZ VCXO Return exchanged v1-052 with v3-S005

ISC-C4(39-3) 79.4MHz PSL VCO Return exchanged v1-093 with v3-S010

These are all return signals, the impact should be minimal.  Work was completed per WP8150

Balun status can be seen here E1900100.