We were thinking about whether we need to increase the bandwidth of the ASC MICH loops during the carm offset reduction sequence, so I measured ASC-MICH while sitting in DRMI (with ALS holding the arms off resonance).  I also re-measured the loops after increasing the gain by a factor of 2.  But, adding the factor of 2 increase into the guardian seems to be a bad idea, since several times this higher gain caused oscillations in the ASC MICH loop(s).  So, it's commented out, although we could still consider adding more gain if we measure near the resonance around 0.5 Hz more carefully.

The plots attached all are the data points for which I had better than 0.6 coherence for both IN1/EXC and IN2/EXC.  I plot measurements of MICH_P and MICH_Y for both the lowGain (original gain), and highGain (with the extra factor of 2).

Following are the consolidated bounce (vertical) and roll modes measured in-vacuum and in-air for the 4 QUAD suspensions.

J. Kissel

After Travis and I increased the PCAL line heights the other day (see LHO aLOG 43676) such that we might be able to see their amplitude in the current level of DELTAL_EXTERNAL sensitivity to get a preliminary and very rough check on the calibration. I've since gathered an ASD of the IFO's sensitivity compared with PCALY in a recent DC readout stretch (2018-08-28 06:31 UTC).

As is rather obvious by the comparison, the calibration -- at least at 331.9 Hz -- is roughly a factor of (6.70948e-17 / 2.27309e-17 = 2.9517) ~~ 3 wrong. This isn't much of a surprise, since we're no where near the O2 level of laser power, but I just wanted to remind folks that we shouldn't yet trust the accuracy of the calibration.
We can't yet resolve (at least in this lock stretch) the 36.7 Hz line, but we can resolve the 7.9 Hz line, which appears to be pretty accurate.

Red curves are "live" (well, the recent Aug 28 06:31 UTC data), green curves are similarly "live" PCAL ASDs, the gold trace shows O2 levels of sensitivty -- where the PCAL lines were at the same frequency, but less amplitude, and within ~2% of the amplitude shown at 36.7 and 331.9 Hz.

J. Driggers, J. Kissel

I was deceived by some red lights on the CDS overview posted to the wall thinking that there were DAQ problems with the H1 SUS ETMX last night while I was gathering data to characterize the highest bounce and roll modes (see LHO aLOG 43713). However, I'm reminded that it's an inconsequential flaw with this version of RCG that the ETMs are running (see LHO aLOG 43039).

Long story short: I was looking at the CDS overview screens and found some filter files that were showing differences, indicating unloaded changes in a few filter banks. The list of front-end models with pending changes, and Jenne and my course of action taken is below.

    H1 LSC -- looks like a precision change of the sampling rate in the filters in MICH1 and SRCL1 banks, otherwise no changes to the actual function. The full filter file was loaded, and there are no longer differences.
    H1 SUS ITMX -- Small changes to the violin mode filter notching on the L2 LOCK stages; like the work of Sheila's finding out of the new yesterday. The filter file was loaded, and there are no longer differences.
    H1 ALS EY -- what looks to be substantial changes to the "ctrlHB" FM10 in the H1:ALS-Y_WFS_DOF_2_P filter bank (see attached screenshot for details). We know that Hang Yu has been working on this (see, for example, LHO aLOG 43516), but he's currently out of town, so we'll (for now) leave it unloaded and ask him about it. 

[Sheila, Haocun, Nutinsee, Terry]

Yesterday we also did a better OMC cavity scan with beams from squeezer.

Compared with last time, we used:

• SEED power input on ISCT6: 54.2mW    <--- high seed power for good alignment
• Green power  input on ISCT6 : 8.8mW     <--- lower green power to get rid of green depletion
• Temp: 51.63C     <--- higher dual resonance crystal temperature to get rid of NLG

The results are shown in the attachment.

• Input power from IR PD: 1.24mW
• Power in the cavity: 0.78mW

     ---> This tells us the propagation and loss from Faraday is ~7.34%. (SRM transmission = 32.34%)

In the Cavity:

• 00 mode: 90.4%
• 10/01 mode: 0.38%  --> The alignment is very good
• 20/02 mode: 8.2%    This is the loss from mode matching. (Keep consistent with the former measurements.)

I'll energize IP1 in a few hours.  We want to "sign-off" on IP1 asap so as to, then, be able to shut down our noisy pumps near HAM3. 

Last night, the Hartmann team turned on the ring heaters on ITMY as well as ETMY. We set the ring heaters to half of a watt for 3 hours.

Based on the HWS ITMY prism trend data and gradient/contour plots (see attached), it appears that this Hartmann beam is well centered on ITMY. We could try to move a bit in pitch if needed. 

The HWS ETMY did not acquire useful Hartmann beam information due to the following error: "ERROR: array must not contain infs or NaNs - State 3D". This error usually occurs when the Hartmann code obtains reference centroids while the green beam saturates the camera.  It appears that the green beam was not shuttered during the measurement (see ETMY trend data attached) so this is mostly likely the case. Will try again tonight if commissioning permits. 

As of 9:45 am local time, the LVEA is laser HAZARD.

Sheila, Georgia, Craig

We are leaving the IFO in the "LARGE_EQ" state because of the Mag 7 EQ in New Caledonia.  

In the morning an initial alignment will be needed.  

We've commented out the y arm states from the initial alignment guardian to avoid confusion . These states align IM4 + PR2 to the Yarm, which is only used when trying to align the input beam to the y arm for example for measuring the Y arm loss.  During our normal initial alignment we want to set these to align the input beam to the x arm, not the y arm.  

Georgia is starting a ring heater test at 5:00:30 UTC.  She put 0.5 W to top and bottom on ETMY and ITMY. for 3 hours duration

Changes are seen on both GigE cameras.  Using the path length to the camera, and the pixel size, I calculated the angle change of the beam on IO GigE 2, and used that to calculate the change at 20m, the distance to MC2, and of course these are approximations, based on the current installation.

Mark D., Tyler G., Chandra R., Kyle R. 

Gerardo M. and I had previously found a leak on the newly installed chevron baffle spool-to-valve flange.  Today I was able to vent IP1 and decouple the leaking flange and inspect it.  At the site of the leak, there was evidence of damage to the spool's knife-edge.  This "feature" is characteristic and typical of a flange fastener (stud in this case) having dragged across the knife edge during the flanging process when the two flanges are close but not in contact with each other.  I can't say for certain but this likely could have been prevented had all of the studs been removed prior to bringing the two mating flanges together.  Another abnormality was that the imprint of the knife-edge into the gasket material was very shallow in the region of the gasket corresponding to that which couldn't be brought "metal-to-metal".  Gerardo will say more on this in his entry.  

I consulted Chandra R. and we decided to attempt an in-situ "stoning" of the damaged knife edge.  John W. had shown me this technique years ago and I have successfully repaired many "dinged" flanges using this approach.  This entails finding a donor flange of the same size in the recycle bin and using it to pre-condition an abrasive stone such that arcs can be worn into it which match the knife-edge of the flange to be repaired.  I then wet-stoned (using IPA) the damaged knife edge which removes the displaced material and restores the cross-sectional profile of the knife-edge (minus the displaced, now removed, material). 

During the re-flanging following the repair attempt, I noticed that the new gasket fit loosely in the gate valve's flange.  I tried a gasket from a different manufacturing batch and it too fit loosely.  It looks like the tolerances are not standard in the valve's machining or are looser in the OD than typical at the very least.  How loose?  I didn't measure but would guess that the center of the gasket could be off as much as 0.040" from the valve's bore in any direction. 

Anyway, I re-flanged, pumped down and helium leak tested the new joint.  It looks good initially but I am leaving the leak detector connected and running overnight and will continue at the next available opportunity. 

{Mark, Tyler, Gerardo, Kyle, Chandra}

Outstanding group effort today in vacuum department. Mark and Tyler installed chevron baffle + newly rebuilt IP6 ion pump (replaced a temporary blank on IP6 isolation gate valve); Kyle, Gerardo, Chandra leak checked.  In summary we think the assembly is leak tight but there was some uncertainty because of leaky vent port on turbo plus multitude of KF joints. We saw a He signal in e-7 Torr-L/s range, but quickly discovered it was coming from the vent port on small turbo pump (vent port o-ring may be wrong size). We sprayed the vent port with instrument air and still saw a rise in He signal from background of 2e-9 Torr-L/s, into low e-8 Torr range, so we tried a few tests by toggling the right angle valve (RAV) on IP6. I sprayed the baffle flanges with the RAV closed (i.e. isolating the leak checker + external turbo from IP6 body) and we observed a rise in pressure. Similarly, while spraying the baffle flanges with valve closed while simultaneously spraying turbo area with instrument air, and then quickly opening the right angle valve to expose IP6 body to leak checker, there was no rise in He signal. 

We finished late, so we will need to resume pumping on IP6 before we can turn its high volts on. It is currently pumped out and valved out. Leak checker was left on mezzanine. We will clean up our mess asap.

A few notes:

• I sprayed the chevron baffle louvers with "top gun" N2 dionizer in LVEA this morning before installation.
• The baffle was installed on IP6 at ground level and then the total assembly was installed on isolation GV via crane.
• IP6 was vented with N2 yesterday to give us a head start since it takes ~30 min. to vent. It was tested for high volts last week.
• Tyler noticed a white goo on the threaded joint (thread to KF adapter) of external He calibration leak so we replaced it with another unit before we started leak checking. Suspicious about this "wet" white substance found on a vacuum component. It was not visible from the inside, but plentiful on external surface.
• Gerardo hack sawed off a stripped bolt on one of the HV feedthru connectors on IP6 and replaced with new bolt. Both HV feedthrus are attached along with grounding cable.

Appreciate everyone's patience today as we used a full maintenance day....and then some. :)

Soon we will have not four, but six ion pumps valved into corner station.

The code looks at H1:SQZ-VCXO_CONTROLS_DIFFFREQUENCY to determine whether or not it's locked and looks at 6MHz RF mon to determine whether or not it's ready to lock. I gave it +-5 Hz wiggle room for lock condition. frequency diff normally wiggles around +-2 Hz.

J. Kissel

The new highest vertical (a.k.a "bounce") and roll mode frequencies for H1SUSETMX, measured at vacuum using driven excitations a.la. methods described in LHO aLOGs 41488 and 40098 are as follows:
    
    V4   9.703 +/- 0.005 Hz

    R4  13.757 +/- 0.005 Hz
I'll update the uncertainty estimate later, once I can download the excited data I mention below.

I've driven up the modes at two times (listed below) in order to gather an estimate of the Q, which I'll do again similarly to the aLOGs cited above once I get three.

             Exc Start                Exc Stop
Trial 1      Aug 28 2018 23:50 UTC    Aug 29 2018 00:02:00   (stopped with 10 sec ramp)
Trial 2      Aug 29 2018 01:05 UTC    Aug 29 2018 01:15:00   (stopped suddenly but turning off TEST output)
I'll want to get one more excitation time (plus an hour of ring-down time data). I'll look for a good target of opportunity in the next few days. It''s already obvious from watching the modes ring down in the sensor ASDs that the Qs will be in the 1000s, unlike ETMY which was unusually low (in the mid-100s) and split.

The template for measuring the ASDs of sensors lives here:
     /ligo/home/jeffrey.kissel/2018-08-28/2018-08-28_H1SUSETMX_M0_V4R4_ModeSearch_BBinj.xml

[Jenne, Craig, PatrickG]

We have a new plotting / calculating routine that will plot a measured DTT transfer function, that has been measured via noise injection.  As EvanH points out in alog 27518, data points with only modest coherence will be biased toward unity if you are just looking at the traditional DTT IN1/IN2 transfer function.  So, now we have a command line tool that will extract the measurement data from DTT, and calculate the open loop gain from the cross spectral densities, and save a plot.  Then, if you want to know what the loop shape would look like if you added some other filters, you can use the second tool to extract filter module info from python, apply those filters to the measured OLG, and plot the result. 

As is often the case with low gain loops (like our ASC right now), we can't push hard enough to excite with noise all of the frequencies that we are interested in.  So, I have measured the CHARD loops with several bandpassed noise injections.  The DTT extractor cannot get info about references, so each measurement must be saved in a separate .xml file.  For example, I have 3 separate .xml files for the CHARD_P measurements that I took yesterday. 

The output from the first plotter shows that CHARD_P, as it was yesterday, had very very low gain - see first attachment. Each color is the data from one of the frequency bands, for data points where the coherence between (IN1 and EXC) and (IN2 and EXC) are both greater than 0.5. 

The output from the second - see second attachment - is the measured TF from above, with the indicated filters applied (here, FM1 and FM3, which is what guardian wants to have turned on for the power up).  You can see that if we were to include both of these filters, we'd be unstable around 9 Hz. 

Things that would be nice to add: a user-chosen coherence threshold, and the possibility of a user-defined gain value, and plotting of the loop suppression, so we can see if there is any expected gain peaking in the candidate OLG.  Farther-future would even be to add error bars to the measurements, given the coherence.  Also, a more generic location for the scripts.

For use right now (in a folder that contains your DTT .xml files):

>> /ligo/home/jenne.driggers/git/dtt_tools/bin/stitchTFs.py *.xml       (In the case that you want all .xml files from this folder.  Otherwise list them explicitly, space separated)  This extracts the measurement data, saves the plot and the calculated OLG. 

>> /ligo/home/jenne.driggers/git/dtt_tools_bin/plotFilters.py filename.h5 FM# FM#    (The previous command saves the .h5 file, so pass that file name to this script.  Then list the FMs that you want to add, space separated)  (ex. filename.h5 FM1 FM3)

(Cheryl V, Gerardo M)

The yellow cover was removed, the viewport already had a viewport protector installed. The lexan cover was LOTOed, and the camera housing installed.

A WATEC camera was installed and pointed at the required component.  Power was set temporarily and will be finalized soon, along with the camera signal to the control room.