Posted below are the OpLev 12 hour trends.

At GPS time 1107854779 [2015-02-13 09:26:03 UTC], H1 appeared to unlock but the ISC_LOCK guardian state didn't change. The attached plot shows the X-arm power and the guardian state to confirm (hopefully).

At the time of writing the guardian is still reporting 'DC_READOUT' as the current state, while there is no circulating power in the arms. The other guardian nodes (e.g. ISC_DRMI) did respond to the lockloss, however, so I presume it's just an oversight in the ISC_LOCK DC_READOUT state definition to check that the other guardians are still nominal, or that the arms are actually locked.

I'm always a step behing the on-site commissioners (good job!), but here is the list of coherence for the 2+ hours lock:

https://ldas-jobs.ligo.caltech.edu/~gabriele.vajente/bruco_1107760396/

Here are my main comments on it:

• In the low frequency range (below 20 Hz) there is coherence with some suspension signals from HAM1, HAM2, ITMY, see the first three attached plot for some examples. In particular coherence is high with HAM1/2 at 15 Hz and ITMY at 30 Hz. I'm not very familiar with those systems, so I probably didn't choose very meaningful channels, but if you look into the full table you'll see many of them.
• It's not high, but nevertheless significant: there's coherence between 10 and 100 Hz with PEM-EY_MAG_VEA_FLOOR_*
• Between 40 and 1000 Hz there is broadband, even if not large, coherence with POP_LF, probably due to intensity noise. However, this is the largest coherence in most of this band
• At 75 Hz there is coherence with the magnetometers at the Y end: I guess it's the well known ring heater line
• There is a "bump" of coherence with IMC angular signals between 287 and 300 Hz. This might be one of the beam jitter resonances. There are similar bumps at about 550 Hz and 820 Hz

All day next Tuesday will be scheduled for maintenance
Model changes planned for next Tuesday

Sheila and Richard will install a low pass filter box between the IO piezo controller and periscope
PCAL was calibrated at end X yesterday
PCAL may have a bad AOM at end X

There is a request to label the analog camera images

model restarts logged for Thu 12/Feb/2015
2015_02_12 00:56 h1fw0
2015_02_12 15:33 h1fw1
2015_02_12 17:14 h1broadcast0

two unexpected restarts (first h1fw0 unexpected restart since 11 days ago). Broadcaster reconfigure for DMT channel addition. Conlog frequently changing channels report attached.

Dan, Jeff, Sheila

Tonight we had some more fun noise hunting.  Turning the power up to 8 Watts was easy, but didn't improve the noise so much since we are mostly not shot noise limited. 

• First, when we got to DC readout we noticed that someone did a nice job damping the ETMY violin mode, reducing the peak in the asd by about a factor of 2.  This allowed Dan to turn on one stage of whitening in the OMC DCPDs, which helped improve the noise floor.  
• We further reduced the gain of the ETMX oplev damping (by a factor of 2 for a gain of -5).  This reduced coherence around 10 Hz and helped the DARM noise floor.
• We have been running all day with the ESD linearization on, this seems to be fine. 
• We had the ISS second loop on for the first long lock of the night, it has been off ever since.  

This gave us a lower noise floor, we sat here from 5:28:10 UTC Feb 13th to 5:41:29 UTC, with the intent bit undisturbed. 

We then made a single destructive attempt to reduce the ESD bias voltage, we'll have to come back to this.  :)

On the next lock we decided to try increasing the input power.  This was suprisingly easy, we have now been sitting at 8 Watts input power since 7:03:43 UTC.  We have been adjusting the BS alignment by hand every 10-15 minutes or at 10 Watts.  We have watched the AS36Q WFS signals while we do this, and can see that they are not always correct (zero does not always give us the best AS90).  BS yaw dramatically improves the intensity noise coupling.  It seems as though the IFO would stay locked like this all night if we sat here and adjusted the BS once in a while.  We are leaving it locked and undisturbed.

In the attached plot, the green trace is earlier today, The blue trace is with 1 stage of DCPD whitening on and the ISS second loop on, which is injecting noise at around 700 Hz.  The red trace is now, with 8 Watts input power, BS YAW oplev damping gain reduced, and the ISS second loop off.  You can see that we are not really shot noise limited, as increasin the power by a factor of 2.8 only gives us a small improvement at high frequencies.  Something is clearly wrong with the calibration.  

I examined the sidebands of the OMC ASC dither lines today.  The excess noise around these lines is disappointingly broad, roughly +/-2Hz.  There are sidebands around the dithers at +/-0.18, 0.45, 0.49, and 1.5Hz.

The attached figure compares the sidebands of the first dither line (middle plot) to the low-frequency DARM noise (top plot).  The bottom plot is the noise around the violin mode.

Sheila, Dan, Elli, Kiwamu, Lisa, Peter, Robert

Here is a calibrated DARM spectrum from today's DC-readout lock, using the new CAL-CS channel described in LHO#16698.

According to GWINC, the BNS inspiral range of this spectrum is 8 Mpc.

H1BROADCAST.ini was modifed to add the signal H1:DAQ-BCST0_CHANS_SAVED to the DMT. The DAQ broadcaster was restarted.

07:00 Karen and Cris in LVEA
08:06 Jeff B. turning off dust monitor 16
08:16 Bubba to LVEA to work on LN2 piping for LTS
08:17 Jeff B. done
08:29 Corey to squeezer bay for 3IFO work
08:35 Rick to end Y to check on Thomas and Sudarshan
08:39 Bubba done
09:12 Jim W. working on SEI BSCs
09:17 Alastair putting 3IFO equipment in cabinets in LVEA
09:47 Corey out of squeezer bay
09:48 Jim W. done
09:54 Robert to IOT2 and PSL enclosure
10:02 Travis back from end Y
10:05 Corey and Andres to LVEA
10:09 Corey back
10:13 Andres back
13:39 Filiberto to end X, not VEA, to read serial numbers and pick up tools
14:18 Filiberto done

Added 889 channels reflecting model changes from yesterday.

Several people reported that the DARM calibrated signals from the OAF model did not look healthy in the past two or three days.  Looking at the data in frequency domain, I confirmed that they have behaved funny. Cosulting with Jeff K, we decided to gradually migrate the functionality from OAF to CAL-CS model (alog 16669) because CAL-CS is going to be the official calibration place in future anyways. We are still in the process of moving from OAF to CAL-CS, but a calibrated DARM signal is now available in CAL-CS. The funny behavior seen in OAF is not seen in CAL-CS so far. The calibration accuracy is not yet examined.

(Bad OAF data)

Looking at the data in frequency domain, indeed they looked funny -- the spectral shape of, for example, OAF-CAL_DARM_DQ was suspiciousely featureless with the noise floor around 100 Hz much higher than it should be by a couple of orders of magnitude. Also it did not show a roll-off shape at 5-ish kHz for some reason. A confusing fact is that it sometimes looks behaving correctly and sometimes doesn't. I did not make a further investigation because we decided to move to CAL-CS.

(CAL-CS)

I copied the DARM-related filters that I had in OAF over to CAL-CS. So it is right now completely a duplication of what we had in OAF. I did not move the DRMI, CARM, or IMC filters yet.

The scripts you need are attached here. The main file is bruco.py

When you call the script, there are some parameters that must be set:

# Command line arguments (with default values)                                                                                                                                                                                                                   

# --ifo=L1                    interferometer prefix                                                                                                                               # --channel=OAF-CAL_YARM_DQ   name of the main channel 

# --gpsb=1087975458           starting time

# --lenght=180                amount of data to use (in seconds) 

# --outfs=8192                sampling frequency of the output results (coherence will be computed up to outfs/2 if possible) 

# --minfs=512                 skip all channels with samplig frequency smaller than this

# --naver=100                 number of averages to compute the coherence 

# --dir=bruco_1087975458      output directory    

# --top=100                   for each frequency, save to cohtab.txt and idxtab.txt this maximum number of coherence channels

# --webtop=20                 show this number of coherence channels per frequency, in the web page summary

1. load a list of all channels saved to the frames at the starting time
2. use only channels with the minimum sampling frequency specified in the options
3. delete from the list all the channels in the exclusion list. This is saved in a file called bruco_excluded_channels.txt. The names in this file can use wildcards
4. bruco loops over all the remaining channels, compute the coherence with the main channels and:
   1. save a plot in the output directory. The top panel of the plot is the coherence in double log scale. The script compute the statistical minimum meaningful coherence based on the number of averages, and plot it a a dashed red line. The bottom panel of the plot shows the noise projection, based on the coherence. Only points above the meaningful threshold are plotted
   2. store the value of the coherence and the channel name, for each bin, if the coherence is within the top values, as listed in the parameters. This information is saved in two files cohtab.txt and idxtab.txt. Although you will never really need to use this, here is how it works. For each frequency bin, cohtab.txt lists the highest coherence values, in descending order. This is a huge nbin x nchannels table. The file idxtab.txt is a matrix with the same size, each entry being a unique id that tells you what channel corresponds to that coherence. The fiel channels.txt contains the list of all channels and their id
5. When all channels have been processed, bruco generates a web page summary, that contains, for each frequency bin, a list in descending order of the most coherent channels. The table is color coded based on the coherence

Here is an example, that I'm running right now with the last night lock:

./bruco.py --channel=LSC-DARM_IN1_DQ --gpsb=1107760396 --lenght=600 --outfs=4096 --naver=300 --dir=/home/gabriele.vajente/public_html/bruco_1107760396 --top=100 --webtop=20 --minfs=32 --ifo=H1 

Beware that the processing of all channels takes many hours, when I run it on one machine in the Caltech cluster.

There might be some options to tune in the script, to set where the data is saved. Mainly in lines 106-109. The script is now configured for the Caltech cluster, but I imagine it should be easy to point it to different places where the data is.

As suggested by Dan and Lisa, we might try to speed the script up by running it in parallel with condor or reducing significatly the list of channels. Now, bruco automatically selects 2390 channels.

Elli, Evan

Summary

After Peter and Kiwamu's deep insights into the in-vacuum POP signal chain (LHO#16691), we've transitioned control of the DRMI DOFs from POPAIR to POP with DARM controlled on RF.

Details

First I drove a 1000 ct, 89.1 Hz line into the PRM in order to phase POP. I set the demod phase for POP9 to 90°, making the Q signal a factor of 50 smaller than the I signal. For POP45, I set the phase to 66°, making the Q signal 25 times smaller than the I signal.

With the line still on, I looked at the TFs of the POPAIR/POP signals. For 9I, 45Q, and 45I, POPAIR is weaker than POP by a factor of 0.15 ct/ct, with 0° of phase difference. So I took the POPAIR matrix elements, rescaled them by 0.15, put them into the corresponding POP matrix elements, zeroed the POPAIR matrix elements, and then transitioned.

Then Elli and I remeasured the OLTFs of PRCL, MICH, and SRCL. The PRCL gain was a bit low, so we increased it by 10%. The SRCL gain was a bit high, so we decreased it by 20%. Now the UGFs should be roughly the same as for POPAIR.

The new matrix elements are as follows:

• POP9I → PRCL: 0.035
• POP45Q → MICH: 0.081
• POP45I → SRCL: 0.08
• POP9I → SRCL: 0.005

This last matrix element is just a rescaling of the matrix element for POPAIR9I → SRCL (which itself is just a rescaling of REFLAIR27I → SRCL), so it should be retuned to optimize the PRCL/SRCL subtraction.

Hanford tells us that they will not be hauling container loads on Friday 2/13 either on day or swing.  Their operations will remain quiet through the end of Monday 2/16.