TITLE: 04/04 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 64Mpc
INCOMING OPERATOR: Travis
SHIFT SUMMARY: one lock loss, relocked and in Observe
LOG:

• 13:02UTC - lock loss
• 14:15UTC - Bubba to EY - in the mechanical room
• 14:15UTC - Apollo at Mid-Y
• 14:19UTC - Observe (Bubba still at EY)
• 14:37UTC - JeffB to PSL chiller room
• 14:43UTC - JeffB back

• etmx - FM1 (+60deg phase) added to ISC_LOCK
• itmy - FM7 (-30deg phase) added to ISC_LOCK
• ISC_LOCK loaded, file checked in to svn, and SDF changes accepted

I noticed a feature in the ITMX GigE camera image when I pulled it up yesterday.  It appears to have a very dark line going through the bright features on the left side of the image.  After taking a snapshot and adjusting the image, I see a to be a dark line at the top and on the right side as well.  I calculated the beam shift using the image, which is fraught with error, but still a decent ballpark measurement, which came up with a beam shift in -X of 12mm.  Will be interesting to see what the high-res images show.

TITLE: 04/04 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 72Mpc
OUTGOING OPERATOR: Jim
CURRENT ENVIRONMENT:
    Wind: 7mph Gusts, 6mph 5min avg
    Primary useism: 0.03 μm/s
    Secondary useism: 0.49 μm/s
QUICK SUMMARY: locked over 12 hours, range is climbing, currently 71.5Mpc

TITLE: 04/04 Eve Shift: 23:00-07:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Observing at 65Mpc
INCOMING OPERATOR: Cheryl
SHIFT SUMMARY:
LOG:

Quiet shift, one PI (27) needed addressing. Not much else to report.

Here is a comparison between O1's cross correlation noise and that of O2.

It is more obvious that noise above 60 Hz is worse than that of O1 by a factor of a few every where (except for a broad peak at 1.2 kHz which is gone in O2). The O2 data is from a long lock stretch from last night. I used the data starting at 03/Apr/2017 2:00:00 UTC with a frequency resolution of 0.1 Hz with 5000 averages (corresponding to ~ 7 hour integration).

TITLE: 04/03 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Observing at 68Mpc
INCOMING OPERATOR: Jim
SHIFT SUMMARY:  One lockloss due to EQ midway through the shift.  Otherwise, no issues to report.
LOG:  See previous aLogs.

GRB alert 22:23 UTC.  Begin 1 hour standdown.

J. Kissel
WP #6557

In order facilitate regular, bi-weekly, CAL measurements, I've added a new state to the ISC_LOCK guardian code called "NLN_CAL_MEAS." So far, the state only turns OFF all calibration lines, as is needed for the frequency dependent sweeps. Upon request to return to NOMINAL_LOW_NOISE, it turns the calibration lines back on, and clears the history in the calculations for front-end calculated time-dependent correction factors (which, sadly, will take several minutes given the impulse responses of some low pass filters in the calculation).

I'll install and test the state tomorrow (4/4) after maintenance is complete.
Thus far, I've only 
    - checked for a clean svn status before I started,
    - written in the change to the ISC_LOCK guardian, 
    - saved it,
    - checked if the graph was OK with the new edges, 
    - commented out all new code
    - committed ISC_LOCK to the svn
so if the code, for whatever reason, gets loaded before tomorrow, there will be no change applied.

I attach the test of the new edges and the representative graph to show how this state fits in the bigger picture.

Peter, Jason, Vern, Nutsinee

WP6556

We thought it's worth to get some most recent ITMY data to compare along side with ITMX to make sure there's no obvious shmud on the ITMY surface. Given the wrong in vac lens and the possibility that the alignment might have changed we might not get anything useful. But if the alignment is good in theory we should still be able to see if there's anything wrong on ITMY surface by subtracting the final gradients after power-up/lockloss from the reference (taken while resonates at 2W).

Because the two cameras can't be run at the same time, I swapped the fiber so that h1hwsmsr machine is now talking to the HWSY camera rather than HWSX camera. Data written to /data/H1/HWS_ITMX starting from 20:19:29 UTC is data from ITMY surface.

Log:

20:08 UTC HWS code stopped

20:15 UTC cable swapped

20:19 UTC HWS code restarted, now writing data from HWSY camera

The HWSY SLED has been turned on also.

No issues relocking after EQ ringdown.  A slight tweak of ETMx was all that was required.

6.5M EQ in Botswana took us down.  LLO was already down by the time we lost lock.

Starting CP3 fill. LLCV enabled. LLCV set to manual control. LLCV set to 50% open. Fill completed in 18 seconds. TC A did not register fill. LLCV set back to 18.0% open.
Starting CP4 fill. LLCV enabled. LLCV set to manual control. LLCV set to 70% open. Fill completed in 102 seconds. LLCV set back to 35.0% open.

Everything appears normal. The chiller plots were generated pretty zoomed and aren't visually what I'm used to seeing but I think the fluctuations are rather normal.

Following up on this earlier observation of a change point in 
DARM combs on March 14 and on comments added by Evan
and Ansel, I tried looking at possible correlations between 
optical lever laser power levels and combs seen in EX and EY
magnetometers.  The bottom line is that there are such
correlations, as documented in detail in the attached 
annotated slides with posted spectra at a variety of power
levels (as measured by H1:SUS-ETMX/Y_L3_OPLEV_SUM_OUT_DQ).

The figures below show EX and EY before/after spectra.
In summary, the spectra show narrow lines before March 14
in combs of ~1-Hz spacing with various offsets (EX worse than EY). During 
periods when the laser powers were lowered, the spectra were contaminated
by broadband noise (motivation for keeping the power high, I believe). 
When the power was raised to higher than it was prior to March 14,
the broadband power dropped again, the lines persisted and were
joined by new lines with 1-Hz spacing offset from integer frequencies
by ~0.25 Hz, a comb observed to be present in DARM following March 14.

The apparent creation of a new comb in DARM from these lasers may
hint that other combs seen for a long time in DARM and in magnetometer 
spectra may also arise from these lasers. Whether the mechanism is 
though the optical lever damping, as suggested previously for glitches,
or through another coupling, such as ESD power supply contamination
from the laser supplies, is not clear to me.

Fig 1: EX magnetometer spectrum before maintenance work on March 14 (Interval "A" on 1st attached slide file)
Fig 2: EY magnetometer spectrum before maintenance work on March 14 (Interval "P" on 1st attached slide file)
Fig 3: EX magnetometer spectrum before maintenance work, on March 15 (Interval "Q" on 1st attached slide file)
Fig 3: EY magnetometer spectrum before maintenance work, on March 15 (Interval "V" on 1st attached slide file)

Attachment 1: Annotated slides showing spectra before / after many change points on March 14/15 of EX laser power
Attachment 2: Annotated slides showing spectra before / after a handful of change points on March 14/15 of EY laser power

HAM 6 pressure spiked at around 23:00 UTC time on 3/29.