Weekly Report for Week Ending June 29, 2006

There will be no LIGO Executive Committee scheduled Monday, July 3, 2006 due to the holiday.  The next Executive Committee Meeting will be Monday, July 10, 2006.

Special Announcements:

The LIGO PAC Meeting is scheduled July 11-12, 2006

Weekly Report Highlights

LSC Issues (Saulson)

The 23-24 June meeting of the LIGO/Virgo Joint Working Group was very successful.  Several teams had succeeded in carrying out a first analysis of the three hours of real data (from LIGO, GEO and Virgo) that had been exchanged just a week before the meeting. The purpose of this project is to ensure that vetoes and data quality information can be properly exchanged, in addition to time-series data itself. The meeting also served as a valuable forum for discussing practical issues for how the two sides will work together after the Virgo/LIGO MOU is signed. We converged on an outline of a transition plan toward fully joint data analysis.

Following up on discussions between Benoit Mours and the LIGO Directorate during the GWIC meeting, we have now finished a first draft of the Virgo/LIGO MOU's Attachment, describing details of the means by which Virgo and the LSC will work together.

LIGO Laboratory Administration (Lindquist)

STATUS OF LSC MOUs (Lloyd)

  • No report.

SITE TELECONFERENCE (Lindquist)

  • There was no site teleconference scheduled Thursday, June 29, 2006.
  • There are currently no open action items.

PROPERTY MANAGEMENT (Luna)

>From: Rod Luna <rluna@ligo.caltech.edu>

  • Provided assistance to the Detector Group (H. Armandula) with packing and shipping of one coated sample from CSIRO to Stanford (R. Route) Account Number LIGO.OPT - 5.4 - NSFLIGO.FY02ON.
  • Created a new Equipment database for Terry Gunter @ LHO.
  • Coordinated the relocation for Xavier Siemens from Milwaukee to Pasadena, CA.
  • Cleared equipment disposal for LHO (B. Riviera) 18 monitors, 3 Computer, 1 Printer and 2 Keyboards.
  • Assisted Gina Salone in processing Invoices for LIGO.

DOCUMENT CONTROL CENTER (Turner, Mak)

>From: Linda Turner - turner@ligo.caltech.edu>

  • No report.

>From: Cleveland Mak <mak_c@ligo.caltech.edu>

  • Processed another batch of revised/new HAM-Optical Bench drawings from Promec.
  • Emptied out a file cabinet located in B. Kell’s office, which resulted in almost another dozen boxes of year's past miscellaneous documentation.
  • Worked on confirming publication list for 2005.
  • Scanning - Progress continues on scanning of contract closeout files.
  • Activity:

Week Ending

06/29/2006

In

Out

Packages

17

9

Faxes

25

14

FINANCIAL SYSTEMS (Cronin, Brambila, Kaufman)

>From: "Cronin, Holly" <Holly.Cronin@caltech.edu>

>From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>

  • Working on the change order to Raytheon for the extension.
  • Working on the new purchase order for the a/c system for Hanford.  The vendor is stalling on the delivery schedule and providing us with their company information, as they are extremely busy.
  • Cancelled the purchase order issued to Cangelosi Ward so that the change order could be added to the subcontract.  Completed the change order to the subcontract for Cangelosi Ward.
  • Placed the order for various Amazon books.
  •  Completed the monthly purchase order close out report. There is one reimbursement purchase order still open which I am following up
  • Triad's change orders #176 & 177 have been completed and re-routed to the vendor.

>From: Florence Kaufman <fkaufman>

  • Received access and training from Criselda so that I can query the P-card system to determine traveler’s name and information on P-card purchases.
  • Determined that the first LIGO summer student was classified to an incorrect expenditure type, which resulted in unwarranted Indirect Cost.  Trying to get this corrected and also to make sure that any subsequent payments are classified to the correct expenditure type.
  • Worked on modifying monthly report to show variances for each Cost Center between projected expenditures for FY06 vs. budget for FY06.  Projected expenditures for FY06 is calculated by taking adding 104% of the expenditures from July to December in the previous year to actual expenditures through June.
  • Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport. (For passwords contact Florence)

SUBCONTRACTS MANAGEMENT (Jasnow, Salone)

>From: Gina Salone <gsalone@ligo.caltech.edu>

  • Nothing significant to report.

>From: Ed Jasnow <jasnow@ligo.caltech.edu>

  • The Livingston Science Education Center (SEC) remains ahead of schedule, with beneficial occupancy still scheduled for the end of August.  The official opening ceremony is scheduled for November 13.
  • Approximately half of the LIGO conference call users have been switched from AccuConference to Intercall, the successor company to Raindance.  This is to provide an alternate teleconference carrier in case one experiences performance problems.

PROPOSALS and REPORTS (Lindquist)

I am assembling an Annual Report for the NSF for LIGO Operations.  I had requested inputs by Friday, June 23.  So far I have received approximately 60 to 75 percent of the material needed.

CHANGE CONTROL/CONTINGENCY MANAGEMENT (Lindquist)

  • There are no open change requests.

HUMAN RESOURCES (Akutagawa)

>From: Cindy Akutagawa <cindy@ligo.caltech.edu>

  • No special activities to report.

Quality/Safety (Tyler)

>From: Bill Tyler tyler@ligo.caltech.edu

Nothing significant to report this week.

LIGO Hanford Observatory (LHO) and Interferometer Operations (Raab)

Summary of S5 Activities at LIGO Hanford Observatory (compiled by M. Landry)

On Sunday 25th, the 22 ton Liebert air conditioning unit compressors tripped off, shutting down LDAS offline analysis, the fb1 diskcache and DMT FsOM.  The power panel for the unit could not be reset.

S5 highlights from the LHO elog are bulleted below:

  • weekend running was good with fast relocking due to the remeasured LA params and less variable range than the previous weekend with the restored darm unity gain frequency
  • weekly updates: range and duty cycle, IFO maintenance budget,
  • in order to have another independent calibration data point, thermally excited violin resonances were considered, but so far the calibration extracted puts us a factor of 2 more sensitive (and thus more work is required here)
  • exremely loud NS binary inspirals (100kpc down to 5kpc) were injected in all IFOs, causing no lock losses, ASPD5 transitions or even actuation chain saturations... they sound great on speakers however
  • new calibrations were performed on all IFOs, using 7h of IFO maintenance time

4K IFO

  • 4k locklosses: the CM and MC are vindicated, leaving the likely culprit as the ISS saturation in the analog chain, or possibly the FSS or laser.  If these glitches do not unlock the IFO, they may impact the range if they are large, or not if small.  Mode hopping in the NPRO has also been suggested as a problem, but this is not established.  The FSS and PMC were unlocked and SLOWDC ramped; the FSS was relocked in a region free of mode hop and the ISS/PSL glitches continued.  Next, temp channels were used to monitor new PDA55s mounted in the PSL.  Later we ran with the ISS off, a test which suggested that the combination of the NPRO/MOPA and the ISS is required in order to produce our pernicious glitches
  • a PEPI shutoff was added to watchdog code
  • angle-to-length decoupling was performed on the H1 modecleaner, reducing the coupling by a factor of 2

2K IFO

  • glitches seen in DARMERR and beamsplitter pitch channels were not due to reflective memory network (RFM) or framebuilder software problems

DAQ

  • DMT monitor trend data was not available for FsOM, and thus the framebroadcaster was moved over from fb1 to fb0 (later, restored)

Outreach (D. Ingram)

LHO welcomes Pasco High School science teacher John Kerr for a summer internship.  John will develop and test materials for LIGO's contribution to the NSF-funded "I2U2" program.

LIGO Livingston Observatory (LLO) and Interferometer Operations (Zucker)

L1 in S5 (Yakushin)

The duty cycle during the week was 82 percent (maybe a record for L1 for a week?) with the inspiral range around 13-14 Mpc.  Today we plan for several hours of downtime to do loud injections requested by Gaby and calibration measurements requested by Brian.

#MZ Late news: Gaby's experiment proved we will not only survive in low noise mode, but can also properly reconstruct a very close BNS event in our own galaxy. Also we verified the calibration is good.

#MZ We also restored our spare MOPA laser to operation in the optics lab, it had an internal mirror mount dislodged in shipping.

Safety and Security (Riesen)

This reporting period we ran a full fire drill in the OSB.  All alarm devices were fully operational, evacuation to the assembly area went smoothly, a head count was taken and the building was repopulated after the O.K. was given.  The laser amplifier experiment in the optics lab has met all laser safety criteria and has been given permission to proceed.  There were no site nor laser safety concerns found this reporting period.

Humongous Educational Kinetic Art Installation (Romie)

Attended the Final Design Review last Thursday. Dropped off the RFP and drawings to all four contractors yesterday and today. There will be a site walk-through on Friday at 11am. Responses are due July 14 at 5pm.

#MZ Today we held the job walk, two qualified vendors attended. The briefing was informative and we are hopeful there will be several competitive bids (due 7/14).

LLO Outreach (Thacker)

  • participated in Teacher PD workshop - Math/Science Partnership project, PRISM=HQT; that 3 week workshop concludes 6/29
  • developing LIGO science education materials for our webpage including: pre- and post- visit teacher packages, concept maps for our exhibits, and more
  • prepared outreach report for LIGO annual report
  • continued development of Slinky demonstration
  • Mentored two RET teachers
  • worked on Opening Event details: budget, invitation list, program details

#MZ adds:  SEC construction is racing ahead, very little heavy traffic any more. Today some of the lobby casework was delivered; internal drywall is almost done, electrical is close behind. Air conditioning will kick on next week, after which we expect to break through the passage into the auditorium foyer.

LIGO Computing and Network Security (Roddy)

  • A switch died in this morning.  Unfortunately it was the switch that all of the servers are tied to.  Presently using temporary spare.  Will replace with a foundry unit this afternoon or evening.
  • Moving the old web files to a USB drive for Bonnie.  Required setting up a temporary server and mounting the old disks there.  USB hard drives on Solaris 10 aren't as difficult as they used to be.
  • Started to set up the V490 until I realized I need two 240 v circuits installed for it.
  • Worked some with the Windows 2003 test servers.  Work in progress...
  • Applied for an ASN from ARIN.  Need to supply them more paperwork.
  • Set up a home for shared Linux applications.  We haven't had to support NIS+ Linux boxes until now, so things are still ramping up.  Trying to make common applications available identically to what is on Solaris. * Working with Shourov on installing a copy of qscan in the above mentioned applications share.
  • Looking into disk space for the users.  We are around 94% full on our home directories. * Still trying to get the results from the Nessus scans pulled together in a useful format to distribute to everyone.

General Computing and LDAS Admin (Giardina)

  • outreach user assistance with network connectivity
  • relocation of servers NATALIA and BASIN
  • setup 2 new Dell PCs, one for HPLF and one for Optics lab
  • security server Windows updates.
  • VPN connectivity assistance for user
  • installed plotter as new printer on two laptops
  • some PCs were unable to obtain an IP address this morning.  dhcpd and network restart on dhcp server did not fix.  power-cycle of switch temporarily resolved DHCP problems.  Other issues arose.  Switch would not work after second power-cycle.   We are using a temporary loaner switch until a replacement is configured hopefully sometime today. 

(With Shannon)

  • eject and import tapes for GC tape library.
  • other usual user requests and support
  • ejected tapes for shipment to CIT, imported and labeled new
  • T3-18 FC controller froze.   /frames became unusable.  fb0 temporarily became primary framebuilder.  Disk2Disk archive scripts were set to use fb0 as primary source.   T3-18 required two power-cycles to free up controller.  FrCheck on /frames frame files leading up to switch-over yielded 8 bad frame files (5 full and 3 second-trend).  These 8 files were removed from their respective directories under /frames so that RDS would not fail.  After cleanup, fb1 was restarted and frames were written to /frames.  Disk2Disk was then set to normal configuration.
  • wrote a simple perl script to run FrCheck on multiple frame files sequentially, something that probably already exists, but I didn't know where.

Data Analysis (Yakushin)

Storage/Condor/LDAS Administration

  • One of LDAS 15 ton Liebert units (the brand new one bought a few months ago) developed a problem.  On Tuesday it was turned off for repairs for several hours. I had to shut down half of the cluster.
  • On Tuesday night T3-18 (which is part of /frames) went down. The control room had to switch from fb1 to fb0. The next morning Dwayne powercycled it and with the help from Ben and Greg, tracked down all the corrupted frames to be removed. By Wednesday night the control room switched back to fb1.
  • 2+1 new 3511s arrived. We need 2 10 m fiber channel cables to connect them to the switch which are shipped from CIT and expected to arrive on Friday.
  • node114 went down. There is no light on the back of the unit at all which seems to indicate the power supply problem. Ordering 6 spare power supplies.
  • Copying V2 h(t) data from CIT to LLO.
  • Noticed the problem with segment replication between LHO and other sites. Duncan restarted the replication and it worked. Not sure why it did not work from the first attempt.
  • Installed Ben's new diskcache-based version of LSCdataFindServer at LLO.
  • ldas-grid developed a problem with automounting /data/nodeN. I tried to kill or restart /etc/init.d/autofs with no success. Eventually, when nobody was running jobs, I rebooted ldas-grid and that fixed the problem.

Data Analysis

  • Working on running simulations with the coherent waveburst on GA2_S5_A and h(t). Unfortunately part of h(t) frames that were on the nodes at CIT were accidentally deleted. Xavi is regenerating them now.
  • Continue working with my SURF student on improving waveburst online pipeline.

Optomechanics Group (Dennis Coyne)

All CIT Optomechanics staff are assigned to work on AdL tasks.  Their work is reported under the AdL section of the weekly report.

Controls and Data Systems (Bork)

No report (vacation)

40-Meter Interferometer (Weinstein)

No report (vacation).

Thermal Noise Interferometer (Black)

This week we opened the vacuum chamber and removed the input mirror from the south arm cavity, in preparation for installing the latest design of ring damper.  Dennis Coyne has simulated about thirty modes with the new-style ring dampers in place, and he finds substantial Q reduction (1e6 to 1e4 or 1e5) in most of them. Scaling arguments indicate that the thermal noise floor should remain unaffected, but we have not done detailed calculations.

The beam profile of the photothermal laser has degraded considerably, and this seems to correspond with a loss in power.  Lee Cardenas returned one of the lasers he was using, and Mike is in the process of measuring its beam profile to see if it will be a suitable replacement.

Matt gave a very good talk on Braginsky's first parametric oscillations paper (1d model), and has started reading Rokhsari, et al. on parametric instabilities in microtoroids.  He will give a talk on that paper in three weeks.  Mike is scheduled to talk next week on thermal self locking.

Greg came up with an idea for Joule heating a ring damper to expand it, in preparation for installing it around an optic, and he has started working on a quantitative design.

LASTI (Ottaway)

No report.

Data Analysis (Lazzarini)

>From: John Zweizig <jzweizig@ligo.caltech.edu>

This week I wrote a perl language version of the Spi (Monitor status page generation) script. The original version of Spi was a tcsh script and made use of pipelines which are heavy users of resources and were suspected of contributing to frame data dropouts. The perl script gives the same results as the original and is much faster (it runs in ~3 seconds rather than 11) but there is no data yet on whether this will help the frame loss problems.

I have also added ASI_CORR_OVERFLOW segments to the automatic data quality segment generation. This flags times where the ASnI_CORR signal exceeds 20000. At LLO the signals were artificially limited at 20000 and glitches were seen in the AS_Q channel where this limiting occurred. At LHO there were no software limits, but the signals tend to be non-linear when they exceed 20000 so they were flagged also.

LDAS Software Systems (Blackburn)

LDAS

Progress continues with porting LDAS to Tcl/Tk 8.4.13. Several memory leaks were fixed at the Tcl layer. This involved modifying the code logic to have nested callbacks to be started using the "after" command of Tcl. Also, the large number of leaked SWIG named objects in the frameAPI has been resolved.

For the past few weeks an infrequent lockup of the diskcacheAPI has been observed. After carefully analysis of the processes stack, a mutex deadlock situation was found to be the culprit. The code has been reworked at the C++ layer to prevent this condition. This change is being tested on the development system.

System testing of LDAS was done using version 1.8.214. The lsync test is still not finding the same set of frames in lsync library. Weekend run results: 163870 jobs 162783 passed (99.33667%) 1087 failed (0.66333%) 0 rejected (0.00000%) APIs rebooted by manager are ligolw and eventmon. eventmonAPI ran into memory corruption and generated a lot of red balls about missing database column. After it was rebooted by hand, the memory corruption stopped. Timeouts and memory leaks are mostly found in metadata and ligolw. Frame has errors about missing channel. Job rate is 2666 jobs per hour with the tclglobus 1.2.0 fix of manager core dump every month.

The system ldas-test has been configured as a system level test machine.  Tcl/Tk 8.3.5 has been configured for this system. It shall be used to verify backwards compatibility of the code. With one night of running, the eventmonAPI has been observed to have a large memory leak that causes it to be rebooted frequently. This condition is under investigation.

TCLGlobus

Delayed the next release of TclGlobus as a result of several issues discovered prior to the scheduled release. These issues are under investigation and more extensive testing is being carried out.

Two GSSAPI unit tests failed under Tcl 8.4 and GCC 4.0.2 (non LDCG).   The unit tests are used to exercise anonymous credentials (GSS_C_ANON_FLAG) and the flag to indicate this option has no effect under non LDCG environment. This issue has not yet been resolved.

Submitted TclGlobus and Caltech sections of the ITR2003 annual report.  This annual reporte is expected to be submitted to fastlane today.

GRID COMPUTING

Work continues with determining the proper tools needed by the LIGO WorkFlow Planner. Condor currently is having difficulty with being installed as the current default glibc for Fedora Core 4 has known issues. Attempts are being made to work around this problem. The major criteria for any solution is that it must be done as non-root.

Met via Telecom with Pegasus/ISI engineers regarding previous request for functionality to update vds-get-sites utility to produce accurate tables tc.data and sites.xml.  Reviewed functionality in the VORS web site and determined that VORS is a candidate for persistent infrastructure to support vds-get-sites information for the OSG.

Contacted John Rosheck, developer of VORS at Indiana University. John has modified the script used to collect information in VORS to support collection of $OSG_GRID needed for determining the location of VDS components on OSG CEs. Script does not yet collect information for all ITB sites. The source of the script error is under study.

Worked with Duncan Brown, successfully produced a DAX script for inspiral hipe on pcdev2. Moved DAX script to OSG submit host and produced a collection of submit scripts for inspiral hipe targeting the local cluster LIGO-CIT-ITB. Submitted the inspiral dags and determined there is a problem with lalapps_tmpltbank executable using the CAL_FAC frame types. Referred problem to Duncan, as naming conventions of some calibration frames have changed and calibration procedures are internal to the inspiral state machine.

Worked with Michael Samidi and Kent Blackburn, on a problem related to the interaction between fedora core 4, some versions of glibc and recent versions of DAGMan that produce fatal errors in DAGMan. Michael is attempting workaround with specfic versions of glibc known to be stable.

Testing of the LIGO work flow planner uncovered numerous bugs, each of which has been added to the PR database.

Testing of the LIGO work flow planner with the inspiral pipe analysis on Purdue-ITaP, UFlorida-IHEPA, and UFlorida-PG have succeeded for the first time this week.

Attended an OSGStorage TG and OSG ITB telecoms.

Reviewed Project Execution Plan for the Open Science Grid proposal and provided feedback to the OSG Executive and Council teams.

Data Analysis Activities (Anderson)

(Igor Yakushin)

  • Working on running simulations with the coherent waveburst on GA2_S5_A and h(t). Unfortunately part of h(t) frames that were on the nodes at CIT were accidentally deleted. Xavi is regenerating them now.
  • Continue working with my SURF student on improving waveburst online pipeline.

(Vuk Mandic)

  • I wrote a new version of the stochastic S4 all-sky paper. The hope is to send it to the LSC for last comments by the end of the week.
  • I ran the S4 H1L1 all-sky analysis up to 1 kHz. This was suggested by the review committee, as a cross-check of the S4 ALLEGRO-LLO analysis. The result is that the ALLEGRO-LLO result is about 50 times more sensitive than H1L1.
  • I calculated the H1L1 coherence over the first part of S5 (up to April 3), in preparation of running the stochastic code on the data.
  • I worked with Xavi on updating the study of accessibility of the cosmic strings models to LIGO. We are working on adding other limits to the calculation (such as BBN, LISA etc).

(Gregory Mendell)

I reported to the pulsar/CW group on work I have been doing with Karl Wette from ANU on parameter estimation from SFTs and its relationship to the PowerFlux method. A technical document will be posted by the August LSC meeting. I am also continuing work on the S4 PowerFlux, StackSlide, and Hough S4 paper, and have run code to produce StackSlide Receiver Operating Characteristic curves for a couple of narrow frequency bands.

(Duncan Brown)

  • Implemented new clustering and better reporting of data quality flags for online BNS glitch trigger pages, as requested by the glitch group. Also upgraded the online BNS search to a new version of onasys at LHO and LLO.
  • Looking at implementation of code to generate waveforms from the spectral coefficients produced by the Caltech/Cornell numerical relativity code SpEC.
  • Benchmarked SpEC on the LIGO cluster. It is only 25% slower at the highest resolutions on our cluster than on CACR's SHC cluster. The slowdown is expected due to our slower GigE node interconnects vs their Infiniband interconnects. The LIGO cluster is still fast enough to get useful work done, however.
  • Worked with SURF student Alex Zaliznyak on his template placement project.
  • Continuing to advise Diego on physical template family development.
  • Continued with my review committee duties for S3/S4 inspiral analysis.
  • Helped David with some issued he had running the inspiral DAX on the OSG, which were due to changes in the calibration frame conventions.
  • Helped Phil clean up a wet cluster room.

(Shourov Chatterji)

(Peter Shawhan)

  • Participated by phone in the LSC-Virgo joint data analysis face-to-face meeting in Orsay.
  • With Erik Katsavounidis, wrote Burst Group sections for the LIGO annual report and the LSC data analysis white paper.
  • The usual Burst Group organizational work.

(Rejean Dupuis)

  • writing/designing a new code for targeted pulsar searches (more efficient and new functionality)
  • investigating an efficient binary pulsar search technique for potential use

Laboratory Computing (Stuart)

LDAS System Administration (Anderson)

Caltech

(Dan Kozak)

  • Did a bit of hunting to find out the status of 4 tapes at LLO (they're unused as best I can tell).
  • Unstuck staging for Xavi.
  • Helped with wedged T3 controller at LLO.
  • Tracked down and eliminated some bad frames (corrupted when written to /archive by D2D).
  • Checked potentially bad frames from GEO as per Steffen's instructions: they were all fine at CIT.
  • Determined that the difference in how long a samfsrestore of /archive took under test conditions as compared to real life was almost certainly caused by the difference between a shared SAM-QFS filesystem and a non-shared SAM-QFS filesystem.

(Phil Ehrens)

  • Most of my week was consumed by the management of the remediation of a water leak in Synchrotron 215A, the home of our computer cluster.
  • Completed scripts for automating addition of GRID users to the CIT cluster.
  • Added new modular components to log_mon.tcl for disk usage alarms and node reboot detection.

(Stuart Anderson)

  • Upgraded Condor at MIT to version 6.7.20.
  • Following up with Caltech facility support on improved monitoring of the computer room water leak detector.
  • Reported a problem with sam-stagerd leaking memory and crashing at Caltech.  We now have a patched version from Sun that will be tested today (Jun 29).
  • Updated the Condor configuration at CIT to support new dagman configuration options and patched condor binaries for additional LIGO specific bug fixes.

MIT

(Keith Bayer)

  • Rebuilding node40 which had multiple problems (disk + cable).

Livingston

(Igor Yakushin)

  • One of LDAS 15 ton Liebert units (the brand new one bought a few months ago) developed a problem.  On Tuesday it was turned off for repairs for several hours. I had to shut down half of the cluster.
  • On Tuesday night T3-18 (which is part of /frames) went down. The control room had to switch from fb1 to fb0. The next morning Dwayne powercycled it and with the help from Ben and Greg, tracked down all the corrupted frames to be removed. By Wednesday night the control room switched back to fb1.
  • 2+1 new 3511s arrived. We need 2 10 m fiber channel cables to connect them to the switch which are shipped from CIT and expected to arrive on Friday.
  • node114 went down. There is no light on the back of the unit at all which seems to indicate the power supply problem. Ordering 6 spare power supplies.
  • Copying V2 h(t) data from CIT to LLO.
  • Noticed the problem with segment replication between LHO and other sites.  Duncan restarted the replication and it worked. Not sure why it did not work from the first attempt.
  • Installed Ben's new diskcache-based version of LSCdataFindServer at LLO.
  • ldas-grid developed a problem with automounting /data/nodeN. I tried to kill or restart /etc/init.d/autofs with no success. Eventually, when nobody was running jobs, I rebooted ldas-grid and that fixed the problem.

(Dwayne Giardina)

  • Ejected tapes for shipment to CIT, imported and labeled new.
  • T3-18 FC controller froze.   /frames became unusable.  fb0 temporarily became primary framebuilder.  Disk2Disk archive scripts were set to use fb0 as primary source.   T3-18 required two power-cycles to free up controller.  FrCheck on /frames frame files leading up to switch-over yielded 8 bad frame files (5 full and 3 second-trend).  These 8 files were removed from their respective directories under /frames so that RDS would not fail.  After cleanup, fb1 was restarted and frames were written to /frames.  Disk2Disk was then set to normal configuration.
  • Wrote a simple perl script to run FrCheck on multiple frame files sequentially, something that probably already exists, but I didn't know where.

Hanford

(Greg Mendell)

  • On Sunday morning a fan motor failed in the condenser for the 22 ton Liebert air conditioner in the LDAS room at LHO. One of the main fuses to the condenser blew, and this caused a loss of cooling to the room.  Email alerts were sent, and a control room alarm was raised.  A thermal shunt tripped off the room power once the temperature rose above ~ 85 F, before anyone was able to respond to the alarm. One of the main power breakers would not reset, and had to be replaced. However, the thermal shunt did what was wanted: no equipment was damaged. Raw, RDS, and trend data generation and archiving are caught up without loss of data. We are operating with 3 out of 4 fans in the condenser unit. The failed fan motor will be replaced next week.
  • (Ben Johnson)
  • Thermal trip tripped in LDAS room due to the 22 Ton AC failing (condenser fan blew fuses for whole unit).
  • Replaced a 3x125Amp breaker. It would not "untrip" after the thermal trip turned it off.
  • Repaired MySQL collection table. It was corrupted due to the power outage.
  • Assisted in the installation of the diskCacheDataFindServer at LLO.  CIT, LHO, and LLO are now all running the new server. Also applied bug fix to handle observatory parameters with or without commas.
  • Investigated data glitches in H2:LSC-DARM_ERR and H2:ASC-BS_P. The conclusion was they were not due to known daqd or RFM corruption mechanisms.
  • Moved LDAS <-> GC connections from an old 10/100 hub to a new 10/100/1000 switch. The GC cable is not liked by the new switch (no link), so I still have to patch the cable through the old hub; i.e., the GC cable is running a half-duplex.
  • Patched Disk2Disk to better handle errors from /bin/dd. It should log an error if /bin/dd dies or /bin/dd encounters an I/O error.

General Computing (Wallace)

MIT:

(Keith)

  • Setup a couple of computer systems for student use.
  • Regular user and server support.

Livingston:

(Shannon)

  • A switch died in this morning.  Unfortunately it was the switch that all of the servers are tied to.  Presently using temporary spare.  Will replace with a foundry unit this afternoon or evening.
  • Moving the old web files to a USB drive for Bonnie.  Required setting up a temporary server and mounting the old disks there.  USB hard drives on Solaris 10 aren't as difficult as they used to be.
  • Started to set up the V490 until I realized I need two 240 v circuits installed for it.
  • Worked some with the Windows 2003 test servers.  Work in progress...
  • Applied for an ASN from ARIN.  Need to supply them more paperwork.
  • Set up a home for shared Linux applications.  We haven't had to support NIS+ Linux boxes until now, so things are still ramping up.  Trying to make common applications available identically to what is on Solaris.  -Working with Shourov on installing a copy of qscan in the above mentioned applications share.
  • Looking into disk space for the users.  We are around 94% full on our home directories.
  • Still trying to get the results from the Nessus scans pulled together in a useful format to distribute to everyone.

(Dwayne)

  • outreach user assistance with network connectivity
  • relocation of servers NATALIA and BASIN
  • setup 2 new Dell PCs, one for HPLF and one for Optics lab
  • security server Windows updates.
  • VPN connectivity assistance for user
  • installed plotter as new printer on two laptops
  • some PCs were unable to obtain an IP address this morning.  dhcpd and network restart on dhcp server did not fix.  power-cycle of switch temporarily resolved DHCP problems.  Other issues arose.  Switch would not work after second power-cycle.   We are using a temporary loaner switch until a replacement is configured hopefully sometime today. 

(With Shannon)

  • eject and import tapes for GC tape library.
  • other usual user requests and support

Hanford

(Christine)

  • Still setting up new computers.
  • Received a phone call from the FBI in Dallas, TX.  They arrested a hacker who had hacked into one of our web cameras a year ago.  The hacker had our web camera IP address in his list of conquests.  The FBI asked me to send them a report on the cost of the damages which will be used to determine sentencing of the hacker.  Working with Fred to determine a cost for hours spent upgrading and patching the camera's OS.
  • I modified the syslog.conf on a new Solaris 10 to post messages to the loghost.  Something went wrong and the computer started spewing out messages to it's own messages file every second.  This caused erratic behavior and finally crashed the computer.  I don't know what went wrong, the same mods to syslog.conf on another Solaris 10 computer worked fine.
  • Created a new account for a summer helper.
  • Misc. user support for SURFs and visitors.

CIT

(Veronica)

  • Taped and processed for streaming the SURF talks on physics of LIGO.  Updates of the Elba website, PAC, roster database, other LIGO-related pages.
  • Updates of LSC-related mailing lists. Working on the website for LSC educational resources.
  • Taped and compressed for streaming the last CaJAGWR talk.
  • Started work on the Project Science-related website.

(Mike)

  • Spent most of this week working in the B/A server room installing equipment and configuring new Sun servers. These are on going projects.
  • Backed up Barry Barish's laptop.
  • Cleared out a workstation from Rana's office. We are now using this workstation as a backup workstation for Surf or Visitor's.
  • Freed up expired IP numbers on the 114 subnet DHCP server.
  • Worked on Spam Filters.
  • Continued loading patches on new Solaris installations.
  • Updated SSHD on a WH server.
  • Other misc. work.

(Christian)

  • Created a backup of Cindy Akutagawa, Julie Hiroto and Irene Baldon workstations.
  • Dorothy Lloyd - Configured loaner laptop for Dot to use on travel.
  • Chris Echols- installed new LCD monitor for Chris.
  • 40Meter- replaced old CRT monitor with a new LCD monitor.
  • Hans Bantilan- Added 1GB of additional memory to visitors' workstation.
  • Created images for the new laptops that are going the loaner pull.
  • Worked on the Spam Filters with Mike and Larry.
  • Other misc.: Continued onsite software/phone support.

(Bruce Sears)

  • iLog Maintenance: (0.5 days), General iLog maintenance (user adds, keyword adds, systems work, etc.)

(Larry)

  • This has been a busy week in the purchasing area. Mostly tools and misc. items for different computers.  Still trying to get Monarch to resolve the server issue.  Spent time looking for a monitor to go into the display case on the 1st floor, for a small project Stan is working on. Still a bit of investigation work to do before making any purchases.  Worked with Gina on a number of different procurement issues and meetings. She has a lot going on right now.
  • Worked a number of DCC issues. There have been a number of documents that are not converting to the PDF format. I am still working on some of them.  Rebuilt the dcc disk again to clear up some problems caused by a cable coming out of the disk drive.
  • Setup a number of SURF accounts and helped many of the SURF students with different problems. Mostly getting them to resources they can use for their projects.
  • Spent a short period of time working at the 40M. I need to spend a couple of days there to get some of the machines back up to par.
  • Helped Mike out on getting Solaris 10 installed, update SSH  and installation of some hardware in the server room.
  • Working on multiple documentation assignments, including the security assessment and FY report information that has been asked for.
  • Installed application s/w on a couple of units.
  • Continual work on going through the spam filters. After checking with a couple of other people here on Campus I don't feel so bad about the amount of spam getting through the system. It seems others are having the same problem. Having the server filter and the users having a filter is definitely paying off on keeping spam down to a minimum.

Mail Statistics for June 22-28, 06

Mail Statistics

6/29/2006

Rejected Messages

42,071

Virus Messages

1,229

Accepted Messages

23,760

Total Messages

65,831

 

Advanced LIGO and Supporting R&D (Shoemaker)

From: Dennis Coyne coyne@ligo.caltech.edu

Advanced LIGO Systems

from Dennis Coyne

See also:

AdL Systems web page

AdL Systems email archives

Records Of Decisions or Agreements (RODA) status web page

  • Started migrating the Systems web pages to the Wiki environment
  • Working on committee review of the documents for the 3rd Preliminary Design Review (PDR#3) for the Quadruple End Test Mass/Input Test Mass (ETM/ITM) Suspension. Documents can be found here. The review will be held July 10 – details here.
  • Held a monthly systems meeting. Discussed optimal diameter for the large optics in the input section which define the aperture to limit diffraction loss and permit decentering for alignment into the FP arms; Assigned actions to resolve (involves potentially increased MMT3 or PRC3 mirror diameters, changes in the RM triple suspension and lower HAM SEI optics table. Also discussed Osamu & Hiro’s modeling of ETM & ITM suspension global control to stabilize photon pressure torque induced instabilities.
  • Issued a technical memo (T050261-00) on finite element modeling of the quadruple suspension (using Ansys and I-Deas). Dave Ottaway and his summer student, Cassie Hunt, are carrying my HAM-SAS dynamic model (created in I-Deas and documented in T060020-00) forward in anticipation of delivery of a HAM-SAS prototype late this summer. One of the modeling tasks that they hope to accomplish is coupling a triple suspension model to the HAM-SAS model to enable study of dynamic interaction and cross-coupling of triples on the same SAS platform.
  • In support of planned TNI tests, calculated the efficacy of adding 4mm square cross-section copper rings to the TNI test mass to reduce the Q of resonances; A reduction in the Q by a factor of about 100 or mode for most modes. Also calculated that the effect on the thermal noise was negligible.

Modeling and Simulation

From: Hiroaki Yamamoto <hiro@ligo.caltech.edu>

LSC and ASC design using AdvLIGO Arm simulation (Osamu, Hiro)

We are continuously investigating a FP cavity with AdLIGO parameters on e2e. Local damping in all DOFs (6, 3, 3 for top, second, third) are implemented, and now only 6 DOFs on top mass are used. Strange oscillation due to radiation pressure kicking disappeared by slower laser power increase.  WFS signals will be connected next.  Osamu gave a summary report at the Aligo_systems monthly telecon, http://www.ligo.caltech.edu/~omiyaka/AdLIGO_ASC/20060628_AdLIGO_meeting.ppt http://www.ligo.caltech.edu/~omiyaka/AdLIGO_ASC/20060628_AdLIGO_meeting.pdf

Mechanical system model for AdvLOGP (Sany Yoshida and SLU team)

Using our AdvLIGO HAM and triple suspension e2e models built recently, performed numerical tests under realistic condition. (With typical ground X and Y motions, this HAM model creates table-top translational motions as required by the AdvLIGO specification and table-top yaw motion based on the same model that we used for LIGO 1 seismic isolation.  The triple suspension model uses the state space matrix created by Mark Barton?s mathematica model.)  When the suspension was freely hanging and the typical ground translational motions were given as input to the HAM model, the resultant optic yaw motion was on the order of 10^-6 (rad/rtHz) at 0.1 Hz, 10^-10 (rad/rtHz) at 1 Hz and 10^-11 (rad/rtHz) at 10 Hz.  Similarly, the resultant optic pendular motion was ~5x10^-7 (m/rtHz) at 0.1 Hz, 5x10^-11 (m/rtHz) at 1 Hz and 10^-12 (m/rtHz) at 10 Hz.  The orientation of the triple suspension was the same as LIGO 1 MC1 (45 deg to the X arm).  More analysis is being made.

Made some progress in the e2e modeling of LIGO 1 Mode Cleaner.  By ramping up the table top motion injected to the suspension point of MC1 and MC3, confirmed that our MC box could be stably locked up to table X/Yaw motion of 10^-5 (m/rad).

AdvVirgo study using e2e (Monica)

The configuration of one Advanced Virgo arm has been tested and locked with radiation pressure on one mirror using both the demodulated reflected signal and the demodulated transmitted signal. Some optimization is needed for the feedback system in order to have the radiation pressure on both mirrors.  The optical response of the sytem is under investigation.

Static IFO Simulation (Hiro)

FP simulation with locking has been finished.

The effect of the orange peel signature, a pattern of a polished mirror surface due to the technique adapted by CSIRO, was calculated using this FP simulation code. A preliminary result shows that the effective loss is ~3ppm or less. More careful understanding of the simulation result is needed to come to a final conclusion.

Bill K and his SURF student are going to investigate PI effect using this code. SIS is coded to accept and produce appropriate data files.

Alfi (e2e front end)

Two new features are being developed. (1) modules with variable inputs and outputs (e.g., FUNC_X or VecSum), and (2) grouping of objects.

Vacuum Compatibility

Vacuum Preparation & Residual Gas Assay (RGA)

See also the Vacuum Bake Lab

Bob Taylor

  • I have received the Hot Stripper and have prepared some QML OSEM wire for testing. I will bake and scan these samples then give them to Lee Cardenas for testing.
  • I am writing a procedure for the new wire stripper for inclusion in the OSEM document. I will send it for approval.
  • I have talked with Riccardo about the FT-IR Spectrometer and I am going over the literature he gave me. I will have some Questions when I am done and I would Like to get together with you and Riccardo maybe next week:  Beam splitter, Sampling window, Frequency range, data output, type of sampling and so on.
  • I have began working on the OSEMs for the controls quad prototype and should be ready to bake them by friday of next week. Brett is sending more OSEMs to me this week.
  • On Monday I started building 5 OSEMs for Nergis and I will get those to UPS today.

High-Irradiance, Contamination-Exposure Cavities

Lee Cardenas, Liyuan Zhang

Cavity # 1: OTF Lab. at W. Bridge: No Change.

Cavity #2: No change

Cavity #3: OTF Lab at Lauritsen Room 38:

The new sample, stepper motor is in the cavity.  The cavity is locked. We were taking measurements every day for absorption and ring down. so far, it looks as it is clean unit!! since the measurements have not changed much.

Scatter/Absorption Test Measurement:

The 60 watt laser is again operational after we fixed the aperture cooling unit leak. Quantronix will send a new unit with the changes that we suggested to them. We have re-installed the aperture cooling unit and re-aligned the optical path and after a quick mode match, we recovered the laser signal and improved power.  The absorption test and measurements for the ITM07 & ITM08 mirrors have been completed.

Seismic Isolation

From: "Joseph A. Giaime" jgiaime@ligo.phys.lsu.edu

Agenda for the weekly SEI telecom

Friday, June 23, 2:00 pm Eastern, 1:00 pm Central, 11:00 am Pacific time

BSC SEI status

  • The gusset that holds top of the stage 1-2 flexure had to be modified to correctly position stage 2 w.r.t. stage 1.  They made the distance between the stage two upper optics table and the top of the flexure top agree with the model.  It is now believed that the alignment pins that hold stage 2 in place are correctly dimensioned.
  • New shim ordered for stage 1-2 blade mount, about 4 mm thicker, to accommodate the blades being too soft.  In addition, the total stage 2 mass is about 300 kg light so that the vertical spacing can be correct.  N.B.: this will require rebuilding the system with another shim and launch angle before the prototype work is complete, in order that the correct payload capacity is demonstrated.
  • Spring tensioning pusher assembly for stage 0-1 appears to be on the edge of unworkability, especially when the new 0-1 blade angle and shim is used.  (We have not yet calculated this angle.)
  • Dynamic test of loaded GS-13 can was carried out incorrectly loaded; the dummy mass should have been on the mount side.  It is probably true that the GS-13 instrument will not significantly affect the resonant frequency of the can reacting against the stage.  This test needs to be repeated.
  • There was a nicked conflat knife-edge the GS-13 pod which needs to be repaired.  Ken will try to increase the number of end caps in the pending order.
  • Ken Watts has modified the in-pod wiring harness for the GS-13.

Stanford: seismometer progress

  • Matt has closed all 12 loops for the re-configured tech demo.
  • There is excess stage 2 noise at 1 Hz, consistent with tilt noise in stage 1 as measured in stage 1's sensors.  Matt is debugging the controller design.
  • Tarm working assembling the new vertical seismometer.

Other news?

  • Brian & Joe met with Corwin to agree on the content of Corwin's final report on his single-stage HAM structure conceptual design study.  The report is expected in about 3 weeks.

From: Ben Abbott <abbott_b@ligo.caltech.edu>

40m

The new QPD Whitening board layout is finished, and it is being made at PCB Express currently.  It should be back sometime next week.

I worked with Sam and Alex to make the XYCOMM switches work for the end station qpd interfaces.

I did some more machining on the DCPD mount to make a place for a beam dump to be mounted.

HAM-SAS

I sent off the new front panel modifications for the Virgo Coil Drivers.  They should be back any day.

I have set up an experiment to test the in-vac Cooner wire's ampacity.  I hope to actually conduct it soon.

I got some cable lengths for the in-vac cables, I will order some from Accuglass if they seem reasonable.

Suspension

From: Ken Mailand kmailand@ligo.caltech.edu

I'm  using the latest information from Ian and working on a lower quad installation arm shop drawings, and a simpler 'conveyor' version to be used at the LASTI site.  The existing 5 axis fixture has been disassembled,  the stainless hardware change out is ongoing, other parts are being modified for stability.

Core Optics

From: Bill Kells kells@ligo.caltech.edu

Rapid progress on the project I am engaging my SURF student on.  Mainly since he is very good ! Many steps in the simulation/calculation pipeline are now done. This simulation will be one of the first to use the preliminary (single FP cavity) version of Hiro's new e2e code. So I have been heavily involved with interfacing Hans (my student ) to Hiro.  We have it up an running.

Prompted by D. Ottaway's recent note on the influence of SRC on PI, I have gone back to more carefully "validate" the code I wrote last October to do this same task. I will have this completely vetted so that I can fully compare to Dave's work when he returns stateside.

Preparation for the AdL TCS reveiw, next week.

From: GariLynn Billingsley Billingsley_G@ligo.caltech.edu

 

 

From: Helena Armandula ahelena@ligo.caltech.edu

Advanced LIGO - LASTI

Have a draft concept/drawing for the LASTI mirror carrier and prototype for Adv. LIGO mirrors. I am in the process of getting some details added before I send it for review.  At this time the cost of this unit is ~$6,000.00.

Advanced LIGO Coatings

SiO2 doped Ta2O5

CSIRO shipped another three-sample coating run (thick, thin and small) this time using a Tantala/Silica mixture which contains ~ 35% silica in Tantala. The refractive index of this mixture is 1.9, approximately the same as the titania/silica mixture sample previously received. The total 30 x 1/4wave layer thickness is thus also the same (~4.8 microns). The samples were also annealed for the same 24 hours at 500C. All substrates looked OK after annealing.

Stanford is measuring the optical absorption on the run's witness sample.

SiO2 doped TiO2

The optical absorption measured at Stanford on the second SiO2 doped TiO2/SiO2 coating showed a reduction in absorption (compared to the first run) after being annealed at a higher temperature.  The absorption coefficient averaged over the whole surface was 0.54 ± 0.04 ppm .

Input Optics

From: David Reitze reitze@phys.ufl.edu

 

 

Pre-Stabilized Laser

From: Peter King pking@ligo.caltech.edu

 

AdLIGO PSL

At long last I have managed to get some output out of the FPGA.  However the DAC output had a lot of glitches, many of which were larger than the input signal that I was digitizing.  Experimenting with both the sample and clock rates did not appear to improve things.  The documentation that comes with the board is not all that enlightening.  The technical help webpage on the Xilinx website was not overly useful either with many broken links leading to a page stating that the page was reached in error and that they were working on it.

Auxiliary Optics

From: Michael Smith <smith@ligo.caltech.edu>

LAYOUT Non-folded IFO

A complete list of optical component data, organized by HAM and BSC tables is being prepared for updating the T010076-01 Optical Layout for Advanced LIGO document. The component data will include their locations in gobal coordinates, their surface normal vectors and the main optical beam vectors.

For additional information about this report, contact Stan Whitcomb or Phil Lindquist