Weekly Report for Week Ending August 11, 2005

The LIGO Executive Committee meeting scheduled for August 15, 2005 is CANCELLED due to the meeting of the LIGO Scientific Collaboration.

Special Announcements:

Weekly Report Highlights

LSC Issues (Saulson)

No report.

LIGO Laboratory Administration (Lindquist)

STATUS OF LSC MOUs (Lloyd)

LSC MOUs and Research Plans and Progress Reports

  • No report this week.

Non-LSC MOUs (Lloyd)

  • No report this week.

SITE TELECONFERENCE (Lindquist)

A site teleconference was conducted on Thursday, August 11, 2005.  The following issues were among those discussed:

  • Property Issues – The annual property report for the NSF is just about complete.  Waiting on GSA vehicle usage data for August.
  • LSC Meeting Preparations – The report from Fred “Yogi” Raab indicates that things a pretty much ready to go, but “Ya don’t know what ya don’t know!”
  • There will be no site teleconference scheduled the next two weeks.  The next site teleconference will be September 1, 2005.
  • The list of assigned actions updated through May 26, 2005 will be found Here.

PROPERTY MANAGEMENT (Chargois)

>From: Ed Chargois <chargois_e@ligo.caltech.edu>

  • Provided assistance to General Computing (L. Wallace) with packing and shipping of warranty Network Switch to Foundry Network Inc., Alviso,Ca.  Shipping was prepaid.
  • Provided assistance to General Computing (L. Wallace) with packing and shipping of LSC support equipment to LHO.  Account Number P204275.
  • Meeting on 8/12 at 10:30am with LDAS (S. Anderson) and Intersect Systems Inc. representative to survey and make final arrangements to transport the LDAS computers and racks to LLO and HLO.
  • NSF annual report is on-going awaiting final GSA mileage report for the month of August, we are ahead of schedule.

DOCUMENT CONTROL CENTER (Turner, Mak)

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

  • No report (vacation).

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

Week Ending

August 11, 2005

In

Out

Packages

18

5

Faxes

16

14
  • Continued to prepare old account charge files for scanning and cd burning.  Another temp was hired this week to assist with this project.  Good progress is being made.
  • Began to process a large set of HAM-Optical Bench drawings from Promec.

COST SCHEDULE CONTROL SYSTEMS (Cunningham, Brambila, Kaufman)

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

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

  • No report (vacation)

>From: Florence Kaufman <fkaufman>

  • Prepared an analysis of the ASI contract.
  • Received approval for 3 new fabrication accounts:
    1. LIGO.HVAC/2.12
    2. LIGO.HVAC/3.12
    3. LIGO.PSLTD/5.3
  • Prepared and sent to Carol Wilkinson a report as of the end of July for Advanced LIGO activities.  This report excludes charges for labor and related charges to permit wider dissemination.
  • Completed report for Visitor program through July 2005.
  • 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>

  • On Wednesday, August 3, bids were opened for the construction of the LLO Science Education Center.  All bids exceed our cost estimate, so various alternatives are being studied to see if we can resolve the disparity without having to go out and re-bid.
  • The Statement of Work for the re-location of LDAS at LLO is being reviewed.  The current schedule calls for release of the RFP on August 17, a job walk on August 25, and receipt of bids on September 2.

SUPPORT (Baldon, Hiroto, Lloyd)

>Irene Baldon

Processed the paper work for four (4) new/revised trips.  At this time there are four (4) trips completed but awaiting the necessary paper work to enter the P-Card system.  Assisted several LIGO people with their travel arrangements using their P-Cards and made several reservations for outside visitors coming to LIGO/Caltech or one of the LIGO sites. Also canceled out two (2) LSC trips arranged by the travelers but who needed my assistance in canceling everything associated with their trip.

Completed twenty (20) Expense Reports and there are six (6)reports yet to be done.  I continue to contact travelers who have outstanding Expense Reports (more than one (1) month old) to ask for their cooperation in sending me their receipts so that these can be closed in a timely manner.  Presently there are five (5) reports more than thirty (30) days old.  I have six (6) reports awaiting signature at this time.  Reconciled thirty-two (32) P-Card charges for the week requiring telephoning hotels and car rental agencies to verify which traveler used my card and for what amount.

Worked on last-minute details associated with the travel arrangements for twenty-seven (27) SURF students and one (1) faculty adviser traveling to LHO from August 11th through August 12th requiring four (4) Super Shuttle vans for pickup from Caltech traveling to LAX, two (2) fifteen (15) passenger vans at Pasco Airport, fourteen (14) room reservations, river cruise arrangements and arrangements for ten (10) students to layover in Seattle for the weekend on the way back.

>Julie Hiroto

No report.

>Dorothy Lloyd

  • No report.
  • Jim continued with data entry in the LIGO database and helping out in the DCC.

PROPOSALS and REPORTS (Lindquist)

The NSF has approved the Annual Report for LIGO Operations.

The Proposal for an Extension to FY 2007 and FY 2008 and Supplemental Funding for LIGO Operations has been submitted to the NSF.  The NSF has requested additional material.  Contributors to the Annual Report have been contacted to provide this additional material.

DCC Steering Committee (Lindquist)

The Executive Committee discussed the change request during the meeting Monday, August 8, 2005.  The committee recommended approval.  Approval is pending additional information in response to questions forwarded through B. Barish by a document management system expert at SLAC.

CHANGE CONTROL/CONTINGENCY MANAGEMENT (Lindquist)

  • CR-050008, LIGO Document Management System, was discussed during the Executive Committee meeting Monday, August 8, 2005.  The committee recommended approval.  Minutes have been prepared.  Approval is pending responses to additional question forwarded through B. Barish by a document management system expert at SLAC.

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 Commissioning Activities at LIGO Hanford Observatory (Landry)

Improvements were made on both interferometers this week such that H1 and H2 posted their best performance since S4:

  • Spot size measurements yielded radii of curvature of the thermally-lensed 4k ITMs, including the new ITMX.  Equivalent scatter in ITMY is less than 3ppm.  This, along with the low scatter observed at the swapped ITMX at CIT suggests that dust may have been responsible for excess absoption seen previously on H1.  Drag-wiping seems to then have helped ITMY.
  • Next, the lower-power 4k thermal compensation (TCS) X-arm laser was reinstalled (as the 30W laser couldn't do both required annular and central heating), realigned and commissioned.  The idea was to run with 6W into the modecleaner, however,  3.2W into the MC and about a 1/4W per arm of TCS made for a promising start (6.5Mpc binary inspiral range).  Later, once WFS gains were modified, the 4k posted 7.2Mpc ("best" link above)
  • On the 2k, a bump at 270Hz and some of the 80Hz hump was mitigated when it was discovered that AS light was landing on a burnspot on ASPD2.  The spectrum unexpectedly showed sensitivity to minimal tailoring of the FSS gain as well, with a 2dB increase in gain impacting some high frequency peaks.

Other recent commissioning highlights include:

  • A first look at photon calibrator mis-centering suggests it is not a problem (later revisited and quantified).  PC-induced displacements happily fall like 1/f^2, an important sanity check.  A comparison to standard calibration procedures gives good agreement for the 4k but a 30% systematic at high frequency to be chased down on the 2k.
  • The 2k REFL port continues to behave unexpectedly: REFL1 and REFL2 were compared, and despite the expectation of lower noise in the REFL2 path, this is not observed.  More work in progress on REFL: an entry on excess noise in REFL1, the FSS is not the culprit, and unexpected intermodulations appear in REFL_I
  • Output from MX froze on weekend;  both the ASC and LSC had to be rebooted
  • On the horizon: new (acoustic-mitigation) meatlockers are expected to be delivered soon, and, an experiment to float the 2k AS output table ISCT10 has begun

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

No report.

Initial LIGO Detector Science & Engineering (Coyne)

CDS

See also the CDS weekly meeting minutes in the commissioning archives

CDS Software

Rolf Bork

  • We've been working on some software to bring front end processors back on line in the proper sequence after a site power outage or similar system shutdown. The new front end software to provide for a proper power up sequence has been installed at LLO. Testing is in progress.
  • Data analysis has seen a problem of a 1/4Hz signal showing up in DARM_CNTRL, but not the other LSC signals. I can't find anything obvious in the front end software that would explain this. I plan to do further investigation and testing on my next visit to LLO this month.
  • Updated and installed new ETM controller software at LHO to acquire the end/mid station STS signals.

CDS Hardware

Rich Abbott

  1. Progress was made on the Reflected Beam Steering (RBS) servo.  Almost all of the work has been completed that can be without the hardware that is due the end of this month.  A check with the supplier, Piezo Systems Jena, verified that they are still on track for delivery.  To date, the Epics filter modules, screens, databases and channel allocation has been done.  Jay has produced preliminary diagrams detailing the interconnections and a block diagram.
  2. The test procedure for the ISS board has been modified and the ISS board tested and received at LLO.  More testing is needed prior to the unit being installed and further refinements will be made to the procedure.  The design has a complete schematic with all modifications reflected in a DCN. Installed and tested ISS at LLO.  Noise is now ~1.5e-8 several hundred Hz.  A report is being prepared to detail the installation
  3. A DCN covering changes to the LOS Bias Modules has been approved and filed in the DCC along with a modified Rev. A4 schematic.  The DCN implements changes to the PA-85 power supply pins to protect against transients on the power leads as well as reverse transients on the output.
  4. Hooked up ASI servo at LLO with Valera.  The servo on ASPD1 and 4 is a bit questionable when you adjust the phase adjustment screws on the front panel.  There are times when the response is jumpy and noisy as viewed in the ASPD dark spectra.  More investigation is needed.
  5. Re-wrote the test procedure for bias compensation on RFPDs.  Have put together the required test setup in the LLO optics lab, and am executing the procedure to setup the diodes.  Once tested, the diodes will be installed as replacements for the damaged AS port diodes.  The shutter threshold has been lowered on the AS port diodes in a effort to reduce the likelihood that they get damaged in the future.  Consideration is being given to increasing the spot size as well.
  6. Reviewed a new design for 1pps locking presented by Daniel.  Reviewers felt the design was well done and that he should proceed pending required information from the Master Timing Synchronizer design produced as a part of the Timing System Re-design effort.

Jay Heefner

LOS Bias Module

  • Upgraded schematics to reflect changes necessary to protect PA85 amplifiers. The DCN and schematics have been submitted to the DCC.

Refl Beam Stabilization

  • Completed the installation of the software and 4116 required for the LLO system. Ken Franzen is now testing the software.
  • Completed the design of the interface board and power supply board required for the system. The layout will be sent to the board house on 8/11 and the boards should be back by 8/18.
  • Chassis components have been received and are awaiting circuit boards.

Ben Abbott

LSC RFPD:

1)      The boxes for the RFPDs have arrived from Front Panel Express.

2)      The parts for the RFPD arrived from Digikey, they were boxed up with the parts that I already had from other vendors, and sent to Screaming Circuits where the boards are being stuffed on a 5-day turn around.

3)      Cables for the inside of the RFPD enclosures will be fabricated by Dave when he's done testing the ISS servo.

4)      I spent some time looking at the tank transimpedance of the RFPDs in order to verify the low values that Daniel had seen.  I found the transimpedance to be roughly twice what he found, and this discrepancy was traced back to the fact that Daniel assumed that the gain of the RF stage was 10, where it is actually 5 after the 50 Ohm output impedance voltage divider.  This increases the calculated tank transimpedance by a factor of two.  The overall transimpedance of the RFPD is lower than we want.  I'm currently trying to figure out if the output resistor on the RF path should be reduced, or the gain of the 4107 OpAmp should be raised, or some combination of both should be employed to remedy this issue.  Rich has asked Vern for tests that he can perform to see the effect of lowering the output resistor value.  Once these tests are finished, the correct course of action should be plain.

DMT

no report

PSL

PeterKing

A small number of minor modifications to the intensity servo test procedure were suggested prior to the entire procedure being re-worked.  In testing board S/N #116, it was observed that for a few AD829s the recommended value of compensation capacitor greatly reduced any oscillation rather than remove it altogether.

Optical Metrology

no report

40 Meter Interferometer (Weinstein)

IFO commissioning:

  • After the adjustments to the ITMX OSEMs during the last vent, ITMX has gone from being the noisiest and most coupled suspension of our 10 suspended optics, to being the least noisy. Now, it looks like ETMX is noisiest, followed in order by SRM, MC2, BS, PRM, ... Now that we know how, we will want to adjust the OSEMS on each of the suspensions, the next time we vent.
  • The X arm is now much quieter, and requires a bit of tickling to get it to lock (as does the Y arm). Osamu modified a script from Rana to push and pull both ETMX and ETMY slowly until both arms are locked. This is now part of the lock acquisition procedure.
  • Osamu has been exercising the lock of the DRMI + DARM + offset CARM; we're back to being able to do this reliably and repeatedly. At night, the arms are so quiet that they need a kick to get them to lock. The switch to a digital common mode servo (feeding CARM to MCL) is now routine and robust. The fast analog path from CARM to MCF is still in progress. The biggest noise source now is a ~16 Hz oscillation seen in the short degrees of freedom, maybe associated with bounce modes of the small optics in the DR Michelson.
  • Error signals in all DRMI degrees of freedom have offsets, which drift significantly over the course of an hour. Osamu finds that he needs to adjust digital offsets every 30 minutes. We need a real solution.
  • Seiji is working with Osamu to explore the behavior of the length control signals as CARM offset is reduced, just before it loses lock, by dithering the CARM offset. Hope they learn a lot!
  • Osamu, Monica and Steve measured the reflectivity of the arms cavities using the DC light at the AP port. XARM has a reflectivity of 0.84 and YARM of 0.86. Then they estimated the power recycling gain, taking the average of the cavities reflectivities as the Michelson reflectivity: PR_Gain = (t_PR / 1 - r_PR * r_FPMI)^2. Knowing that T_PR is 7.7 percent, the calculated power recycling gain is 6.1; undercoupled. The design is 16.5, overcoupled. The arms have an anomalous source of loss.
  • David and JuLi are exploring the possibility that light power buildup during the reduction of the CARM offset is sufficient to create an optical spring that hinders the path to full resonance. He is thinking about experimental measurements to evaluate this.
  • Ryan and David continue to measure seismic noise and coherence at various places along the arms, and are preparing a paper on the results. There is a clear 20% difference in seismic velocity, the velocity along the X arm is slower than along the Y. The coherence between 40m baselines falls at 5 Hz along the x arm and ~6 Hz along the Y arm (right in the vicinity of our stack resonances), from which one can expect a small difference in the differential motion between the two arms.
  • Ryan and Davis are also measuring the coherence between the OSEM signals of the ITM and ETM. For they Y arm, the coherence is strong at low frequencies, and falls at the same frequency that the seismic coherence falls. But the X arm is quite anomalous, showing no or small coherence. This has improved since the ITMX adjustments during the vent several weeks ago, but it still is much worse than the Y arm. It looks like ETMX needs adjustment. We will wait before we vent for that.
  • Ryan and Rana continue to develop the noise budget code.
  • Ryan and Rana put EPICS monitoring channels of POS,PIT,SIDE for OSEMS from all 7 optics into trend frames.
  • Rana & Monica adjusted the oplev gain and filtering to reduce the noise on the SRM.
  • Osamu added bounce and roll mode bandstop filters on all the oplev servos.
  • Osamu and Steve moved the SP optical spectrum analyzer from the SP table to the (much closer) AP table, realigned, and now gets a much stronger signal. We still have a QPD & a CCD on the SP beamline on the SP table; everything else is on the AP table.

IFO modeling and DC detection development:

  • Mike Smith has completed a detailed parts list, costs, and purchase reqs for all the optics and other devices in the DC readout in-vac beamline. This is now under review. Mike will be on leave for a few weeks, starting next week.
  • Mike has drawn up all the mechanical parts needed for the output mode matching telescope, etc, for the DC readout beamline. He is sending them out to machine shops for quotes.
  • Monica has merged e2e control plant code from Matt Evans into her 40m optical plant model. She's working on studying the CARM offset signals.

Electronics, controls:

  • Dan and JuLi have been working on diagonalization of all the suspensions, adjusting the coil gains and adding filters to reduce the pos2pit and pos2yaw couplings. So far, all 4 test masses are done, they'll do the other optics next (PRM, SRM, BS, MC, and maybe MC1 and MC3, except these don't have oplevs to sense the coupling). They'll also work on reducing ang2pos couplings.
  • Dan fixed the PSL frequency reference cavity ion pump HV power supply (actually, the 24V PS inside). We still have no digital monitoring of the ion pump HV; we just have to watch the front panel.
  • Ben changed the LSCaux2 database for the addition of the Common Mode Servo, but some work remains to debug the system. As it stands, there is a problem switching the 4116.
  • Ben found some discrepancies with the End rack wiring and the latest drawing, He has redrawn the wiring diagrams for the 1X4 and 1Y9 racks. They are updated on his website.
  • At Rana's request, Ben looked at the QPD Whitening board because its whitening filters weren't switching. He found that the XYCOMM 220 was not switching. It looks like aproblem with the front end code. He will talk to Alex about this.
  • The c1lsc processor failed 5 times, requiring reboot. After a reboot, the cpu is running at ~ 45 usec, but it creeps up to > 60 usec after a while. This requires work by Rob to find the problem and/or speed up the code.
  • The c1susvme2 cpu is still flaky; it falls out of synch. Alex is trying to diagnose the problem, it may be due to the heat load in the crate.

Lab Infrastructure:

  • Steve reports that the pumpdown 2 weeks ago proceeded smoothly, but only 2 of 3 roughing pumps were used, because the third has a leaky valve. Steve will get a replacement.
  • Steve tested the ion pumps. Although they pump effectively, they "sweat", leading to a 10x worse peak at AMU 41 (when the turbo pump is off). Steve will leave the ion pumps on, but pumping on closed gate valves. SV ion pump is running hot, it looks like the gate valve is partially open, despite the readback "closed". Steve is watching the situation.

Thermal Noise Interferometer (Black)

This week we finished aligning the arm cavities, closed the chamber, and pumped out. We were never able to get very good visibility with the new coatings, with NAC at 65 percent and SAC at 55 percent, but we decided to go ahead and try to measure the noise anyway. We closed on Tuesday and locked on Wednesday, with first data with the new coatings taken Wednesday night. Today (Thursday) we have spent analyzing the data.  Results will be posted when we have finished the analysis.

LASTI (Ottaway)

No report.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)

E2E Weekly Physics Meeting

Nafis Jamal, SURF student of Sany Yoshida presented his work on E2E simulation of alignment sensing and control of LIGO I Mode Cleaner.  Viewgraphs available at http://www.ligo.caltech.edu/~e2e/ME2ET/Minutes05/050811/

Mechanical simulation

(Mark Barton) Following up on a request by Hiro, I added support for wedged optics to my pendulum models. This required changes to each model (alas not the toolkit) to allow for off-diagonal components in the moment of inertia tensor. The result is that a vertical wedge creates a small (second-order) and probably insignificant coupling between yaw and roll, and a horizontal wedge produces a similar coupling between pitch and roll. See G050337-00 for details.

FFT study

(Biplab)--Found a bug in MIT FFT code, which never got discovered probably because nobody ever used the code to study the effects of transverse shifts of beam and/or optics. Hiro helped to eliminate this bug and after that consistent results are obtained for various cases of transverse shifts.

 -- Used my Matlab FFT code for Advanced LIGO cavity to study diffraction losses for fundamental TEM00 and 1st order Leguerre Gaussian Mode.

Analyzed results with David Blair. Numbers seem to be reasonable. More cross-checking is going on.

Simulation of 40m Advanced Interferometer

(Monica) Solved bugs related with the use of new alfi bundles, bundlers and channels extraction. Inserted new e2e primitives (like "VecScaMerge" and "mscal_out") in e2e package for 40m in order to obtain a better organized code for data-collection and channel-extraction from bundles.  Simulation tests for the lock acquisition procedure are in progress.

Alfi

(Bruce)

  • Port centering work completed (PR 325).
  • Fixed junction delete bug (PR 489).
  • Adding comments to Macros, and implementing in the parser (PR 435).

Data Analysis Activities (Lazzarini)

Chatterji:

Attempting to reproduce 1 Hz comb feature observed in high resolution spectra using <ahref="http://ldas-pcdev1.ligo.caltech.edu/~shourov/1HzLines/">a Matlab based algorithm</a>.

Continued work on implementing a coherent network analysis for unmodeled bursts with Sutton, Stein, Lazzarini, Tinto and Searle.  Succesfully applied simple analysis pipeline to both white noise and simulated detector noise data using the Caltech computing cluster.  Migrated to Matlab version R13 for improved performance.

Continued work with Hormoz on Q pipeline based consistency tests for collocated detector data.  Working on incorporating into existing Q pipeline code base.

Developing Q pipeline based tool to scan multiple auxiliary and environmental channels for significant events in proximity to a time of interest.

Mandic:

I continued the stochastic analysis of the S4 H1L1 data. In particular, I investigated different quality cuts to eliminate bursty data, based on the behavior of the variance over 60-sec segments. After applying such cuts, the coherence seems very clean, and its histogram follows the expected exponential behavior. I started coherence calculations over the whole run (with higher resolution) with the goal of searching for 0.25 Hz and 1 Hz harmonics.

I also started working on recovering the remaining S4 hardware injections.

Mendell:

The occurance of 1 Hz harmonics in S4 H1 and L1 SFT data, probably due to problems with DARM_CTRL, has led me on to look for all the StackSlide outliers in the S4 SFT data. (A StackSlide outlier is an event with power greater than expected by gaussian noise, given the number of templates used in a search.) A report on these in the S4 H2, H1, and L1 data between 55 and 615 Hz is posted on the S4 Investigations page under the General Catagory here:

http://www.lsc-group.phys.uwm.edu/cgi-bin/enote.pl?nb=puls4general&action=view&page=6

Searle:

Preparing talks for LSC meeting.  The process has clarified quite a few aspects of the analysis.

Shawhan:

Since returning from vacation, I have worked with my SURF students and done some reviewing of pulsar analyses (besides catching up with my email).

Sutton:

I've been working on the coherent network analysis project.  I had several meetings with Chatterji, Lazzarini, Searle, Stein, and Tinto as we try to understand different facets of the problem.  I've made some small improvements to the simulations code.  I'm currently studying the statistics of the xpipeline output, attempting to figure out how to set thresholds for detection, false alarm rate.  In other news, I submitted the LIGO-TAMA bursts paper to PRD.  I've also been working with Principe on estimating the efficiency of a network given single-detector efficiencies, for her Network Simulator project.

LIGO Data Analysis System

Software Systems (Blackburn)

LDAS

The fix that was introduced last week to remove the disappearing threads didn't correct all instances of the dissappearing threads. Additional debugging has been added to try and better understand the source of the problem, but since the new debugging statements were added, not instances have occurred.  As a  result we have decided to proceed with the LDAS 1.8.0 development without a 1.7.1 release to specifically address this issue.

Decided that the TclGlobus package is mature enough to begin integration of X509 certificate based authentication into LDAS. We plan to target support for this in the 1.8.0 release of LDAS as an optional means of authenticating. We will delay making this the default mechanism until the community has had time to adjust.

Began porting LDAS software over to the Solaris 10 and Fedora Core 4 Operating systems. These are targets for the S5 Science Run. Fedora Core 4 introduces an issue with compiler upgrades so we are exploring the stability and other issues associated with the GCC 4.0.1 compiler shipped with FC4. Currently LDAS is at GCC 3.3.3. Have identified one unit test failure with GCC 4.0.1 in the data- ConditionAPI at this time. No integration or system tests results yet.

Began setting up the infrastructure for adding certificate support to the LDAS controlMonitorAPI. This included installing Globus 4.0.1 on a client side work-station for integration with TclGlobus and the CMonClient. All LDAS developers now have a DOE certificate which will be needed for testing the new authentication interfaces.

Testing: ran system test for ldas 1.7.1 and weekend testing. Adjusted loop delay time for added/archive S1 S2 and S3 mount points to dev for weekend test.

Upgraded ldasbox2 and suntest1 to DB2 8.2.2. Brought up suntest1 and check some metadata jobs with the new db2. Copy db2 include and libs from suntest1 to suntest2.

TCLGLOBUS

Continued fixing segmentation faults resulting from calling the SWIG wrapped globus_gass_transfer_send_bytes. Fixed two mistakes that might contribute to this problem:

a. I didn't have exception handler code in the SWIG-wrapped implementation.

b. I didn't use last_data flag to indicate client sending the last n bytes of the data and this caused globus_gass_transfer_request_destroy didn't close the socket connection.  But, the problem still persists. Based on Kent's suggestion, there might be an issue with data type coversion from Tcl layer to C/SWIG layer. I'm still investigating this issue.

Upgraded tclglobus box with Globus Toolkit 4.0.1 and all TclGlobus testsuite passed with the latest version of Globus Toolkit.

Began working with the LDAS group to understand the usage model LDAS will have for TclGlobus.

GRIPHYN/IVDGL/OSG

Upgraded the testbed cluster to Linux 2.6.12-1.1372_FC3smp via a Yum repository mirrored from the production cluster.

Determined that GridCat job measurement uses a single point sampling of the job queue rather than use historic statistics.

Determined that GridEx requires the FQDN for the Central Manager node of each OSG RP or GridEx 's Condor-G engine can not submit to the jobmanager- condor and reports an error that accumulates in GridEx statistics.

Determined that a missing $WNTMP field in GridCat is not a configuration error.

Make contact with a developer in the USATLAS1 VO (Xin Zhao) and have plans to run ATLAS condor jobs on the LIGO test bed cluster.

Changed the default running time of the test bed cluster to 48 hours to sustain long job runs.

Working  from the documents generated by Duncan Brown for an OSG demonstration of Pegasus workflow, built and installed the LSC Algorithm library (LAL), the LALApps, and the Grid LSC User Environment (GLUE).

Attempted to download a binary tarball from the VDS site, found the release had a missing file and reported the damaged release to the griphynligo mailing list.

Hardware Systems (Anderson)

Caltech

(Stuart Anderson)

  • Working on Condor problems when applications use too much memory.
  • Working on schedule to ship cluster nodes from Caltech to LLO.
  • Replaced 2 failed cluster node disks.

MIT

(Keith Bayer)

  • Rebooted node70 as it became unresponsive (except for pings).
  • Removed cross-mount script as it was still trying to hardmount cluster nodes (unsuccessfully) and make cluster node reboots take ~5minutes or so.
  • Received 2 new 3ware ATA cards #7506.
  • Rebuilding pcraid#3 with new 7506 cards.
  • Received 5U rack mount pcraid server from ASA.
  • Add another 5-6 condor users.

Hanford

(Greg Mendell)

  • Preparation for S5 continues with quotes now received for purchasing tapes to cover archival of data for the first year of the run.

(Ben Johnson)

  • 3 T3s spontaneously powered off this past Saturday (Aug 6th 2005).  Replacements for the resulting 6 corrupted files were grabbed from the backup frambuilder.
  • T9940B drive 42 was replaced this Tuesday (Aug 9). That slot has not reported any of the scsi: errors that were occurring.
  • Working with summer student on disk monitoring script.
  • Moved and notified all but 3 users from the /dso-test IDE-RAID. The final 3 are actively using the cluster. I will move them after the LSC meeting.
  • Turned on swap (32GB) on gateway; effective immediately. Also configured dataserver's /tmp filesystem to be "swap"; dataserver's change will take effect on the next reboot.
  • Found bug in way SAM-QFS handles mmap() for offline files which don't have their tapes in the library.
  • Ordered twelve replacement drives, ultimately to replace several modes' "sick" HDDs.

General Computing (Wallace)

MIT:

(Keith)

  • Working on XP pro x64 printing for CAD computer
  • Cleaning up from power outage on Tuesday afternoon (NW22)
  • Investigating backup mailserver LSI scsi card issues

Livingston:

(Shannon)

  • Setup a firewall rule to fix problems with the NDS proxy for CDS.  This will have to be given some thought before it becomes a "production" change for Livingston, and later maybe LHO.  It will have to go through the approval process for CDS system changes.  It was a stopgap measure to allow NDS data retrieval since the NDS proxy kept croaking at LLO.  Still working on a better dhcp system.  I have most of it in place now, but it is not quite ready for a switch over to it.  One difficulty is that I do not have a simulated environment set up yet to test it fully.  I will likely need to set up a NAT router, etc. in the lab to fully test it.
  • Spent last week at the USENIX security conference.  I will be passing around some notes to the sysadmin mail list concerning a couple of things that were discussed there.
  • Bell South caused a network outage yesterday by unplugging my core switch accidentally.  This was during a phone service upgrade.
  • I have been having some weird drop-outs on local network traffic.  I have not narrowed down the cause yet, but I have some logs of some of the symptoms to work with now.  I will continue to look into this.   One change that was done yesterday may have fixed the problem, but only time will tell since these are random dropouts.

Hanford:

(Christine)

  • Network usage can be seen at http://www.ligo-wa.caltech.edu/~christin/mrtg/198.129.208.1_198.129.78.122.html
  • For the LSC: Set up meeting rooms with conference phones, projectors and laptops.  Set up new wireless access points.  Set up a NAT router for the Auditorium wireless.  Compiled a list of the MAC addresses provided through the web registration.
  • Modified the firewall access list several times trying to allow Szabi to have access to the Trinet computer from Columbia University.  So far web access is still not working.
  • Wrote a memo outlining the choices for the backup network provider.
  • Purchased supplies for the LSC and several items requested by users.
  • Dealt with a problem caused by having two wireless access points too close to each other.  No one could get a DHCP number because their computer couldn't distinguish which access point it was connected to.  The access points have been separated, testing continues to determine optimum distance between access points.

CIT:

(Mike)

  • Ken Mailand: Worked on a Solid Works issue for him. This included trouble shooting a performance issue. I am still currently working on this for him.
  • Printing supplies: Took a supply inventory for all CIT printers, and placed an order with Gina Salone. I am also working with Gina on getting a refund on old toner cartridges, which belong to old printers that ended up being surplus.
  • W/B Server Room: installed four 2U rack mount servers, plus a edge switch.  In order to install these units we had to move other servers around to make room for these units.  Worked on configuring daisy chaining two KVM switches, for the NTSRV's. I have more work on this.
  • 3rd Floor W/B: Swapped out our backup edge switch, with a replacement switch that came in for the switch that we were having problems with. I also took some time and straitened out the network cables in that network closet.
  • Ben Abbott: Swapped out his old CRT monitor with a Flat Screen.
  • Spam Filters: Continued work on Filters searching for false positives.
  • Other user onsite/phone support.

(Veronica)

  • LSC:  Updates to the August meeting website.  Did VRVS testing in preparation for the meeting.  Modifications of the Papers under Review database, updates of its frontend webpages.  LSC website updates.
  • LIGO:  Wrote up a primer on the use of cascading style sheets for Stan and Julie.  Updates of the LIGO website.  Updates of the roster database and various mailing lists.  Installed latest security patches on the Windows servers and the SCR machines.  Dealt with my desktop intrusion.  Trying out Fedora core 4 installation.
  • CaJAGWR:  Videotaped and captured the latest talk for web posting.  It will be available as soon as I have it compressed.  User support.
  • Project Science:  Finished the setup of the website and backend for the next workshop.  The registration application is up and running.  Updates to the website.

(Christian)

  • Working on Riccardo's laptop.
  • Setup a laptop for Mike Smith.
  • Setting up a couple of new notebook computers.
  • Repaired and restored multiple notebook computers.

(Larry)

  • Worked through a number of purchases. Worked on P-card reconciliation.  Getting quotes and information on multiple cpu dual core systems.  We are looking at both AMD and Intel based systems.
  • Assisted Mike on working a number of PC related issues. Including the e-mail problem with one of the PC's.
  • Made a few more modifications in the server room and updated patches on a couple of the server boxes.
  • Working with David M. in trying to provide a backup location for some of his work with LDAS.
  • Assisting Christine and others on getting things ready for the LSC meeting at LHO.  So far things are looking pretty good.  We are working on testing out another video conference scheme that may be helpful for future conferences. It has worked well in the past for small groups and now we are going to see how well it will work with a larger group (if there is time).
  • Assisted a number of SURF students with various issues. Mostly, application modifications for their specific needs.
  • A number of new accounts and modifications to the mail alias tables have been made. Still have a number of items that need to be cleaned out.
  • Made a few more backup tapes from a couple of the application servers.
  • Worked a number of mail issues including the mail filter.

Mail Statistics for Aug. 04-10, 05

Mail Statistics

August 11, 2005

Rejected Messages

22,469

Virus Messages

694

False Positives

142

Accepted Messages

15,420

Total Messages

37,889

 

Advanced LIGO and Supporting R&D (Shoemaker)

Systems and Management

No report this week.

Seismic Isolation

From: Jay Heefner <jay@ligo.caltech.edu>

AdL SUS

Started system design for the LASTI quad controls using LIGO-like systems.

AdL SEI

  • Testing low noise AC couple amplifier to be used for low frequency noise measurements of the Capacitive Position Sensors.
  • Talked with Brian Lantz about the design of the GS 13 pre-amps.

Suspension

From: Janeen Romie <romie_j@ligo.caltech.edu>

Helena and I were in Glasgow last week (she's there this week as well) working with Caroline and others on a number of issues related to fibers, ribbons, ears, bonding and welding. It was a very successful trip.

Rich Mittleman from MIT has joined us to prepare the 2nd MC for cleaning, baking and shipment to LASTI. He's been working with Mark and Mohana on the quad electronics as well as giving us a hand with the quad build work. Norna has also joined us for the day. We have re-suspended a quad chain.

We thank Rich and Norna for joining us. Bob Taylor has also joined us for the build work. As soon as we get the MC parts separated, Bob Taylor will clean and bake them. We hope to get them to him tomorrow.

The MC spacer, to bring it to quad beam height, was sent to Dave Ottaway this morning.

From: Ken Mailand kmailand@ligo.caltech.edu

Working on a layout for the photon calibrators for beam clearance area around baffle structure assembly.  Making simplified S/W drawings of the Adv. LIGO optical component assemblies for import to Zmax, for Mike Smith, and converting Mechanical Desktop LHO full site layout to S/W

From: Rolf Bork <rolf@ligo.caltech.edu>

Continued work on porting software to a dual AMD64 processor computer for use in the quad suspension controls. As a test, the present LSC front end code was ported and tested. On the AMD64, it runs in a nominal 31usec (as compared with about 48 on the present VME 2.2GHz processors at the sites).About 40% of the time is used in PCI accesses, so some work still needs to be done there in optimization. But at least the hurdles of porting from 32 bit architecture to 64 bit and the new PCI calls have been overcome. We are also able to lock these real-time threads onto one of the CPUs, leaving the other free to run EPICS and other Linux tasks. Alex is continuing work on porting the EPICS part. He is having a bit more success with the latest version of EPICS ie the EPICS core is running now. Work still needs to be done to get the EPICS sequencers functioning.

From: "Mark Barton" <mbarton@ligo.caltech.edu>

AdvLIGO SUS: With Mohana I finished debugging the revised cabling for the quad prototype at Caltech. (A lot of minor changes were needed to accommodate the new more efficient allocation of 4 OSEMs to each 25-conductor cable instead 3 as in LIGO-I.)

From: Helena Armandula ahelena@ligo.caltech.edu

Acquired a good understanding of how the pulling-welding machine works.

Observed the pulling and strength testing of ribbons, also, the detrimental effect that mishandling the ribbon has on strength. We measured a ribbon up to about 28 kg load. This was thinner than an Adv LIGO ribbon, the dimensions couldn't be accurately determined since the non-contact optical profiling machine is being refurbished.

Spent time looking at the different function/space issues.  Questions still remain on how the welding process will be done at the sites (local to each BSC or at a central location). The options will be brought up for consideration during the PDR review in October.

Saw how a laser weld was achieved. Was very pleased seeing how the laser can be controlled during the process resulting on a precise and clean weld.

Made some silicate bondings on the test ears. I understand the process and I am able to repeat it very well.

The design of the ears is being reviewed to reach the correct compromise between optimum strength and desiderable features to make the welding of ribbons during installation and repair easier.

Adv. LIGO Coating Development

Had a teleconconference with Gregg, Jim and Sheila to discuss issues to be talked about during the LSC meeting since I'll not be able to attend.

Core Optics

No report this week.

Pre-Stabilized Laser

From: Peter King pking@ligo.caltech.edu

An estimate of the shipping and labor costs for the PSL was made.  In doing so I had to assume some parcel sizes and weights in order to get a good handle on the freight costs, which were all based on rates published by UPS.  The labor costs were generated based on experience with Initial LIGO.  I could not generate a reasonable estimate for the AEI/LZH labor costs as I do not know their labor model.

Auxiliary Optics

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

OPTLEV

Aabeg measured the QPD output voltage as a function of input beam angle for the 13meter EFL receiver lens. The receiver lens has an angular resolution < 100 microradians. The interaction between displacement and angle was minimized by focussing the first element of the lens system.

SLC/BRDF APPARATUS

Shasta obtained BRDF data from the oxidized, polished stainless steel baffle material that is being considered for the advanced LIGO baffles and beam dumps. The measured BRDF  is 1E-3/sr^-1at an incident angle of 35 degrees.

Other Laboratory R&D

From: Riccardo DeSalvo desalvo@ligo.caltech.edu
Juri

I attended the “Optics and Photonics” conference in San Diego where I gave an oral presentation on the status of our Mesa beam experiment. I used the measured map of the input and folding mirrors in my FFT simulation program to study the impact of the mirrors deformation due to the mechanical mounting, on the resonant beam shape. The preliminary results indicate that these deformations are a critical issue also with the new ring-mounting. Studying a mathematical model for the hysteresis of the GAS blade.

John

We finished the new mirror mounts and after a brief and inconclusive experimental test we mapped the mirror surfaces once again. The flat mirrors showed some improvement but are still not optimal.  The cavity was then re-aligned with these new mirrors. A spectrum has been taken which shows the highest peak to be the fundamental. Experimental work is in progress.  I have also been studying the theory and implementation of FFT code with help from Juri.

Barbara

Writing Amaldi Mexican Hat paper.

Marco

Presented LIGO talk, writing thesis, preparing LSC presentation.

Nicky

I have been gathering information and data so that I can write a paper on the results from my oven.

Chiara

I was in San Diego for SPIE conference.  Now I'm assembling all the mechanical parts for the test of horizontal attenuation system with four inverted pendula.

Maria Paola

I have been still working on transformers, and I have built many of them in order to have various turns ratio (1:2, 1:5, 1:10) and tested them with a signal generator and an oscilloscope. The gain obtained obviously depends on the input impedance of the oscilloscope that can be chosen between 50 Ohm and Megaohm, and the last one produces a better gain in all the cases. The main problem is that I still get a low voltage, and it's probably a problem of impedance matching: my next step will be to find the best load for the photodiode to produce a good output voltage.

Francesca

I took some measurement of the output LVDT with different gain, in order to find the response of the system and the behavior of my Lab View program. I found that the program runs correctly. I expected, for voltage vs. frequency, a parabolic trend. I found this trend using gain 1 and -1, but I didn’t find it for higher gains. Probably this problem is connected to hysteresis effects.

Anamaria, David

Mounting readout system for smthe GAS blade test.

For additional information about this report, contact S. Whitcomb or P. Lindquist