Weekly Report for Week Ending May 5, 2005

The LIGO Executive Committee Agenda for Monday May 9, 2005 will be:

(Meeting time: 10:30 am Pacific Time)  

  1. Announcements
  2. Comments on Weekly Report
  3. LSC Issues (Saulson)
  4. LIGO Lab Operations
      • Administration (Lindquist)
      • Sites (Raab, Zucker, Shoemaker)
      • Commissioning (Fritschel), Detector (Coyne)
      • Campus Research Facilities
        1. 40 Meter (Weinstein)
        2. TN, ( Libbrecht)
        3. LASTI  (Shoemaker)
      • Data Analysis and Computing (Lazzarini)
  5. R&D and Advanced LIGO (Shoemaker)
  6. CHANGE CONTROL BOARD/TECHNICAL REVIEW BOARD SESSION AS NEEDED

·        CR-050004 Adjusting FY 2005 Budgets to Reflect Actual Costs through First Half

Special Items:

Special Announcements:

Weekly Report Highlights

LSC Issues (Saulson)

no report

LIGO Laboratory Administration (Lindquist)

STATUS OF LSC MOUs (Petrac)

LSC MOUs and Research Plans and Progress Reports

(LSC Attachment updates (for Feb. 15, 2005 through Aug.15, 2005) and Progress Reports through Feb. 15, 2005)

  • Attachment updates for all active LSC groups are in LSC/LIGO sign-off/Saulson and Barish
  • Progress Reports for all active LSC groups are in DCC for web posting.

SITE TELECONFERENCE (Lindquist)

A site teleconference was held on Thursday, May 5, 2005.  The following issues were among those discussed:

  • Budgetary Issues -- Expenditures through April total approximately 15 million which effectively represents half way point due to time lag.  Still running less than same period last year mostly due to HEPI expenditures last year.  Labor running at about 85 percent of budget to date.  Difference is primarily in contract labor.  Projection to end of year is $30 million plus non-recurring charges (Amaldi,
  • Advanced LIGO, special Observatory maintenance items, et.)
  • Action #130 -- TAX Tree is done, but has not yet been sent.
  • Action #131 -- Convenience Check limits have been increased to be more consistent with P-Card limits.  Action CLOSED.
  • Safety Audit at Hanford -- Completed.  Considered productive.
  • The list of assigned actions last updated May 5, 2005 will be found Here.

PROPERTY MANAGEMENT (Chargois)

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

  • No report.

DOCUMENT CONTROL CENTER (Turner, Mak)

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

  • No report (Broken Foot).

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

  • packages: in - 16, out - 6
  • faxes: in - 25, out - 28
  • No special projects to report.

COST SCHEDULE CONTROL SYSTEMS (Cunningham, Brambila, Kaufman, Salone)

>From: Esther Cunningham <esther@ligo.caltech.edu>

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

  • Completed change order #32 to Excel and submitted it to the vendor and received the signed copy back.
  •  Completed the Purchase Order to TPS for the convection oven for the 40M Lab.
  • Completed the Purchase Order to Datum for the face plates for Calum.
  • Followed up with the vendor on the cancellation request of the laser purchased for Laser Mill which had already shipped to Caltech. The vendor is still in the process of getting back to me on the request for order cancellation. The order has been billed to my credit card.
  • Completed change order #15 to Sydney Meshkov and sent out the revised consulting agreement.
  • Submitted the monthly Purchase Order closeout report for April 2005.

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

  • Nothing significant to report.

>From: Florence Kaufman <fkaufman>

  • Completed and sent out the reports for the following Awards as of the end of April 2005:
    1. FY2005 Operations
    2. Visitors Award
    3. Outreach Award
    4. MIT.GRID
    5. Low Noise
    6. DIA
    7. AJL.INT
  • Prepared schedule for budget modifications for unfilled positions for the second quarter of FY2005.
  • Prepared Cost Transfer to correct travel expenditure in the 40 Meter fabrication account.
  • Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport. (For passwords contact Florence)

SUBCONTRACTS MANAGEMENT (Petrac, Jasnow)

>From: irena@ligo.caltech.edu (Irena Petrac)

  • No report.

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

  • Evaluation of the exhibits to be provided by the Exploratorium is being conducted to divide them into two groups.  This is being done because the LLO Science Education Center will not be ready to have the entire complement installed.  When the quantity configuration has been agreed to, a contract modification will be issued reflecting the change and the payment terms.
  • The SEC architect, EDR, has compiled a list of eight contractors to receive qualification statements.  Five of the contractors have worked previously with EDR, and three were recommended by an architect in the area, Holly and Smith.  When the qualification statements are received, they will be reviewed and the contractors to receive the RFQ will be selected

SUPPORT (Baldon, Hiroto, Lloyd)

>Irene Baldon

  • No report.

>Julie Hiroto jhiroto@ligo.caltech.edu

  • Planning and reconciling many challenging trips for Barry.

>Dorothy Lloyd

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

PROPOSALS and REPORTS (Lindquist)

We “pushed the button” on the Visitors Program Annual Report.  The NSF has accepted the report and approved our funding for the next year.

As mentioned during the Executive Committee meeting on Monday, April 11, 2005, LIGO must submit an Annual Report for Operations by August 1, 2005 and this annual report must be accompanied by a request for a two year extension (FY 2007 and FY 2008) as well as a justification for why Caltech/MIT should be continued in the role of management of LIGO.  I have proposed and distributed a general outline and assignments.

DCC Steering Committee (Lindquist)

The document management system steering committee met on May 5, 2005 to regroup, evaluate where we stand with regard to the systems that have been demonstrated, and to plan where we should go from here.  We are currently favoring two vendors, Synergy and Agile.  Agile is significantly pricier than Synergy but offers engineering oriented features that we may or may not be able to use effectively.  The next step is to contact some references and run further hands-on tests of the two.

CHANGE CONTROL/CONTINGENCY MANAGEMENT (Lindquist)

  • I have distributed change request CR-050004 adjusting the FY 2005 budgets to reflect (primarily) actual costs through the first half of the fiscal year.  I propose to discuss this request during the executive committee meeting scheduled for Monday, May 9, 2005.

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

Completed a successful safety audit of the LHO.  Several important agreements were achieved, including a new "policy" for use of Work permits at the LHO and the definition of an "eye-safe" power "threshold" for "scattered-source" beams.  Also, the remaining open items from the 2004 safety audit were discussed and "closed" at the 2005 audit meeting.  However, the latest audit did generate a new list of action items, which will be distributed next week along with a draft audit report.

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

Summary of Commissioning Activities at LIGO Hanford Observatory (Landry)

Commissioning highlights from this week in Hanford are bulleted below, pertaining to general, 4k and 2k issues, respectively.

  • Selected epochs of Conlog data have been archived (excluding science and engineering runs).  This saves disk space but may impact selected queries on times older than six months.
  • Astrowatch: data are archived periodically based on the IFO "up" bit, an indicator that the IFO is in near-science mode ("everything but setting the science mode flag has been done").  Recently the up bit has not been set, owing to the PSL out of nominal status.  We have to fix the PSL and ensure the robustness of the up bit so as to not unnecessarily limit Astrowatch data.  Later, the up bit definition was temporarily modified to exclude the PSL status field.
  • A document on S4 correlations between 2k and 4k AS_Q channels, and correlations between AS_Q and PEM channels, was posted.  Coupling observed between IFOs is predominantly acoustic.
  • The spare PSL laser was sent to LLO as the main laser there spontaneously dropped in power and needed service

4K IFO

  • Photodiodes on ISCT4 were serviced.  ASPD2 was replaced.  ASPD4 was problematic in a way different from other PDs: in sweeping the alignment of the beam onto the photodiode, good and bad regions were not found, as was observed in PDs with burn spots.  While the dominant noise was shown to be upstream of the demod, the anti-aliasing board was also suspect.
  • The new 30W CO2 TCS laser has arrived, and preparations are underway for installation, alignment and commissioning.  Upcoming tasks regarding the upgrade from 8W to 30W are bulleted in the following work permit
  • A comparison of data from various (old) 4k locks, with different thermal compensation status are summarized here

2K IFO

  • After the current shunt was replaced last week, the next task on upgrading the 2k PSL was to install the latest version of the ISS.  This was started in the latter half this week, and necessitated modifications and alignment to the optical layout on the PSL table (including revamping the MC power-in PD path), so the task is taking a little longer than initially expected.
  • Work started on the IOO WFS: phases and setup of the PDs was varied and different from other configurations that it was suspected something was either wrong with the detectors or simply misset (now appears to be the latter).

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

no report

CDS Hardware

LSC RFPD

Ben Abbott

1)      The boxes are on order, and should be in about 7 weeks from now.

2)      Todd or I will test the PDs that he has made to date, and we will ship them to the sites soon.

3)      Vern, Rich and I are talking tomorrow morning to discuss possible changes to the ASI input on the PD, and how these may affect the PCB.  Once this is resolved, I will send off for another PCB run.

FO Link

Sander Liu

  • The parts for LLO are being packed up and will be shipped today. Installation is scheduled for next week.
  • The source of the excess jitter may be the optic receiver used. Further testing is needed to confirm.

Peter King

I have been going over the numbers given to me by Jay for the timing jitter requirements. Sander Liu and I measured the timing jitter of the existing system in the 40m Lab.

[Note by D. Coyne: Measured to be larger than expected, but within requirements.]

DMT

John Zweizig

This week I continued to benchmark the Quad Opteron box as a potential platform for online DMT use. I was able to run 29 monitor processes in parallel on a test box borrowed from ldas. The monitors read frame data from shared memory (using the DMT online data distribution mechanism) and there was no apparent contention for resources. The 2.2GHz Opteron processors are ~3 times as fast for the DMT applications as the 750MHz UltraSparc machines (this is a little surprising given the differences in the architectures).

PSL

PeterKing

I had left a message and e-mail with Dan Feehan of Lightwave with regard to repairing S/N #108 under warranty.

TCS

Mike Smith

I revised the ACAD drawing for the 30W TCS system.

Optical Contamination Cavities

Lee Cardenas, Liyuan Zhang

OTF Lab. (W. Bridge) 

Cavity # 1

We have prepared and cleaned one set of 70ppm REO mirrors.  We introduced this new mirror into the cavity, made new laser alignment and locked the cavity.  The laser power is now 134mw instead of the 180mw with the previous ATF mirror.  The ring down is 21microsecond instead of 22 microsecond.  We are setting up the equipment to measure the absorption.

Absorption Test Measurement prototype -- in progress

We have made the frame to hold the new HEPA in place.  Now the new HEPA that goes on top of the enclosure is in place and running.  New alignment to bring both the HeNe and Nd:Yag laser beam to meet at the same point on top of the mirror surface is in progress Scatterometer system -- in standby

The Quantronix 60 watt laser -- In progress

Preparation for the large mirror absorption test:  We have found another leak from the aperture holder solder seal.  Very tiny leak from the solder seal.  We took apart the laser cooling and removed the aperture holder and cooling and re-soldered the seals and after several trials the leak has stopped completely.

We have re-assembled the cooling system and run the laser again and recovered the alignment and the power is ~32 watt @ 30 amps the mode might be OK, but we are working on it as now to improve it.

OTF Lab at Lauritsen ROOM 38 -- No Change

Cavity #3 

The contamination test for (20) pieces of Glenair Micro-D-connectors still in progress.  Cavity is locked and we are taking measurements  for absorption and ring down for contamination loss every day. So far so good. It needs another month for the test to be completed.

Cavity #2 -- no change

40 Meter Interferometer (Weinstein)

IFO commissioning:

  • Osamu continues to work on improving the DC-transmission locking of the arms. He measured the rms motion of the arm lengths at 1:30am, and got (5-8)*10^-8 mrms above 0.01Hz for YARM and maybe 10 times higher for the XARM. He worked on optimizing the damping gain on ITMX, but it didn't help much; ITMX still doesn't damp very well. If the servo doesn't catch on the first fringe, it kicks the optic, so that it takes many more fringe sweeps to catch again.
  • Rob continues to develop plans for making a smooth transition from DC-transmission locking of the arms to CARM and DARM, using Finesse modeling as a guide.
  • Rob and Dan measured the sign of the LSC signals for the ETMs, in order to be able to know how to hand control off to CARM and DARM.
  • After losing the mode cleaner on Thursday (see below), Osamu re-aligned everything upstream: the PMC, the frequency reference cavity, and the Mach-Zehnder; in all case, significant improvements in transmitted beam were acheived.
  • Steve and Osamu did some re-layout of the MC reflected beamline, replacing some 1" optics with 2". They would have replaced more mirrors, but CVI sent the wrong mirrors (for green, not 1064; they're out of stock on the 1064 mirrors).
  • So far, we have re-laid-out almost all of the essential beamlines in the last couple of weeks: MCR, AP, POB, TRX, TRY.
  • Because of the MC problems after the AC test, Dan postponed the relocation of the SP RFPDs from ISCT-SP to ISCT-AP. He will do it by the end of this week.
  • Monica has been taking data with arm locks in order to calibrate the DARM and other LSC signals.

IFO modeling and DC detection development:

  • Virginio has finished a coarse design of a suspension point interferometer for the 40m, and is writing it up.
  • Monica continues to work on e2e simulations of error signals sweeps for the 40m, with artificially short arms, to make quasi-DC sweeps for comparison with Twiddle. Osamu wrote a Mathematica routine to export Twiddle sweep data for direct comparison with e2e. There are still some normalization problems to work out.

PSL:

  • Steve and Dan measured the laser power with the calorimeter, at the first power-limiting polarizing beam-splitter after the main beam exits the MOPA box. They measured ~6 watts dumped and ~3 watts transmitted. There's also a known loss of ~1.0 watt betwen the MOPA and the beamsplitter (measured a year or 2 ago). Thus, there's 10.2 watts out of the MOPA box. The monitoring PD in the MOPA box reports 11.2 watts, so the EPICS database was modified to report the measured value of 10.2 watts.
  • Steve then measured the PSL power at the periscope where the beam exits the PSL table and enters the vacuum chamber. The uncalibrated Newport power meter was measuring 1.45 watts there. He measured 1.26 watts with the calorimeter.
  • The ISS continues to saturate frequently. This seems to be associated with the Mach-Zehnder PZT running out of range, sue (it seems) to a drift in the MZ arm length difference due to temperature changes. This might be due to temperature-dependent OPL changes in the Pockels cells; one arm has 2 Pockels cells, the other has one, so we are thinking about putting a compensating crystal in the other arm. But, meanwhile, we need an alarm or alert when the MZPZT drifts out of range, so that we can break MC lock, re-center the MZ PZT, and re-acquire lock of the MZ and MC in a controlled manner. Dan is working on that.
  • Dan found and fixed a bug in his ISSautoreset script: when you turned the ISS servo off (eg, to measure PSL power) the script would turn it back on after a minute. Now if the user turns off the loop switch, the script will not reset it.

Electronics, controls:

  • Last Thursday, Steve turned off the air conditioner for ~2 hours in the IFO hall (which warmed up by ~2 degrees C), to see what effect it had on the MC RF frequency difference (we monitor the 33 MHz RF transmission and reflection from the MC and see evidence that the MC length is changing by microns). Well, it made a big mess, and we didn't learn much... But the MC wouldn't come back up, afterwards. Turned out that *many* op amps on the MC demod board and servo board were burned out (on the MC Servo board, one of the LT1125s blew, a nd all three of the LT1128s blew. On the I&Q Demod board, an LT1125 blew.) Ben and Osamu spent several days finding and fixing them. This probably was due to an unknown problem with the Sorensen power supply or supplies), which may or may not have anything to do with the AC. Ben will check the power in that rack.
  • We have been monitoring the 33 MHz RF transmission and reflection from the MC, in terms of the frequency difference between actual and resonant (MC_FREQ_DIFF). We see large (~100 Hz) changes between MC locks, smaller (~20 Hz) daily variation, and smaller (~7 Hz) variations with 20 minute periods. We believe that these are all from different causes, including both MC length drifts and PSL frequency drifts. The 7 Hz variation seems to be correlated with the PSL FRC temperature MINCOMEAS. We're thinking about ways to correct this.
  • Virginio is working with Rana on a coordinated effort to diagonalize the MC suspension controllers and coils.
  • Osamu reports that the DC-offset-lock of the arms is not easy, because of excess noise on the QPDs at the ends. There seems to be a grounding problem with these QPDs. When the ground shield is connected to the rack, the C1:LSC-TRX_IN1 signal becomes around an order of magnitude noisier. Jay and Ben did a search for the causes of this noise, but we haven't located it yet.
  • Ben has finished testing the 166 MHz RFPD for the SP beamline; he will deliver it it tomorrow.
  • Todd has finished working on the 166MHz I&Q demod board for the new 166MHz PD, has tested it and delivered it to Ben. He will bring it over with the PD.
  • Ben fixed a bad LEMO cable on the ITMX LL OSEM. Osamu thinks the ETMX UR OSEM may also be bad; Ben will check it.
  • We expect 4 new Rev B coil drivers (for PRM, SRM, BS, spare) from Todd in the next week.
  • Rob found that the ETMX CPU was running around 58-59 usec (out of 61 usec). No timing delays were evident, but nonetheless a reboot of the CPU fixed this "phase delay". It may be due to a problem in the fiber optic timing link; under investigation.

Lab Infrastructure:

  • Our 6 month old HP2550LN color laserjet printer broke and Larry and Mike replaced it with an old, noisy HP4500N unit. Larry is trying to get the HP2550LN repaired and returned to us.

Bake oven Lab:

  • Carpenters are tearing up walls in the South Annex to allow for the installation of a new large air bake oven for AdLIGO suspension development.
  • Bob is assembling new PDs and LEDs for new AdLIGO OSEMs.
  • Bob is working with Calum on building optics table legs for the Synchrotron suspension lab,
  • Bob is working with Jay, Janeen and Calum on OSEM cables for quad suspensions.

Thermal Noise Interferometer (Black)

Nothing significant to report this week.

LASTI (Ottaway)

Laurent Ruet reports:

I have adapted the mathematical model of the one dof modal control/estimator to n dof. I had to use a very different method and use state space format instead of transfer functions matrices but it works now a lot better. I can now study the stability of the loop and see how a bad modeling affects it. I think I will be able to test the control loop on the pendulum next week.

Richard Mittleman reports:

We have been working on the BSC support structure ground noise amplification problem (BSSGNAP). There are a number of entries in this weeks ilog. So far we haven't isolated the problem, but there are many hints.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)

Weekly Physics Meeting

Hiro presented a preview of the SimAdvLIGO package being developed by Matt.  After that he presented some FFT results for simulation of the g-factor measurement by Rick Savage et al at LHO. Viewgraphs available at http://www.ligo.caltech.edu/~e2e/ME2ET/Minutes05/050505/

Commissioning Support

(Hiro) The effect of the mirror phase maps on the g-factor measurements by Rick, Keita, et al., was studied using FFT.  The summary is available at http://www.ligo.caltech.edu/~hiro/talks/g-by-FFT.pdf.  The phase map affects the g-factor measuments in two ways.

(1)   Without mirror tilts, the average ROC(ITM) is shifted by 30-50m due to the furface aberration. The average mirror curvature is determined  by fitting the phase map measured in 7.5cm radius, while the beam sees smaller area. This difference introduces this difference.

(2)   When a mirror is tilted (in the simulation, ETM was tilted), the  ROC changes by 30-50m for a tilt of 1 micro radian. When the surface is smooth, the mirror tilt does not affect the g-factor measurement.

The magnitude of the measurement t0 measurement variation in the  g-factor measurement is several 10m, so it can be explained by this mirror  surface aberration.

(Biplab) As reported by Rana in LLO elog, implementation of a proper Gouy telescope for WFS1 remedied a number of problems encountered earlier.  Probably some more adjustments in other WFSs would also be necessary (according to FFT simulation), even though no immediate reasons for doing so has arisen yet for LLO WFS system.

SimAdvLIGO

(Matt Evans) Matt has delivered SimAdvLIGO, a framework of the advanced LIGO simulation package.One major guide line is that the software organization is the same as hardware organization. One table holding sets of optics is represented by one box, cables carrier, digital or analog signals are represented by one connection lines, etc.

It has many place holders and empty boxes, but the major structuring was finished. Matt tested the closed loop behavior and confirmed that the loop is working. Hiro summarized the products and the infrastructure in the e2e weekly meeting.

Matt will come back to Caltech on June 3rd, and the model will become fleshed out and hopefully will be usable for simple, yet important tasks, such as the linear or alignment actuation study.

Simulation of 40meter interferometer

(Monica) The implementation of a new Twiddle function made by Osamu allows writing raw data of the error signal sweeps. After having collected all the Twiddle results for the sweeps in the resonant point region, the comparison with sweeps (DC response) simulated with e2e package for the 40m can be done. The comparison of the error signal sweeps requires a  DC response from the e2e package, so a 4m arm long is used in the e2e simulation instead of 40m as a good compromise between simulation speed and adiabatic case.

2-photon model in e2e

(Biplab) Have gone through relevant literature to make a tentative list of what we need to do and take care to implement 2-photon model in e2e for studying squeezed states. Started correspondence with Yanbei Chen to understand and clarify various issues.

modeler

(Hiro) In order to support the new SimAdvLIGO, modeler code is being updated. Major changes are

(1)   different time step support - modules like digital filter between  ADC and DAC sees proper digitized time step automatically.

(2)   merging of setting and macro

(3)   fast dual recycling module

(4)   better modal model to simulate more complicated system  - like different losses seen from AR or HR sides.

It will take several months to complete these modifications.  The priority will be determined by the progress of the SimAdvLIGO package by Matt.

Alfi

(Bruce) Reworking the implementation of bundle generated connections.  Both a speed and a modular design issue.

(Melody) Continuing on fixing existing Problem Reports.  Currently working on the user interface for allowing the user to  choose the display for nodes: graphic, node name, or a multiline comment.

Data Analysis Activities (Lazzarini)

Charlton:

This week I have been working on matlab code to compare calibrated strain data with the (frequency-domain) calibrated strain produced in the stochastic background analysis pipeline. The purpose of the exercise it to look for errors or differences in the strain calculated by the two methods.

Creighton:

I started looking into a bug in the LAL 2d mesh tiling routines, whereby they occasionally get trapped in an infinite recursion.

Mandic:

I worked on understanding the features in the coherence between H1 and H2 strain channels during S4. I observed that the H1ASQ-H2ASQ features can be found in the coherences between the ASQ channels and the microphone or ISCT-accelerometer channels. On the other hand, such features could not be observed in coherences of the ASQ channels with magnetometers or with the mains-monitoring channels. Hence, the coupling of the features observed in the H1ASQ-H2ASQ coherence is likely acoustic or mechanical in nature.

Mendell:

I have added to the CW/pulsar S2 StackSlide investigations page: http://www.lsc-group.phys.uwm.edu/cgi-bin/enote.pl? nb=puls2stackslide&action=view&page=last showing the results of a study of the power spectrum of exact pulsar signals vs. the power spectrum in the approximation that the amplitudes and frequency of the signal are constant during each SFT.  This approximation is widely used by the CW group. Conclusion:  The mean relative difference between the approximate and exact signal power is near zero, though the mean absolute difference may tend to be positive, indicating some bias towards the approximation over-estimating the power. However, this conjecture comes from a few outliers, the statistics are not good enough to say for sure. If true, this would only have an effect on the predicted StackSlide power and maybe the predicted F-statistic. The approximate form of the signal also predicts the bin with maximum power incorrectly about 4% of the time, but the actually power loss due to this is much smaller, since this happens when the frequency is between two bins, and the power in neighboring bins differs by only a few percent from that in the bin with maximum power. Thus, this should have an effect on StackSlide, Hough, and F-statistic at much less than the 1% level.  I am also working on further validation of the StackSlide code to prepare for the generation of S3 and S4 results.

Shawhan:

  • Reviewed existing data quality information for the S4 run.
  • Made a catalog of lock losses during the S4 run, and defined PRELOCKLOSS data quality flags.
  • Studied dips in the stored light in the arm cavities during S4, which  tend to be associated with glitches in AS_Q.  Many of these occur near  the end of a locked stretch, so it seems prudent to omit the last ~30 seconds of each locked stretch.  I am currently working on  defining  LIGHTDIP data quality flags.
  • Generated segment list for re-running WaveBurst on S4 data.
  • Critiqued an analysis in progress to look for gravitational waves associated with the giant gamma ray flare from SGR 1806-20.
  • Checked the total observation time of the S2 binary black hole  inspiral  search.

Lazzarini:

Working with Tinto and Sutton on an extension of the Gursel-Tinto algorithm to develop a detection and estimation algorithm for an arbitrary network of detectors. This will yield a model-independent estimator of source h-plus, h-cross, and source location. It is essentially a extension of the r-statistic concept to an arbitrary network of global detectors.  Interested readers may find the following movies I made interesting (NOTE: these are very large files!; some PCs  may not be able to view them - caveat lector):

http://www.ligo.caltech.edu/~lazz/distribution/LSC_DataDir1/LIGOmovies/ GT_ReconstructvsSNRPosition.avi (<- supposed to be PC compatible)

http://www.ligo.caltech.edu/~lazz/distribution/LSC_DataDir1/LIGOmovies/ GT_ReconstructvsSNRPosition.mov (<mpeg)

http://www.ligo.caltech.edu/~lazz/distribution/LSC_DataDir1/LIGOmovies/ GT_MovieAitoffv2.avi (<- supposed to be PC compatible)

http://www.ligo.caltech.edu/~lazz/distribution/LSC_DataDir1/LIGOmovies/ GT_MovieAitoffv2.mov (<mpeg)

This is still a work in progress....

Weinstein:

  • Reviewing S2 BNS analysis and paper
  • Working with Lisa and Duncan on ringdown search

Yakushin:

Data analysis:

1)      Tuning quadruple coincidence LIGO-GEO waveburst script;

2)      Modifying waveburst to run on the new S4 MDC frames.

LIGO Data Analysis System

Software Systems (Blackburn)

LDAS:

  • Work continued on PR 2825 to address the issue of not properly reporting time ranges where there are frames files that should be reported. Currently have a fix that is being tested on tandem-V.
  • Worked on PR 2724 concerning the modification of resource variables and PR 2617 which is related and concerns the headers for these resource variables.  Adding the ability to source columns with a single mouse click in the  controlMonitorAPI's client.
  • Successfully completed system and integration testing for LDAS 1.5.60 and updated the results on the webpage.

TCLGLOBUS:

The first alpha release of Tcl/Globus is nearing announcement. The tarballs have been ready for a couple of days but a request for rpms has delayed the announcement. Also a few issues with the documentation on how to build the alpha release have been discovered and are being addressed today.

There is significant pressure in the GRID community to migrate to the new Globus GT4 release which came out last week. This was discussed that both the GriPhyN and OSG meetings. A review of the changes to Globus components used by  Tcl/Globus between the current 3.2.x version and the new GT4 version suggests that this should be a nearly transparent port for Tcl/Globus.

Fixed FTP client, completign the callback SWIG wrapper function. The function had mis-matched arguments when passing the value back to the TCL layer. Updated chapter in thePDF document for Globus FTP client.

Created a debugging model for all test cases so its easier for  end-users to know what cases to implement in the new model

GRIDCOMPUTING:

Attended the GriPhyN all-hands meeting at Argonne National Lab.  Presented LIGO talks for status and planned use of technologies that have come out of the GriPhyN/iVDGL Projects.

Discussed the formation of collaborative activities with Univerity of Chicago, ANL, and ISI to evolve LIGO's current usage and to help guide the development of the virtual data language based on experiences.

Attended the OSG Consortium Council face-to-face meeting at UW Madison to discussOSG by laws, governance procedures, reports from integrations, deployment and operations activities as well as the interim executive board. Funding agency prospects and stategies for overseeing proposal activites were also discussed.

Hardware Systems (Anderson)

Caltech

(Dan Kozak)

  • Did some research about modification/changed/creation times in SAM-QFS.  Determined that the problem we'd noticed (modification times not updating on the MDS for files being appended to by a shared client) has been fixed in 4.3.
  • Got STK to look at drive 57 (CIT).  Diagnostics reported no problem and the drive has been working since.
  • Made an entry in releaser.cmd (weight_size = 0.0) at CIT to get the releaser to ignore file size (and hence implement an LRU algorithm for releasing files).
  • Working on getting new(!) S2 RDS frames to tape.  Also using this opportunity to clean up CIT archiver.cmd.
  • Continued to reinstate dual copy at CIT.  Ongoing.
  • Finished copying some "missing" (at CIT) E10 rds data from LHO using tar/globus-url-copy.  Working on getting this to tape at CIT.

(Phil Ehrens)

  • Various minor tasks in support of LDAS.
  • Ongoing analysis of applicability of DOE/GRID proxy cert use with Subversion/Apache. Determined that security level of proxy certs is incompatible with security level of full user cert challenge protocol. Even when the certificate chain is fully visible to openssl, there is a problem with the proxy cert being issued by a user and not a CA.  Elevating the authority of user certs in the same context where authoritative CA certs are used is probably not good. A way of separating the authority level is probably possible. The inconsistency in the use of certificate nomenclature between various software suites is the primary difficulty.
  • Installed Solaris 10 on the "samtest" machine in anticipation of testing a new Sun 10 Gbit Ethernet card.

(Stuart Anderson)

  • Further adjusting of Condor configuration to handle a large number of jobs.
  • Ran failed test of 10GigE host-to-host test between Solaris/Suse quad-Opteron machines in Millikan/Bridge. The next step is to run between Solaris/Solaris in the same building.
  • Running LDR to catch up on Astrowatch data.

MIT

(Keith Bayer)

  • Added new scsi drive to dmt machine lancelot (this drive will be backed up).

Livingston

(Igor Yakushin)

  • node70 failed, probably bad drive.
  • Transferred h(t) S4 data from CIT to LLO.
  • Transferred SG6_S4 MDC frames from PSU to LLO.
  • Discovered a strange behaviour of LSCdataFind when a client on ldas-grid@LLO queries the server on ldas@LLO through GC interface: if the queries are followed one after the other, some of them fail, but if there is "sleep 1" between queries, they succeed. When client is running on my desktop or at CIT and using the same server at ldas@LLO there is no such problem, if client is running at ldas-grid@LLO and queries server at CIT, again there is no such problems. Most likely there is some hardware network problem in GC that affects the communication between ldas@LLO and ldas-grid@LLO.

Hanford

(Greg Mendell)

Several enhancements to the createrds.tcl and createrdsGUI.tcl scripts have been completed

1)      Added options that allow the merging of data from different frame types and IFOs.  This will be used to merge GEO data with LIGO data.

2)      Added a better gap checking algorithm (should only get "short frames" now when a true gap in the input data exists) and an option to align output frame start times with the output framefile length.  Thus, a rule can be used to find any normal (i.e. non-short) frame file based on its GPS start time, which will be a multiple of the duration of data in that file.  Changes have been checked into CVS: CVS/Root=:pserver:anonymous@gravity.phys.uwm.edu:2402/usr/local/cvs/ lscsoft CVS/Repository=dsorun/contrib/createrds/scripts.  In this directory see README.createrdsAndcreaterdsGUI for further details and createrds.rsc for an example resource file.

(Ben Johnson)

  • Dusting off old S3 data corruption notes. The stochastic group asking questions about 16Hz.
  • Touched last of "old" A4 data.  All old A4 data should be at Caltech by now.
  • Working on new AstroWatch archiver GUI/system. It should allow for better comments, auto-elogging of intervals+comments, and a auto-archive timout in case the button doesn't get pushed.
  • Attempting to get the certificates updated for the various servers; ldas-grid, ldas-gridmon, and ldas-pcdev1.
  • Sent off "1 Year + Option" Sun Service Contract P.O. to Dorothy Lloyd May 4 afternoon.
  • Getting h(t) frames from Caltech. H1/H2 frames completed. L1 frames transferring today.
  • Publishing S3 L1 LLO RDS data on the cluster. The original copy+publish process was not completed for some reason.
  • Helping diagnose why the conlog_up script does not consider recent data to be "UP". The issue appears that the PSL status bit is bad, the reasons are murky. Possibly due to bad EPICS code.

General Computing (Wallace)

MIT:

(Keith)

  • Continuing LDAP work (when time permits)
  • Setup accounts for new grad student (Nickolas Fotopoulos)
  • Installing darcs software on cadwalader

Livingston:

(Shannon)

Busy with a network installation for CDS. See Livingston report.

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
  • Two scanners seem to have suffered from the power outage last week.  Both are visible from the PC, but will not scan and will not communicate with the scanning software.  The manufacturer has been contacted for support on one which is still under warranty.  The other I will try a complete reinstall of drivers and software.
  • A SCSI cable on the tape drive used to backup user files is reporting errors.  The main file and e-mail servers will be powered down in order to replace the cable.
  • Tom Villani of NoaNet dropped by for a visit on Tuesday.  Tom is promoting an alternate WAN connection between Richland and Seattle which we are looking at.
  • Spent some time trying to figure out why a Sun workstation would not recognize one of the printers.  Finally ended up putting the printer in the hosts table on the Sun and then it worked.  Still don't know why communicating with this one printer was different than the other printers.
  • Downloaded some recent versions of software from Caltech ITS.  Ordered some software upgrades.  Renewed the maintenance on several Matlab licenses.  Purchased the latest version of Partition Magic along with 10 more licenses.  The number of dual boot Windows/Linux computers is growing.
  • Other misc. user support.

CIT:

(Mike)

  • Worked on the swapping out a printer from the 2nd floor of W/B, and moving it to 40 Meter. This took most of the day cleaning up this printer, and adding an additional network card to enable users to print from both,  113 & 115 subnets.
  • Continued to work on two visitor workstations, loading XP and working on setting security and loading GC software.
  • Worked on wiping two laptops to give to PMA. Ed Chargois is aware of this, for these are surplus laptops.
  • Helena Armandula: Her laptop died. After trouble shooting this issue, I called Dell's tech support to get another other board for this unit.  This issue has been resolve, and user is back up and running.
  • NTSRV's: Ran end of month ghost backups.
  • Ghost Server: Burning critical images to DVD, and removing them from the server afterwards. This will allow additional disk space for additional backups.
  • Continued to work on a NTSRV that is loaded with 2000 server. This server is to be use for an apache webserver.
  • Rerouted many network cables in the server room.
  • Other user support that included the usual support: Printers, Software, E-mail, and some networking issues.

(Veronica)

  • LSC:  Website updates. Installed the website for the upcoming meeting.
  • LIGO:  PAC website updates.  High-resolution images for the NSF and a popular-science publisher.  Compressed the video of last week's software demo.  Working on changes to the roster database utilities to update the authorship list.
  • Project Science:  Moved the files over to TMT.  Waiting to hear whether our help is needed with installing and configuration.
  • CaJAGWR:  Website updates and user support.
  • Initiated work on the new LIGO MIT web site.

(Larry)

  • Worked on a number of purchase items. All items have been received and delivered.  Received quotes for a number of computer systems to be installed in the server room. Still working on a SUN quote.  Renewed the license for the FEMLAB package.
  • Worked with Mike in replacing the printer in the 40M. Still in the process of getting their unit repaired. There were a number of logistical issues that had to be worked out with different companies. We are now waiting for HP to supply a shipping container to have the unit shipped to their repair facility.  We still have not been able to find out why they will not just send out a fuser kit and have it replaced in the field like they do with the other printers.  Moved a number of computer monitors around to new locations.
  • Helped Stuart with some testing of a 10GigE board. This opportunity allowed the OS upgrade to be done on the sandbox, so even though the board did not work something positive did come out of the time spent.
  • Working on getting power to the computer room for next weeks scheduled power outage. West Bridge and Bridge annex are scheduled to be down all day on May 14.
  • Working on recabling in the computer room. Punched a new hole in one of the racks and started moving cables. Mike has swapped a number of them over. Now we just need to install a couple of cable trays to get the cables organized.
  • Performed the monthly backups on the user accounts. Started the full backups on the servers. Just 12 more to go.  Restored a couple of file systems for different users.
  • Worked on cleaning up a number of user accounts. Cleaned off a couple of Gigabytes of disk space. Presently, generating a new list of user accounts, no longer active, to get approval for removal.
  • Working on a couple of documentation items. A new Computer Usage Policy and NIS+ setup for Linux are two that are in process.
  • Worked on getting the mail servers back on-line. The end-user server was down for a couple of hours with a disk problem. This outage showed that there is more dependency between the e-mail servers than there should be.  Discovered that backup server for the mail servers no longer exists, so that is now a new project to get going on.
  • Working on a couple of new hardware items for travelers. If they work it should facilitate their being able to communicate electronically.
  • Worked on a number of enduser machines to get them back on-line.
  • Between Mike and Larry over two hours a day is spent on the spam filters, checking for false positives. Difference between spam e-mail and legitimate is getting grayer.

 

Mail Statistics for April 28 thru May4, 05

Accepted Messages                              17,958

Rejected Messages                                23,598

Virus Messages                                       1,245

False Positives                                           373

Total Messages                                     41,556

Advanced LIGO and Supporting R&D (Shoemaker)

Systems and Management

From: Carol Wilkinson <wilkinson@ligo-wa.caltech.edu>

Development Progress Updates

Progress updates through April 30, 2005 on development of advanced LIGO subsystems have been completed using a new process for collecting the update information. Updated files will be available on May 10 at the usual web site:

·         See http://www.ligo.caltech.edu/~tfrey/index.html for a complete listing of all project related cost and schedule data.

Advanced LIGO Planning:

Work is progressing on recalculating bottoms-up estimates for the costs and shedule for the Advanced LIGO construction project. Work to date has concentrated on SEI fabrication and the Facility Mods and Installation subsystems. Fabrication and installation costs and schedules for the remaining Advanced LIGO subsystems will be scrubbed in the next two months. Subsystem leaders should be prepared for a planning session in that time window. The goal is to have the new overall project costs and schedule by the end of June

The web pages with advanced LIGO planning information, including the cost book, will be revamped and password protected to make the information more accessible to planners and less accessible to the general public. The cost book itself is being recreated in a different format.

Seismic Isolation

From: Ken Mason kmason@ligo.mit.edu

no report this week

Suspension

From: Janeen Romie romie_j@ligo.caltech.edu

Advanced LIGO Suspensions

Checking drawings for Calum. Working on a tablecloth dummy for the ETF quad.

Participated in the installation fixtures review this morning.

Met with JoAnn Hasbach and Calum in the Synchrotron yesterday to give JoAnn a heads up about the clean room that will be installed there in a few weeks. Discussed a few safety related issues and other small things.

Met with Ric Paniagua about the tablecloth details. Will include on the drawing the mounting holes for the ECDs and mark up the top plates of the top mass to include the vertical stops.

Compiling input from a number of SUS team folks for design progress (formally Primavera updates) for Carol and Dwight

Provided updated budget info to Carol, Dennis and David.

Participated in a meeting on charging of optics, hosted by Gregg Harry, during the regular SUS weekly slot. Representatives from Moscow Univ., the Univ. of Glasgow, Stanford, Caltech and others reported on research.

From: ctorrie ctorrie@ligo.caltech.edu

 

From: Ken mailand kmailand@ligo.caltech.edu

I have worked with Helena and purchase order was released for an air bake oven with 3’x 4’x 5’ internal dimensions. The delivery is promised in 5 to 6 weeks.

I have worked with Bob Taylor re. the installation at the 40 meter site, with an oven dimensional drawing, the location set up, and electrical and venting prep work has begun.

I have requested quotes on a cleaning station, and bake oven combination, these quotes is expected w/in 2 weeks, this system will have the capacity for all the large Advanced LIGO parts.

A completed preliminary articulated arm fixture was discussed at the Wednesday meeting, and comments and suggestions are being incorporated into a revised arm design.

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

This week I finished coding the dSpace controller software which will be used to run the quad prototype being assembled at Caltech, and started working with Mohana and Paul to interface it to the electronics.

Core Optics

From: Gregg Harry <gharry@ligo.mit.edu>

Flavio Travasso has joined us at MIT to work on our thermal noise effort.  This week, we sent the CSIRO coating with intentionally poor stoichiometry to Stanford for annealing.  In its place, we hung a sample with silica/tantala coating from CSIRO with the substrate commercially polished instead of superpolished.  The lowest two modes, the butterflies (or farfalla as we now call them), have been found and rungdown.  The Q's are 3.92 +/- 0.04 E5 and 3.55 +/- 0.09 E5, compared to 4.4 +/- 0.1 E 5 and 4.5 +/- 0.1 E5 for the superpolished sample.  This difference is probably not due to the polish, but further data is needed to be sure.

From: Helena Armandula ahelena@ligo.caltech.edu

CSIRO

Last Friday, CSIRO visited Caltech. We had a telecon with several members of the Coating Development Group. We used the opportunity to discuss coating issues related to CSIRO's work.  We decided to anneal the poor stoichiometry coating sample at Stanford to evaluate any "Q" changes. GariLynn will make interferometer measurements on the coated part before the annealing, absorption will be measured also.

CSIRO is finalizing their research to produce a low stress coated sample.

Next, they will coat 2 "Q" substrates of each type with the regular 30 layer design; the samples will be used for different annealing experiments.

LMA

Jean-Marie will start the coating calibration runs next week in preparation for coating the TNI mirrors. The fixtures were promised now by the end of next week. Still is quite possible to get the mirrors finished by the end of May-1st week in June.

There is a "Q" measurement system now at LMA being calibrated. So far, they have a very close correlation with our measurements, agreement is not with an specific "Q" number, but with  the percentage of improvement seen between different coatings.

Jean-Marie will look into the coating design suggested by DeSalvo/Pinto, he is in the process of refining coating simulations.

Adv. LIGO SUS

The indium cold welding to join flags and magnets works in principle, but, the procedure needs to be up-graded to be deemed reliable.  I am looking into possible improvements, I am considering to enhance the indium bonding with a silicate bonding?

Pre-Stabilized Laser

From: Peter King pking@ligo.caltech.edu

The newest hardware revision of a high power photodiode was tested at the 280 mA level.  The heatsinking appeared sufficient, as the case is just slightly above ambient temperature.  Testing at the 500 mA level will start once some connector components arrive which will enable completion of another photodetector.

Auxiliary Optics

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

BS DIFFRACTION LOSS

I recalculated the BS geometric and diffraction losses using ZEMAX, with the input beam from the power recycling (PR) mirror centered on the BS HR surface; this minimizes the loss from the signal recycling (SR) mirror direction and also minimizes the total loss. The loss was studied as a function of BS diameter and thickness, for the case of no lensing in the reaction mass (RAM) in front of the ITM. I completed the loss calculations for an alternate BS model that incorporates a lens element into the reaction mass (plano-convex, with R = 9m); this reduces the beam size at the BS from 60mm to 46.7mm. This change allows a smaller BS for the same amount of loss obtained with no lensing in the RAM.

The total loss (in Watts) with 1 W input from the PR mirror and from the SR mirror for two possible choices of BS dimensions, with the standard configuration and the alternate configuration are summarized below:

Table 1 : BS Loss Comparison for Two Dimensions

BS DIMENSIONS                     PR LOSS TOTAL           SR LOSS TOTAL        D/t

t = 60mm, D = 350 mm             3.94 E-4                        1.01 E-3                     5.8

t = 60mm, D = 370 mm             1.61 E-4                        4.18 E-04                   6.2

 

Table 2: Alternative Design, BS Loss Comparison for Two Dimensions

 

t = 40mm, D = 260 mm             5.69 E-4                        1.18 E-3                     6.5

t = 50mm, D = 300 mm             9.10 E-5                        2.18 E-04                   6.0

Other Laboratory R&D

From: "Erika D'Ambrosio" <ambrosio@ligo.caltech.edu>

Marco (3May)

I placed a 10 degrees wedge on the transmitted beam from the folder mirror. The acquired beam shapes don't shown any astigmatism. Noise is dominating for TEM00 profile, while TEM10 couplings are strongly affected by misalignments.  I also took many spectra by cavity length scans: after several comparisons it seems there is some not-understood "drift effect" that made non-linear the spectrum sampling.  In order to identify each peak seen on these spectra today I assembled a dithering servo loop to lock the cavity on the Airy peak: for a while I wasn't able to lock the cavity.  End mirror PZTs cables were fixed and their work was good.

Creep experiment: long run at 40 C: average position stable at 6.97 mm almost every day.

Juri

I completed the analysis of the expected beam shape using the LMA Mexican Hat mirror maps 5008 and 5009 and the results are available as "Beam shape analysis for the mirrors 5008 and 5009"in http://www.ligo.caltech.edu/~jagresti/ . I used my program for eigenmodes calculation in the case of non-symmetric cavity to find the transverse modes frequency distribution for our FP cavity (with spherical end mirror); the results are in perfect agreement with the theoretical analytical formulas for Laguerre-Gauss modes. This was a necessary step in order to test the procedure to extract the frequency spectrum from the eigenvalue problem because in the case of Mexican Hat mirrors we don't have an analytical formula for that. The comparison with the experimental spectrum is still under investigation.

Justin

Currently making ring down experiments for q-factor determination of the steel flex joints. After realizing a slight miscalculation, a correct resonance of the flex joints has been found.

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