Ref WP#3489

Prior to this work, EY H2 Coil drivers were disabled to prevent any damage occuring to ETMY and TMSY (Vern Sandberg and Jim Batch).

Various H2 front ends were rebooted by Alex. The failure of all front end data as detected by the DAQ data concentrator was eventually reproduced, along with the spontaneous crash of h2dc0 when the front end recovered.

Alex spent most of the day with the H2 DAQ down while he tried various upgrades to the MX_STREAMS drivers and code on the concentrator. 

By 4pm the H2 DAQ was returned to operational. At around the same time the H1 DAQ was restarted with a complete set of VACUUM and FMCS channels, so that when the H2 DAQ is taken down over the weekend these channels will continue to be trended.

Alex will continue this work through to Monday. The aim is that the bug whereby a single front end restart/reboot  invalidates all front ends data for many minutes at a time will be resolved.

Ref WP#3492

Due to today's H2 DAQ work, I reconfigured H1 DAQ to also acquire the VACUUM and FMCS EPICS channels. The process was:

h2ldasgw0 and h2ldasgw1; stop acquiring data, tar up the raw minute files for all the VE and FMCS channels. These tar files were 62GB and 48GB respectively, and have 277 and 382 number of channels respectively.

compess the VE and FMCS tar files, size reduced to 18GB and 8GB respectively.

copy the h2ldasgw0 files to h1ldasgw0, copy the h2ldasgw1 files to h1ldasgw1. Uncompress and install the VE and FMCS files into the minute_raw trends dirctory on h1ldasgw[0,1]

Reconfigure h1fw0 and h1fw1 to acquire VE and FMCS files. The H0EDCU_VE.ini and H0EDCU_FMCS.ini files were copied from H2 DAQ to H1 DAQ.

The H1 DAQ was restarted. VE and FMCS data in full frames, second frames and minute frames started from scratch. The raw minute trend frames were then appended with the latest data, and all previous minute trends are available back to the VE and FMCS epoch.

In addition, the H1 TIMING EPICS channels from the h1ecatc0 IOC were added to the frame. These channels cover the status of the H1 Timing Master in the MSR. The new channels have the prefix H1:CDS-TIMING_

The new ini file is called H1EDCU_TIMING.ini

In prep for a possible move next week, we've installed the cookie cutter, cleared a path on the ISI table, removed a bunch of the QUAD dog clamps, and disconnected all of the QUAD cables such that the ITMy is ready to swing 90deg into it's new position.

After 2 days and some heroic efforts by a few guys, whose identities shall remain incognito, the formerly-known-as-FMy full suspension has been removed from under the test stand and parked just outside of it.  It is strapped to a leg of the test stand now and will be relocated to the BSC2 ISI in another week or after the ISI finishes a few rounds of testing.  The maneuver was particularly difficult since the QUAD was still under the ISI, taking up a lot of real estate.

HughR, GregG, JimW

 Between yesterday afternoon and today, the stage 2 of this assembly had its cap sensors aligned, gapped, then the stage was floated and leveled. After more cps fine tuning, a first pass at cable routing and a round of locker tuning I think stage 2 is pretty well dialed. On Monday we will do more of the same on stage 1, then one more pass at cabling. After that, this ISI should be ready for its first round of testing.

Chub, Gary, Fil, Richard
We all spent the week getting the RF system in the front end of the IFO setup.  Specific accomplishments include:

1.  Powered up all RF components in the CER to propagate signals out into the LVEA.
2.  Made up all in-situ RF cables in the CER.  Verified all cables and RF components are functioning properly at the right magnitude and frequency.
3.  Measured RF losses for each part of the RF distribution system in order to know what value attenuators to add to set the signal levels appropriately.
4.  Made up all RF in-rack cabling in the ISC IMC racks out next to the input HAMs.  Final QA testing is ongoing.
5.  Verified up to Local Oscillator inputs, that the correct RF signal levels exist for locking the IMC length and for proper operation of the IMC WFS detectors

This work will continue next week as the last checkout items are completed.

Rich

I started the x1ioptstsus1 and x1susquad models on the bscteststand2 computer in the staging building.  Manually started awgtpman for iop after correcting an error in the testpoint.par file.

Jim Batch and I have turned off the following coil drivers and AI chassis at EY to allow Alex and company to test and examine the front-end code in search of "mx-streams" issues on the H2 system.

SEI Rack    3 EA    SEI coil driver chassis
SUS Rack    3 EA    AI filter boxes
TCS Rack    1 EA    Ring Heater Driver, 1 EA AI filter box

    These should be re-enabled on Monday morning or after Alex has finished his investigations.

Attached are plots of dust counts > .3 microns and > .5 microns in particles per cubic foot. The data was taken from approximately 6 AM to 6 PM. Also attached are plots of the modes to help show when they were running. I moved dust monitor 4 in the LVEA into the clean room over the test stand holding the cartridge assembly. I put it in a narrow space between one of the test stand legs and the clean room curtain.

I noticed on the H1SUSMC2_GDS_TP screen a "red" status under the first (of four) elements for the "RT NET STAT". All other SUS models are "green".

Reported to Jim Batch to look into...

After HAM1 needed a large  movement  to rectify some measurement issues the springs had come into hard contact with the side of the housing collar. I re-centered the springs and loaded the weight back to the previous values. This brought the support tubes to +- 25 thousandths of an inch. During the process of bringing the support tubes to nominal one of the springs started to bind, and after removing the nut one could see several demolished threads. Tyler G. expertly repaired them and I replaced the spring. 

The support tubes are now sitting within +- 5 thousandths although with the amount of activity around the dial indicators some of those values may be adulterated. After I locked the HEPI feet down I re-zeroed the dial indicators and hopefully placed them in better positions for when IAS can shoot the table.

To enable the more automated Matlab TFs to be taken on the LHO triple test-stand, I have added the following channels (and corresponding data rates) to the channel list:-

X1:SUS-HXTS_M1_TEST_L_OUT_DQ 2048
X1:SUS-HXTS_M1_TEST_T_OUT_DQ 2048
X1:SUS-HXTS_M1_TEST_V_OUT_DQ 2048
X1:SUS-HXTS_M1_TEST_R_OUT_DQ 2048
X1:SUS-HXTS_M1_TEST_P_OUT_DQ 2048
X1:SUS-HXTS_M1_TEST_Y_OUT_DQ 2048

These have been added to the *.ini files of both models i.e. "X1SUSHXTS05.ini" and "X1SUSHXTS27.ini".

After logging into the test-stand front-end "ssh tripleteststand", the frame-builder has been restarted by switching models using the script "useModel27", waiting a 60 seconds, then finally returning back to the required model "useModel05". 

Running the "LHOTripleTFs.m" Matlab script from the "susSVN/trunk/Common/MatlabTools/SchroederPhaseTools" directory, now successfully extracts the above channels from the frame data.

Worked on cable installation in electronics room racks ISC-C3 and ISC-C4. While powering up electronics in ISC-C4, connected ±18V backwards to either:

D1002735 Harmonics Generator Power Supply - SN S1000791 or
D1000124 RF Distribution Amplifier - SN S1000604

Units seem to be powering up, but will look closer once testing begins.

Filiberto Clara

I was notified of the following activities:

• Michael, Rick, Cheryl, and Rodica to PSL work, they still working in there.
• Richard and Ken to Y-end station.
• Mark Lubinski to visit all outer stations.
• Patrick to LVEA to move dust monitor.
• Apollo to LVEA, ACB parts and BSC floor removal, dome and door installation fom BSC08.
• Fred Raab to LVEA.

Received several (4 of them) dust alarms from the dust monitor moved by Patrick to the cleanroom for the test stand (West bay, north most test stand).

Two PSL alarms.

First thing this morning, the Apollo crew raised the chamber cleanroom back up to its usual working height and removed the lower portion of the leg jacks. Next, they craned the dome back over to the BSC. The walking plates were removed and the dome was re-installed with all bolts in place. Once the dome was in place, Thomas supervised the removal of ACB parts from the chamber to a large C-3 covered work table. The ACB parts were covered with C-3 and transported to the HAM cleanroom for dis-assembly. After much discussion, the BSC flooring was removed from the chamber, preserved clean, and transported to the HAM cleanroom for Wrap-Bag-Tag. The door was returned to the chamber. (Not bad for a day's work!)

While Mark L. was in-chamber to retrieve the BSC flooring, he surveyed the chamber bottom for detritus after the flooring was removed: this is our SOP when doing ICC. This chamber has the potential to be quite instructive WRT contamination issues because we have very sharp data points for length of exposure and type of activity. The highlights (or low-lights, as the case may be): two relatively large pieces of UHV foil, at least three human hairs, and lots of Al fines. See a few pix below.

Checked DB25 connectors for IO QPD , ISC QPDs and IO/ISC picos and they're all fine. Measured from outside, no shortcircuit  between pin13 and vacuum chamber.

Firstly, a set of MC2 (HSTS) M1-M1 TFs were taken, with damping loops OFF, for all DOFs (see 2012-10-03_H1SUSMC2_M1_ALL_TFs.pdf), which show good agreement with the model. These TFs were then repeated, but with damping loops ON.  

Both damped and undamped TFs obtained have been compared to all other LLO HSTSs at the equivalent test Phase 3a, (see allhstss_2012_10_04_Phase3a_MC2_ALL_ZOOMED_TFs.pdf). 

Power spectra have been taken with damping loops both ON and OFF for each stage (see 2012-10-04_1000_H1SUSMC2_M*_ALL_Spectra.pdf).

Power spectra data, with both damping ON and OFF can then be compared to LLO MC2 during Phase 3a testing (see allhstss_2012-10-01_Phase3a_MC2_ALL_Spectra_Don.pdf and allhstss_2012-10-01_Phase3a_MC2_ALL_Spectra_Doff.pdf). n.b. the purge-air was off for the LLO measurements. 

In addition, power spectra for specific degrees of freedom (L, P and Y) can be more conveniently compared across multiple stages (M1, M2 and M3) of the same suspension in the final plot found below (see allhstss_2012-10-01_Phase3a_MC2_H1SUSMC2_M1M2M3_Spectra_ALL_Don.pdf).

Finally, all data, plots and scripts have been committed to the susSVN as of this entry.

This should now provide sufficient data to be able to complete Phase 3a testing of this suspension

As requested by Jeff K, I have updated the compensation filters for H1 HXTS suspensions. A similar update previously been carried out at LLO by Joe B (see LLO aLOG entry 4417). 

The process was first tested with the SR2 (HSTS) suspension, using the prepare script, located in "/ligo/svncommon/SusSVN/sus/trunk/Common/MatlabTools/", with the following Matlab command:-

prepare('H1','SUS','SR2','HSTS')

Updated foton filter files were then visible from the GDS_TP screen for h1sussr2. The "Coeff Load" button was selected to import the new filters. A cursory check over the medm screens indicated that new filters were in place (see note below).

This was then repeated for PR3 (HLTS), PR2, PRM, MC3, MC2, and MC1.

The process initially failed (hung-up) for MC2, as this was found to already have in existence h1susmc2.txt.bak and h1susmc2.bak.txt, which prevented the process from completing. Once these were deleted, prepare scripts could complete.  

N.B. it was observed that the following models did not contain any DAMP filters: SR2, PR3, PRM. I shall confirm if this is intended with Joe B or Jeff K.