Late last week, Deepak and I precariously swapped the problematic and AOSEM brackets described in my previous alog:

https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=5499

I use the word precarious because the operation meant that tools, small fasteners, and hands were wedged inside of the suspension above the optic and extremely close to fragile wires in order to replace the 4 brackets.  This did not make me happy.  See some pictures below, but the collection is also posted  ResourceSpace: https://ligoimages.mit.edu/?r=23901

HAM3-ISI level 3 controllers were designed on Jan 16th 2013. Controllers are shown in the first .pdf attahcement of this post, and saved in the SVN under: /seismic/HAM-ISI/H1/IMC_Test/Isolation_Filters/HAM3/

Those controllers were designed on in-air measurements. In-air VS In-Vacuum comparisons show that the major resonances we account for during ISO loops design are not affected by switching from in-air to in-vacuum.

cdsfs0 was reset. I verified from the control room that I could recover this system "remotely" via the administration port. If it fails again before Tue I'll try to remotely recover it, but for now it appears to be a Saturday problem.

h1seih16 When I got to the control room at 15:30 I noticed that h1seih16 had been frozen from 13:15. I took it out of the Dolhin fabric and tried restarting the models, but no I/O cards are detected. I took a quick look at the IO Chassis and found it without power (front panel switch is ON, 24V power strip has +24V LED ON). It looks like  a power supply issue so I'm leaving this till Tuesday. I'm running the models to "green up" the EDCU, but no useful data is coming from HEPI HAM1.

For the second saturday in a row the /ligo file system at LHO has issued a scsi error and switched to read-only mode. I'll drive to the site later this afternoon to reboot it. We will work on replacing the scsi controller card next week to see if we can fix this problem.

The error occurred at 08:06 today, at at 00:05 last sat.

This morning all the valves that could be closed were closed to isolate the piping for tracking the very slow downward trend in the pressure.  Since Wednesday afternoon there has been a drop of 3psi.  This is way too much to be temperature related.  Hopefully it will just be a loose fitting; wish me luck!

Attached are plots of dust counts > .3 microns and > .5 microns in particles per cubic foot requested from 5 PM Feb. 14 to 5 PM Feb 15. Also attached are plots of the modes to show when they were running/acquiring data.

Data was taken from h1nds0.
1440.0 minutes of trend displayed

Attached are power spectra of PRM for Phase 2b checking.

-Damping seems to be working looking at 2013-02-14_1500_H1SUSPRM_**_ALL_Spectra.pdf

-Comparison between PRM L1 Phase 2a (Jan 2013) X1 (Sept 2012) and H1 (Feb 2013) has been plotted for damping on and damping off

(allhstss_2013-02-14_Phase2b_H1PRM_ALL_Spectra_Doff.pdf and allhstss_2013-02-14_Phase2b_H1PRM_ALL_Spectra_Don.pdf )

**Frequencies of the main DOFs for M1, M2 and M3 are matching the model and L1 power spectras.

**An extra peak appears at ~5Hz on several curves from L1 (ex : allhstss_2013-02-14_Phase2b_H1PRM_ALL_Spectra_Doff.pdf --> M1_DAMP_L_IN1 on page 7). Noisier environment there ?

**The amplitude of some power spectra from H1 PRM is higher than the ones from L1 (ex : allhstss_2013-02-14_Phase2b_H1PRM_ALL_Spectra_Doff.pdf --> M1_DAMP_Y_IN1 on page 12), and most of them are higher at low frequencies.

A reason would be that the PRM was uncovered during measurements (figured out afterwards) and "wind" could have made it moving. Will run power spectra again to check this, now the cover is back on.
 

• See the entry by Kyle and Gerardo about their MX vacuum gauge changeout
• The HAM 2 east door appears to be off
• Rick and Kyohei operated the 10W laser in the LSB lab
• Three alarms from dust monitor 4 occurred mid-morning while Cris was cleaning the work platform and the floor in the BS test stand clean room
• Betsy:  Changeout of PR3 SUS brackets
• Sheila:  Worked on picomotors at EY
• Justin:  Heavy lifting at clean and bake
• Keita, Kyohei, Gregorio:  Inspected the OMC in the H2 PSL enclosure
• The CDS channels listed in the morning entry remain yellow
• Work permits signed by ops today:  3717 (HAM 2 cleanings), 3718 (10W laser), 3719 (MX vacuum gauge)

Weekly trends. Next week we may need to increase the current on pump diodes 3 and 4 as they've dropped to 90% of their nominal value, and the output power of the 35W laser has dropped below 95% of its nominal value.

Kyle, Gerardo 
Replaced PT343B gauge head -> The 1.5" metal angle valve (in series with the 1.5" O-ring valve) was found to be over torqued (closed) as its depth indicator was "off scale" combined with the observation that nominal applied torque resulted in no perceivable stem rotation -> We couldn't demonstrate that this valve's over torqued knife-edge seal leaked appreciably after replacing PT343B -> the rate of pressure rise indicated by PT343B when isolated from the BT and the pump cart (while rising through the 10-6 - 10-5 torr range) did not change noticeably when the O-ring side of the re-closed metal valve was vented versus when it was under vacuum 

We helium leak tested all joints of X1-9 prior to opening to the BT (baseline 1.5 x 10-9 torr*L/sec)  

Note: the indicated pressure value is not calibrated -> we will use neighboring gauges/history to adjust new gauge when time permits

During the last few days of vent activity in BSC1 late last month for the ITMy alignment, I parked the BSC1 particle counter inside the chamber with us.  Attached is a plot noting the events that the particle counter witness (thx, Justin!).  Things to say about the plot:

1) The particle counter has small spikes to ~15 counts/cub.ft during the quiet nighttime hours while no activity was taking place anywhere in the LVEA.  Note, we'll take a second monitor into the chamber during our next study such that we can confirm counts.

2) The highest counts seem to be during the soft cover activity.  My guess is that the settled particulate in the chamber gets quickly stirred up by the soft cover (or has even collected in the soft cover since that is the direction of "out") and gets re-tossed around in the chamber.

3) There is a longer than expected ringdown of particulate counted for the hours after people have left the chamber and the soft cover had been restored to the door.  It takes many hours for the particulate to settle back to the floor or horizontal surfaces after getting kicked up from traffic.

4) The overall counts of ~100 cnts/cu.ft are in line with what Robert reported in his in-chamber particulate measurement last month:

https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=2071

For the record, I wiped the flooring in the chamber as I exited the chamber (not shown on the plot), pulling from the far side towards the door with many IPA saturated wipes.

Apollo got the doors off HAM2 and I followed with locking the ISI--no difficulties.
I took a few photos, mainly focused on the cabling coming off Stage0 to the electrical feedthrus.  Lot's of work to do in this department; the NW corner is pretty good but the others, not so much.  The last image is from the East (right) side and shows the PR mirror's dummy masses that will need to be removed.

Minor administrative changes to hpi pump controller h1hpipumpctrll0, configured resolv.conf to allow other computers on network to be found by name, and started the ntp daemon to keep the time up to date.  Time was reset to current time (it was off by a few minutes).

There were no unacknowledged alarms ringing this morning.  Channels that were in alarm status were the following:

• H!SUSB123-SUS-ITMY-CPU-MAX (yellow)
• H!SUSB6-SUS-ETMY-CPU-MAX (yellow)
• H1:IOP-SUS_B123_ITMY_WD_OSEM5_LPOUT (yellow)
• H1:IOP-SUS_B123_ITMY_WD_OSEM6_RMSOUT (yellow)
• H1:PSL-OSC_XCHILALARM (red)

After a restart of the alarm handler, the two CPU-MAX alarms and the two IOP-SUS_B123 alarms returned (yellow).

late entry

As part of WP3715 I added a DAC output part to h1pemmx and restarted the model. This removed the "DAC not in use" flag but added in the "18bit DAC does not have an AI chassis connected" flag. Since the DAQ status went to 0x2000 (ini file change) I reverted the model back to the original for now rather than restart the DAQ. We will install the AI chassis for the PEM to green-up these systems.

Not a show stopper, but darn unfortunate for a variety of reasons.  After installing the glass PR3, I noted that you cannot easily adjust the middle mass AOSEM  positions.  After a call to LLO to confer, I discovered that they reworked the brackets just before installation of their PR3!  Fortunately they made 1 set extra, so it was overnighted our way.  The unfortunate part is that I need to remove 8 silver plated screws from the awkward position on the inside of the cage over the optic, rework the bracket assemblies, and then reinstall them.  The attached picture shows the "old" brackets - a circle indicates the captured adjutment screw.

We pulled the soft cover off the North side of HAM3, and took a series of pictues in the chamber. 

These pictures will be used to asses the state of the ISI payload when we measured the ISI transfer functions for the Viton Pads Efficiency Investigation.

Deepak and Bubba report they're done for the day, so commencing Matlab TFs on PR3.