Weekly Report for Week Ending November 8, 2001


 Exec. Comm. Agenda
Highlights
LSC
Administration
Hanford Observatory
Livingston Observatory
MIT
Caltech
Detector
40 Meter
TNI
LASTI
Data Analysis
LIGO II/Adv. R&D
Past Weekly Reports

The LIGO Executive Committee Agenda for Monday  November 12, 2001 will be:

 (Meeting time: 10:30 am Pacific Time)

Open meeting 10:30 - 11:30

  1. Announcements
  2. LSC Issues (Weiss)
  3. Comments on Weekly Report
  4. WBS 1 LIGO I Construction (Lindquist)
  5. WBS 2 LIGO Lab Operations
  6. WBS 3 and 4  Advanced R&D and LIGO II (Sanders)
Executive Committee only 11:30 - noon   Topics:
 

Special Items:


Special Announcements:


Weekly Report Highlights

Rai gives a lengthy discussion of the LLO seismic disturbance issue
 


LSC Issues (Weiss)


No report


LIGO I Construction/LIGO Laboratory Administration (Lindquist)



 

LIGO Operations--Administration



LIGO Weekly Site Telecon (Lindquist)

There was no site teleconference on Thursday, November 8.

The list of current actions revised to reflect the status of open actions assigned through November 1, 2001 may be found at ACTION LIST.


PROPERTY MANAGEMENT (Chargois)

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


DOCUMENT CONTROL CENTER (Turner, Mak)

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

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. . .

ACCOMPLISHMENTS

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

ACTIVITY

Packages Faxes
In 40 20
Out 6 37

Press here to access the DOCUMENT CONTROL CENTER WEB PAGE.


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

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

Press here for ACCOUNTS PAYABLE HISTORY DATA.

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

From: Florence Kaufman <fkaufman@ligo.caltech.edu>

SUBCONTRACTS MANAGEMENT (Petrac, Jasnow)

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

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

SUPPORT

 
Irene Baldon Dorothy Lloyd Rita Torres

Advanced LIGO (Frey)

From: Thomas Frey <tfrey@ligo.caltech.edu>

Progress Period from 11.02 to 11.08

Accomplishments:

Schedule 11.09 to 11.15:

Reports (Lindquist)

Nothing to report.



Change Control/Contingency (Lindquist)

The following change requests have been distributed to the Change Board:
 

CR-010011 
Revision C
Construction Adjust LIGO Construction Budgets to Reflect Actual Costs for Completed Tasks as well as REU Expenses, Accounting Adjustments, and 7-LIGO (old work order) costs. P. Lindquist
CR-010012
Revision B
WBS 1.4.4.1 Closeout Construction Budgets for Initial Computer Equipment Complement at the Sites (Larry Wallace will argue for spending these funds on network connections) P. Lindquist
CR-010013
Revision A
WBS 1.4.3.3 Realign budgets for LIGO Data Analysis and Computing.  This is a zero-sum change request. P. Lindquist
CR-010014 WBS 1.1.4 Additional budget to complete Livingston Staging Building. G. Stapfer/
P. Lindquist

Press for the latest Contingency Needs Projection.


Human Resources (Akutagawa)

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


Quality/Safety (Tyler)

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

No report this week.
 


LIGO Hanford Observatory (LHO) Operations (Raab)


No report.


LIGO Livingston Observatory (LLO) Operations (Coles)


Detector: (Jonathan) Optics and Installation: End Transmission Monitors were reworked last weekend by Mike Zucker and crew, to accommodate the recycled/recombined throughput. We moved ISCT3 back into position. We then aligned the beam on the table and removed the pick off for the BSPD. We also removed the pickoff for the CCD. We tried to align the scanning Fabry Perot, but there was not
enough light to do so easily. We aligned the New Focus PD1811. It has about 10-120 mV DC when the interferometer is in state 2. ISCT3 is the Y pickoff.
Removed permanent power to the PSL laser safety enclosure and installed a temporary feed in anticipation of installation of the Acoustic Enclosure. Also moved the entire enclosure aft ~2" to accommodate the acoustic structure. Fabricated special fixtures to safely lift cast-iron grates in the parking lot.  Programming of the Laser Safety Interlock system is in progress.

CDS: (Chethan) Trying to get Easy Channel access commands ezcawrite and ezcaread working.   Changed the master.config file to add a few more slow channels for Rana. Looking  at the web interface for medm screens program to fix the problems associated  with it. Added the Dust monitor databases to the vacuum system and modified the  dust code to add these new dust monitors.

LDAS:  (Shannon) Looking into purchasing a new computer for Igor.  He has requested a dual processor Xeon.  Have been discussing with Tom Evans what to do in case of a temperature problem in the Mass storage room in case I am not present or reachable.

GC:  Received  a replacement for the Penguin rack mount that was given to Jonathan for the laser safety computer.  Installed Office and fix some problems on Allen Sibley's computer.


Detector/Technical Support (Whitcomb, Coyne)


 
Installation& Commissioning:
Hanford
Livingston
Other Science/EngineeringActivities:
Design/Analysis/Fab
Issues/Concerns
See also the Installation web page

1.1 LHO INSTALLATION & COMMISSIONING

2km Commissioning

Nergis reporting:
Much of the activity this week has been to get the 2km interferometer running in power-recycled mode again. The ifo has been plagued with locking problems due to a poorly understood ring-up of the 334 Hz violin modes of the large optics. This violin mode ring-up seems to be both alignment dependent and at times is seen to depend on the servo loop gain as well. A full realignment and recentering of the beams was done with the hope that better centering on the optics would reduce the coupling. That was not the case.
Several other things were fixed/tweaked also: Despite all these ministrations, the violin modes persist in keeping the interferometer from locking.

Meanwhile tidal compensation using the fine actuators on the ETMs is now operational on the Y arm as well. Both arm cavity compensation is running in single-arm mode.

4 km Commissioning

Nergis reporting:
Here too most of the effort is directed towards power-recycled operation of the interferometer. To this end, arm cavity at a time and State 2 have been locked. New (and better) servo compensation filters were loaded to account for some filtering in the DSCs. With different States locked...

1.2LLO INSTALLATION & COMMISSIONING

4 km Commissioning

Rai Weiss reporting
We have learned a great deal this week. There have been technical and operational successes but also we have affirmed and gained significant new information on the seismic noise which has enormous impact on how we can approach the development and operation of the interferometer at Livingston. To be frank with readers of this weekly, I consider it an emergency for the project and recommend that we give the solution of the problem the very highest priority. It needs a multi pronged effort with a practical solution working in the fall of 2002.

The technical and operational successes:
The recycled/recombined interferometer can be locked (and holds) at will when the seismic noise velocity spectrum between 1 to 3 Hz is smaller than 0.3 micron/sec/sqrt(Hz). Such amplitudes and smaller occur at Hanford any day of the week and all day long. At Livingston they occur most nights between 6 PM to 6 AM except for two trains each night that take an accumulated 1 hour. The good conditions are also obtained on Sunday during the day. If nothing changes the duty cycle of the interferometer would be no better than 0.6.

At night the interferometer runs continuously but the lock needs to be broken to avoid the tides (and probably also the temperature changes of the laser reference cavity) from causing the maximum control currents in the test mass drivers from being exceeded. It is safe to assume that should the tidal compensation and the microseismic feed forward system be fully functional, the instrument would stay locked continuously between 6PM to 6AM except for the hour dedicated to the trains. The tidal compensator and the feed forward system are under active development. In particular, the microseismic feed forward system is working well enough so that within the next two weeks it will become a standard part of the interferometer. The remaining effort is to reduce an undesired excitation of the isolation stack near 2 Hz.

Many small improvements made in the past two weeks have contributed to the continuous locking. Among these improvements are the application of the differential mode wavefront sensor, the readjustment of filters in the length sensing servo to deal with the ambient seismic spectrum and simply better operating knowledge.

A single arm cavity (either the x (the easy one) or the y (the hard one)) can now be locked under the worst seismic conditions we experience in Livingston. The lock is not pretty. The cavity optic axis dances around riding the seismic waves, some people here compare it to bronco riding at the rodeo, but the cavity will hold typically breaking lock after 10 minutes and sometimes after an hour. These times are long enough to get some real work done. It allows the test of wavefront sensors, it has enabled a test of the common mode servo which has functioned well enough to leave only a remaining 10% of the cavity control to the test mass servoes. This is a significant step in unloading the common mode cavity length controls when running the main interferometer and may yet play a role in improving the chance of locking the full interferometer during the day.

The lock of the single arm cavities during the day is only possible due to the increased dynamic range controllers which are being used on all the test masses and, by this evening, will also be installed on the beam splitter and recycling mirror. It is now clear that the longitudinal servoes for the arm cavities have enough dynamic range to deal with the frequency fluctuations of the mode cleaner/stabilized laser and the seismically induced arm cavity length fluctuations under the worst conditions. The remaining noise preventing a 4km interferometer of any kind from functioning at Livingston during the day are angular fluctuations for which we do not currently have the proper control and which, yet, may be induced by the incompletely orthogonalized motions of the test masses as they are pushed hard to maintain the cavity length.

The seismic noise and operating during the day.
We have not been able to lock either the recombined or the recycled (the full) interferometer during the seismic activities of the day.

The initial thought has been that it would be easier to approach the control problem of locking a recombined system during the seismic noise of the day. The short term goal was to learn to lock the recombined system during the day and perfect the lock of the full interferometer at night. Once night time operation was secure to begin in earnest on the real problem of LIGO associated with the noise in the system which is guaranteed to be from the electronics that is common to Hanford and Livingston.

We have begun a systematic study of what happens to the recombined interferometer when the seismic noise starts at about 6AM. It is now trivial to lock the recombined system by 5:30AM and wait for the world to wake up. With this strategy it is still possible to watch the lock drop out and reaqcuire and watch again until 6:30 when currently moderately close logging activities begin. The data from these lock loss events is still being analysed, however, it is clear that the primary symptom that leads to the breakdown are antisymmetric angular motions (one cavity beam goes left the other right or one goes up the other down). The angular motions are large amounting to a few microradians. The motion causes the overlap of the beams to be incomplete and the fringe at the antisymmetric port goes from dark to bright. The system with the current servo configuration and locking sequence cannot recover from this and drops out of lock. (If one gets really desperate, it is possible to think of acquisition programs that could deal with this but it is certainly better to avoid the breakdown in the first place.) Much of the information comes from video frames that show the breakdown, signals from the antisymmetric wavefront sensor confirm the beam motions and the photo current at the antisymmetric photodetector also shows the change from dark to bright as the system goes into the breakdown. Where there is still more work is in trying to understand what drives the angular motions. The breakdowns do not correlate with activities in a host of other signals that are being looked at which include the optical levers, the OSEM sensors, the mode cleaner frequency control, the interferometer control signals, the coil drive signals. There is only a broad correlation with the seismic noise, in that no single event is seen to trigger the breakdown sequence, rather, it is the general seismic increase within 5's of seconds before the breakdown. A candidate for the cause is incomplete orthogonalization of the test mass drives which is expected with the analog controllers that cannot easily adapt the control matrix for the drive frequency - a principal argument for the digital controllers. We do not have evidence that this incomplete orthogonalization is the cause.

Clearly, more work is needed here and there may well be pay dirt in engaging the wavefront sensors in an active control loop. At the minimum such loops will help hold the lock, we might be lucky and find that they actually help acquire the lock as well since the individual cavities seem to be able to deal with seismic noise.

Insight into the source of seismic noise
The close logging activities to the y end station have provided more precise information than before for the size of the signals to be expected from logging. Broadly, logging at 1km from the y end station provides seismic motions as large as the trains but with an even more ragged statistic between rms and peak to peak motions. The peak motions at the base of the chambers is about 8 microns/sec near 2 Hz with the rms velocity in the 1 - 3 Hz band being about 1.5 microns/sec. The motions at the point of support of the test mass pendulum is expected to be 3 to 5 times larger depending on the location of an isolation stack resonance.

When LIGO was built at Livingston the expectation was that the rms velocity around 2 Hz would be about 0.1 microns/sec and this is approximately the value achieved during the night. It is still about 10 times worse than Hanford but, hopefully, will be good enough not to cause excess noise by non linear couplings in the interferometer.

The logging activities as well as almost all other surface noise sources will propagate as Rayleigh waves and attenuate as the 1/sqrt(distance). To bring the rms logging noise down to the level of the night noise would require the closest logging activity be further than 225 km away. To bring the noise to the level of allowing the acquisition code to work ( 1 micron/sec at the suspended mass and using a stack amplification of only 3) requires the closest logging to be 20 km away.

In a visit with the Weyerhauser chief logger in the Livingston area (George Tully  225-686-2278), he informed me that it is guaranteed that there is some logging activity within a 10km radius of LIGO at any time in the year. It happens that the fall is most intensive since the weather is cooler, there is less rain and Christmas funds are empty. He told me of the their cutting schedule around the planned E7 run when they intend to cut well within 3 km of LIGO near the LVEA and then along the y arm. He explained that Weyerhauser is also in the business of selling land. The value is about $3K/acre so a 10km radius area would cost about $230M. An ill advised investment for LIGO at almost any price since there are other noise sources to deal with even though it now seems that logging is the source of the bulk of the noise here. One of the nuisances with the logging is that though they harvest once every 20 years, they have small patches not large tracts and they come into the patches three times in twenty years. The first visit is to thin out the growth, the second to remove lower branches and then, finally the third, to harvest the wood.

We have no alternative other than making a significant investment in a crash program to reduce the effect of the seismic noise on the interferometer at Livingston. We need to look at several possible solutions, we should be involving industry and making a real shift of our engineering and scientific manpower into attacking this problem. It is, along with the excess noise in the interferometers, our critical problem.

2.0 OtherEngineering and Scientific Activities

2.1 Design/Analysis/Fab

Modeling

Erika D'Ambrosio
William Kells and I have been discussing a lot about sidebands' imbalance, but we have also focused on our pictures independently. I am not entirely sure the imbalance can be completely cancelled so we decided I am reading his notes in a few days but I still want to develop my model.

Indeed I can see clearly from my Mathematica notebook that for Guoy phase not too small everything works, as the change in the ITM mirrors' radii of curvature is compensated by adjusting their position. This is also in agreement with simulations but for the typical Guoy phase of the recycling cavity the mix of the modes is such that there is no simultaneous solution for all the effects; I made some carrier runs to make this point clear in my mind and my mathematical frame is becoming very simple.

Bill Kells
Continued work writing up the detailed analysis of SB imbalanced in LIGO like ifos (beyond the brief synopsis completed for the now draft complete MELODY paper)

Finish basic note on the Braginsky/Vyachanin mechanism for parametric instability of TM acoustic modes.

Core Optics

GariLynn Billingsley
All transmission, absorption and phase map data for the WA2K are now available by clicking the "Data" link under the  from the COC optic index page.  Access the index page at: http://www.ligo.caltech.edu/~gari/COCAsBuilt.htm You can also get to this page by following the link from the LIGO home page through the Research Bulletin Board, Core Optics Components, Optic Index

Bill Kells
Begin trying to get operational contamination test stands in OTF (with L. Cardenas, and consulting with Jordan).

Lee Cardenas
I have completed searching for all those turbo pump stations at the PSL lab, bridge & optic lab, the bake room by the 40mlab.  there are a total of (7) complete pump stations included one new PFEIFFER  which is at bridge in the optic lab and I believe Phil Willems is putting together, from those seven units, only (4) works and two separate turbo pump.

In terms of the contamination chamber, still pumping and I did not opened it because Bill Kells suggested until We'll find out from Jordan's response.  I will build back the other chambers from scratch and I will need  two NPROS. I will take care all of that with time.  In the meanwhile, I am gathering and ordering all the items needed in the lab.

I went to see the enclosure and checked it completely and took pictures.  I gave the go ahead to have it shipped today and it will arrive on Nov. 16th. at MIT.  I am in process and schedule to leave to MIT on Nov. 26th for two weeks as Kent Mason wanted.

IOO PERISCOPE

Mike Smith, Ken Mailand
The third periscope is at CIT awaiting vibration testing before being shipped to LASTI the week of 11/12.  Drawing documentation packages for the LLO prototype and the production model periscopes are in the process of being released. A preliminary technical note describing the results of the vibration testing and the frequency noise measurements at LHO and LLO using the new IO periscope was published.

PSL

Peter King,
Another NPRO laser is back from a repair job at Lightwave.

REO have agreed to coat another batch of substrates since they got the last coating run wrong.  The intention was to have the coating run performed by the end of this week, or early next week.

AdvLIGO PSL:  I contacted Lightwave about a meeting to see what they think is necessary in upgrading the 10-W laser, on loan to Stanford, to 20 W.  To date they have not returned my emails or phone call.

Rick Karwoski
PMC  spares: I Completed testing the spare for LLO and I am in the process of testing the newly layed out PMC for LHO.

ISS: I have also been concentrating on getting a third ISS unit ready for the Dec engineering run.

Sander Liu
Completed testing two more FSS Servo Card

CDS/DAQ

Rolf Bork
After a number of CDS problem calls last weekfrom LHO, opted to travel  to LHO to investigate:

1) DAQ controller had VME bus error problems, attributed to new reflected mem board installed.  Dave could get the system to run with unit borrowed from end station.  Found that new board had been installed in what apparently is a bad VME slot in the crate. When we moved the new board to another slot, bus errors went away and system operates properly.

2) When we went to put the borrowed reflected mem board back in the end station, the DAQ controller would indicate that the end station was going in and out of timing synchronization.  Using a VME bus analyzer, found that every tenth or eleventh cycle was 10msec too long. Various component changes, timing clock tests, could not find a problem. Finally, put bus analyzer into another system not showing sync faults, and bus analyzer still showed this timing anomally.  Added code to read GPS time on each interrupt cycle, and timing was correct.  The VME bus analyzer appears to have a problem.  Went back to troubleshooting end station sync problem, and it went away when Dave/Richard moved the fiber connection off of the reflected memory autobypass unit.  Problems with this bypass unit will be investigated further as operations permit.

3) Received a number of calls over the course of last week with problems on 2k system: LSC controller performance meter pegs, ASC crashes, LVEA ADCU problems, etc, all requiring constant VME crate reboots to correct. Put diags into system which showed that all 2k LVEA systems would report timing errors at the same time.  Found that problems had started roughly when GPS rcvr w/4MHz clock and clock fanout unit were moved from a DAQ crate to a suspension controller crate to allow test of the DAQ crate and still permit the 2k to operate.  We moved the GPS rcvr, etc., out of the suspension crate into its own 3 slot VME crate.  The reported 2k problems have not reaccurred since.

Calibration

Luca Matone
We are presently developing an automated calibration for the ITMs. The method consists in locking a simple Michelson on dark fringe with one ITM and taking the control signal mean value. Inverting the gain sign forces the Michelson to go from dark to bright fringe and, if the transition lacks disturbing transients, the Michelson length will change by lambda/4. The mean control signal difference between dark and bright fringe will then correspond to lambda/4, resulting in the calibration factor.

Michael Landry has written a tckl script that would change the loop sign and stand-by for an arbitrary amount of time before changing the sign again. I wrote a C program that goes through the data, takes the different lock stretches, estimates the various means and standard deviations and gives the calibration factor.

We have tested a first version of the program with the 4k simple Michelson. Consistent and precise results (down to ~1%) were obtained, however, systematic errors (such as electronic offsets) still need to be addressed if a calibration accuracy of ~1% is desired.

Digital Suspension Controls

Peter Fritschel
I have completed the design of the digital filters for the LOS and SOS suspension controllers, and a document describing the design. The document can be found at (and soon in the dcc):

http://www.ligo.caltech.edu/~pf/T010140-00.pdf

The design includes low-pass filters to suppress local sensor noise, and output matrix filters to provide frequency-dependent cancellation of the force-to-pitch angle and pitch axis torque-to-position couplings inherent in the single loop suspension. I will begin making filter coefficient files for the LHO 4k optics, but they will rely on some modifications to the suspension controller front end code that are pending.

Jay Heefner
A new proposed output chain for the MC SOS was presented at this week's engineering meeting. The output chain incorporates the 0.1/1.0 pole/zero needed by the MC length controls and also includes bias offsets for each coil and a fast input to be used by the LSC. Comments on the design should be sent to Jay by 11/12.

LSC

Mike Zucker
Mostly working on our main length sensing photodetectors: Jay Heefner Rich Abbot AdvLIGO

EO Shutter Controller

Sander Liu
Received Serial 001 Controller from LHO. In the process or repairing and updating the unit by adding a daughter board. Except for the front panel and the VMOS switches, most components are now in and we are in the process of building up five mother board and five daughter boards.

Data Monitoring Tool

John Zweizig
This week I have started to prepare for the E6 run and the PEM deep mine effort. I have installed the new version of the DMT code on the DMT machines and am verifying that it works well.

AdvLIGO Suspensions

Janeen Romie Mark Barton
This week I finished debugging the thermal noise part of my Mathematica quad pendulum model and have been packaging it for release. I've added functions for exporting state-space models to Matlab and "usage" tags for all the functions and variables (so that they show up in the online help). I've arranged that all time-consuming calculations save their results to files, so that in subsequent sessions the results can simply be reloaded. I've reordered the code so that utility functions, the model specification and the model calculation are all cleanly separated. I've also started work on a manual.

Calum Torrie


40 Meter Interferometer (Weinstein)




Thermal Noise Interferometer (Libbrecht)


no report


LASTI (Zucker)


LASTI (Bayer, Fritschel, Goda, Harry, Laliberty, MacInnis, Mason,
Miller, Mittleman, Ottaway, Overmier, Phinney, Rollins, Shoemaker,
Traylor, Zucker)
=======================================================================
 
BSC SEI INSTALLATION:
Harry Overmier and Gary Traylor came up from LLO to help Myron, Bob and
Fred get the BSC stack installed.  Despite the extensive rigging and
cleanroom workarounds needed to accommodate our tight lab space, the
downtube and stack elements went in without a hitch Monday and Tuesday
morning (#MZ: so fast I missed it!). After also helping us R&R a bad
BSC bellows seal and miscellaneous other fixes, we ran out of jobs so
the pros spent Tuesday afternoon taking in the aquarium. Thanks Harry
and Gary! (and thanks to all of LLO for doing without them a few days,
we appreciate it).
 

We now plan on installing our test cavity SOS's into HAM13 along with
associated wiring, feedthroughs and viewports before doing an overall
leak check, tentatively the week before Thanksgiving.
 

HYDRAULIC ACTUATORS:
Ken is still iterating the mechanical design with feedback from the
Stanford group. The finite element models continue to reveal
detailed nuances of the unusual bidirectional coil springs,
including a problem with stress concentrations in a small zone of the
coil ID and a rotational interaction between the formed ends of the
coils where they are attached.  Normally these things could only be
detected by testing; with the high tooling charges to make even one of
these custom springs, working it out in the computer is a great
advantage.
 

Joshua is trying to figure out why the hydraulic pump is not achieving
the design flow rate or pressure head (although it's worth noting that
even so, it's still more than adequate for what we have planned).
Cavitation has been all but ruled out by tests. Josh is consulting with
the manufacturers and with Brian Lantz at Stanford (who, by the way, will be
visiting us next week along with Dan DeBra).
 

CDS/DAQ/GDS:
We received and are installing a spare RAID unit sent from LHO by Dave
Barker.  This has the old (obsolete) firmware revision which is
known to communicate OK with the GDS Sun. Dave and Jay are still
negotiating with the company what to do with the newer unit we procured
(of the same nominal type), which appears fundamentally
incompatible with Suns.
 
PSL:
The new PSL enclosure was completed and inspected at the manufacturer's
plant in CA by Lee Cardenas.  It will be shipped to us this week for
installation the week before Thanksgiving.


Data Analysis and Computing (Lazzarini)


Simulation and Modeling (Bhawal)
E2E Physics Meeting
-------------------
Ed Daw talked about ground motion at LLO and the microseismic feedforward system
currently being tried out at LLO.

Lock acquisition in mode-mismatched LIGO
----------------------------------------
(Biplab & Matt) Encountered problem in acquiring lock when modes are not matched
in Han2k model. Working on it.

Code maintenance
----------------
(Hiro) worked on low level coding and debugging of modeler.

Alfi
----
(Bruce)
- Container tree updating properly now.
- Fixed design flaw in connection bookkeeping object and a related parser problem.

(Ed Maros)
Started to investigate 3 problems ( 2 in Alfi, 1 in ALParser) reported by Hiro.

(Melody)
- Worked on more member node widgets functionality for alfi5.
   Started on the display of the node parameters.

LIGO Data Analysis System

Software Systems (Blackburn)

This week was primarily spent adding functionality to LDAS to support user
requested data sets with gaps. The frameAPI now recognizes requests over
ranges of time for which there are no frame files and if the user has
specified that frame data gaps are allowed the data is passed to other
APIs (dataConditionAPI or lightWeightAPI) of a form which captures the
user requested range and the actual data that exists for that range. If
no frames exist for the requested range an exception is noted.

The dataConditionAPI has had additional functionality added to it to
support these gaps. Upon ingestion, the gaps are automatically filled
with zeros. A user action fillgap() has been added to allow the user
to specify a potentially more meaningfull gap filling method. This date
can then be used in search codes running on the BEOWULF. The periodic
mock data challenge scheduled for the week after Thanksgiving will use
this new functionality.

The new functionality for adding database query information to user jobs
through the four new options {-dpquery, -dbspectrum, -dbqualitychannel,
and -dbntuples} which was only partially tested prior to the last release
now has a complete suite of tests to test nominal and failure modes. A
dozen or so minor bugs in these options have been identified with these
new tests and will be fixed today.

A new set of tests for the frameAPI were also developed over the past
week to more completely test request of ldas to produce reduced data
products from frame data. These have identified several bugs in the
frameAPI which are now fixed.

The new naming convention for frames which has been discussed extensively
across several groups within LIGO has been integrated into LDAS. The older
formates for frame file names are still supported but with great pain and
lose of efficiency.

The dataConditionAPI and the eventMonitorAPI are now capable of two
types of data product organization for outputing frame files. The user
can now specify all data products appear in a single output frame or
that each data product gets a unique output frame file.

A new release of LDAS will probably happen this Friday. This is necessary
to support the Super-Computing 2001 demo of the joint LDAS GriPhyN software
prototype, the PEM Deep Mine Project, the E6 engineering Run, and the
Periodic MDC. A lot is riding on this release of LDAS and it is still only
alpha code, not yet having all the APIs in place.

Hardware Systems (Anderson)

Caltech
-------
(Dan Kozak)

Assisted James Patton with connection of STK9940 tape drives in LIGO
silo to HPSS mover nodes.

Compiled and installed hsi 2.6.2 on Linux hosts.

Worked on HPSS data migration to 9940 tapes (ongoing).

Restarted HPSS many times.
 

(Al Wilson)

Put ldas-pcdev3 back online with the new RH 2.4.9-12 kernel. Also have
included the new kernel in the rpmsync links. The test system is updated
to this kernel also. Working on including the rpmsync functions in the
kickstart file.
 

(Stuart Anderson)

Starting to request quotes for the next small Beowulf cluster: a 32 node
rack mount system for the initial ldas-cit system, i.e., LDAS Caltech
production system.

Continuing to identify and order the necessary accessories for the
recent order of Sun servers, e.g., racks, UPS, KVM switch, ...

Obtained an additional 20 AIT-2 tapes for LDAS use at Caltech.

Positive test results on RedHat 7.2 stability so far.
 

MIT
---
(Keith Bayer)

retrieving QFS license
 

Hanford
-------
(Greg Mendell)

1) Worked on plans for the Known Pulsar Demod Mock Data Challenge
(MDC).  Wrote a planning document for the MDC, and initiated the plans.

2) Prepared T3s at LLO and LHO to archive the E6 data.  At LLO we are
all set to go, exactly as for E5.  At LHO I've upgraded the T3 directory
structure and tapecontrol script configuration to handle frames from
fb3, which are written to a QFS file system.  I am currently testing the
new configuration. I've also set up a backup set of T3s to archive data
from fb0, if needed.  (I am also running a quick test at LLO, to make
sure nothing has broken there since E5.)  The control of the tapecontrol
script will be the same as for E5.

General Computing (Wallace)
MIT:
(Keith)
-Installed new matlab software / licenses
-Upgraded Administrative Assistants PC to win2k
reinstalled software
setup printers etc...
-Experimented with wireless coverage to bldg next store.
It appears that this system will work for a couple of offices that
will be occupied by LIGO people.
-Moved all active users to new RAID disk
-Upgraded remaining sun nodes to OpenSSH_2.5.2p2
-Investigated missing NIS+ credentials for users.
-Investigating laptop for postdoc.

Livingston:
Nothing to report.

Hanford:
(Christine)
- Met with an ergonomics expert and safety person from Caltech
concerning purchasing computer equipment for people with disabilities.
Also, discussed computer equipment to prevent repetitive stress
injuries.  It was very informative and worth the time.
- Still troubleshooting a problem with Framemaker saving the file every
minute on certain computers and not on others.
- Installed more freeware software on our application server.
- Looking into a couple of problems with the new NIS+ server.
-Resolved a P-card problem for a number of users. The issue is that those with
older versions of the s/w on their PC's need to have a few libraries upgraded or
they will not be able to login to the main system.

-Richard and Larry have made contact with NoaNet and are working on getting
more information on their network backbone. Presently, they look like a viable
resource for expanding the Observatory WAN connection.

CIT:
(Mike)
-Complete rebuilds of a laptop and PC workstation.
-Installed a new hard drive on one of the PC's then loaded the OS and General
Computing software; I am still finishing up loading the other PC. Laptop is
completed and ready to put back into the pool.
-Windows explorer shortcut some how ended up corrupted, to where the users
laptop was pretty much dead.
I logged in as administrator and deleted the users desktop shortcuts. This
fixed the problem.
-Backed-up users laptop personal files on a cd and created ghost images of C:\
& D:\
-Software problems, exceed was corrupted, I uninstalled exceed and error
message went away. This has happened multiple times this time I left this
software off this computer.
-Help Larry resize a server rack, to make room to transfer a Blade 1000.
-Daily administration all servers look okay after reviewing log files.
ANTARES I had to compact the database.
-Onsite and phone support that included software reloads, software fixes,
troubleshooting hardware problems and printing problems

(Lisa)
- Loaded winNT on Janeen's sunpci 2.  This has completely resolved all of the
problems she was having with running engineering packages on the sunpci card.
The winNT drivers for sunpci behave differently than the win98.  I've been
flipping back and forth between the 2 and I think NT is far superior in this
configuration.
- Looked into incompatibilities with reply-to addresses for people using
alternative mail clients.  Basically, anyone using something other than eudora
or dtmail needs to configure the appropriate reply-to address. Contact me if
you need instructions for a particular mail client.
- Found a solution for Riccardo's data transfer needs in the synchatron.  I am
going to put up an ultra 1 with a secure samba server. This will allow him to
save data directly to the ultra1 then use ssh or sftp to get at the data.
- Created and tested an access control list for the wireless access points.
This will restrict access to our wireless network by mac address. Anyone who
has not submitted their mac address to me will not be able to connect to the
access points once this is deployed.  This will be an important thing to
remember when we have visitors and review board folks around.
- Usual round of account maintenance and user support.

(Veronica)
- Continued work on the database and scripts for the Project Science
website. Mike and I have added the Project Science webpages as another
website to the Amaldi server. Working on the webpages for the January
workshop as well as the general contents of the website.
- Updated various parts of the main LIGO website.
- Compressed the video of the talk by a JPL team that they gave at the
CaJAGWR seminar. The video will be posted at the CaJAGWR website as soon
as the clearance procedures are finished. Planning an upcoming CaJAGWR
workshop. Updated the "seminars" page.
- Communicating with Marie Woods, an administrative assistant at MIT LIGO,
on helping her update the MIT LIGO website.
- Participating in planning the upcoming Elba GWADW Conference, as a
member of the Organizing Committee and as a webmaster.

(Larry)
-Fixed a number of PC issues. Including s/w reloads but most of the PC
issues were fixed by cleaning up some files or reloading a driver.
-Worked a number of procurement issues. Resolved a number of problems with SUN
and working on purchases for various groups. Still working on getting FOUNDRY
equipment(sent to us by mistake) sent back. Also, resolved a few P-card issues
for different people.
-Continued working on the server room so it will be able to accommodate the new
network and server hardware. Once the network hardware is up and running the
FORE equipment will be distributed to the Observatories. With the new
arrangement I was able to move a couple of servers out of my office.
-Still working on the e-mail server virus-scan s/w. The filter pkg. appears to
work OK but the virus-scan for existing files is still having problems.
-Working on computer policy documentation and related issues.
 


LIGO II/Advanced R&D (Sanders)


no report


For additional information about this report, contact sanders@ligo.caltech.edu