Ed, Jason, Dave:

It looks like the Beckhoff OPC computer in the diode room is not accessible over the network. This machine is called h1pslctrl0.

At first we wondered if the CER network switch had failed, but the digital cameras are working, only the diode room computer is down.

Jason is investigating remotely and Ed is going into the diode room to check on the computer.

Camilla, Sheila

We went back to ISCT6 during the commisioning time this afternoon.  We repeated the measurement done here: 54566

We also changed the lens in the path to the temporary AS air 1811 diode.  After realigning this we found that we could get a much larger signal out of the 1811.  Camilla measured 147mV of DC power on the 1811, with a transimpedance gain of 1V/mA and responsivity of about 0.75A/W this is about 196uW, enough that we could be saturating the diode.  Last time we measured the power in the path to this diode,  54636, we saw only 130uW, so there is a bit of an inconsistency there.  Not suprisingly, the RF spectrum out of this 1811 is dominated by 45MHz and harmoincs, perhaps so much that we are also having RF saturations. After demodulating at 3.125MHz, we see a large noise signal at 270kHz.  

At this point we decided to quit for today and go back to triple coincidence.  

TITLE: 01/23 Day Shift: 16:00-00:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Observing at 118Mpc
INCOMING OPERATOR: Ed
SHIFT SUMMARY:

There was COMMISSIONING time (1hr51min) today.

SEI_DIFF has been in the CORNER_DIFF_CPS state, but Jim noted that since microseism signals are below the 90th percentile, we could switch SEI_DIFF  to FULL_DIFF_CPS...with low microseism, this would mainly help us ride through and acquire better with higher winds.
LOG:

• 16:05 Clearing tumbleweeds for Anemometer move at EX (Scott, Chris)
• 16:51, 19:42 MX trip (Vanessa)
  • 18:00-18:47 MY cleaning (Karen 
• Noticed on DIAG MAIN:  BRS_CHECK:  BRS X C code has stopped
• 18:10 - 21:05 Anemometer move (Scott, Hugh)
  • 19:08:31 (gps 1263841729) Time when Wind Sensor #5 (direction sensor) plugged in by Hugh.
  • 19:31:39 (gps 1263843117) Time when 2nd Sensor plugged in.
  • 20:19:14 (gps 1263845972) Time when 3rd sensor plugged in.
• 21:32 EQ Alert (5.5 N. Argentina), will monitor.
• 21:34 - 23:25 COMMISSIONING time begins; also coordinated with LLO.
  • 21:34 DRMI Guardian reloaded (Camilla, Jenne)
  • 21:36 SEI_DIFF transitioned to FULL_DIFF_CPS since we have low microseism & this can help us with winds.
  • 21:37 - 23:25 Out on floor for SQZ Mode matching/alignment (Sheila, Camilla)
  • 21:58 - 22:35 Forklifting items from LSB Receiving to Fire Water Tank Sea-Container. (Tyler, Scott)
• 22:45 -22:47 Benton Sheriff vehicle (SUV w/ lights flashing) sped into site & turned around at LSB Parking lot.  No contact made.
  • 22:57 - 22:59 Hanford Patrol also entered (both vehicles hung a right toward SR 240).

S. Karki, J.Kissel, S. Dwyer, C. Compton

We made two set of sensing function measurements during a routine calibartion measurment time on Monday 2020-01-20. We made the first measurement in normal configuration and the  second with A2L gain turned off.  The measurements, analysis scripts and the results are saved in the appropriate location in the cal svn as listed below:

Measurements:

svn/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2020-01-20_H1_DARM_OLGTF_LF_SS_5to500Hz_15min.xml

svn/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2020-01-20_H1_PCALY2DARMTF_LF_SS_5to500Hz_10min.xml

svn/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2020-01-20_H1_DARM_OLGTF_LF_SS_5to500Hz_15min_A2LGainOff.xml

svn/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2020-01-20_H1_PCALY2DARMTF_LF_SS_5to500Hz_10min_A2LGainOff.xml

Analysis scripts: (First script collects all past sensing function measurements and compares against these new two measurmnets and the second script compares the new two measurements by themselves)

svn/aligocalibration/trunk/Runs/O3/H1/Scripts/FullIFOSensingTFs/process_sensingmeas_collection_20200120.py

svn/aligocalibration/trunk/Runs/O3/H1/Scripts/FullIFOSensingTFs/process_sensingmeas_collection_20200120_A2LGainOnvsOff.py

Detail results can be found inside the following folder (svn/aligocalibration/trunk/Runs/O3/H1/Results/FullIFOSensingTFs/) but the main plots are attached below.

The first plot contains the comparison between the two measurements made on 2010-01-20 and the second plot compares them against all past sensing function measurements made during O3B.

The goal was to see if the sensing function is different at low frequency between these two configurations and results show that they are. Next step is to create a working model that describe these results. 

Managed to get through the tumble weeds to layout the wind sensors.  Attached is 6 hours showing the direction sensor and the two speeds starting back up a few hours ago.  Speed_4 is on top of the building near the PEM sensor and has continued to run while all the others were off line.  I'll put in a comment with an updated location map but I'll say now that all the senors are 39' +Y of the Wind Fence line.  Maybe not exactly free-stream but that was all the cable we have.  I hope we can extend this to get further +Y and higher altitude.  The direction sensor and speed2 are 17' above ground level (AGL) and speed3 is 12' AGL.  They are about 21 to 29 feet -X of the Wind Fence X position center.

Today Jeff K. pointed out that the results from the final position survey of the NCal at EX were never posted to the alog; this post rectifies that.

We surveyed the NCal in 2 ways: we measured the vertical position of the top, center, and bottom of the NCal rotor w.r.t. WBSC09=0.0; we also measured the X, Y, and Z coordinates of a reference punch mark that Timesh placed on one of the support struts for the NCal housing.  This punch mark was roughly placed in line with the center of the NCal rotor.  Please recall that the H1 ETMx sits at -80.0 mm w.r.t. WBSC09=0.0 (from E1400205).  And now, the results:

• Rotor height w.r.t. WBSC09 = 0.0
  • Top: -52.3 mm
  • Center: -70.9 mm
  • Bottom: -96.2 mm
• X,Y,Z coordinates for reference punch mark
  • X: 3998931.4 +/- 0.7 mm
  • Y: -1243.1 +/- 2.9 mm
  • Z: -71.1 +/- 0.05 mm

Please keep in mind that the above coordinates are only for the reference punch mark, not for the center of the NCal itself.  Timesh will have to provide those coordinates, as that is a calculation he made that I do not have access to.

• Anemometer installation at EX is ongoing this morning.
• Planned COMMISSIONING out-of-observing time is slated for 1:30pm local time.
• H1 is humming along at 10.5+hrs of lock.

I did a quick look at the DetChar pages to see if I could get a sense for the utility of the new wind fences. We do not have the free-stream anemometer running again at EndX, so for now I'm just using the roof-top PEM anemometers to get a qualitative sense.

The 'flip-book' style set of pages below shows that the fences work pretty well.

Before the fence, when the wind is blowing, the histograms of the roof-top sensors are all about the same.

After the fence, the wind speeds at the two ends are clearly lower that the speed at the corner station. By eye, the wind speed is not quite a factor of 2 smaller, but close. The slab tilts at the end are much smaller. (flip between pages 4 and 5). There is a lot to see in these plots, and hopefully they can be used to ask interesting questions that a REAL study can try to answer.

These are also in the DCC at G2000122.

Similar to this alog in Nov 2019 by TJ, noticed DIAG MAIN quickly flashing the message:  "BRS_CHECK: BRS X C code has stopped".

Looking at the DIAG MAIN log, we've had a flurry of these errors over the last 24hrs (generally lasting about an hour).  A symptom is that during these errors the H1:ISI-GND_BRS_ETMX_VEL channel drops to 0.  Attached are three periods of these errors/drops over the last 24hrs.  I notified Hugh and he mentioned this could mean a restart of the code would be needed.

TITLE: 01/23 Day Shift: 16:00-00:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Observing at 117Mpc
OUTGOING OPERATOR: Cheryl
CURRENT ENVIRONMENT:
    SEI_CONF state: WINDY
    Wind: 4mph Gusts, 3mph 5min avg
    Primary useism: 0.02 μm/s
    Secondary useism: 0.28 μm/s

Microseism has continued its downward trend from yesterday (just above 50th percentile).  Winds are calm.
QUICK SUMMARY:

• H1's been locked 6.75 hrs since the 6.2 Alaska/Bering Sea earthquake, and has a range which looks closer to 119Mpc (thank you, lower microseism!). 
• COMMISSIONING Time (Sheila):  @1:30pm PST later today down for 1-1.5hrs coincident with LLO.
• Jenne is the LHO Site Advocate.

TITLE: 01/23 Owl Shift: 08:00-16:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 117Mpc
INCOMING OPERATOR: Corey
SHIFT SUMMARY: relocked first try, one break in Observe, ETMY violin modes 12, 18, 20 damped
LOG:

• 09:31:17 - H1 in NLN
• 09:41:23 - done testing ETMY mode 20 new filter
• 09:46:16 - done handling SDF diffs, H1 in Observe
• 11:00:39 - I clicked ETMY mode 12 -30deg phase filter, FM2, and that took H1 out of Observe, unmonitored FM2/3/4/5, -30, +30, -60, +60 deg
• 11:01:30 - H1 in Observe
• 14:45 - JeffB, Tyler, Chris, to mechanical room to move a chiller, rollup door at recieving
• 15:00 - chiller move complete

New Violin Damping: ETMY mode 20 - needs to be monitored by OPS

• current gains for ETMY modes 12, 18, and 20, which all effect each other
  • mode 12: gain = 1
  • mode 18: gain = 10
  • mode 20: gain = 1, the new filter, in FM7, is currently damping, the old fiter is in FM1
    • if there's a need to switch, ramp the gain to zero, switch, and raise the gain
  • these are updated from my alog, alog 54667
  • I believe that the filtering will be reset to nominal at relocking, since those switches are called out in the Violin guardian's lscparams file, but would be advisable to make sure
• new filters in ETMY modes 1 and 6, and ITMX mode 2, none are tested

With a new damping filter for ETMY mode 20, I have damped modes 12, 18, and 20, with a great deal of monitoring, so be aware that the settings are not tested, have only worked tonight, and not always well, so they're available, but not to be used unattended.

Current Gains for ETMY:

• mode 12: gain = 10
• mode 18: gain = 10
• mode 20: gain = 1
• gains that are higher will work, however, with higher gains, when there's a glitch, anywhere from one to all 3 modes will ring up

I added 4 new damping filters and one monitor filter into the violin mode filter banks for 5 modes, saved and loaded.

• Damping filters added to ETMY modes 1, 6, and 20, and ITMY mode 2.
• Monitor filter added to ETMY mode 7.

I took 10 minutes to try the new ETMY mode 20 filter, after reaching NLN, and it looked promissing, but I wanted the option to use either the old or new filter, so I unmonitored FM1 (old filter) and FM7 (new filter) in SDF.  In order to put a new filter in FM7, I removed an ancient 1.5K violin mode damping filter.

To clear SDFs I accepted ETMY mode 7 changes, in the attached snapshot, and I unmonitored ETMY mode 20 filters FM1 and FM7.

Later, switching the phase on ETMY mode2 caused H1 to be kicked out of Observe, so I unmonitored the phase filter banks, FM2, FM3, FM4, and FM5.

Attachments:

1. ETMY modes 12, 18, and 20 - current damping, 30 minutes, showing gains used to damp all 3 modes
2. ETMY modes 12, 18, and 20 - full history from the beginning of the current lock
3. ETMY mode 20 filter bank snapshot, in NLN, in the current damping configuration, though the gain may be different
4. ETMY mode 1 filter bank snapshot, in NLN, untested
5. ETMY mode 6 filter bank snapshot, in NLN, untested
6. ITMX mode 2 filter bank snapshot, in NLN, untested
7. SDF diffs for ETMY mode 7 monitor filter and gain changes, where gains match all other monitor filter gains, and I created the mode7 monitor filter
8. SDF diff for ETMY mode 20 filter bank, which had the new filter enguaged, but zero gain on the filter bank, FM1 and FM7 are not not monitored
9. SDF snapshot of unmonitored filters in ETMY mode 12, FM2, FM3, FM4, FM5
10. power spectra showing the peaks for ETMY modes 12, 18, and 20, from the beginning of the lock, before using the new filter on mode 20, and current, as of this alog

H1 locked in Observe.

TITLE: 01/23 Owl Shift: 08:00-16:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Earthquake
OUTGOING OPERATOR: Ed
CURRENT ENVIRONMENT:
    SEI_CONF state: EARTH_QUAKE
    Wind: 3mph Gusts, 1mph 5min avg
    Primary useism: 0.16 μm/s
    Secondary useism: 0.29 μm/s
QUICK SUMMARY: unlocked due to EQ, relocking asap

• NOTE: Heavy fog on site.  Less fog near Richland.

I'm submitting an early summary as re-locking may not be possible before my shift is over

• 04:07 Intention bit Commissioning
  • Squeezer unlocked
  • called Dave B regarding the timing error in H1IOPASC0
    • Diag Reset cleared the error
• 04:09 Intention Bit Observing
• 05:59 6.2Mag EQ from Japan
  • H1 to EQ Mode
• 06:00UTC Lockloss - 6.2 EQ  from the Tanaga Volcano area, Alaska (not so much Japan)
• 06:03 Holding IFO at DOWN until ground motion settles

I am experimenting in trying to manually multiply the all the Kappas and set the cavity pole as at the time of measurement on 20200113, as is being discussed in the last two comments in this aLOG thread.

I used ndscope to acquire Kappas for: UIM, PUM, TST, Optical Gain, and Cavity Pole (not really Kappa, just altered the actual value to the one at time of measurement).

Attached here are two plots:

• The first, is where I have taken the response ratio at the time of the measurement and divided it by the underlying model. - You can see some of the wobbles are well described, but there is something going very wrong.
• The second is where I have manually reduced Kappa_PUM to something like 1.005, instead of 1.017. From fiddling with the numbers I came to realise that it is the Kappa_PUM that impacts the frequencies that look way off in the first plot - hence this alteration.

I am beginning to suspect that the estimate of the Kappa_PUM might be grossly exaggerated? Perhaps the frequency where it is being sensed is not a great choice?

[M. Wade, J. Kissel, A. Viets]

Maddie and I have produced new GDS filters for the calibration model update described in LHO aLOGs 54269 and 54473.

I restarted the primary, redundant, and testing calibration pipelines on the DMTs around GPS time 1262990593.  Data seems to flowing normally.

The filters are found in revision 9137 of the calibration SVN here:

aligocalibration/trunk/Runs/O3/GDSFilters/H1GDS_1262900044_no_response_corr.npz

They were produced using the run script

aligocalibration/trunk/Runs/O3/H1/Scripts/TDfilters/H1_run_td_filters_1262900044_no_response_corr.sh

Plots of the frequency response of the filters are attached, comparing them to the frequency-domain model.  Note the line at ~4kHz in the resudual corrections filter.  This is actually in the control correction model, but it shows up the residual corrections filter plot because we normally apply control corrections above 1kHz in the residual path, since the control path is sampled at only 2kHz.  It is surprising to something this large coming from the actuation at such a high frequency.  Moreover, modeling this accurately would require a much longer filter sampled at at least 8 kHz, which we do not currently have the computational power to do.  Given our skepticism, these filters do not model anything in the actuation above 1 kHz.  We have an opportunity tomorrow to update the filters again should we decide that it is a good idea to model this the best we can.

Jeff did a Pcal broadband injection just after the pipelines got running, so once the C00 frames are available, I will add GDS results from that injection.  I also plan to test these filters on real data to see how well the model the response function once enough data is available.