Report from the Software Coordinator
---------------------------------------
The software change control board has decided on a single formulation
for the discrete Fourier transform to be used in all algorithms. The
convention is that used in the detector diagnostic and monitoring programs.
(From email this week, it looks as though the issue is still being
debated.)

The first day of the LSC meeting is dedicated to the work of the
upper limit groups. The morning of the meeting is for a plenary
session on common issues. The afternoon is for individual upper
limit group meetings. At the end of the day there is a second
plenary session in which status reports will be given by the upper
limit group chairs.

Discussion of the future of the ASIS committee
----------------------------------------------
Bruce Allen lead a discussion on the future of the ASIS group.
To focus the discussion he formulated the possible directions
as:

(1) Organizing systematic testing of the scientific analysis codes that
have already been written.

(2) Regrouping to tackle theoretical problems that may be important in
searching for astrophysical sources, but are not being agressively persued
by the broad theoretical community.

(3) Carrying on with further development and refinement of data analysis
methods and algorithms.

(4) Porting data analysis codes to a Grid environment, where they can take
advantage of national and international computing resources significantly
larger than those currently available to LIGO and the LSC.

The executive committee suggested that the ASIS committee take up this issue
at the next LSC meeting. The comments from the committee ranged from giving
importance to all the directions raised to a focus on the near term tasks
associated with validation and testing. A suggestion was made to divide the
ASIS committee into subgroups each with one of the listed tasks.

In the context of the ASIS discussion, the question of how the LSC should
operate was raised. Should it continue with groups that have a vertical
structure (groups with the tasks of dealing with a specific technical issue
such as the current ASIS and the technical development groups) or convert to
horizontal structures (groups which have a range of technical expertise applied
to a specific scientific problem such as the current upper limit groups)?
Strong arguments were given to retain both types of activities The arguments
for the horizontal structures are based on coherence while those to retain
vertical structures are for efficiency.

Planning for the LSC meeting
-----------------------------
The initial draft agenda was modified to accomodate a more extensive presentation
on the requirements being developed for advanced LIGO. Time was allotted
for presentations on new LSC usable facilities at the University of Wisconsin
at Milwaukee and at Pennsylvania State University.

Report from the LIGO Directorate
----------------------------------
Discussions are taking place within the Laboratory on the timing of the E6
run. The latest schedule for the run and the interferometer configurations
available for the run will be discussed at the LSC meeting at Hanford in August.

The LIGO renewal proposal for funding operations and the research and development
will be presented to the National Science Board August 8 and 9.

Next LSC Executive Committee meeting
-------------------------------------
At the LSC meeting in Hanford Wednesday Aug 15, 2001 at 7pm.

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

There was a teleconference on Thursday, July 26. The following were among the items discussed:

Web pages for the DCC give simple how-to's for document numbering, easy access to the latest on-line documents, and search capabilities for the DCC database. Take a look. . .

Irene Baldon

Processed the paper work for eighteen (18) new trips (including Advance Checks written and hotel/car rental authorizations filled out and FAXed to appropriate vendors in various locations). I have seventeen (17) trips in various stages of completion before ticketing can be done. Worked on several trips that were changed or modified and assisted several more travelers with questions regarding their existing or potential travel.
I worked on two (2) that were problems and clarified a few more that Travel Audit or the traveler were having trouble with. I completed twenty-three (23) reports and there are thirty-nine (39) reports to be done. I'm holding one (1) report that needs a check.
Reconciled twenty-two (22) items on my P-Card which required telephone calls to vendors who do not enter the traveler's name on their charges. Assisted a few travelers with their reconciling and/or data entry.
Worked on the preparation and distribution of the Travel/Vacation Itinerary for July 23, 2001. Performed normal recording and filing associated with Travel and Reimbursement. Also performed miscellaneous duties as requested by various members of the LIGO Project here at Caltech as well as from members of the staffs of each of the two (2) sites. I continue to do MIT's travel to the sites for installation activities and also to assist them wherever possible.

Dorothy Lloyd

Processed the usual requisitions, invoices and receiving on-line. For more detail, see "Cost Schedule Control Systems" report by Esther Cunningham.
Tracked and followed up on invoice problems.
Reviewed and recorded payments processed by Esther the weeks of July 16.
Continue to monitor contract and blanket order encumbrance and notify task managers when supplements are needed.
Managed to start storage files for DCC by pulling closed open orders and 1998-99 contract invoices, to make room for FY02 files. As storage boxes become available and time permits, I will continue with the 1999 standard purchase orders.
Jim continues with data entry in the LIGO database and supports Irene (travel) in the mornings and Linda and Cleveland (DCC) in the afternoons.

Rita Torres

Completed updating the LIGO Personnel roster containing our home information; updated with as much information as people provided. Distributed the roster to the administrative staff only. Took a pleasant walk to Sponsored Research to obtain a signature.
Looking for an alignment tool, and am gathering vendor responses to my request.
Placed a few Pcard orders, and accepted packages for folks not here this week; kept the packing slips to help with reconciliation. Canceled an order with PC Nation when they informed us of a handling fee of almost $60 for the item. Placed the LDAS order instead with a vendor that Larry Wallace recommended. Followed up with vendors on some long lead items, one item is 16 weeks.
Arranged with AT&T for a standing reservation for the Advanced LIGO Systems teleconferences. These are scheduled for a couple of times each month.
Updated our laser training and eye exam records. Did site trip updates.

Elizabeth K. Wood

Spent a few days cleaning up LIGO’s voicemail lists, getting names with the proper phone numbers, changing names, clearing out phone numbers that have been assigned to other departments, and giving the phone office the correct PTAs to charge. We haven’t had any more strange phone bills this month.
If you are an international employee, postdoc, graduate or undergraduate student and you want to go home for any reason, please please please contact the international scholars’ office BEFORE you make your travel arrangements. They will advise you on what you need to do before you leave, and they will make sure you have the proper documentation to return to the United States. Your contact with the international scholars’ office may be as simple as a phone call [(626)395-3435] or email (elaine.sommers@caltech.edu).
I can no longer make payment requests (check requests) for purchases over $100.00. Such purchases will have to made either on a pcard or through a purchase order. I can still send you across the street to get a petty cash reimbursement for greater than $100.00 but less than $200.00, however, this will involve taking a walk to 363 S. Hill, getting the voucher approved by the Bursar, and then walking back to the cashier’s office in Spalding. If you plan on making frequent purchases for parts and supplies at the same establishment and you don’t have a pcard, please consider having us set up a purchase order for you.

Advanced LIGO (Frey)

Quality/Safety (Tyler)

There will be telecon on Friday, 27 July, to review/discuss the updated/modified proposal for the laser access system. The discussion will include information on site system integration impacts and identifies differences, if any, in the access system that are required by each site.

The LIGO Vehicle Usage Policy was approved by Gary Sanders for distribution to the site management for their concurrence.

Regarding the conference room projection systems, all of the hardware, cabling and wireless "mouses" are finally on-hand at Caltech-LIGO. A new Service Request as been submitted. A schedule for conference room down time will be worked out when Caltech support (electrical shop) is available.

LIGO Hanford Observatory (LHO) Operations (Raab)

LIGO Livingston Observatory (LLO) Operations (Coles)

Detector: Optics and Installation: Alignment and OSEM replacement of the vertex is complete. Ham and BSC doors are reinstalled and the pumpdown is in progress. At the X-end, tidal compensation Piezos were installed and operated Wednesday, OSEMs have been replaced and the optic has been drag wiped. ETM-x is being aligned this afternoon and we should have the door reinstalled and pumpdown by the weekend. In-vacuum installation is almost finished, and we'll be resuming commissioning next week. Jonathan Kern

Vacuum: The vacuum system in the corner station is being slowly pumped. Pressure is about 100 torr and no dust on the witness plates. We should be able to get the turbo pumps on line by tomorrow. (Allen Sibley)

CDS: Tested upgraded version of vacuum software and graphic displays. Performed LLO1 shutdown test. Performed a 1 hour test of UPS in Mass Storage Room. Completed installation and testing of power and communication to seismometer array. Completed satellite amplifier upgrade. Completed installation and testing of power and communication to new RGA location on the south arm. Verified operation of conversion software for FMCS. Installed power,communication and camera's for internet viewing of CDS racks. Built new and installed new cross connect for mode cleaner servo upgrade. (Rus Wooley, Mike Fyffe)

Setup 3 more axis web cams for Rolf and Rus. The cams can be found at http://130.39.245.206/ http://130.39.245.207/ http://130.39.245.209 http://130.39.245.210/ http://130.39.245.211/ Still waiting on the drive enclosure for the new disk space. The FedEx says it should be here tomorrow. Installed another access point to take care of a dead zone in the office area. Working on a web page so that Rolf can pull up one page and have a view of all of the racks via the web cams. Investigating some software that would allow automatic failover and load sharing for the web server and mail server. The Firewall came in this week but I have not had a chance to look at the documentation as of yet. Tom has installed it in the rack. (Shannon Roddy)

Computing: Bell South is in the process of installing our second T1 line to LSU campus, and should be finished this afternoon. LSU is checking specifications for the new router we need to enable the second line, and is estimating installation in about two weeks.We have received our Cisco PIX firewall, and are installing it.We are beginning preparations to support the E5 run. (Tom Evans)

Operations: The E-5 operator schedule is posted on the LLO home page under control room. Implemented a new site cleaning schedule which is posted near the
janitorial closet. Preparing a preventative maintenance and cleaning schedule for the sites 3 weather stations. Transferring past completed "paper" checklist data into our electronic checklist. Nearing the completion of Checklist trending software development. Finalizing the newest changes to the checklist and operator training log. (Rich Riesen)

PEM: Installed the set of accelerometers for BSC1, so all accelerometers are now mounted, although a few still have cabling issues. Reinstalled the microphone at the Y-end station, it is operating normally. Will return the LVEA seismometer to its proper location, it has been located under HAM3 for the past 2 months for student projects. Posted a prototype web page to help visitors understand the environmental conditions at LLO. The first page is devoted to LLO's seismic activity. This is an electronic version of the PEM notebooks that we have in the control room. The hope is that this web based PEM information can help with the analysis of the interferometer data. The link is http://www.ligo-la.caltech.edu/PEM_Ref/seisref/seisref.htm. Please feel free to give me feedback on the page. (Doug Lormand)

LDAS: Still resolving some issues with the BigBrother application and looking at some scripts to integrate tripwire. Looking at ways of integrating the temperature monitor into BigBrother. Getting some documentation on different administration tools similar to cfengine. Started looking at the UPS software today. I hope to have something going by the end of the week with this.
Networking: As a result of a question from Albert, I contacted Qwest to see if we could get some pricing on OC-3 connections for the future. I have given them our requirements, and had a conference call with an engineer and a salesperson. I should know something by next week. One interesting note - They do have some programs where they will make arrangements to partially fund some connections to universities etc. The interest in Qwest came about because LSU just switched some of there networks to Qwest. (Shannon Roddy)

Detector/Technical Support (Whitcomb, Coyne)


Installation& Commissioning: Hanford Livingston	Other Science/EngineeringActivities: Design/Analysis/Fab Issues/Concerns

See also the daily electronic logs for the installation and commissioning activities:

Hanford Detector Log

Livingston Detector Log

See also the Installation web page

1.1 LHO INSTALLATION & COMMISSIONING

2km Commissioning

Nergis reporting: (Stan W, Fred R, Michael L, Daniel S, Richard M, Josh M, Rolf B, Nergis M)

Since early success with locking the power recycled interferometer, we found that we were experiencing unusually large and frequent alignment drifts. This was traced to a malfunctioning coil driver on ITMX, which we repaired. Subsequent interferometer locking went very well. With laboriously careful alignment we have been able to lock the interferometer with a power recycling gain of 24. At that point more than 80% of the light is coupled to the interferometer. We do not as yet have a calibrated noise spectrum, though a first look shows significant contribution from frequency noise in the 10 to 100 Hz band. The periscope resonance at ~95 Hz is of course the dominant feature. As we work more with the ifo we have found that bad locking behavior is strongly correlated to increased laser frequency noise as measured on the mode cleaner frequency control signal (MC_F).

The lock acquisition parameters are the same as we have been using since a week ago and seem to be pretty stable.

Rolf has made changes to the LSC front end code so that we can attach up to nine test points. (Recently, we had been using up a large fraction (100%) of a sample cycle if we attached test points to the LSC.)

We found an impedance mismatch problem in the mode cleaner length path where an output network in the MC servo amp was unable to drive a differential receiver stage (a Pomona box kluge from the early days). This has been replaced with a properly instrumented differential receiver stage (one of the new rack mount chasis) and now functions well. The problem was discovered when Rick S and Hugh R were using the MC length control signal to verify the tidal actuator on the PSL reference cavity and could not reconcile the calibration.

Rolf Bork

Modified the LHO 2k LSC front end software:

LSC FE rejects any requests for more than a total of nine (9) TP/EX from GDS. This is to ensure LSC does not exceed time budget. In the process of testing this using the GDS tools, noted that some tools do not release test points even after exiting the GDS program. This could be the GDS tool itself or the Network Data Server in the FrameBuilder. This will be investigated further. (Dataviewer always released test points in this test. )
Changed LSC time meter such that it does not read GPS timing module (to save I/O time). It now estimates based on "free" cpu cycles from end of a calculation cycle to data ready in ADC for next cycle.
Changed the way the front end sends/reads data from EPICS supervisor to save a couple of microseconds (every microsecond counts here!).
Noted that makefile at LHO does not set compiler opitimization to level 2, whereas makefiles at Caltech do. Turns out that code compiled at Caltech then runs a few microseconds faster than same code compiled at LHO. Makefiles will be fixed at LHO.

4-k Interferometer

Dennis reporting:

The work to re-align the MC, MMT (subsequent to their removal as a result of the errant digital control system) continues. The alignment tweaks of the ITMs and the RM are also underway while the vertex volume is vented.

Mark Barton

This week I spent at LHO helping with the servicing of the 4k small optics. In fact there was relatively little to do because all the stuck optics came free during the vent as soon as the pressure reached around 300 Torr. (This is a little surprising because the Paschen minimum for breakdown voltage tends to be around 5 Torr, and if it was static charge as suspected, one might have expected it to dissipate at some pressure on the low side of the Paschen minimum.) The two optics that we removed for direct inspection were generally healthy, although we made a few mechanical adjustments: rotating MC3 on its wire and adjusting its upper OSEM holder, and resetting the lower OSEM positions on MMT1. (I suspect the optics may have been bumped on the way out of the chamber – the positions of the lower OSEMs were way off in opposite directions, which doesn't correspond to the problem observed in vacuum.)

Data Acquisition

Rolf Bork

Installed an additional two modified clock driver modules in LVEA ADCU. We are now trying to get the code which analyzes the timing ramps to run on the DMT such that we can 1) compare timing of these units to those systems which were not modified and 2) determine if timing shifts occur when ADCU are reset. We also have diagnostics code within the ADCU which does the same analysis which we want to test.
Modified the DAQ controller software:

1)Reports to EPICS screen/alarm handler when the controller itself falls out of sync (previously just printed the message to a terminal). This will be expanded to have the controller resync itself.

2)Automatically sends a resync command to any ADCU which falls out of timing lock with the system. (ADCU with pentiums already had code to accept this command. Barker today modified MIPS cpu code to respond in the same fashion. This will be tested tomorrow.

We are looking more closely into the errant gain settings we are seeing on the h2 PEM ADCU on some channels.
The LVEA H2 suspension ADCU was reconfigured to read all suspension signals (had been limited to reading three of the five ADC modules before the MIPS cpu was replaced by a Pentium III). This brought the present full frame size up to about 6.8 MByte/sec (~5.7MByte/sec of data + frame overhead), with total DAQ network traffic at 9.3MByte/Sec (with test points and other none recorded data). DAQ controller (present limiting factor in DAQ, limit being total DAQ net traffic) is now running at 88% capacity. Only channels yet to be added, that I can think of, are the 45 H1 suspension coil readouts in the LVEA (45 x 2kHz x 2 bytes = ~ 180KBytes/sec additional). The present 6.8MBytes/sec frame is likely too much data for the LDAS tape system to keep up with for the E5 run, so we may have to drop a number of H1 channels during that time.
Brought the X end station ADCU back on line. Only LHO DAQ system fault indicator is now the Y end station, which is turned off while the bakeout is going on in that building.
The present frame data rate is limiting our framebuilder full frame storage time to ~13 hours. This is an inconvenience for the commissioning crew, as data from the previous day's run is not available for review the next day. It is unclear when the LDAS T3 system will be available. Therefore, we are in the process of adding a 500GByte RAID to the primary framebuilder to bring our capacity back up to about 30 hours of on line data. Hopefully, this will be completed tomorrow.

1.2LLO INSTALLATION & COMMISSIONING

OSEM Replacement

Dennis Coyne reporting:

The MC length was measured (in air and corrected for vacuum conditions) to be 12.2444 meters compared to the spec length of 12.2454 meters. So the Mode Cleaner is 1.0 mm too short. The tolerance on this length is +/- 1.4mm which arises from matching to the theoretical recycling cavity length and to minimize the conversion of phase modulation into amplitude modulation.
The ITMx was drag wiped, all chamber exit checks/documentation performed, the spool and all doors replaced and torqued in the vertex volume, ready for pump down.
The ETMx OSEM replacement is underway. The re-worked flourel tips have been replaced, the optic has been drag wiped and will be aligned 7/27.

PSL

Peter King

The mode-matching lenses to the reference cavity were adjusted. The cavity visibility increased from approximately 66% to 94%. The frequency servo settings were tweaked up to yield an in-the-loop noise spectrum consistent with the requirement. Thus far the laser has remained locked for over a week, except for when the loop was deliberately broken.
Intensity noise data after the pre-mode cleaner was acquired for use by the E2E group.

2.0 OtherEngineering and Scientific Activities

2.1 Design/Analysis/Fab

Optical modeling

Erika D'Ambrosio

Evaluation of the perturbation induced by tilt on the stability of the flat top beam cavity using an expansion formalism. Predictions are compared with data and the output field grids obtained by a batch of FFT-code runs have been carefully analyzed in order to understand the composition in terms of unperturbed portion of the field and its excited dipolar component. At the dark port the odd and the even part agree with perturbation techniques predictions.
For an analysis of the diffraction loss effect entangled with the tilt perturbation a full analysis of the field grid inside the Fabry-Perot cavity must be finished. Since this effect can have an impact on the power associated with the dipolar component, I want to check those simulation results with the model I used to interpret the dark port output.
Many other checks have been finished and they agree well with analytical predictions. I am also working on the evaluation of the diffraction loss through the beamsplitter using the full formalism of electric field propagation since I learnt the clip approximation is satisfactory only as an order of magnitude and according to the characteristics of the field the diffraction effect can be two or three times larger.

PSL

Peter King

Another partial shipment of super polished substrates from Mindrum Precision have arrived. So far only half the order has been filled.

Rick Karwoski, Paul Russell

LASTI PSL Installation W/O 7/29: Major installation items and equipment (x-connect, power supplies/housings, cabling, connectors, etc.) received @ MIT. Custom electronics has been tested in the 40m. system.

ISS: Development of the digital interface for ISS servo parameter control is in process.

Sander Liu

Tested one set each of the PMC and FSS boards designated for LASTI using 40M facility with Dennis Ugolini. All boards functioned as expected. Completed testing the PMC Frequency Reference Board and the Servo Board designated for GEO in the lab.

Lee Cardenas

The new bake oven chamber for PSL reference cavity parts is being completed except for some odds and ends. (RGA unit not available yet, 40m is using it).

LSC Electronics

Rich Abbott, Mohana Megeswaran, Flavio Nocera

Flavio is preparing additional old style clock driver boards for Rolf to use.
Mohana is ready to start testing with the Crystal Laser unit sent from MIT to be used in the Photon Recoil Calibration system. She must first prepare an SOP for the laser, and obtain the necessary approval for its operation.

LSC characterization:

Luca Matone

The Anti-Aliasing filter from the schematics was compared to a measurement. The Anti-Image filter from the schematics was recorded (seems to differ from some measurements, 5th order elliptic not 4th order). Bank of digital filters doing the inverse of (1) the Whitening filter and (2) the De-Whitening filter was recorded.

A first comparison of the theoretical/measured tf between the Whitening board and the Anti-Image board was made (with the Whitening filter off and AA on). Good agreement found but the stopband filters have not been included in the theoretical tf yet. Once the digital stopband filters have been nailed down and included in the model, I'll begin to compare a set of measurements on the LSC system with a theoretical tf.

Digital Suspension Controllers

Jay Heefner

Module production continues, but production of several modules is on hold for delivery of LEMO connectors. It turns out that the connectors are being held by LEMO pending the clearing up of some Caltech credit problems. Hopefully it will be resolved soon, or we may experience some schedule slips.

Module production is as follows:

Module	Needed	In Fab	Complete	Tested
PD Interface	29	34	0*	0
Anti-Image	12	0	8	7
Universal Dewhite	22	25	0	0
LOS Coil Driver	14	18	0	0
LOS Bias Module	14	17	17	0
SOS Dewhite	11	16	0	0
SOS Coil Driver	15	0	24	24
Optical Lever Interface	15	13	4	0

Sander Liu

Received 50 100Hz low pass filters from Frequency Devices. All were shipped to LHO.

ADC/DAC

Jay Heefner

The RFP for development of an ADC module should be out by the end of this week. A specification and RFP for the DAC will be worked next.

AC OSEM

Rich Abbott

Stuffed and completed preliminary testing of the prototype for the OSEM stabilized oscillator and it is running. I am still evaluating its AM noise performance.

DMT

John Zweizig

This week I have been preparing the DMT software for the E5 run by collecting and installing a bunch of new monitors and other code submitted by LSC members including:

Histogramming package (M. Ito)
eqMon - an earthquake monitor (R. Rahkola and M. Ito)
MultiTaper line finder/follower (A. Ottewill)
Rayleigh monitor (P. Sutton)
Improved Operating State code (D. Chin)

Ed Daw

This week I have written DMT code to look for peaks in filtered data. The code finds the n biggest peaks per t seconds using a continuous peak search routine in the data monitor tool environment. The filter operates on the data before the search algorithm can be FIR, IIR, or any other filter type that inherits the pipe base class. First, I will use a 1-3Hz IIR bandpass filter to look for transient peaks in the 1 - 3 Hz frequency bands in the seismometers at Livingston.

Integrated Layout Dwgs

Mike Smith, Ken Mailand

The D970308 integrated layout drawings (six sheets) for the LHO interferometer are being revised to reflect the current as-built design configuration. A current layout of the PSL components was received from Dave Ottoway and is being incorporated.

IO Periscope Re-design

Mike Smith, Ken Mailand

A re-design concept for the periscope has been completed. A finite element analysis will be done of the final design. Vendors have been contacted to determine feasibility of various fabrication options. A decision was made by Stan and Dennis to purchase PI piezo tilt stages with strain guage sensors instead of the untested PiezoJena alternative, which would require a redesigned mirror mount as well as modifications to the electronic driver circuit. The design will enable LLO to use Optosigma mounts and LHO to use Newport mounts as before. It will accomodate the bottom steering mirror to be placed either on the PSL table or on the periscope.

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

Seiji Kawamura is back in town this week, and with his help we have been
continuing our investigations of the TNI lock. We now believe that our
stable locks in the recent past were not to the carrier, but to a sideband.
This does not affect our sensitivity estimate, since that was made based
on a calibration peak produced by a broadband Pockels cell. We are able to
lock to the carrier, but the locks are short -- 10-20ms -- and we are
trying to improve that time. (We can tell this is the carrier by looking at
the dip in reflected power.)

A few months ago, at Gary Sanders' request, we ordered some spare lasers
for the LIGO community to use as needed. These lasers have now arrived and
are available for loan. We have two 700mW Lightwave lasers available
immediately, and we have two Innolight Mephisto 800mW lasers. The
Innolight lasers do not currently have PZT actuators installed, and we
would like to install PZT's and characterize their performance before
loaning them out.

LASTI (Zucker)

Data Analysis and Computing (Lazzarini)

"E2E for LIGO I" monthly meeting
--------------------------------
The monthly E2E/LIGO-I meeting took place on Tuesday. Hiro, Luca, Biplab,
Virginio, Robert, Dave Reitze, Peter King reported progress in their
activities. Details are available in http://www.ligo.caltech.edu/~e2e.

Mode Mismatch
-------------
Biplab continued validation and improving understanding of the mode
mismatch problem in a recycling cavity. Results obtained so far agreed
with analytical approximation. Some differences in details are being
sorted out.

PSL Frequency Noise
-------------------
Biplab and Rick improved their calculations and considered some realistic
cases. The estimate of FSS noise in presence of higher order modes, for
example, in case of a 3 beam-width PD whose response function looks like a
tilted surface with 1% variation across it and which is moving transversely
with 1e-8 m/Sqrt[Hz] motion, is about one order of magnitude lower than the
LIGO design requirement ( 1e-2 Hz/rtHz at 1 kHz). In other studied cases
this level is lower and, especially for longitudinal motion of PD, it's of
the order of the shot-noise limit (4e-5 hz/Sqrt[Hz]) or below it.
Rick is trying to get more information about PD nonuniformity in its
responses.

Alfi
----
(Bruce)
- Working to interface new Alfi5 classes with the application
skeleton Melody has prepared.

(Melody)
- Worked on the development environment for the new Alfi (getting the
Makefiles' dependencies and packaging the classes to Java Archive
(JAR) files working properly).
- Looked at the JGO code to see how to extend the classes for Alfi.
- Worked on a design document for some base classes. Modified one of
the JGO programs to display the class hierarchies and used the
drawings for the document.
- More discussions with Bruce about integrating his design with the
base classes which have been developed.

Continued stress testing of LDAS software after pushing system onto the new
hardware at Hanford and Livingston for the first time revealed a condition the
possibility in the managerAPI, controlMonitorAPI and the mpiAPI (possibly
others) in which these APIs could exhaust all available file descriptors (used
to open files and sockets). This has been traced to nested vwait statements
in some of the TCL source codes. A possible solution is being tested out on
the development system. However, the development system is far less likely
to fall into this state due to its not being attached to nightly rsyncs and
routine frame file copies.

Further testing of the dataConditionAPI to stimulate the thread bug on the
Sun systems with an attached debugger have failed to produce a useful
trace into the problem. The dataConditionAPI is working well on Linux,
though it does fail every hour or so under heavy use. This is currently
thought to be inherited behavior found in the 2.95.3 GCC's standard
template library. A more stable version of this library was released with
GCC 3.0 but we will not be able to migrate to that release prior to the
3.0.1 release in mid August.

The eventMonitorAPI is now able to produce ILWDs for "Processed" frames
using the frame's FrProc structure. These have been successfully turned into
frame files and verified against as legitimate frames with correct content.
The only remaining step to make this functionality available to the data-
Pileline is to add a bit of TCL code to the frameAPI to handle the request
to translate eventMonitor ILWD as part of the dataPipeline command.

The UWM group is having problems building a standalone wrapperAPI
other than the 0.0.17 version (we are currently releasing 0.0.19). Attempts
to reproduce the problem here have shown no problems at all. We are
working with UWM to understand what it is about their steps in building
and running the standalone wrapperAPI are causing the problems that we
do not see here.

Hanford and Livingston both had problems with the framebuilder codes
running at the respective sites this week. These problems were tracked
down and solved by Alex. Alex is also testing an new version of the
framebuilder software at Hanford this week.

Key LDAS team members met with key GriPhyN team members at the
USC-ISI facilities this week. A course was laid out for integrating
several grid technologies into LDAS as a prototype/demonstration
effort. Details of this planned activity will be presented at next week's
GriPhyN meeting in at University of Chicago.

Several meetings took place between LDAS and LSC software development
teams to discuss a less bumpy approach to carrying out integration tests of
new LAL and LALwrapper code under LDAS. The key issues being those
of configuration control and utilization of available manpower. Two actions
will be taken to improve the platform for development. The first is to have
more frequent LAL and LALwrapper tags on that CVS repository. The second
is to have a meeting of all LSC institutions that are in a position to
configure working versions of LDAS for the purpose of providing the collaboration
with more LAL and LALwrapper development/test environments.

LDAS Caltech (Dan Kozak, Stuart Anderson, Al Wilson)
----------------------------------------------------

Run of the mill HPSS stuff (unjamming tapes, restarting servers,
repacking tapes, adding users, etc.).

Testing of a locally attached QFS filesystem with data on a single T3
RAID unit shows read speads in the 80-85Mb/sec range for blocksizes
above 8Mb and write speeds in the 75-88Mb/sec for blocksizes from 512K
to 512Mb, with the peak at an 8Mb blocksize (88Mb/sec).

Testing of QFS vs. UFS when exported over NFS concentrated on write
speeds, particularly of 47782450 byte files (the current frame
builder's output file size at LHO). QFS+NFS showed write performance of
14-31Mb/sec for these files (the variability seems to be connected to
metadata performance, i.e. it's more stable if you're rewritting an
existing file of the same size; this is an area for further
investigation). UFS+NFS showed write performance of 10-26Mb/sec.

Since experience with the framebuilder suggested that file deletion may
be a bottleneck, I tested that as well: QFS+NFS showed 66.67
unlinks/sec, UFS+NFS showed 2.3 unlinks/sec.

Started a major upgrade of all LDAS Intel computers to RedHat 7.1 using
an semi-automated scheme that encourages tighter version control.
Concurrent with this is upgrade is the beginning of the switch for
all LDAS machines to use OpenSSH as their primary authentication mechanism.

1) Conducted further test of the tapecontrol script and ldas monitor
client GUI
to prepare for E5. The control and monitor GUI has been debugged and is
now
working. Operators will be given training on how to use the new
features on the
interface next week. Test of the tapecontrol script indicate that data
can be
read from the fb0 disk cache and written to tape on average at 6 MB/s.
There
is talk that the LHO frame size maybe reduced to less than this for E5.

2) Dell replaced the memory and motherboard on LHO beowulf node6. The
burntest
was re-run on this box, and it passed.

3) The disk I/O errors reported on LHO beowulf node11 disappeared once
the box was brought to my office. The box also passed Dell's HD
benchmark test,
and the burntest. A memory test is currently runnin on this box.

4) Contractors have finished walling in the LHO LDAS area in the staging
building.
Otto has brought more A/C into the area, and the temperatures seem
stable. However,
we are montoring the situation. A dedicated A/C unit for this area
shipped last
week, and a concrete slab for the compressor was poured this week.
Photos are at
this URL: http://apex.ligo-wa.caltech.edu/~gmendell/LDASwall_pics/

6) Started writing an LDAS DSO to run search code for known pulsars
using the LALDemod function.

Still resolving some issues with the BigBrother application and looking
at some scripts to integrate tripwire. Looking at ways of integrating
the temperature monitor into BigBrother.
Geting some documentation on different administration tools similar to
cfengine.
Started looking at the UPS software today. I hope to have something
going by the end of the week with this.

As a result of a question from Albert, I contacted Qwest to see if we
could get some pricing on OC-3 connections for the future. I have given
them our requirements, and had a conference call with an engineer and a
salesperson. I should know something by next week. One interesting
note - They do have some programs where they will make arrangements to
partially fund some connections to universities etc. The interest in
Qwest came about because LSU just switched some of there networks to Qwest.

Following up with MIT facilities on a/c troubles and monitoring LDAS
machine room temperature.

Pursuing MIT network 'sluggishness', e.g., it took 6-7 days to mirror
the initial 44GB of LDAS related software to the new LDAS gateway machine.

MIT:
-Continued LDAS installation work
-Ordered wireless network setup for MIT in support of
MDC in September
-In process of bringing new file server on-line
-Ordered virus software for PCs
-Installing virus software on PCs
-Attended laser training safety course at MIT (required)

Livingston:
(Tom)
-Bell South is in the process of installing our second T1 line
to LSU campus, and should be finished this afternoon. LSU is
checking specifications for the new router we need to enable
the second line, and is estimating installation in about two
weeks.
-We have received our Cisco PIX system, and are installing it.
-We are beginning preparations to support the E5 run.

(Shannon)
-Setup 3 more axis web cams for Rolf and Rus. The cams can be found at
http://130.39.245.206/ http://130.39.245.207/ http://130.39.245.209
http://130.39.245.210/ http://130.39.245.211/
-Still waiting on the drive enclosure for the new disk space. The FedEx
says it should be here tomorrow.
-Installed another access point to take care of a dead zone in the office
area.
-Working on a web page so that Rolf can pull up one page and have a view
of all of the racks via the web cams.
-Investigating some software that would allow automatic fail over and load
sharing for the web server and mail server.
-The CISCO PIX security system came in this week but I have not had a chance to
look at the documentation as of yet. Tom has installed it in the rack.
-As a result of a question from Albert, I contacted Qwest to see if we
could get some pricing on OC-3 connections for the future. I have given
them our requirements, and had a conference call with an engineer and a
salesperson. I should know something by next week. One interesting
note - They do have some programs where they will make arrangements to
partially fund some connections to universities etc. The interest in
Qwest came about because LSU just switched some of there networks to Qwest.

Hanford:
(Christine)
- The user directories have been moved to their final (?) resting places
on the new disks.
- The file server, rainier, and tape robot will be replaced the weekend
after the LSC, 8/18 and 8/19. Continuing to document what will need to
be moved/restored from the old server to the new server.
- Loading MS Office 2000 Pro on several borrowed computers for the LSC
mtg. More wireless network setup. Purchased a wireless pointer mouse
to be used with the PC and projector.
- Prepared a computer for a new hire to start Mon. Helped other users
with various problems and "How to's".

CIT:
(Wendy)
-Finished installing Cadence and Solaris with Lisa
-Cleaned software room--software is now in folders in the file cabinets
- Learned to do usfdump at 40m
- learned to set up (configure) Unix box for general computing
-Learned to flash printer firmware
-Inventoried computers in Wilson House and Lauritsen
-Installed floppy drive
-Assisted on the setup of a Blade1000
-Loaded a few PCs
-Flashed prom on Ultra10
-Updated inventory list

(Lisa)
- Did a PO to purchase additional licensing for the Sun Solstice Backup
software.
- Finished up the Cadence installation. The new version of Cadence can run on
Solaris 7 but really prefers Solaris 8. Sol 7 needs quite a few patches.
- Rebuilt alcor for Al Wilson.
- Finished building br40m for the 40meter. (Mick and Wendy helped out with these
rebuilds).
- Worked on some minor problems with ftpanon and acrux.
- Fixed a printer that got the red worm virus. Showed Mick how to make sure the
other printers were not vulnerable to it.

(Mike)
-Deliberated with ANSYS people on issuing us another license to uninstall
from the current License server that was running on win98; then used the
new license to run on NTSRV4.0. This is now complete.
-Updated service packs on our NTSRV's due to the current viruses that are now
out. Microsoft recommends that all NTSRV's running IIS should have the
latest service packs installed, which is service pack 6a and then apply the
red worm virus patch.
-Loaded NAV on all NTSRV's and scanned for viruses.
Onsite user support, which included updating, patches for OS and NAV, then
encouraging users to update their NAV using the live update plus walking
them threw the process giving them step-by-step instructions on how to work
this.
-Connecting users to networked printers. Cleaning users computers from viruses
-Salvaging multiple computers and peripherals.

(Barbara)
- Loaded new WBS dictionary and SEI data into the Costbook. Refreshed the
PDF reports. Reviewed the 4.02 report in detail. Wrote up a cheat-sheet
and automated the procedure to generate WBS levels in case we load the WBS
dictionary again.
- Updated LDAS home page and maps to reflect LDAS MIT. Am waiting for word
from Stuart to install.
- Set up database for upcoming LSC transparencies and created script for
generating the web pages. As soon as information from prior meetings can
be loaded, all LSC transparency pages will be generated via the script.
- Resumed work on generating LIGO roster from Roster database.
- Revisited all the antares and pherkab backup jobs to bring up-to-date.
- Refreshed public DCC search tool.
- Made several web site changes for Publications pages, added car rental
link to internal site. Began to update the About LIGO pages.

(Mik)
-Flashed the PROM on alcor, fixing shutdown problem
-Ran MU on sadal, and alcor
-Finished compiling security audit data
-With Wendy, inventoried Wilson House
-Flashed BIOS on PC, fixing shutdown problem; then reloaded software
-Installed NAV on virus-infected user computer and performed clean
-Updated the firmware on three HP printers with the HP Download Manager
-Set up ORiNOCO AP for west side of Millikan, and another with Lisa in
Liz's office
-Set up Linksys AP in Dennis Coyne's Office (354A), and installed an
ORiNOCO wireless card in his laptop
-With Wendy, inventoried labs in Lauritsen
-Continued to cross-check inventory data with active IP numbers and
resolve discrepancies

(Larry)
-The big issue for this week has been virus problems. A number of people did not
have the latest definitions file for the virus scan and were infected because of
this. Mik and Wendy have been assigned to go through and make sure certain
people have the latest files on their system. This is something the PC users
need to do on a weekly basis.
Mik and Wendy are also going to be backing up a number of the PC's on to CD for
a number of users that are not familiar with buring a CD.
The vulnerable areas on the PC's for this last virus should all be taken care by
now.
-Working on installing some Blade1000 units to augment and replace some of the
servers being used by the General Computing community. Sargas is to be replaced,
the SUN450 unit is being turned over to the LDAS group for a need they have at
the CACR facility.
-Spent time with the MIT group. Mostly working on the LDAS equipment
installation which was a good learning experience. Also, assisted Keith on a
couple of General Computing issues and changes that he is implementing.
-Besides repairing PC's with virus problems, worked on a number of PC's that had
h/w issues and OS configuration problems. This past week definitely was a work
on the PC's week.
-Placed an order for more memory upgrades. A number of SUN boxes and PC's that
were not upgraded on the last go around will be taken care of on this one.
-Resolved a couple of procurement issues and working with Gina on a procurement
issue she has with SUN.
-Finally, received a couple of quotes on getting the network moved to a GIGE
based setup. We now have some numbers to work from.
-Worked a couple of minor server issues.
-Worked a few network issues and assisted on some of the wireless setups.

LIGO II/Advanced R&D (Sanders)

Adv LIGO SEI
Rich Abbott
· Finished testing the new capacitance position sensors to be used in the seismic isolation test setup. The resultant noise floor of the initial version is 2.5e-10m/rtHz. There is to be a second generation of this setup that the manufacturer claims will have almost an order of magnitude improvement in displacement noise.

Update from last meeting
Alessandro, Riccardo
got first sapphire flex joint made from Faimond, surface looks with few micron
size flakes, making more flex joints with finer passes to reduce the flakes
dimensions and then will try polishing techniques. Will make samples to be
tested both in ICRR and our cryostat (if approved).

Alessandro
Accelerometers tested on inverted pendulum in synchrotron lab. Working fine. Improved electronics : BW 60 -> 140 Hz, phase margin 20 deg -> 40 deg. Likely to fly to Japan next weekend.

Hongo activity Akiteru, Giovanni & Florian
Good news, one accel ok, two need more gain, Alessandro's modifications will do some good.

Lia
Fixed LVDT signal data taking and analysis. Fan crosstalk eliminated. Some variation over 36 hours found, need further checks. One LVDT channel seems broke, will need repair.

Andrea
Work on 40 m. In absence of Dennis Ugolini, working on matlab programs for frequency analysis. Prepare to measure slope of error signal to measure laser frequency stability.

Mathieu
Looked at Frederic files. Listed models and boundary conditions. Working on silica bond fibers, nice drawings on FEM first calculations. Seems to be RAM-size limited. in touch with Lisa Bogue.

Charlotte
Simple model of flex joint, waiting for license of drafting program to feed ANSYS with more complex designs.

Riccardo
Back in Europe. Investigating hydrojet cutting for sapphire, progress on ultrasound cutting machine, sapphire tests in our cryostat approved.

Francesco
Resume acoustic emission measurements. Looking into totem spring mounting for creep data taking. Sorting out small computer problems.

Weekly Report for Week Ending July 26, 2001

Special Announcements:

LSC Issues (Weiss)