Weekly Report for Week Ending June 16, 2005



There will be no LIGO Executive Committee meeting scheduled Monday, June 20, 2005.  We will resume as usual on Monday June 27th.


Special Announcements:


Weekly Report Highlights


LSC Issues (Saulson)


No report.


LIGO Laboratory Administration (Lindquist)


STATUS OF LSC MOUs

  • No report.

Non-LSC MOUs

  • No report.

SITE TELECONFERENCE (Lindquist)

  • No site teleconference was scheduled for Thursday, June 16, 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>

  • No report.

DOCUMENT CONTROL CENTER (Turner, Mak)

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

  • No report.

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

  • packages: in - 16, out - 9
  • faxes: in - 25, out - 10
  • Worked on gathering data (funding sources, cite references, etc.) for all 1999-2005 publications.

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

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

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

  • Completed the purchase order to Piezosystems and faxed to the vendor. There should be some changes coming through on this order.
  • Completed change orders #161, 162 & 163 issued to Triad and routed them for approvals.
  • Attending an all day training session on TechMart to prepare for implementation next week.
  • Working on pulling purchase orders and subcontracts that are on the sample lists for the PWC auditors.
  • Working on completing the new order to Yamilet Gardens for landscaping for LHO.
  • Working on completing the two change orders to the Exploratorium..

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

  • Nothing significant to report.

>From: Florence Kaufman <fkaufman>

  • Requested set up of new fabrication account for Stabilization System for ISC Tables at Livingston.
  • Updating monthly reports.
  • Met with Carol Wilkinson to discuss ways of reporting and tracking Advanced LIGO costs and budgets.
  • Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport. (For passwords contact Florence.)

SUBCONTRACTS MANAGEMENT (Jasnow)

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

  • Stan Whitcomb and Ed Jasnow met with Tom Schmitt, Caltech Associate Vice President for Human Resources, to discuss placing LIGO contractor staff in the HR data base in order that they may use the new TechMart system being introduced by Purchasing Services. After some discussion, Tom agreed to put the LIGO contractors in the data base.  Seven contractors have been given specific purchasing dollar limits, and forms to fill out to get their information into the Oracle system.  We will expand the contract staff in the data base if required.
  • We have not yet received the 100 percent drawing package from the LLO Science Education Center architect, Eskew - Dumez - Ripple.  The drawings were originally due on Monday, June 13.  This may delay the bid opening scheduled for the middle of July.

SUPPORT (Baldon, Hiroto, Lloyd)

>Irene Baldon

  • Processed the paper work for sixteen (16) new/revised trips.  At this time there are six (6) new trips that need to be completed and ticketed before the paper work can be completed (reservations made and prepaid by my P-Card and advance checks made out.  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.
  • Completed twelve (12) Expense Reports and there are sixteen (16) 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 is only one (1) report more than 30 days old.  I have no reports awaiting signature at this time.

>Julie Hiroto

  • Assisted with SURF student orientation and incoming student requests.
  • Processed reservations for three new trips, including international faxes and phone calls; procured rental cellular phone for Barry to use in Japan. For other travelers, “Cellular Abroad” seems to offer the best rate for long term visits.
  • Redesigned LIGO Activities bulletin board, outside 102 E. Bridge.

>Dorothy Lloyd

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

PROPOSALS and REPORTS (Lindquist)

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.  We are beginning to get some contributions.

DCC Steering Committee (Lindquist)

We met to review status and define a future course of action.  We have an action to look at the demo pages of another system, LibertyNet, to determine whether or not they are worth a serious look.  We are beginning the process of comparing the attributes of various document management systems against our requirements list and my attempt to set up an additional discourse with the vendors for a selected few.

CHANGE CONTROL/CONTINGENCY MANAGEMENT (Lindquist)

  • No open change requests.

HUMAN RESOURCES (Akutagawa)

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

  • A Staffing Committee Meeting was held Monday, June 13, 2005.

Quality/Safety (Tyler)

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

  • Arranged and coordinated appointments for laser safety eye exams for SURF Students and LIGO personnel.
  • Provided safety briefing for new LIGO SURF students.

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


Summary of Commissioning Activities at LIGO Hanford Observatory (Landry)

  • Seismic upconversion appears to be caused by stack motions, and not simply optic motion induced by LSC or oplev injections.  Shaking the ground to produce side motion in the optic does not induce upconversion.
  • better compensation filters were made by first measuring the hardware transfer functions on the floor, fitting this as-built filter with Matlab code, inverting the fit and sticking it in Foton format/LSC filter banks as compensation
  • 108Hz line: it was shown that the line observed in the VME crates in fact originates with the fan power supply
  • The noise budget code received some mods to improve its reliability, plus new wrapper code adapted from LLO.
  • just your run of the mill plague of locusts, well, grasshoppers
  • More on the broadband coherence study observed between AS_Q and magnetometers: these two experiments suggest that the magnetometers are indeed seeing noise that is entering into AS_Q, instead of being fooled by the AS_Q signal itself

4K IFO

  • The 4k input pointing was realigned, and the position on test masses modified, restoring a good deal of the binary inspiral range that we had lost over the past few weeks: we're above 7Mpc, so, better than we've been running, but not consistently achieving 8Mpc.  Later, at high frequencies, the H1 displacement sensitivity grazed the SRD.  Low frequency noise at the time was abysmal.
  • The AS4I_CORR sign flipped is now understood to have flipped two weeks ago when cable was added 
  • A new photodiode and preamp for the ITMX TCS laser was installed.  Low noise modifications to TCS preamp were summarized
  • More photon calibrators were installed, employing a new alignment-friendly layout
  • 50% less light now lands on WFS5; scaling to high power operation of 6W on the mode cleaner would have corresponded to 600mVpp RF on the diode.  The diode was also moved to expand the beam landing on it. 
  • Seven hours at 6W: a long lock was registered on the 4k.  A WFS centering servo is argued for in this elog.

2K IFO

  • WFS work was the dominant commissioning effort during the week.  The WFS1 Gouy phase telescope was further adustedAnother sensing matrix was measured and inverted, yielding similar results to recent work in which it had been observed that one could reliably close loops in pitch but not in yaw.  This problem was later mitigated with an increase in the WFS bandwidth to better balance the gain hierarchy.
  • There was some nice progress with the 2k bull's eye detector.  Installed on ISCT10 POY, the detector was commissioned and then used in common and differential TCS servos.

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


From: Brian O'Reilly

Report Highlights:

Tropical storm Arlene and nearby logging have interfered with IFO operation for the past few days. We are back in low noise mode after last weeks PSL work, but at a lower power than previously.

Many of the reports include items from last week. Due to network problems at LLO they did not make it into last weeks report. The current status is that we should not have any more protracted outages, but the root problem remains to be fixed (see Shannon's report on General Computing for details).

Most of our remaining SURF students arrived this week and are getting dug in on their projects.

This week we welcomed Lisa Bogue is the latest addition to the LLO staff. Lisa is our new CDS SysAdmin.

IFO Commissioning (Valera Frolov)

The storm and unusually high seismic activity from earthquakes and nearby logging lead to low interferometer commissioning duty cycle this week. Nevertheless the interferometer low noise operation was recovered with 3.5 W into the Mode Cleaner, and the work is in progress to increase the power into the interferometer to the maximum available level of 8 W.

The noise coupling from the input beam jitter and the laser frequency were measured.  It was found that these sources do not limit us at the current sensitivity. The frequency noise is a factor of three below the noise floor in amplitude and the beam jitter is two to three orders of magnitude off.

The Mode Cleaner length was measured relative to the main sideband modulation frequency and it was found that the Mode Cleaner is about 300 microns longer than it was during the S4. Changing the MC length back to the S4 value using HEPI HAM2 position offset did not change the noise spectrum. This indicated that the oscillator phase noise is small enough so that the change in coupling did not affect the ifo noise at the current sensitivity level.

Last weeks report:

  • The pre-stabilized laser optical table rework was the main interferometer commissioning activity this week. The power into the Mode Cleaner was increased to 8 W. The phase correcting electro-optic modulator was rotated by 90 degrees to follow the polarization flip on the optical table. The bandwidth of the frequency stabilization servo can now be set to the design value in the range 0.5-1.0 MHz.
  • The dark and shot noises were measured in situ for the recently installed anti-symmetric port photo-detectors and the noise budget was updated. The angle to length couplings for the test mass were re-optimized (this update reflects the last coil driver modification).
  • The measurements of the angular noise and the Michelson and power recycling cavity degrees of freedom noise coupling to the dark port are now automated for the use in the noise budget.
  • The test mass suspension wire violin mode ring downs were measured in various conditions.   Different interferometer power levels, test mass angular offsets, and lock to lock variation were investigated.
  • The measurement of the light scattered by the test mass was performed. The data is being analyzed.

General Computing (Shannon Roddy)

From last week which didn't make it into the weekly:

The fiber connection between LLO and LSU has been having issues for approximately two weeks.  Before then, we would see occasional errors on the receive side of the LLO router.  LSU was not seeing any problems, therefore it had to be on our receive side of the fiber.  The errors would increment at a rate of around 2-6 per day.  There were occasional outages less than one hour (usually around 5-10 minutes) and then the circuit would recover.  On May 30th, the circuit went down and stayed down much of the day.  Bell South was notified and they started working the problem, but the circuit came back before they could diagnose any real problems.  The following Monday, June 6th, the circuit again went down for much of the day.  Both outages followed strong rain in Livingston the day before.  Then on June 9th, the circuit went down at 2:39 AM and stayed down.  This outage also followed a thunderstorm in Livingston.  Bell was showing an alarm at one of the repeaters in the Baton Rouge Main CO.  They replaced a multimode jumper and the circuit came back.  The correlation with rain is likely a coincidence, but last year it turned out that our outages correlated with the afternoon temperatures and that time was not a coincidence.  The correlation with rain made me suspect the same fiber splice as last year.  The circuit has been up since approximately 4:40 PM yesterday (6/9) and I have not had any problems.  Since then I also have seen zero receive errors on the router.  The problem is hopefully fixed.

Setting up six PCs in the outreach area for John Thacker.

Working on rounding up a Linux workstation for a student arriving  Monday.  Unfortunately I may have to purchase one in town since we are out of spare workstations.

Working on a summary of security issues for Albert for the yearly report.

Working on documentation of the LAN topology.  I will have to work with LHO in the next couple of months to try and get Hanford and Livingston as close as possible.

Working with Solsoft on pricing on a software upgrade.  I need to look at the budget and determine if this is possible.

Discovered a problem in our LAN/WAN configuration due to the Bell  outages.  I will have to implement a piece of equipment between the main router at LLO and the router at LSU.  Unfortunately when the WAN connection goes, so does our LAN routing right now.  The Fujitsu equipment keeps a link signal at our router interface which causes the router to believe that the circuit is still live.  It then tries to store and forward all packets destined for the WAN and then runs out of memory when it cannot forward them.  At that point the router more or less shuts down and stops forwarding for the local nets also.  The real fix would be for Bell's equipment to terminate the link if there is not a round trip path for the packets, but short of that I will have to implement something else.

For this week:

Almost all of this week (and weekend) has been spent working on the Bell connection to LSU.  Several things have been done since last week.  Saturday I was on the phone with Bell most of the day and we "rolled fiber" (their terminology) to an unused pair between the Goodwood C.O. and the Denham C.O.  This didn't fix the problem.  Sunday we "rolled fiber" between Denham and Livingston, again did not fix the problem.  Monday we decided to replace the interface cards in the Goodwood C.O., however they had to be ordered.  They shipped out Tuesday and were received on Wednesday.  They were replaced at ~4:30 AM central time.  At this point all I can do is watch my logs and see if this fixed the problem.  Also, Boyd (LSU) and I worked a couple of solutions on his router.  The long term outages should not really happen again unless we have lost light totally on the circuit.  After 5 minutes, Boyd's router should auto-recover from the errors that we have been seeing on the circuit.  If we continue to have issues on the circuit, they are now talking about flying a guy in from Atlanta with some test equipment.  This would mean some significant downtime that would have to be scheduled.

Set up a student PC running Fedora Core 3.  Core 4 would not install - might have been my CD image.

Rearranging the computer lab.  Installed a UPS for my testing servers.  Had to move a lot of items around in the lab to accommodate the rack, etc.  Ordered a PC for a student since we are basically out of spare machines.

Working on setting up the router for remote IDS when I can squeeze in time.

Installed tcpdump, etc. on it.

Looked into an issue where it seemed the interfaces on the router were coming up at the wrong speeds.  It appeared that they were coming up at 100 mbit, but they were in actuality coming up at 1000 mbit.  I do think that the Cu interfaces are coming up at the wrong speed though.  The Syskonnect cards are fine though.

LDAS (Igor Yakushin)

Since last week weeklies did not make it in time due to the network outage at LLO, this weeklies include both.

Data storage/Condor/LDAS admin:

1)      The redundant controller in 3510 failed and was replaced. However, there seems to be a mismatch in the firmware versions. The firmware on the secondary controller needs to be upgraded during the next downtime (according to the documentation, upgrading firmware might cause a few minute interruption in the work of 3510).

2)      The batteries in all T3s were replaced.

3)      Right before I left for the LSC meeting, Condor started behaving strangely: it would allow to run jobs only on about 28 nodes although nothing seemed to be wrong with the other nodes. Restarting Condor fixed the problem. Yesterday, I had this problem again. Again restarting Condor (couple times) fixed the problem. It is probably connected to the new configuration files rsynced from CIT or the new version of Condor since I never had such a problem before. I also had to define LIGO_PREEMPT in /usr1/condor/condor_config which was not defined in the new configuration files, as discovered by Ben, and resulted in Condor not running on the nodes.

4)      Cleaned drive 03 in the tape robot that displayed "Need cleaning" message.

5)      The failed drive in the spare T3-13 was replaced however failed to work. It turned out that three more drives had to be replaced although fru stat or fru list did not show any problem with them. Now all the disks in T3-13 seem to be working fine.

6)      Found some strange problem with Condor@LHO: the jobs in the dag that could grab CPUs at the beginning ran successfully but the rest of the jobs were reported as finished though they were never tried. Again, it most likely has something to do with the new configuration files rsynced from CIT.

Data analysis:

1)      Attended LSC meeting and burst face to face meeting in Michigan.

2)      Made a presentation on the status of the S4 untriggered burst search at the LSC plenary session. This presentation is a rehearsal of the one I shall make at Amaldi.

3)      Worked with Michele to process LIGO-VIRGO simulated data with waveburst. Michele presented the preliminary results in face to face meeting in Italy.

4)      Compared waveburst triggers on h(t) and AS_Q and reported the results on this week burst telecon: http://www.ligo-la.caltech.edu/~igor/S4/HT_vs_AS_Q

5)      Working on finalizing Amaldi6 presentation which now turned into a poster since the burst group got only one talk out of four that were originally submitted.

6)      Discovered a problem with V2 calibration used in SG16_S4 MDC: the injection into H2 were twice stronger than they should be (in AS_Q only, DARM_ERR channels most like were not affected, h(t) were not affected).

7)      Ran waveburst on DARM_ERR, V3 calibration, both production and SG16_S4 simulations. Processing the triggers.

8)      Working on processing more simulated Virgo data with waveburst so that the results can be presented at Amaldi6.

9)      Published waveburst S4 production and simulation triggers on h(t) and AS_Q: http://www.ligo-la.caltech.edu/~igor/S4/TRIGGERS

HPLF etc. (Ken Franzen)

1)      HPLF news: The laser has continued to work without problems. Rupal Amin and I have been testing CVI high power thin film plate polarizers at 100 W. So far the data indicates that the degree of polarization is lower than specified. We have also observed stronger thermal lensing than expected, probably originating in some of the Thorlabs 2 inch lenses being used. We are now looking for some better high power optics to test.

2)      The PZT tip/tilt platform systems to be used in the REFL port beam stabilization system was ordered (by Rich Abbott). Unfortunately, a mistake was found in the original quote which needed to be corrected. Have been explanining the requirments of the mirror holder to Luke Williams at the University of Florida, who will base his drawing on a design by Mike Smith at Caltech. Some differences are that we need to add black glass after the mirror in order to reduce back scattering into the vacuum and also use different 2 inch mirror mounts.

Safety and Security (Rich Riesen)

  • Completed IR scan on all 5 laser tables. Found no stray beams.
  • In process of Laser/Site Safety requirements for Lisa Bogue, Bonnie Schneider, and Ian Duke.
  • Found no safety concerns on site safety tour.
  • Working on the LSS message log data.  Data archived will be broken down in monthly folders and made available on the DCC and elsewhere for convenient browsing.
  • Checked all LLO site safety emergency lighting, all units are functional.
  • Audited all LLO site fire extinguishers. All are properly tagged with current inspection requirements.
  • Pulled a GSA pick-up from service due to operational safety risks. Awaiting further instructions.

Ashfaq Khan

1)      Working on a CDS Diagnostics system that includes mapping of the CDS signals and the Path into the IFO. It also has a detailed PSL layout that shows placements of different objects on the PSL table.

2)      Helped Doug in wiring the Photon Calibration and Mic Signals.

Sanichiro Yoshida

The local damping mechanism has been added to a 3D suspension e2e box which takes into account the suspension wiresÂf violin-mode motions (3DsusViolin.box). To model the violin mode, the 3DsusViolin box uses a state-space representation (ABCD matrix) computed by a Maple program developed by Virginio Sannibale. This state-space representation is based on a violin mode model also developed by Virginio. This model describes the violin motion by dividing the wires into several segments; each simulates the dynamics by a pair of massless string and bob. The current version of 3DsusViolin.box uses five segments to describe the wiresÂf dynamics (i.e., up to the fifth order of the violin mode) and an additional bob to describe the suspended optic for the pendular, pitch and yaw degrees of freedom. With a white noise input to the suspension point and without local damping actuation forces, this box has demonstrated two violin-mode peaks at 290 Hz and 430 Hz, each is split into two fine structures due to the length difference between the two wires.

The box file with the local damping mechanism (LOS_violin.box) consists of the same lower level boxes as the LOS.box used in the current SimLIGO, except for the 3Dsus primitive has been replaced by the 3DsusViolin.box. The LOS_violin.box has been tested with a realistic suspension point input and various OSEM gain settings. The resultant power spectrum for each degree of freedom (pendular, pitch and yaw) seems to be reasonable, showing that as the OSEM damping gain is increased the peak at the corresponding natural frequency decreases and instead the peak at the table stack resonance increases. The violin mode peak at 290 Hz is barely seen in the yaw and pitch spectra when the local damping is turned on, whereas none of the violin peaks is visible when the optic is freely hanging. More detailed investigation on the violin peaks is under way.


Initial LIGO Detector Science & Engineering (Coyne)


from Dennis Coyne

CDS

see also the CDS weekly meeting minutes in the commissioning archives

CDS Software

no report

CDS Hardware

Rich Abbott

1)      Finished the LSC frequency distribution panel and shipped it to Peter Fritschel at LHO.

2)      Changed the frequency distribution system block diagram to reflect some recent changes.  A sampling splitter was added to the ASC path, and a request was received to add a jumper path for the RF signal to be available on the front panel of the box.  This jumper path is intended to allow simple switchover to an external frequency source for testing.

3)      Pushed through the purchasing order for the Piezosystem Jena Refl beam steering servo.  Also put in a change request for a noted change.

4)      Working with Dave Grimmett to get the ISS boards tested and shipped to LLO.  Possible that the first system will ship 17 June 2005 to LLO.

5)      Gathered changes together to write a DCN on the LSC whitening filter.  Still more work here before I am confident that I know what all the changes are.

Ben Abbott

LSC RFPD: The long-lead capacitors have arrived today, which means that a total of 8 RFPDs will be ready for test by the end of this week.  Three of these will be 29.5MHz, and 5 will be 24.5MHz.

DMT

no report

PSL

PeterKing

The beam propagation of an NPRO laser was measured a number of times.  Whilst the beam profile had a nice gaussian profile and the propagation fitted nicely, there was a large discrepancy between the measured spot sizes and locations with the calculated values.  That made mode matching a trifle tedious.  At present most of the ellipticity in the beam has been removed by 3 cylindrical lenses.  Work will now continue with the rest of the optical train..


40 Meter Interferometer (Weinstein)


IFO commissioning:

SURF students Marcus Ng and Ryan Kinney have arrived. Marcus will work with Rob and Mike Smith on the design and construction of the output mode cleaner for the DC detection experiment; he's busy reading Siegman. Ryan will work with Rana and everyone on measuring noise spectra and the noise budget, in various IFO configurations; he's busy reading Rana's thesis.

IFO commissioning:

  • Rana continues to develop more IFO autoalign and IFO locking scripts. Steve exercises the IFO lock scripts every morning.
  • Rana worked with Ben to get the periscope PZT mirror (steering the input beam from the PSL into the mode cleaner) to be controlled by the IO front end, with EPICS control of the DC offset. This is in preparation for using the PZT mirror in an autoalignment procedure. He found that the PZT HV driver board was broken; Ben promptly fixed it. After these changes, Dan adjusted the mirror steering to maximize MC transmission.
  • Also for autoalignment: The in-vac PZT mirrors that steer the beam from the MC to the main IFO used to be controlled thru EPICS slider. Rana and Ben wired them up so that they can be controlled through the ASC FE code. We need Rolf and/or Alex to make the necessary mods to the ASC code.
  • Steve adjusted the IPANG QPD to center the beam, which is a bit high and is clipping on a steering mirror in the ETMX chamber.
  • An earthquake knocked all the suspensions out on Sunday, and an even closer one today (Thursday) did it again. After today's earthquake, the PRM was stuck in a strange place. It was severely pitched and moving the pitch bias didn't seem to do much of anything. Rob pushed the slider back and forth, and apparently knocked it loose, and after calming down it appears fine. A few ideas of what it was stuck to, but nothing conclusive. The SRM has shifted alignment a great deal, but appears to be damping fine. The mode cleaner needed some serious re-aligning to acquire lock. Maybe it was shifted back into the old position, antecedent to last week's earthquake. The earthquake knocked loose the PENTEK OUTPUT cable responsible for driving the ITM positions. We are in the process of recovering.

IFO modeling and DC detection development:

  • Monica is measuring OLTFs in her e2e model to calibrate error signals. She closed a control loop and locked one arm. She talked with Matt about his e2e simulation and will try to incorporate his model into hers.
  • Rob's Finesse simulations of DARM and CARM with CARM offset seem to accurately illustrate the problem we are having with reducing the CARM offset. Measurements of the OLTFs in the real IFO show RSE peaks in the same place as in the simulations. This means, amongst other things, that our signal mirror is correctly positioned, to within 1/2 mm.
  • Matt reproduced the lock acquisition sequence that we have been using for locking the full interferometer, in e2e simulation. He saw the same effect that we see in the real IFO and that Rob has been exploring with Finesse: as we reduce the CARM signal offset to bring the arms into resonance, the coupled cavity pole crosses the CARM UGF and lock is lost. He implemented a filter combination which allows a zero to move in tandem with the cavity pole, as arm resonance is approached, and was able to hold lock all the way to the full resonant configuration. Rob will work on implementing such a filter combination.

Electronics, controls:

  • At Rana's request, Ben bypassed the 850 Hz AA filters for some 16kHz channels in the IO data acquisition system.
  • Rana and Dan found a flaky RF cable from the POY RFPD. Bob and Steve replaced it and checked it out.
  • Rana and Dan found significant couplings between electronics modules in the LSC rack, specifically, couplings from the QPD signals to the RFPD signals. After some searching, Jay and Ben traced this to the +15V power supplies going into current limit mode. The current limit of the +15V supply was set at a level where the extra current drawn when the QPDs were powered up dropped the supply voltage. This supply voltage drop then caused the LSC whitening boards to act up. They adjusted the supplies to stop the behavior, but we really need to redo the power for the LSC rack, and supply each crate with a separate set of supplies. Rana suggests that we go back to switching power supplies, just keep them in separate racks (ideally, in a separate room).
  • Alex came in to upgrade the suspension controllers' code, but discovered that sus1 and sus2 controllers are too slow when running the latest code. He is working on this problem on the test stand. Code changes to the suspension controllers were undone, and we are in the process of recovering.
  • Rob made some improvements to the LSC control screens.
  • Dan has compiled a summary of the spectra he took of all 10 optics swinging freely. He has good measurements of the POS, PIT, YAW, SIDE and bounce frequencies for each optic, and has compiled a (dismayingly long) list of anomalies.
  • Dan ran Vuk's coil balancing script on several optics, to decouple POS->ANG and ANG->POS at frequencies above and below the pendulum peak. This work continues.
  • Steve measures >= 0.3 Mohm to ground on all the ISCT tables, except for the PSL and SP tables, which are dead shorts.
  • Bob has been checking for unlabeled or poorly-grounded or un-terminated cables around the lab, every morning.
  • Bob has ordered spare RF filters; they will be here in two weeks.

Lab Infrastructure:

  • Steve found mice droppings on top of a rack. He sealed up utility holes into IFO hall, to stop the mice from getting in.
  • We need to get gutters installed on the 40m building. Steve awaits a response from PMA.
  • Steve gave Marcus and Ryan the lab safety tour and training. Both students are reading the safety and SOP documents.

Thermal Noise Interferometer (Black)


Since our last report, we have locked the bond-noise shadow sensor under vacuum and done a preliminary noise breakdown. We are now working on reducing the noise and calibrating the instrument.

Helena contacted Jean-Marie, and he says our mirrors are ready to ship.

Jay has framebuilder boxes up and running with a signal generator serving as the clock. This is not an ideal situation, since the signal generator and the cpu clock will get out of phase after a few days of operation, but it is adequate to take preliminary data.

SURF students Chinyere Nwabugwu, Royal Reinecke, Richard Kirian, and Kate Dooley have arrived and have started on their projects.

Congratulations go out to former TNI group member Michael Zhang, a Caltech undergraduate who worked with us during the summer of 2003.

Michael was recently awarded Caltech's upper-class merit award, a prestigious, merit-based scholarship awarded to Caltech's most academically-talented students. Michael made invaluable contributions to the TNI during his time with us, and he shared authorship on one of our published papers. We congratulate Michael on his merit award, and we wish him the best in all his future endeavors.


LASTI (Ottaway)


From: Richard Mittleman

Laurent reports:

I have designed the modal control + estimator loop for the 6 dofs for the MC suspension. The results in damping and noise transmission are very good and I manage to have a lot less sensor noise transmission than a classic feedback method.

I have tested the loop on the triple pendulum in LASTI

I have tested the damping to an impulse response and checked the 10s settling time and it works perfectly

I have tested the noise transmission by injecting a very large amount of artificial noise in the sensors and measured the amplitude of the top mass motion...This is not a perfect test since I don't have inertial sensors and after 10 Hz I am mostly measuring frame's motion, also I don't have enough sensor on mass3 to check the bottom mass. But the results on what I have match the simulation perfectly, which is a good sign.

I am writing a 4th document about his, hopefully I will be able to merge the 4 documents into one after that.

LASTI Infrastructure:

Had a visit from clean room systems representative will quote us the class 100 clean room that the suspension and seismic isolation system will assembled under.

LASTI Suspensions:

We have received the parts for the solid space that will fit between the HEPI and the Quad controls prototype. We will have this cleaned and air baked soon for preparation and installation into LASTI for a fit check in August.


Data Analysis and Computing (Lazzarini)


Simulation and Modeling (Bhawal)

Weekly E2E Physics Meeting

Matt Evans talked about the lock acquisition of advanced LIGO and 40m.  Viewgraphs available at http://www.ligo.caltech.edu/~e2e/ME2ET/Minutes05/050616/

Commissioning Support

(Biplab) Collected all parameters for optical paths for all Wave Front Sensors (WFS) of Hanford 2Km interferometer and analyzed. Conclusions:

(i) Gouy phase for WFS1 path is as bad as it was (=187deg) in both 4Km interferometers. This needs to be corrected to 90 deg.(elog June9)

(ii) WFS 2,3,4 are not at the theoretical optimal position but those are not at a very bad location either. By tweaking the optical path the signal can be improved but we cannot expect more than a factor of 2 improvement.

Locking in Advanced LIGO simulation

(Matt) Worked on length control and locking in Advanced LIGO simulation.  Following the approach used at the 40m lab, I was able to lock at the operating point.  The final step, moving CARM (Common Arm mode) to the operating point, has not yet been performed at the 40m, but this simulation suggests that their approach is viable. Much work is yet to be done, but these results also suggest that this approach will work for locking Advanced LIGO.

Simulation of 40 meter interferometer

(Monica) Calibration of Error signals has been done by calculating open loop transfer function of the complete model of 40m interferometer.  First attempts of locking one arm with the 40m e2e package: all the optics are blocked except the X-arm input mirror(ITMX) and end mirror (ETMX) which are shaken with white noise.

New binary release

(Hiro) I installed the version of modeler needed to run Matt's advLIGO package in:  /home/e2e/LinuxHOME and /home/e2e/Linux64HOME.  The first LinuxHOME is the version for 32bit machine, and Linux64HOME is the one for AMD 64 bit machine. It runs on computer homam, but does not run on deneb, the Intel 64 bit machine.  Source codes are still under modification and there is no tar ball release yet.

Alfi

(Bruce) - Implemented renamable parameter links (box settings.)

  • Implementing new caching container list for the main object,ALFINode in order to improve overall Alfi performance.   (These lists are queried a tremendous amount, so caching such returned results between non-list-changing method calls will improve performance.

(Melody)  Continuing with fixing the Problem Reports (PR).  Fixed PR 473: Duplicated item placed outside of view.  Currently working on PR 482.

Data Analysis Activities (Lazzarini)

Charlton:

Working with Xavi I found a bug in my Matlab script for comparing band-limited RMS in AS_Q vs. h(t) which occasionally caused it to use data where calibration lines had dropped out for a short time. This appears to have caused ~10 of the large differences I observed (out of around 180 over the whole science run). After fixing this bug I re-ran the comparison using the H1 DARM_ERR instead of AS_Q as I had done previously. There are still a few large differences (34 at 5*sigma and 219 at 3*sigma) but overall the match seems to be better. The standard deviation of 0.45% is an improvement on the 0.80% I observed when using AS_Q.

This is not yet a final result since Xavi's h(t) used the V1 calibration information and I'm using V3. I'll need to re-run the comparison when Xavi has created the next version of h(t).

Mendell:

As reported at the June 2005 LSC meeting, the StackSlide pipeline for producing upper limits is complete (the previous pipeline produced estimated upper limits that appear good to within 10%.)  Tests by Joe Betzwieser and myself using software and hardware injections shows that the core StackSlide function, updated last January by Virginia Re, is working. Some tweaking of the code is still needed, but it is ready to

  • Lots of time spent reviewing pulsar analyses.
  • Improved conlog code to check for clean locked segments in AstroWatch data.
  • Working with SURF student Sarah Caudill.

Weinstein:

  • Heaps and heaps of reviews of Inspiral talks and posters for Amaldi.

Yakushin:

Data analysis:

1)      Attended LSC meeting and burst face to face meeting in Michigan.

2)      Made a presentation on the status of the S4 untriggered burst search at the LSC plenary session. This presentation is a rehearsal of the one I shall make at Amaldi.

3)      Worked with Michele to process LIGO-VIRGO simulated data with waveburst. Michele presented the preliminary results in face to face meeting in Italy.

4)      Compared waveburst triggers on h(t) and AS_Q and reported the results on this week burst telecon:  http://www.ligo-la.caltech.edu/~igor/S4/HT_vs_AS_Q

5)      Working on finalizing Amaldi6 presentation which now turned into a poster since the burst group got only one talk out of four that were originally submitted.

6)      Discovered a problem with V2 calibration used in SG16_S4 MDC: the injection into H2 were twice stronger than they should be (in AS_Q only, DARM_ERR channels most like were not affected, h(t) were not affected).

7)      Ran waveburst on DARM_ERR, V3 calibration, both production and SG16_S4 simulations. Processing the triggers.

8)      Working on processing more simulated Virgo data with waveburst so that the results can be presented at Amaldi6.

9)      Published waveburst S4 production and simulation triggers on h(t) and AS_Q: http://www.ligo-la.caltech.edu/~igor/S4/TRIGGERS

Zanolin(MIT):

Prepared material for a f2f in Italy (June 9 and 10) with the LIGO Virgo collaboration.  Submitted a revised version of the GWDAW9 proceedings on the presentation on comparison of burst search methods applied on simulated single interferometer ligo and virgo data.  Started the validation process for the matlab stand-alone parameter estimtion.

LIGO Data Analysis System

Software Systems (Blackburn)

LDAS:

The last error message for IFrameStream::ReadFrameH is now available on the LDAS Development System through the C++ method GetLastError(). This resolves Problem Report 2019.

Work has started on resolving the difference between the FrProcData elements and the underlying FrVect data elements that has been identified through the Problem Report 2823.

About 70% of the C++ code has been written for supporting the "-no-metadata- check" option for the createRDS command associated with Problem Report 2824.

Added seed.cmd to the controlMonitorAPI's client to initialize the database after a database reconstruction. This is needed to successfully run the conditionData tests.

Fixed a test wrapping problem when viewing the resourse files. The display of multiple lines in the resourse file is now fixed as well.

Continued to fix "interim results" problem report associated with the running of all the user commands from a single submit in the controlMonitorAPI's client. This required minor changes to the nodeUsers test and the createLDASdb script.

Ran system and integration tests on LDAS 1.6.10. Identified a problem with the dataConditionAPI leaking objects. Conducting similar tests on the LDAS 1.6.0 release on LDAS-Test to see if the problem was introduced by recent changes to the frameAPI's C++ code.

Set up the LDAS-DEV system to simulate a beowulf cluster using a single host. The remaining nodes from the LDAS-DEV system will be donated to the OSG integration testbed project.

The sinstall rules for rsync (PR2826) and wget (PR2841) have been modified for the latest versions of these packages.

TCLGLOBUS:

The issue of memory corruption observed in the Globus Toolkit 4.0 port is now understood. By default, the Globus Toolkit 4.0 port was compiled multi- threaded which revealed a short coming in the Tcl API of accurately detecting which thread was running the code. As a result, any asynchronous callback was causing the Tcl interpreter to execute in two concurrent threads causing the code instability. The pthread library is now being used. The code still needs one more modification to support Tcl's Thread extension package.  This will be done early next week.

Modified XIO asynchronous client/server test case to be able to run with threaded or non-threaded Globus

Fixed SWIG-wrapped asynchronous vector write function. The fix is to check if size of vector buffer is 0, then set the pointer to NULL to prevent any memory corruption.

Worked with Ed fixing segmentation fault caused by asynchronous test cases.  Have been running the test cases with the fix since last night and there is no segmentation fault and no memory leaks.

Continue to improve the Tcl/Globus website and documentation associated with the next release of the code.

OSG/GRID COMPUTING:

In preparation for initial installation of OSG 0.2.1 release of middleware on a LIGO test bed (osg-itb.ligo.caltech.edu) performed the following:

Obtained a system cert for the future head node of the test bed cluster Tested my personal cert on an LSC site and verified the cert is good.  Reviewed the installation requirements and instructions for installing  a computing element node as specified in the OSG CE Install Guide.  Met with the hardware POC (Stuart Anderson) and coordinated planning  on preparing the hardware for the OSG test bed.  Began reviewing documents on the GUMS privilege system and its interaction  with other security components.  Reviewed OSG overview docs including "A Blueprint for the Open Science Grid".  Investigated resource allocation for shared volumes on the osg-itb test bed.  Reached  agreement  on the external name for the test bed: "LIGO-CIT-ITB".  Began monitoring several OSG mailing lists relevant to tasking.

At this week's LSC Comp Comm meeting discussed plans for setting up a LIGO- wide VOMS server to securely distribute authentication information onto the OSG grid. This will be established at PSU. PSU will also be filling out the registration form for this VOMS service with the OSG. This is expected to be completed by the end of next week.

Also discussed plans for migrating the LSC's GRID3 clusters at UWM and PSU onto the OSG. UWM expects to have this done within a few days to a week.  PSU will follow shortly after they have established the VOMS service.

Hardware Systems (Anderson)

Caltech

(Dan Kozak)

  • Continued CIT /archive cleanup work.
  • Continued prep work for LHO /archive upgrade (making sure all files were on tape).
  • Worked with Ben on LHO upgrades:

-         combined /archive & /cluster.

-         converted /frames & /archive to use md devices.

-         SAM-QFS/FS 4.3.11 on gateway/dataserver/fb1.

-         SAN-4.4.2 on gateway/dataserver/fb1.

-         built new RAID sets in 3511s, now 10+1+1 hot spare.

(Phil Ehrens)

  • Cloned hard drive for ldas-test node3 from ldas-test node2   using usb-ide cable. Cloned drive failed SMART at startup   in spite of having been badblock tested for 30 hours.
  • Finished configuring hardware on m27 - this is now a 2GHz P4 box with 1 Gb of ram. The cd-rom drive fails to read a cdrom that is used to boot the box (weird), so I will have to load the hard drive via usb-ide cable from another box.
  • More tweaking of the LIGO desktop Linux box config script.  The script now successfully configures the box to the point where yum can be called against the local mirror, with only the NFS+ startup requiring handwork. Began incorporating the condor config requirements into the script.
  • Installed Fedora Core 4 on my desktop machine, Tarazed, after a kernel bug trashed my filesystem.
  • Installed Solaris 10 on ldas-suntest2 and configured it for a local build of LDAS and for use as part of a tandem system with ldasbox2.
  • Created a rescue disk with memtest86+ versions 1.27 and 1.55.5 and with the LIGO desktop config script and other installation notes.
  • Various cluster maintenance tasks, including replacing one IDE raid drive and rebooting nodes 289 and 236.

(Stuart Anderson)

  • Supported the upgrade/reconfiguration of SAM-QFS at LHO.
  • Moved a few more clusters users to SAM-QFS for their cluster home directories.
  • Installed and started testing FC4 on a laptop.
  • Upgraded Condor to version 6.7.8, but this unfortunately only fixed half of the problem associated with users opening files at a high rate.

MIT

(Keith Bayer)

  • Investigated NFS trouble on pcraid7 increased number of nfs threads from default (to 50).
  • Rebooted node62 due to memory exhaustion seemed to fix the problem.
  • Received 2 new power supplies and installed into pcraid3.
  • Added cluster accounts for LSC people.

Livingston

(Igor Yakushin)

  • The redundant controller in 3510 failed and was replaced. However, there seems to be a mismatch in the firmware versions. The firmware on the secondary controller needs to be upgraded during the next downtime (according to the documentation, upgrading firmware might cause a few minute interruption in the work of 3510).
  • The batteries in all T3s were replaced.
  • Right before I left for the LSC meeting, Condor started behaving strangely: it would allow to run jobs only on about 28 nodes although nothing seemed to be wrong with the other nodes. Restarting Condor fixed the problem. Yesterday, I had this problem again. Again restarting Condor (couple times) fixed the problem. It is probably connected to the new configuration files rsynced from CIT or the new version of Condor since I never had such a problem before. I also had to define LIGO_PREEMPT in /usr1/condor/condor_config which was not defined in the new configuration files, as discovered by Ben, and resulted in Condor not running on the nodes.
  • Cleaned drive 03 in the tape robot that displayed "Need cleaning" message.
  • The failed drive in the spare T3-13 was replaced however failed to work. It turned out that three more drives had to be replaced although fru stat or fru list did not show any problem with them. Now all the disks in T3-13 seem to be working fine.
  • Found some strange problem with Condor@LHO: the jobs in the dag that could grab CPUs at the beginning, ran successfully but the rest of the jobs were reported as finished though they were never tried. Again, it most likely has something to do with the new configuration files rsynced from CIT.

Hanford

(Greg Mendell)

As announced to DASWG on May 31, 2005, the createrds.tcl and createrdsGUI.tcl have been updated to improve RDS generation in several ways and to take advantage of new options in the 1.6.0 release of  LDAS.  Improvements, tests, and new options are:

1)      Alignment of output frames with the output frame length.

2)      Improved the algorithm for determining that a gap exists.

3)      Tests of generating L3 data directly from raw data; this works.

4)      Tests of LDAS v1.6.0 options to merge of input frames of different types; this works though some changes to merge GEO frames with LIGO frames are still needed.

5)      Astro-watch RDS generation started using the new code on May 27, 2005.  The code was tested extensively over a 4 day period as well, and all tests passed.

6)      The scripts are available here: CVS/Root = :pserver:anonymous@gravity.phys.uwm.edu:2402/usr/local/cvs/lscsoft, CVS/Repository = dsorun/contrib/createrds/scripts

General Computing (C. Barker for Wallace)

MIT:

(Keith)

  • Attempting to roll out first FC3 box on gc using LDAP backend.
  • Adjusted emvogil-3 SSH daemon. Put tcp wrappers on box and used old emvogil-3 ssh keys.  Fixed path to sftp daemon.
  • Updated nw22 print spool (lprng) to default to duplex.
  • Spec'd out dell desktop GX280 for scientist.

Livingston:

(Shannon)

From last week which didn't make it into the weekly:

  • The fiber connection between LLO and LSU has been having issues for approximately two weeks.  Before then, we would see occasional errors on the receive side of the LLO router.  LSU was not seeing any problems, therefore it had to be on our receive side of the fiber.  The errors would increment at a rate of around 2-6 per day.  There was occasional outages less than one hour (usually around 5-10 minutes) and then the circuit would recover.  On May 30th, the circuit went down and stayed down much of the day.  Bell South was notified and they started working the problem, but the circuit came back before they could diagnose any real problems.  The following Monday, June 6th, the circuit again went down for much of the day.  Both outages followed strong rain in Livingston the day before.  Then on June 9th, the circuit went down at 2:39 AM and stayed down.  This outage also followed a thunderstorm in Livingston.  Bell was showing an alarm at one of the repeaters in the Baton Rouge Main CO.  They replaced a multimode jumper and the circuit came back.  The correlation with rain is likely a coincidence, but last year it turned out that our outages correlated with the afternoon temperatures and that time was not a coincidence.  The correlation with rain made me suspect the same fiber splice as last year.  The circuit has been up since approximately 4:40 PM yesterday (6/9) and I have not had any problems.  Since then I also have seen zero receive errors on the router.  The problem is hopefully fixed.
  • Setting up six PCs in the outreach area for John Thacker.
  • Working on rounding up a Linux workstation for a student arriving Monday.  Unfortunately I may have to purchase one in town since we are out of spare workstations.
  • Working on a summary of security issues for Albert for the yearly report.
  • Working on documentation of the LAN topology.  I will have to work with LHO in the next couple of months to try and get Hanford and Livingston as close as possible.
  • Working with Solsoft on pricing on a software upgrade.  I need to lookat the budget and determine if this is possible.
  • Discovered a problem in our LAN/WAN configuration due to the Bell outages.  I will have to implement a piece of equipment between the main router at LLO and the router at LSU.  Unfortunately when the WAN connection goes, so does our LAN routing right now.  The Fujitsu equipment keeps a link signal at our router interface which causes the router to believe that the circuit is still live.  It then tries to store and forward all packets destined for the WAN and then runs out of memory when it cannot forward them.  At that point the router more or less shuts down and stops forwarding for the local nets also.  The real fix would be for Bell's equipment to terminate the link if there is not a round trip path for the packets, but short of that I will have to implement something else.

From this week:

  • Almost all of this week (and weekend) has been spent working on the Bell connection to LSU.  Several things have been done since last week.  Saturday I was on the phone with Bell most of the day and we "rolled fiber" (their terminology) to an unused pair between the Goodwood C.O. and the Denham C.O.  This didn't fix the problem.  Sunday we "rolled fiber" between Denham and Livingston, again did not fix the problem.

Monday we decided to replace the interface cards in the Goodwood C.O., however they had to be ordered.  They shipped out Tuesday and were received on Wednesday.  They were replaced at ~4:30 AM central time.  At this point all I can do is watch my logs and see if this fixed the problem.  Also, Boyd (LSU) and I worked a couple of solutions on his router.  The long term outages should not really happen again unless we have lost light totally on the circuit.  After 5 minutes, Boyd's router should auto-recover from the errors that we have been seeing on the circuit.  If we continue to have issues on the circuit, they are now talking about flying a guy in from Atlanta with some test equipment.

This would mean some significant downtime that would have to be scheduled.

  • Set up a student PC running Fedora Core 3.  Core 4 would not install - might have been my CD image.
  • Rearranging the computer lab.  Installed a UPS for my testing servers.   Had to move a lot of items around in the lab to accommodate the rack, etc.
  • Ordered a PC for a student since we are basically out of spare machines.
  • Working on setting up the router for remote IDS when I can squeeze in time.  Installed tcpdump, etc. on it.
  • Looked into an issue where it seemed the interfaces on the router were coming up at the wrong speeds.  It appeared that they were coming up at 100 mbit, but they were in actuality coming up at 1000 mbit.  I do think that the Cu interfaces are coming up at the wrong speed though.  The Syskonnect cards are fine though.

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
  • Cleaned up the OSB Computer User's Room and set up a color printer and two more computers for the SURFs.
  • Helped the SURFs get settled in with computer accounts, software and getting their personal laptops on the network.
  • Continuing to set up a dual boot laptop for the loaner pool.  I purchased the latest version of Partition Magic which I am finding is not as user friendly as the older version.  I've spent more time trying to get the disk partitioned than installing the OSs.  I still can't get Fedora Core 3 to install properly.
  • Put a new hard drive in a desktop and am installing the software on it.  This computer will be used in the Staging building for the visiting teachers.
  • Continued talks with PocketiNet about network options.  We are not in line of sight with their Richland tower, but can see their Red Mountain tower which is about 11 miles southwest from our site.  The connection cost to the Red Mountain tower is more than double the cost to connect to the Richland tower.  PocketiNet's lowest bandwidth is DS3 which they can scale back to 10 Mb if necessary.  They do offer free IP addresses and free BGP set up for multi homing.
  • Other misc. user support.

CIT:

(Mike)

  • SolidWorks 2005: Upgraded all SolidWorks users from the office version to professional; this includes the PDMWorks plug-in. I also did two complete reinstalls of SolidWorks, one on a visitor's workstation, plus one for General Computing.
  • Loaded two more pc workstations for surf students. These were very time consuming due to hardware issues. I was able to use parts from other non-working pc's.
  • Millikan: Cleaned up two pc workstations that were turned on from Wilson House. These are now assigned to Millikan. Larry and I are loading these boxes as LINUX workstations.
  • Surf Students: Setup users on their assigned computers with local ligo accounts, and printing services. One workstation had to be replaced due to a motherboard failure. I have swapped this out with another computer.
  • Much misc. other user support.

(Veronica)

  • LIGO:  Website updates.  Roster database updates.  Mirror website at MIT:  Installed the files, need the input from MIT on the content.  Advanced LIGO:  Installed password protection to the website, posted updates on the subsystem progress.  Working on the website makeover.  Working with Carol and Dwight on the redesign of the cost information database.
  • LSC:  Installed the website for the August LSC meeting.  Working on the online credit card registration form.  Posted the presentations from the June meeting.  Updates to the 'papers under LSC review' pages.
  • CaJAGWR:  User support.
  • Project Science:  Transfer of the website to the ITS and continued web support.

(Larry)

  • Worked more procurement issues. Presently, getting quotes for a new modem pool  system and a new notebook system.
  • Rebuilt the E2E machines again. This time the disks were replaced with new ones  and things are working a lot better.  It took 14 hours to transfer 43GB of data from one of the corrupted disks to a new disk. The corrupted disk won't format because of the h/w errors on the disk.
  • Spent time setting up things for the SURF students. So far about half have their accounts activated.
  • Worked a couple of documentation items. One still has a little more work to be completed.
  • Still working with the air-conditioning people. Most of that should be wrapped up in a couple of weeks and then the focus will be seeing how much more air we will need for the computer room.
  • Assisted a couple of people in getting things ready for their travel and presentations.
  • Assisted Mike in trouble shooting a couple of PC systems.
  • Cleaned out a few more accounts.
  • Started working with Mike on testing out the Linux installation. There are a few modifications needed in the procedure in order to accommodate the GC group. As for setting up a GC-LDAS box the procedure is pretty complete. The script that Phil E. wrote has been a big help in automating the setup.
  • The usual user support, file restoration and printer fixes took place.

Advanced LIGO and Supporting R&D (Shoemaker)


Systems

from Dennis Coyne

See also:

AL Systems web page

AL Systems email archives

Records Of Decision or Agreement (RODA)

See also the RODA status web page

  • nothing new

Requirements

  • nothing new

Interface Issues

See the "Interfaces" section of the AL Systems web page

  • Issued three Interface Control Documents (ICDs) as drafts for comment in support of the SUS Preliminary Design Review (PDR) #1:
    E050159-00, ICD: SEI and SUS/UK
    E050160-00, ICD: SUS/UK and SUS/US
    E050168-00, ICD: SUS/UK and COC
    E050169-00, ICD: SUS/UK and AOS
    All are available here

Vacuum Compatibility

Residual Gas Assay (RGA)

See also the Vacuum Bake Lab

Bob Taylor

  • I have shipped out the turbo pumps for repair and am awaiting a estimate of cost for repair.
  • I have re cleaned and reassembled the plumbing for oven C , while it is down for turbo pump repair.
  • I have received word that the new large air bake oven will be shipped on Friday the 17th or Monday the 20th.
  • I have been working on the OSEMs for the Quad suspension. I have four more to build.
  • Lee has picked up the Emitter and Photo Diodes from the bake lab for testing in the optical contamination cavity.
  • Helena and I have finished writing the air bake oven Qualification Document and submitted it to Dennis for his approval..

 

High-Irradiance, Contamination-Exposure Cavities

Lee Cardenas, Liyuan Zhang

Glenair Micro-D connector test report is pending; passed vacuum compatibility testing.

 

Cavity

(Location)

Material/Item

Start

End

Comments

Cavity #1

(OTF Lab, Bridge)

Will place Nd-B-Fe magnets into the cavity next (stepper motor removed & back into the queue)

~6/8

TBD

Replaced 70ppm REO mirrors. Power is at 155 mW.

 

Last week: Power dropped from 175 mW to ~25 mW after introducing the stepper motor sample, and continues to decrease.  It is very hard to keep cavity locked. The stepper motor may have contaminated the mirrors.

Cavity #2

(OTF Lab, Lauritsen)

NA

NA

NA

no change

cavity is close to being ready for samples

Cavity #3

(OTF Lab, Lauritsen)

OSEM emitter & photodiode (Dennis Coyne, SUS )

~6/10

~9/10

Started the optical contamination cavity testing on these OSEM devices. So far the ring down & absorption measurements indicate that there is no change.

Queue

Priority 1

OSEM Flexi-circuit cable, qty ~ 45

(Helena Armandula, SUS)

To be supplied by Univ. Birmingham (Stuart Aston); Helena is finding out date

TBD

TBD

DuPont Flexible Circuits. The whole part to be constructed of 'flexi' i.e. no 'rigid' sections. (Stuart Aston,Univ. of Birmingham, SUS/UK subsystem)
- Start Laminate: Kapton (LF8515)

(document link: http://www.dupont.com/fcm/products/H-73244.pdf

Coverlay (x2): Kapton (LF0110)

(document link: http://www.dupont.com/fcm/products/H-73245.pdf

DuPont Pyralux Series - Kapton / Acrylic Adhesive system.

(document link: http://www.dupont.com/fcm/products/H-73246.pdf

 

Queue

Priority 2

MMG nickel plated Nd-B-Fe magnets (Helena Armandula, SUS )

TBD

TBD

Vacuum bake & RGA completed

Comprised of sintered 24% Neodimium, 75% Iron, 1% Boron by weight and may contain traces of fully alloyed Cobalt and Dysprosium. The coating is electroless nickel, 10-15 um thick on top of a thin copper layer.

Queue

Priority 3

Stepper Motor (Riccardo DeSalvo, possible SUS or ISC use)

TBD

TBD

 

 

Large Item Cleaning Plan/Facility

Ken Mailand

  • With Helena and Bob, completed the 40m large air bake oven qualification spec. The oven is in testing and will ship Friday or Monday via ABF trucking, which is about a 5 day service. The exhaust HEPA filter is on back order and should arrive in two weeks.
  • Adv. LIGO large part oven / cleaning station spec. has been modified, and 2 of three responses received another expected later this week re. rom price and delivery estimates.

Meetings & Reviews

  • Held the first of many SUS Preliminary Design Reviews (PDR). This one was focused on requirement updates for the BSC suspensions (UK scope)
  • Future near term planned meetings & reviews:

 

Date

Subsys

Review

Topic(s)

Enabling event(s)

Schedule motivation

 

June, 05

SEI

BSC Critical Design Review 3

review basic requirements, interfaces & dynamic coupling

available analysis/reports

timely decision on proceeding with SEI/BSC prototype for LASTI for integration with the SUS quad prototype

 

~Aug, 05

SEI

HAM Critical Design Review

Recommendations w.r.t. HAM prototype development based on ETF results

Completion of SEI/BSC critical design reviews; LSC review of ASI HAM configuration design

timely decision on proceeding with SEI/HAM prototype

 

15-Jun

SUS

PDR, Review 1

Requirements update

completion of the DRD update
fleshing out interface & generic requirements

ensure RAL effort working on proper baseline

done

Jun 29,30?

ISC/40m

40m DC Readout Review

DC readout experiment

 

 

 

Jul 11-13

SYS

SYS Mtg

CDS Infrastructure & HAM Isolation Requirements

 

 

scheduled

12-Jul

SUS

PDR, Review 2

Electroncis req & design; Focus is on the front end electronics (UK) -- limited Digital controls/electronics (US) review

 

 

scheduled

~Sep

SUS

PDR, Review 3

Quad design

Completion of the quad controls prototype assembly;

timely transfer, to RAL & UB efforts, of lessons learned from the controls prototype

 

~Nov

SUS

PDR, Review 4

Quad Installation

Completion of installation at LASTI

Inform the UK final design & noise prototype design effort ASAP

 

~Dec

SUS

PDR, Review 5

Triple design

Available SUS/US staff

Enable SUS/US final design phase

 

~Feb

SUS

PDR, Review 6

quad controls prototype test results
ribbon process/design

completion of LASTI testing

timely incorporation into final design effort on the noise prototype

 

TBD

SUS

PDR, Review 7

BS, FM/ITM SUS design
RM design
non-cavity SUS

design work completion (has yet to start on FM/ITM, not mature for RM)

 

 

Sep

AOS

Stray Light Control, DRR/CD

 

SYS PDR?

primavera late finish 6/15/05

 

TBD

AOS

Thermal Comp., DRR/CD

 

SYS PDR?

 

 

~Oct

SYS

PDR, Review 1

Engineering & Implementation ('generic') Requirements;
Interfaces
Revised Optical Layout
Optomechanical Layout

completion of generic requirements definition; completion of first draft of ICD; revision to optical layout; establish integrated opto-mechanical equipment layout

timely system level definition enables/helps define subsystem reqmnts & design

 

~Dec

SYS

PDR, Review 2

CDS Infrastructure
Stable Recycling Cavities
Lock Acquisition
Modulation Scheme
Power Induced Instability

Sufficient CDS requirements & concept work (also 7/11-13 mtg)
E2E Modeling for AL
40m Progress on Acq. & Mod.

CDS Infrastructure is key to subsystem electronics req.
Stable cavity is key to IO MMT design

 

~Sep

IO

PDR Review 1

Faraday Isolator

SYS PDR?

 

 

~Jan

IO

PDR Review 2

Electro-Optic Modulator

 

 

 

~Mar

IO

PDR Review 3

Mode Matching Telescope

Determination of whether a stable recycling cavity will become part of the AL baseline

 

 

TBD

COC

PDR

 

SYS PDR?

 

 

 

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

AdL CDS

  • Continued noise measurements of the AD7679 show that the noise is less than 100nV/rtHz for 10Hz<freq<8KHz.The actual noise floor may be lower as we may be limited by the noise of the input amplifiers.

Seismic Isolation

From: Ken Mason kmason@ligo.mit.edu

SEI Structure:

We initiated talks with ASI to contract them to soften the stage 0-1 and stage 1-2 springs. They will be supplying us an estimate of time and costs next week. Our plan is to supply to ASI the approximate blade geometry. They will calculate launch angles, interface shapes, stiffness matrix, and make modifications to the parts. At that point we will take these design changes and update the detailed drawings.

Dwight Carter, Carol Wilkinson, and I have had several meetings redefining the schedule for the prototype and production isolation systems.

The design and fabrication drawings for the test stand used to assemble and test the advanced ligo seismic isolation and quad suspension has been completed. The drawings have been sent out to fabrication shops for quote.

Rich Mittleman and I are putting togther a plan for alignment and leveling of the test stand at LASTI.

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

Agenda for the weekly SEI phonecall Friday, June 10, 2pm Eastern, 1pm Central, 11am Pacific time

Announcements

  • UK SEI PDR next Wednesday.

Electronics (Jay)

  • Jay looked at Ken's wiring diagram (originally from Larry) that shows the wiring scheme.
  • He caught some inconsistencies and potential errors or non-optimal choices.  Jay will change the pin assignments for (at least) the GS-13s; Ken and he will update the diagrams.
  • Accu-glass doesn't have male large-gauge 3-pin D connector in its catalog, only the female, for some reason.  Jay will work out how we handle this, perhaps getting some custom pieces made.
  • At LASTI, we already have the D25 cabling in clean storage.
  • Ken Mason should determine the positions of the clamps, and verify that the cables can be run that way without interference or excessive mechanical constraint of the stage motion.
  • We might want to reconsider changing to the L-4C bayonet connector version, and stay with the multi-pin connector we use now.  Jay will consider this, from the point of view of overall cost and trouble.
  • Displacement sensors:  ADE reproduced the 24 Hz problem, and attributes it to an instability due to the larger probes that we use.  They made some component changes and eliminated the problem.  Jay has send back the rest of the sensors for rework.

BSC amplification (Rich, Ken, Pradeep) (see cool pictures in the lasti ilog )

  • Rich has posted a set of magnitude/phase displacement contour maps of the floor around the base of the pier, when shaking the BSC flange.  The motion seen corresponds to the pier pivoting about its base in the vicinity of 9 - 12 Hz, with probably two directions (and frequencies) seen.
  • Ken is making progress in the Nastran FE model construction, and will soon be able to compare against Rich's models.

LASTI test stand (Ken)

  • At the moment it looks like this will be needed in September.
  • It is necessary to level and make coplanar the mounting bosses (on the support-tube area), and also to level the entire structure on a bumpy floor.  Ken is working on a concept for this.

Modeling for softer springs (Brian L.)

  • Brian is using the up-to-date masses spring stiffnesses, and moments from the ASI DVD.  (an Excel file dated Feb 18, 2005).
  • Brian got Ken Smith's natural frequency design tool going.  This allows varying spring and mass parameters (a GUI-driven program in Matlab).
  • It turns out the the stage 0-1 springs in the ASI design are not that much stiffer than the 1-2 springs.  Brian is exploring variation in the 0-1 spring length (between 14 - 17") while keeping the stress constant.  In another parameter set, he uses longer 0-1 springs, the same length 1-2 springs, but increases the stress to 41% of yield (105 ksi = 724 MPa).

 

From: Jay Heefner jay@ligo.caltech.edu

AdL SEI

  • We have received the first 4 capacitive position sensors back from ADE. This week we will put them back into the test setup and determine if the 24Hz noise peak has been removed.
  • Preliminary in vacuum cable drawings should be ready for review at this week's SEI meeting (Friday). I received a quotefrom Accuglass for the male version of the 3 pin power connector we will need for the actuators. They are ~$150 each with a 6 week lead time.

Suspension

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

Advanced LIGO Suspensions

Working on the Universal SUS DRD in preparation for the SUS DRR Update today. Made viewgraphs for various team members and posted documents and viewgraphs on my web page.

Will meet with Carol tomorrow morning about cost book information and organization. Will hold a telecon with Caroline and Helena prior to iron out some details. Worked this week on preparing for this.

Yesterday, visited Datum Control and Nelson Grinding in Orange County with Calum to check on the progress of the catcher parts and the penultimate mass. Took exciting pictures. Asked Datum Control a number of QC related questions as they are a new vendor/machine shop to us. All answers were satisfactory.

From: ctorrie <ctorrie@ligo.caltech.edu>

Advanced LIGO Suspensions

SolidWorks and PDMWorks

We have now all of the add-on's associated with SolidWorks 2005 and PDMWorks 2005 working. All of the Caltech licences are SolidWorks (Research) Office Professional Seats.

Quad Suspension

Work on the upper structure is progressing very well. reference the June photograph section http://www.ligo.caltech.edu/~ctorrie/pictures/QUAD-JUNE-2005-01/QuadJUNE2005-01.html  The clean room has also been installed in the lab. Helena, Janeen and I plan to suspend one chain from a fixed top mass to further investigate the clamps associated with the upper intermediate stage.

Design Meeting

http://www.ligo.caltech.edu/~ctorrie/QUAD_ETM/quad_etm_setup_page2.html

Visits

Janeen and visited the 2 machine shops in Orange County that are working on the lower structure face plates. Work is progressing well and all of the parts should be with us before the end of June.  Tim Hayler, Ian Wilmut and Joe O'Dell will all visit Caltech in July to support the build of the Controls prototype quad suspension. Russell Jones will visit in August.  Other visitors will be arranged at next weeks Suspension meeting.

Drum End Wires

As part of our on going research into drum ended wires. The company it Italy sent us a "clean" wire without some of the characteristics we observed on the first batch of prototypes. This wire is currently being baked for 100 hours at 435 degrees C.

From: Helena Armandula ahelena@ligo.caltech.edu

Adv. LIGO SUS

Building of the clean room in the Synchrotron Lab. is complete. By Friday, Caltech's electrical maintenance personnel will make the electrical connections for the lights and fans to the Synchrotron's electrical lines. Also, the optical tables will be moved back in place, allowing for the assembly of the Quad to move forward.

Core Optics

From: Bill Kells <kells@ligo.caltech.edu>

Besides ongoing work on understanding the TM HR mirror quality (see my report to Fred), I have got swept up in the new surge of interest in the "parametric instability", particularly w.r.t the rather dire proclamations of the Peth group. I believe we must conduct some characteristic experiment to understand this phenomenon (the fact that AdL will nominally attain the instability threshold is undeniable). But the situation is more complex and subtle than the Perth group outlines, to the extent that I am not yet convinced that simple, passive, design precautions can obviate this issue. We will all meet (apparently) in a few weeks to "summit" on this (some Perth people will be at CAltech, I understand). I am reviewing and augmenting my old calculations on this to sharpen where I agree and disagree with their various now printed claims. An interesting aspect of this (now that FS has been chosen) is whether we should "apodize" the coatings for the AdL TMs. A tight coating boundary goes a long way to eliminating the possible opitcal HTMs which can participate (I believe that they have not addressed this, nor even used the correct nominal dimensions in their articles).

Also, I am writing code (actually expanding existing Mathematica algorithms I had already developed) to answer a recent question: What should be the PRC/SRC optical surface tolerances (esp. just ROC) for AdL optics?  I realized that a simple modal model should suffice for getting overall orientation on this (with, ultimately a full FFT to detail). Then, there are a host of similar questions that can concommitantly be resolved with this approach. Results soon: its simple/easy to use!

Pre-Stabilized Laser

From: Peter King pking@ligo.caltech.edu

No report this week.

Auxiliary Optics

From: Phil Willems <willems@ligo.caltech.edu>

Advanced LIGO Thermal Compensation:

We have begun assembling the equipment to for the preliminary design of the carbon dioxide laser compensator.  The laser and optics are here, the AOM and photodiode are on order.  For now we are using a spare TCS chiller which we are plumbing into Caltech's site chilled water.

We have also extracted a Texas Instruments DLP micromirror array from a display projector and tried to remove the glass window to test its utility as a spatially modulating reflective mask for 10.6um radiation.  The first attempt shattered the window, likely making the array inoperable, but we can now reverse engineer the construction to see if we can get the next window out cleanly.

We investigated an alternative technology, a "grating light valve," that works like a micromirror array, except that rather than tilt tiny mirrors it forms tiny gratings, diffracting unwanted light.  It would be an attractive technology but the versions offered by the one company who makes them work at 1um, not 10.6um (gratings, unlike mirrors, care a lot about wavelength).

From: Michael Smith smith@ligo.caltech.edu

BS DIFFRACTION LOSS

A revised doc, T-50066-03 BS Diffraction Loss (AdLIGO), was completed. It will be distributed to the DCC shortly.

SURF

My SURF students have washed up on the campus. Aabeg and Shasta are learning how to use Matlab and the ABCD matrix method for calculating Gaussian beam propagation. All of the needed equipment is available for the BRDF experiment; Peter King has an 80 MHz AOM driver. I am awaiting help from Ben Abbot for the QPD readout for the Oplev receiver experiment.

Other Laboratory R&D

From: Riccardo DeSalvo desalvo@ligo.caltech.edu

Marco

We were able to lock the cavity with the Mexican Hat mirror 5008 on higher order modes such as TEM10 and TEM11. Until now we haven?t seen a stable fundamental mode (Mesa beam) but I?m working on the alignment and the input mode matching in order to improve the coupling with the driving input Gaussian beam. I?m using Juri?s simulation program to tests different configuration of mirror tilt and input beam in order to reproduce the experimental data. I took a spectrum of the cavity and it shows a considerable high coupling with the TEM10; this is in agreement with the easy lock on the TEM10 mode.

Juri

I made some graphs for the Amaldi6 presentation, showing the results of the calculation of  the thermal noise in finite size fused silica substrate test masses: they show the noise reduction we expect in using a Mesa Beam instead of a Gaussian Beam. I started working with Bench2 in order to calculate also the improvements in the expected range for NS-NS.

I worked with Marco on the computation of the theoretical spectrum of the cavity and on the simulations with the real map of the mirror we are using.

Chiara

After the introduction of cylinders under the plane which supports the experimental set-up we have found the new balance of the system working on the soft springs attached under the table. Now I?m starting the new set of measurements with the accelerometers using the spectrum analyzer. Still taking classes in the Machine Shop.

John

After the SURF meetings and orientation, I started working with Marco on the alignment of the output optics for the Mexican Hat cavity and I’m studying the problem of input mode matching.


For additional information about this report, contact Phil Lindquist