Several times over O3 operators have made note that they see SDF diffs for H1:IMC-MCL_FM_TRIG_THRESH_ON and H1:IMC-MCL_FM_TRIG_THRESH_OFF, and sometimes these are the only differences that prevent us from immediately going into Observe.  But, it doesn't happen all the time, which has made it somewhat confusing to track down.

I think I've finally grok-ed what is going on.  These thresholds are set in the IMC_LOCK DOWN state.  For cases where the IMC goes through its DOWN state while we're still ramping down in power after a full IFO lockloss, the DOWN state would set those thresholds assuming that we wanted to try to acquire lock at high power.  If the IMC caught lock, and then stayed locked until we next arrived at NLN, we would see an SDF diff.  However, if we ever lost the IMC and it had to run through its DOWN state again once we were already at 2W, then it would set the thresholds as if we wanted to reacquire at 2W. 

But, we seem to not have needed the filter that is triggered, regardless of input PSL power, for several months.  (The filter exists, and used to be triggered, to help the IMC relock.)  So, since it seems that we are doing fine without this filter, and anyway we nearly always reacquire IMC lock at 2W, I have just set these threshold values to be the 2W values, no matter the PSL input power.  This should prevent these from showing up as SDF diffs anymore.

Note that this reload of IMC_LOCK is the first one since June 4th, so IMC_LOCK only knew about the lscparams file as of June 4th.  Sheila and Keita did a lot of sleuthing (Keita is writing an alog), but it seems that the IMC_LOCK guardian thought that our final PSL power was still 35W since we hadn't reloaded the IMC guardian in so long.  See Keita's alog about his and Sheila's investigations for more details.

It seems that the PRM position for DRMI ASC needs to be reset to better match initial alignment.  Right now, the initial alignment sets the PRM in a place that is pretty close to where it should be for full IFO (this is about -0.6 on the POP_A QPD pit_inmon signals, which looks like -0.8 at the PRC1 error signal since the POP_A QPD has a digital offset of -0.2).  But, the DRMI ASC wants to put the PRM to a place where the PRC1 error signal is zero.  This makes DRMI ASC a little hairy, since having the PRM move so much so fast is difficult for other dofs to follow (last night we decreased the PRC1 P gain by a factor of 10 by hand, today I've put into the guardian to reduce the gain by only a factor of 5).  This then makes the full IFO ASC (including soft ADS loops) slow, because they have to move the PRM back to where the initial alignment wanted it. 

When we had the DRMI ASC on, I moved the PRC1 offset to see if the buildups were okay if the PRM were closer to its initial alignment location, but the DRMI buildups didn't seem so happy.  I didn't wait for full ASC convergence, and I didn't go all the way to where the initial alignment location for PRM was, so we probably need to revisit this next time we are at DRMI ASC (especially now that I see that the full IFO PRM place is quite close to where initial alignment leaves it).  For now, so that we survive the DRMI ASC, I've lowered the PRC1 pit gain in the guardian to 2, from its typical value of 10.  We'll probably want to speed it back up once we reset the setpoint.

Separately, for PRX initial alignment:  Last night Corey had to increase the gain of PRCL by hand, but I think that is just because I had forgotten to hit 'load' on the ALIGN_IFO guardian after last night (alog 51044) re-reverting the gain to the high value.  After loading the guardian today, PRX locked.  However, the ALIGN_IFO guardian didn't think it was locked, since the PRM position (which had been set most recently by offloading DRMI ASC during the previous lock) was so far wrong.  The guardian wanted POP_A_LF to be greater than 1.0, but it was hovering around 0.85.  I lowered the 'locked' threshold for PRX in lscparams.py to be 0.5, and then the ALIGN_IFO guardian was able to align PRX as normal.  Probably once we fix the DRMI PRM setpoint to match the initial alignment and full IFO positions, POP_A_LF will start out much closer to its usual locked and aligned values, so we can put the locked threshold back to its old value of 1.

FRS13367

Following sw-lvea-aux network switch problems Saturday 3rd August, several digital video cameras were not working. Their MEDMs had frozen data, existing image viewers were frozen and new image viewers could not be restarted.

h1digivideo2 was rebooted at 08:50 PDT.

h1digivideo0 was rebooted at 12:55 PDT.

h1digivideo1 was rebooted at 12:58 PDT.

   End-Y - HEPI fluid level is 9 0/16 unchanged from last check. No new leaks observed. Changed out some of the absorbent pads that had become saturated. 

   End-X - HEPI fluid level is 8 8/16, unchanged from last check. No new leaks observed. 

   Corner Station - HEPI fluid level is 8 6/16, down by -1/16 from last check. 
   The leak between the pump housing and the motor's output shaft may be growing a little. Cleaned up the area and changed out the absorbent pad. Will put down a drip pan to get an idea of drip amount. There is another small leak at the other end of the pump station near the pressure gauge for the accumulator rail. wiped up and placed a clean absorbent pad.        

   All three dust monitor vacuum pumps pumps (EX, EY, & CS) check out good. 

   Vacuum pressures needed a couple of small tweaks and the temperatures were all well on the low side of 150f.  

Kara and I swept the LVEA.

The LVEA was left in laser HAZARD

J. Driggers, J. Kissel

Philosophical discussion -- for ISC-related models, we only are really using safe.snaps when things are going wrong. And when things are going wrong, we kind of want to know that every single setting is in the same good state that's ready for the guardian to take over. This is true for ISC-related computers and Beckhoff systems (e.g. ASC, LSC, OMC, TCS, SQZ, EX/EY/CS_ECAT_PLCs), and the ever-blurry suspension systems. We worry -- when it comes to unmonitoring individual filter banks that are currently controlled by guardian that at some point in the future we may take them out of guardian.  Seismic and PSL systems are different. CAL, IOP and AUX monitor computers are different still. We didn't resolve this today, but this is a constant debate.
Further, it seems that some suspension alignment offsets have been monitored, and some has not. This, again, depends on the person reconciling, the philosophy they're adhering to, and how they feel that day. How I felt today is indicated below -- but I ended up unmonitoring 4 of these suspensions, which mankes them now all consistent.

We've taken the following action on safe.snaps:
SUSITMX -- accepted and *unmonitored* OPTICALIGN_OFFSETS
SUSBS -- accepted and *unmonitored* OPTICALIGN_OFFSETS
SUSITMY -- accepted and *unmonitored* OPTICALIGN_OFFSETS
SUSETMX -- accepted and *unmonitored* OPTICALIGN_OFFSETS
    H1:SUS-${OPTIC}_${ISOSTAGE}_OPTICALIGN_${DOF}_OFFSET
   (OPTIC = [ITMX, BS, ITMY, ETMX], ISOSTAGE = [M0, M1], DOF = [P,Y])
   See philosophical discussions above -- today, my philosophy is that suspensions that regularly touched during alignment of the IFO should not have their optic align offsets monitored.

SUSITMX, SUSITMY 
    -- Accepted new spot position A2L gains on ITMs (as per "finalized" gains mentioned in LHO aLOG 50972; ETMs had already been accepted).
    H1:SUS-ITMX_L2_DRIVEALIGN_P2L_SPOT_GAIN 
    H1:SUS-ITMX_L2_DRIVEALIGN_Y2L_SPOT_GAIN 
    H1:SUS-ITMY_L2_DRIVEALIGN_P2L_SPOT_GAIN 
    H1:SUS-ITMY_L2_DRIVEALIGN_Y2L_SPOT_GAIN
    -- Accepted A2L TRAMPS on ITMs to be 3.0 sec since it doesn't matter, and OBSERVE.snap wants it to be 3.0.

SUSETMY 
    -- accepted L2_COILOUTF_LL TRAMP as 3.0 -- it was the only one set to 10. Don't underdstand why.

SUSSRM 
    -- accepted (but not unmonitored) M2 LOCK_L TRAMP and GAINs in such a way the ISC_DRMI likes in the DOWN state
SUSPRM 
    -- M3 coil output filter state accepted to match the coil driver state request (2.0). The state request is in the correct configuration, so there's no reason that the filter banks should be accepted differently since they're front-end forced controlled to be correct given a state request number.

SUSIM (specifically IM4) 
    -- accepted (but not unmonitored) the INPUTs to the M1_LOCK_P/Y filter banks as OFF, as requested by the ISC_DRMI in DOWN state.

SUSBS 
    -- M1 LOCK L gain value accepted (but not unonitored) -- confirmd to be in ISC_DRMI in DOWN state.
    -- M1 OPTICALIGN TRAMPS accepted 

SUSETMYPI 
    -- now defunked for observation, but used for testing occasionally -- channels related to identification of ETMY drumhead mode used last on June 26 by Jenne / Georgia (see LHO aLOG 50213):
    H1:SUS-ETMY_PI_DOWNCONV_DC8_SIG_SWSTAT 
    H1:SUS-ETMY_PI_DOWNCONV_IN_MTRX_8_2 
    H1:SUS-ETMY_PI_DOWNCONV_IN_MTRX_8_4 
    H1:SUS-ETMY_PI_UPCONV_OUT_MTRX_2_8
accepted current settings (already accepted in OBSERVE.snap).

ASC  major work here
    -- WITH THE "MONITOR SELECT" MASK ON
    accepted ASC LOCKIN OSC1 channels in the OFF configuration; used for occasional testing but irrelevant for observation / lock acquisition:
    H1:ASC-LOCKIN_OSC1_MTRX_1_6
    H1:ASC-LOCKIN_OSC1_MTRX_2_4
    H1:ASC-LOCKIN_OSC1_MTRX_3_7
    H1:ASC-LOCKIN_OSC1_MTRX_4_8
    H1:ASC-LOCKIN_OSC1_DEMOD1_SIG
    H1:ASC-LOCKIN_OSC1_FREQ

   Accepted or reverted the following channels such that in safe, the start off as ZERO / OFF.
   ASC-[PRC / SRC / INP]_P/Y_SMOOTH_ENABLE channels confirmed to be in PREP_FOR_LOCKING of ISC_LOCK guardian. Accepted -- but NOT unmonitored.
   ASC-[C/D]SOFT_P/Y_SMOOTH_ENABLE -- reverted. 

   Accepting ADS PIT/YAW 10 oscillators to be consistent with MICH BRIGHT ALIGN (initial alignment state, only used occasionally).

   Accepting ASD PIT/YAW Sensing, Output, and Oscillator Matrix "10" values used in MICH_BRIGHT_ALIGN, in the state that the guardian expects

   Now all going to be ZERO -- OFF. 

   Reverted (but not unmonitored)  ASC-PRC1_P filter setting -20dB gain as OFF. See discussion LHO aLOG 51044. Will need to fix this issue with DRMI ASC later in guardian, but for now accepting as though "last night didn't happen."

   Accepted (but not unmonitored) the SRC1 Y offset as ON, because we confirmed that the entire filterbank is controlled by guardian. Note that the *value* is currently ZERO, as per 

   Accepted (but not unmonitored) SRC2 P/Y SMOOTH limits as is, because we confirm they're controlled/set by guardian.

   -- WITH THE "MONITOR SELECT" MASK OFF -- checking DIFFs between the 606 unmonitored channels (there were only ~50 or so)

    TRAMPS are still there.
    Everything else has been accepted to match what guardian wants.

LSC 
    -- 5 diffs reduced to 2, the diffs made consistent with what ISC_LOCK guardian wants in PREP_FOR_LOCKING

OMC 
   -- accepted new OMC QPD offsets, for better optical gain (LHO aLOG 50973) -- though note that these might change due to excess glitching (see LHO aLOG 51034)
   H1:OMC-ASC_QPD_A_PIT_OFFSET
   H1:OMC-ASC_QPD_A_YAW_OFFSET
   H1:OMC-ASC_QPD_B_PIT_OFFSET
   H1:OMC-ASC_QPD_B_YAW_OFFSET

   Accept (but not unomnitored) ADS YAW 5 and ADS PIT 3 in current filter states as is -- confirmed to be consistent with what guardian wants/ controls

ECAT_PLC_2 
   -- accepted LSC_REFL SERVO things that are  consitent with ALS_COMM down state.

ECAT_PLC_4
   -- accepted the LO_SERVO_COMBOOST as 0.0, and LO_SERVO_IN1EN as "off" (or 0.0) because it is set as such in the SQZ_LO_LR guardian in its DOWN state.
   -- accepted the SQZ_VCO_CONTROLS_ENABLE as 1.0 or "on" because it is set as such in the SQZ_LO_LR guardian in its DOWN state.

SQZWFS
   -- accepted SQZ-ASC_WFS_SWTCH as OFF (0.0), because this is requested in the SQZ_MANAGER DOWN state, and via the sub-function called turn_off_sqz.

LSC
   -- H1:IMC-MCL_FM_TRIG_TRESH_ON and OFF. These have been problematic for a while. Here's why: 
      These trigger thresholds are set by the ISC_LOCK guardian depending on what the power is going in to the IMC we when the IMC_LOCK goes through its DOWN state. Occasionally, when the IFO loses lock -- including the IMC 
   -- we'll still be ramping down the power from 37W. So the thresholds will be set for the higher power that's not 2W. Right now the guardian is setting these thresholds by checking the current power. One option to fix this would be to check the requested power instead of the measured power. Another option is to just leave at the 2W value.

    Jenne will post a separate aLOG about this.

CALINJ
   -- Unmonitored the H1:CAL-INJ_TINJ_OUTCOME. This is readback channel written to by the TINJ guardian.
   -- Accepted H1:CAL-INJ_TRANSIENT_TRAMP and H1:CAL-INJ_TRANSIENT_GAIN as 2.0 sec and 0.0 respectively. They've been this way for the entire run, save for brief stints around July 23 and 25 when the power failed. 

Continued work on power cables for NCAL installation. Today, a junction box was placed on the FAC rack in the VEA. Junction box helps transition the 8 awg field power cable to our standard 3W3 power connectors. As of now, no power supplies have been installed in the VDD rack.

To make room on top of the VEA rack the old laser kill circuit box was removed.

F. Clara, P. King

WP8391 h1isiitmy model change

Jim W, Hugh, Jeff K, Dave:

h1isiitmy was restarted to fix some ADC->IPC connection issues. No DAQ restart was needed

DNS issues

Carlos:

ns0 and ns1 were rebooted to fix a DNS issue.

The monthly (starting this month) charge measurement results for the ETMX and ETMY is attached below.

For the ETMX, the effective bias voltage has not risen much since last measurements (3 weeks ago), and the long term trend for all 4 quadrant for the pitch and yaw is below 50V.

For the ETMY, except for the 3rd quadrantfor the Yaw and pitch mode which is now sitting at 50V and around 40 Volts respectively, all other quadrants are well below 50V. The long term trend looks stable and doesn't require a sign flip this time.

The slider values were restored after the measurements were complete and I made sure that there were no SDF differences.

{Gerardo, Chandra}

From roughly 9am to 10:30 am local, we core drilled the three accessable holes at vertex turbo station on OMC tube near HAM5. Interferences prevented us from drilling the remaining two marked holes at XBM turbo station. We left the 100' long water hose in squeezer bay temporarily until this exercise is complete.

Following on from last week's Round 2, at Keita's request I took a series of beam profiles between ALS_M11 and ALS_M12 (ALS_M12 is the bottom periscope mirror, so this is the green beam as it goes into WBSC10 (as close as I can get to it, at least)).  Using the ALS_M11 mirror as the reference point, I took 4 profiles along the path; the profiler sensor was at 0° for these measurements, and the distances are from ALS_M11 to the profiler's sensor:

I've attached a beam profile of the beam at the farthest extent from ALS_M11 that I could get the profiler without bumping ALS_M12 (this is the final data point in the above table).  All of the profiles taken had this shape to it.  To my eye it looks like the top half of the profile is compressed versus the bottom half.

In addition, I rotated the profiler sensor by +45° at each data point to check for any gross off-axis ellipticity.  I've attached an example of this, taken at the same location as the first attachment.  The beam gets more round when the sensor is rotated, showing no gross off-axis ellipticity; all sensor-rotated profiles showed this shape.

Today yellow CAUTION tape and signs were put up around the Green Spool cameras to keep us from inadvertantly bumping them./

All PSL and TCS SDF safe.snap files have been reconciled.  There were no differences for PSL or OpLev.  There were four (4) TCS differences relating to the TCS-SIM; these were changed on 7-25-2019 and have not been touched since, so I accepted them (see attached).

I reset both PSL power watchdogs at 15:08 UTC (9:08 PDT).  This completes FAMIS 10722.

TITLE: 08/06 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Calibration
OUTGOING OPERATOR: Patrick
CURRENT ENVIRONMENT:
    Wind: 3mph Gusts, 2mph 5min avg
    Primary useism: 0.01 μm/s
    Secondary useism: 0.04 μm/s
QUICK SUMMARY:

!4:53 Re-Running PEM injection script as it may have been terminated prematurely.

During the current lock and after being at NOMINAL LOW NOISE on the order of 45 min, all of a sudden have begun to see a broadband bump on DARM (attached is a screenshot showing instance where the bump is seen on the DARM spectrum & you can also see how often it is occurring on the DARM BLRMS striptool).  It is also seen on DMT Omega.  The range has only taken a small hit of about 5Mpc (down to 112 Mpc from 117Mpc).

Only difference I accepted for this lock was the enabled filter for ASC PRC1_P FM1 (-20dB). 

Will start taking a look at the H1 Summary Pages to see if I can glean anything from them.

Note: 

• I increased the Y-axis on DARM BLRMS for the 500-1k & 1k-2k axis from 0.100 to 0.300 so this noisy period could be viewed on the StripTool.
• From 12:37 to now (~8min) it has not been noticeably observed.