|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday March 11, 2002 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items: S1, S2, and S3 run schedule and "shutdown" planning
Subject: Notes from LSC Executive Committee meeting 2/22/02
to: LSC executive committee
from: R. Weiss March 3, 2002
concerning: Notes from the LSC Executive Committee meeting February
22, 2002
The LSC Executive Committee meetings are being permanently moved to
the
fourth friday in each month.
Plans for on coming LSC meetings
--------------------------------
The Source simulation workshop will begin in the late afternoon of
Wednesday March 20 at Livingston. Supper will be catered.
Bev Berger will present the "View from the NSF" at the meeting
The next engineering run, E8, will take place over a weekend in
mid April.
Data analysis and "upper limit" group activities
-------------------------------------------------
Alan Wiseman reports that the LSC wide data analysis phone call
fulfilled the Collaboration's expectations for disseminating information
about the efforts of the "upper limit" groups. We will discuss
the value and continuation of such meetings at the LSC meeting.
The LSC and detector commissioning group have come to a consensus on
the
channels to be included in a reduced data set. The decimation of the
data
will first be carried with the data conditioning API.
The mechanism for data exchange with GEO requires a more formal protocol
than
what has happening in the upper limit groups. The exchange will be
carried
out with knowledge and agreement of the project Directors.
TAMA and LIGO/LSC will formulate an MOU for a joint coincidence run
and
data analysis. Al Lazzarini will be working with N. Konda, who is currently
visiting LIGO at Caltech. The MOU will be presented to the LSC Executive
Committee for approval. Konda will make a presentation about the joint
run to the LSC at the March meeting.
Planning for LIGO Science Runs
-------------------------------
Gary Sanders presented the Laboratory's current thinking about the
first
LIGO science runs and expectations. The proposal is:
S1 be scheduled for two weeks at the end of June and beginning of July, 2002.
S2 be scheduled for a month beginning around Thanksgiving and ending
at New
Years.
S3, the first long science run, would occur in mid-2003 and last for
three
to four months.
The seismic retrofit would begin in Livingston after S2.
Further discussion of the K. Riles shift proposal
-------------------------------------------------
The discussion of the science runs made the resolution of how shifts
are to be manned even more cogent. Nevertheless, there is still no
consensus on the committee. Two things were decided:
* the MOU of collaborating institutions with LIGO will now include
a section on service being provided to the Laboratory by the institution
* I will pull a small committee together to formulate a position at
the
oncoming LSC meeting.
Action on Advanced LIGO funding in Britain
------------------------------------------
Ken Strain reported that PPARC is about to make critical
decisions on the funding of the GEO contributions to Advanced
LIGO.
Next LSC Executive Committee meeting
-------------------------------------
Friday March 22, 2002 at Juban's Restaurant, Baton Rouge
LIGO Operations--Administration
A Site Teleconference was held on Thursday, March 7, 2002. The agenda included the status of FY 2002 costs-to-date, of NSF funding and the cooperative agreement, the establishment and budgeting of FY 2002 accounts, the FY 2002 Financial Reporting format, as well as the status of the construction at the two sites.
From: Ed Chargois <chargois_e@ligo.caltech.edu>
>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
ACTIVITY
| 03/07/02 | Packages | Faxes |
| In | 65 | 35 |
| Out | 06 | 37 |
From: Esther Cunningham <esther@ligo.caltech.edu>
Press here for ACCOUNTS PAYABLE HISTORY DATA .
From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
From: irena@ligo.caltech.edu (Irena Petrac)
SUPPORT (Baldon, Torres, Lloyd, Tischler)
Progress Period from 03.01 to 03.07
Accomplishments:
We are still doing a standard Annual Report for the Construction Project as of the end of November. Actually it has been completed and is ready to be printed to be sent to the NSF.
Assembled pile of documents (Annual Reports and Work Plans) to support incurred cost meeting with the DCAA scheduled Thursday, March 7.
The following change request has been submitted:
| CR-010012
Revision B |
WBS 1.4.4.1 | Closeout Construction Budgets for Initial Computer Equipment Complement at the Sites | P. Lindquist |
Press for the latest Contingency Needs Projection.
From: Cindy Akutagawa <cindy@ligo.caltech.edu>
No report this week.
-----------------
FACILITIES
-----------------
The construction of the new building is on schedule. The steel framing
is
complete and so is the roof decking. Currently the wall studs are being
installed. We have three changes in process.The HVAC unit was specified
incorrectly and that needs to be changed, we are adding a power and
communications outlet by each chair in the auditorium, the LDAS room
sound
insulation was omitted in the original contract and that is being added
in.
We do not expect the original contract price to increase since we had
some
cost savings in the beginning.
Gave the A/V package for pricing to Chervenell construction and to
MediaTel
Inc. in Seattle.
Interferometer: The PSL glitches were investigated. It seems
that they are not seen in the direct intensity output of the laser, but
are rather occurring in the frequency of the laser.
The mode cleaner transfer functions were checked in order to get good
frequency dependent calibrations of the monitor channels. This in turn
will be used to measure crossover frequencies and laser frequency noise.
(Kovalik, King) We have updated the commissioning schedule of activities
to be accomplished prior to S1 to include those CDS tasks agreed upon during
the teleconference led by Dennis last week. This list and schedule, which
is still evolving, is posted at http://www.ligo-la.caltech.edu/~coles/Commissioning/CommSched020305.pdf
It appears that the week of March 18 will be a key week for CDS related commissioning activities because of the overlap of completion by LLO staff of preparations for equipment scheduled for installation at that time and the presence of many visiting staff who have committed to assist with various installation tasks.
CDS: Developing a test station for VME Boards. Worked on Pentek
and Xycom test code.
Tested a Xycom 220 VME Board. Configured the Apache web server for
CGI scripts.
Installation of a new Ultra 60 for london successful. Upgraded memory
on Frame
Broadcaster LLOfb1 to 256 MB. Ordered more memory for the same machine.
Found a
bad disk on LLFb0 and replaced it. This disk was spared by the hot
spare pool
and hence no data was lost. Alex found some corrupted minute trend
files that
could not be read and recovered them. Forte C/C++ compiler installed
on CDS
machine control 3. Udated weather stations medm screen. Working on
getting epics
security gateway machine setup. (Chethan)
Optics and Installation: The Livingston Laser Safety Interlock
system is
finally on-line, and for the next week or two we'll be debugging the
system.
Access cards have been distributed. Still piecing together a DC photodiode
and imaging system to monitor scatter at the cavity optics. Met with
Dr
Weiss to design a system to detect shorter wavenlength harmonic generation
on the large optics. Staged material and assisted Joe Giame in measuring
the
transfer functions of BSC piers. Thanks to Harry and Gary for packing
this
material up and returning to MIT. Prepared a few renderings for Mark
and
Gerry of building options to house the ROBIE telescope. (Jonathan Kern)
LDAS:
* Finished reading DB2 Replication Guide.
* Set up test replication configuration between two databases on the
same host. Currently working on setting up replication configuration between
two databases on different hosts. Next thing to test after that would be
whether BLOBs are supported by replication tools.
* Installed more RAM into beowulf nodes.
* Running burn test on them.
(Igor)
General Computing: This week has been spent preparing the new building for the LSC meeting (networking, audio visual, etc.). (Shannon Roddy)
Other: Preparation of the new building to host the LSC meeting is nearly complete on the renovated side of the building. The new portion of the building still has considerable work left to do and will not be available for breakout sessions. The areas that will be available are the auditorium, the upstairs office/LDAS area, and the large open lab/shop area behind the auditorium. The available areas are now nearly ready for use, although a lengthy punch list of work to complete following the LSC meeting is being prepared. The audio visual eqpt contractor is now installing equipment with the expectation that this work will be complete near the end of next week. The roof shelter for the patio area is also being erected and should be complete by 4/10. Next week will be a very busy week as we take care of many small tasks required to make the building useable for the meeting - installing temporary power and network connections in the staging building open area, testing the network link from the new to the old building, installing mini-blinds in the areas used for breakout sessions, testing the network, testing the sewage lift station, testing the A/V, and setting up screens, tables, chairs, and dividers as necessary of the meeting.
| Installation&
Commissioning:
Livingston |
Other Science/EngineeringActivities:
Issues/Concerns |
See also the Installation web page
The EO shutter on IST10 (antisymmetric port) was reinstalled and checked out. Once we gain some experience with it and confidence in its operation, we will be able to increase the amount of light on the AS photodiode and start to decrease the noise above 1kHz.
Mike Landry and Paul Schwinberg have continued to work on the WFS. They have completed their electrical checkout of WFS 2-4 and have stared the optical tests. They used the Beamscan to measure beam waists and checkout the Guoy phase telescopes and to find and eliminate some clipping on IST7. With those changes they have started the dither tests on WFS2-4 with better reproducibility and agreement with what they expect. Closing the loop with WFS4 is next.
Bill Kells
Trying to get in an analysis of the "scatterometer"
work I have started at LHO.
Arm cavity buildups of 1200 have been seen and the spots are very stable on the optics. No noise spectrum yet, but we are ready to start working on that.
Richard and Josh have been measuring the noise at the end stations. They have characterized the dewhitening boards, and are working at understanding a MHz oscillation in the anti-image boards. They reduced the 60Hz pickup in the suspension coil driver by moving the 180V power supplies to the other rack.
The test was carried out at only the end test mass of the y arm cavity and the control signal to hold the cavity resonance was reduced by a smaller factor than the reduction in Q! since some of the cavity longitudinal motion arises from the input test mass motions. The technique looks promising as part of the solution to reduce the influence of the seismic motion at Livingston. The Project is being asked to seriously consider the installation of fine actuators on the input test mass isolation stages.
Wavefront sensors: The inhomogeneity of the signals in the wavefront sensor quadrants is due to wavefront distortions. The distortions are most likely due to the electro-optics shutters.
Common mode servo: Work continues on measuring the parameters of the frequency control servo and laser to incorporate in a model of the common mode servo.
Conversion of angular to longitudinal noise: The work to determine the optimal location of the beam on the test mass optic to reduce the noise coupling continues. The displacement of the center of area from the center of mass due to the wedge in the optics is being studied.
The team is assembling documentation for the pending Design Requirements and Conceptual Design Review. Early next week the date for the review will be set (sometime after the LSC meeting).
Joe Giaime and Rana Adhikari (see elog entry; username reader, password readonly) were able to suppress motion at the first two BSC stack resonances using the ETM fine actuators in the existing seismic isolation system. We are looking into installing fine actuation and control systems onto the ITMs as well, as an interim measure before the seismic upgrade project is installed.
External Pre-Isolation
Hydraulic External Pre-Isolation (HEPI)
Further testing with the DYP-2S differential hydraulic servo-valve
at Stanford indicates that turbulence cannot be tripped (with shock or
vibration) with the newly designed valve orifice for flow rates <= 20
ml/sec. The nominal design point is 5 ml/sec. Turbulence can be tripped
by introducing bubbles into the flow stream (a condition that is prevented
in the design). In this case the transition from turbulence back to laminar
flow occurs at 12.8 ml/sec. We feel we now have a workable servo-valve
from the stand point of maintaining laminar (quiet) flow. One potential
problem remains with the valve. There is a 550 Hz oscillation that can
be heard (with a stethoscope). This is probably OK, but could limit life
through fatigue. Not sure yet if this is a mechanical or an electrical
cause.
Recent testing led us erroneously to think that the aluminum cylindrical slug introduced into the bellows volume was pushing down the actuator resonance. Further testing shows that this is not the case. This is good news since we plan to use this volume to mount an L4C geophone.
The hydraulic bypass network will be available from the machine shop this week and will be added to the new actuator on the Stanford test stand for characterization.
Electronics & control requirements for the pump station have been drafted and are under review by the team.
SEI Hydraulic Pump Station (Ken Mailand)
CONSTRUCTION: Materials and Components have arrived [95%] for the assembly of the test pump station. Filters, valves, fittings and heads are here, assembly has started, accumulators are on hold from jeremy, a dimensional description ref. part number is now known.Electro-Magnetic External Pre-Isolator (MEPI)LAYOUT DRAWING: I have started a Solid Works layout drawing of the location of the components on the granite base, and configuration of the plumbing. I have an isometric pencil drawing showing all the components.
MANIFOLDS: Station noise filter, and test load design have been started, configuration tbd.
LONG TUBE RUN: Assembly has started on the support frame Test taps at the 60 foot mark [shortest run] and 180 foot mark [Longest] one test load valved to two lengths.
Internal Stack Damping
Final part detailing is underway by mechanical deigners in Italy. Fabrication
of the parts will begin soon.
Szabi Marka
Shape Memory Alloy (SMA) actuators (for the STS-2s in the active isolation):
I succeeded to produce a sample of torqueless spring, which eliminates
the potentially harmful effect of twisting the ribbons of the locking mechanism.
It seems that the new spring is as strong as the simple springs we used
before. I ordered SMA wire material and I am preparing for the durability
tests.
There is no instability for coupling TEM00->TEM00 because in this case sidebands are excited both for higher and lower frequencies. Such a symmetry is broken for other couplings TEM00->TEMmn and if the spacing in the frequency domain between these two is the frequency of the interaction that caused the excitation there may be instability when the excited TEMmn is at lower frequency than the pumped mode TEM00.
I also started reading the cure Prof. Braginsky brought us.
Bill Kells
Erika and I have begun explaining the results we have on SB imbalance
with a wider audience :e2e group and with P. Fritschel who we will meet
with Friday to get at AdvLigo design issues pertaining to this. We are
outlining a true paper on these issues.
The MELODY paper has been wrestled out of review and we are discussing final touches to get it into Journal review.
Lee Cardenas
Contamination Cavity # 2.
Still locked. We are putting 120 milliwatts of power into the cavity and having 75milliwatts of transmitted power. Quick ring down measurement made, signal lasted ~32us. Laser frequency shift measurement in progress.Contamination Cavity # 1
This cavity has been being upgraded and It is locked as right now.
Helena Armandula
The DI water unit in the OTF lab has been serviced.
Lee Cardenas
I have received a complete and new 126-MOPA 10 watts laser from
Lightwave. Misc tasks (more orders placed, list of future needs and maintenance
items & cost for PSL & OTF lab on going)
Hongyu is debugging the new EDCU; hope we can start putting it thru serious testing next week.
Rolf Bork
Primarily Lori and I spent the last week writing the new version of
ASC code. This includes the new standard filter modules and addition of
WFS5. A version which would run on LHO2k is complete, less coordinated
switching of hardware/software whitening/dewhitening filters. I am presently
adding the optical lever servos needed for LLO4k, to be installed week
of Mar. 18. Software for both LHO2k and LLO4k should be complete by Friday,
with testing all next week.
New pentek timing modules with delayed DAC clocks and interrupts should be ready for testing with software tomorrow. Lori will work on this, testing the board and then developing software functions which can be plugged into present systems which use Pentek modules.
In addition to the RAID system, Stuart Anderson has installed an ATM interface which will allow high speed access to the HPSS system and a tape robot. The price for this extra connectivity is that saiph will be used for tests of streaming data from HPSS, but this should have limited impact on availability and provide direct benefits to the saiph user community. With the large disk cache and a high bandwidth connection to the LIGO raw data archive, saiph should become a primary detector characterization platform.
I have also tagged a new release of the GDS software. This includes many recent improvements as well as Daniel and Masahiro's event analysis classes and a portable version of the diagnostic test tool. This provides more complete standard version of the GDS software and will be available for testing in the "new" directory.
The joint LIGO/VIRGO document describing the results, conclusions and
recommendations will be available early next week from the DCC: LIGO-T020036-00-D,"Measurement
of the time offset between the LIGO and VIRGO Data Acquisition Systems"
Alain Masserot, Benoit Mours (LAPP/VIRGO) and Szabi Márka (LIG0)
In response to widespread requests from the community, we are currently
reworking our schedule at the TNI. Our original plan was to install
sapphire optics early this year, then debug the instrument and report
on
our measurements. However, the response to our review of Feb.7
showed
that there is significant interest in thermal noise measurements in
fused silica, so we decided to postpone installation of sapphire.
We
now plan to debug the instrument with fused silica in place and install
sapphire later. We are putting together a formal schedule that
reflects
this change in direction.
Many thanks to the review committee, Joe Giaime, Gregg Harry, Sheila
Rowan, David Shoemaker, and Stan Whitcomb, for providing useful and
interesting feedback on the project at this stage. We would also
like
to thank Bill Kells for some very insightful comments.
Eric spent two weeks at the NAOJ in Tokyo, working on an RSE experiment
with Seiji and his group. He had a wonderful time and learned
a lot.
Many thanks to Seiji and company for hosting him!
We have also identified a major systematic noise source in the photothermal
experiment, arising from back-reflections of one of the beams through
the Pockel's cell. We plan to fix this by adding an extra Faraday
Isolator. The process of making space for the FI gave us an opportunity
to recheck the mode matching.
We found that it was extremely difficult to locate the first beam waist
outside the laser, as beam is too powerful for the beam scanner, and
it
is ergonomically unfeasible to put ND filters between the scanner and
the laser. It should come as no surprise that reducing the power of
the
laser from 500 mW to 10 mW also seems to change the beam profile, so
we
didn't follow that avenue very far. Lee Cardenas suggested that we
make
a telescope in front of the laser with two lenses, move them around
until we see a well-defined waist, and then use that waist in the
starting point in our calculations. This worked.
We are presently re-building and realigning the photothermal experiment.
While doing the mode calculations, we verified and documented our
group's mode-matching calculator (LIGO-T020035) spreadsheet, which
has
been submitted to the DCC. We think this will be generally useful to
the
LIGO R&D community.
no report
Simulation and Modeling (Bhawal)
E2E for LIGO I meeting
--------------------------
This meeting between hardware experts and simulators was held on Tuesday,
5th March.
Erika and Bill Kells presented their FFT & Melody work on sideband
imbalance. Biplab
talked about the E2E studies of transition from cold to hot state;
The conclusion is
that the IFO remains locked as the transition happens. The details
of the intermediate
states are being analysed. Matt reported progress in development of
the LIGO I e2e model
with details of subsystems included. Hopefully it can be used during
the upcoming E2E school
at LLO. Hiro reported about inclusion of seismic noise data in LIGO
model; Biplab will
coordinate the data taking with the help of Robert at LHO and Sany/Ed
Daw at LLO. Luca
talked about matlab and e2e model for LSC and use of e2e in calibration
studies at Hanford.
LIGO I simulation system
--------------------------
(Matt) I have finished construction of all essential sub-systems, but
testing awaits
some fixes in modeler. I have begun documenting the system and
listing needed parameters.
Code maintenance
-----------------
(Ed Maros)
1. Worked on compiling e2e software for CACR systems.
2. Worked on documentation.
Alfi
----
(Bruce)
- Completed rework of graphics update messenging from the bookkeeping
to have a more clear and robust methodology.
- Added version to help menu.
- Bulletproofing in the parser to deal better with poorly formed
box
files.
(Melody)
- Implemented grid placement for node widgets for alfi5. Also
worked on
several
other issues, such as the nodes' size and the settings
dialog.
LIGO Data Analysis System
Software Systems (Blackburn)
Kent and Phil met to rework the design of the tree structure used to
capture the state of the disk cache for the diskCacheAPI. The previous
design did not allow a one to one mapping between information in the
cache and files on disk. This required searching the file system for
filenames under most circumstances. The design proposed by Kent has
eliminated the need to search the disk.
There is a new issue with the frameAPI running on the new Sunfire
server's integrated into LDAS system over the past couple of weeks.
The
code will randomly exit on this hardware platform even though this
behavior has not been seen on the Sun 450 servers. There may be a bug
in our code of a thread nature which does not manifest itself except
on the faster Sunfire hardware, though I haven't eliminated the
possibility that there is an issue with the newer hardware platform.
Several core files were produced by the dataConditionAPI this week
which clearly identified a source of some of our thread safty issues
in that process. We are currently investigating the possibility of
rewriting the string class found in the standard library to remove
the
thread issues which were identified in the core files. These core
files were produced when roughly 5000 jobs were went through the
development version of LDAS in one night.
The dataConditionAPI was also modified datacond API to use a new output()
action and removed all references to intermediate() action. The final
product of the dataConditionAPI must be specified now using this
output().
controlMonitorAPI and utilities: Fixed coreWatch email problem; Fixed
tape control procs to support running tape control on any of the ldas
machines; Fixed problems with graphic display of job and database
statistics.
metaDataAPI and eventMonitorAPI: Added debug statements and updated
the
metadata macros to ensure a return to the manager; fixed eventmonAPI
handling of invalid -mddapi (multDimData) option.
Began slight rework of handling of the detector geometry information
within the dataConditionAPI. This will provide a single implementation
solution instead of a continuous evolution require by the previous
approach.
The documentation links continue to be cleaned up. We only have about
half a dozen open link issues to solve in the LDAS build and install
scripts.
Nightly builds have become much more stable this week. We are back
over 50% success in the nightly builds after a spell of several
weeks of problems associated with our migration to beta release
structure and support for the new diskCacheAPI.
The support for the detector geometry data objects in the wrapperAPI
have been completed. All unit tests are passing in stand-alone mode.
Once the dataConditionAPI is supporing detector geometry, we will be
in a position to perform system tests.
Had to make a slight change in the lightWeightAPI to support the new
FrDetector data objects in ILWD and XML objects. This has been unit
and system tested.
We perform several tests to evaluate the memory expense of starting
up all the lamd processes at once with LDAS starts up. The ~350KB
per search user is minimal on the Beowulf so will will make changes
to the mpiAPI to make this the standard usage model for lamd and
thereby reduce start up times for search codes by several minutes.
Continued working on interpolating filter for datacon API. As per Sam's
suggestion have changed this to a class - necessary because for
successive
interpolations of the same data the interpolator needs to remember
it's
state - it's state is nothing more than the state of a linear filter,
however it's probably best to specialise to an "interpolation state"
because only very special linear filters are suitable for interpolation.
Time-shifting by whole numbers of samples will be handled by exact
means.
Philip met with John Whelan and Ed Maros to discuss communication of
physical units from frames to wrapper DSO's via datacon API. Draft
proposals were circulated to the datacon API mailing list for comments.
Philip met with Phil and Isaac to discuss extending the time-range
syntax in -framequery option to include nanosoconds, in the form:
secs_0:nanosecs_0-secs_1:nanosecs_1
Began implementing hierarchical archiving of log files to avoid problems
with extremely large numbers of logs accumulating in the single LDAS
archive directory.
Added diagnostic to 'reattach' function of generic API to detect
malformed
return values from API's as early as possible. This was done to speed
up
the development of new user command macro code and to detect unhandled
return paths from LDAS API's, which were formerly difficult to detect
(producing the 'unexpected socket closure class of bugs).
Modified startup code of API's so that ::RUNCODE is now derived from
::LDAS_SYSTEM instead of being a resource variable.
Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* Continued HPSS migration from Redwoods to 9940s.
* Working on procedure (with written policy statement) for closing HPSS
accounts.
* Have left messages with Sun trying to get some movement on the Ultra
10 issue.
So far haven't heard back from anyone.
* LIGO SAN testbed:
* Set up a single fabric
with both switches in it (finally!).
* Interswitch links work.
* Now working on zoning issues and educating myself about SANs.
* Helped Al Wilson with some Big Brother/sendmail problems.
(Al Wilson)
* Adding the final tweaks for the release of Big Brother to the sites.
* Testing the changes on kitalpha for the cfdef5 script. Also looking
what
cause the def file to break under the new script.
(Stuart Anderson)
* Upgraded the main LDAS server (ldas-sw) and all of the Solaris machines
in the LDAS-DEV system to the latest and last Solaris
8 Maintenance
Updated (sol8mu7).
* Verified that Fiber Channel works over a 650m run of 50um fiber at
Caltech.
* Reduced the size of /ldcg by 50% on the LDAS-DEV system.
* Explicitly disabled the use of SSH version 1 protocol in the LDAS-DEV
system.
* Furthered the configuration of the GC SB1000 (saiph) machine at Caltech
as a DMT/frame-archive/reduced-data/GRID server:
* Installed and configured
an ATM OC-12 connection to the
Caltech High
Performance Network (hpn.caltech.edu) for
high-speed access
to the HPSS archive.
* Moved the Caltech LDAS
30-slot AIT-2 robot from the LDAS-DEV system
to saiph.
* Installed and found to be not very interesting the recent beta version
for the next RedHat Linux release.
* Diagnosed a 40MByte/day memory leak from picld on the new SunFire
V880
servers. Sun already has a patch (110849-10) which has
been verified
to solves this problem.
MIT
---
(Keith Bayer)
* Applied patches to Sun boxes.
* Added ldas-mit alias as list for ldas related activities at MIT.
* Added webpage/area on emvogil-3.mit.edu for mit ldas related documents.
Livingston
----------
(Igor Yakushin)
* Finished reading DB2 Replication Guide.
* Set up test replication configuration between two databases on the
same host.
Currently working on setting up replication configuration
between two
databases on different hosts. Next thing to test
after that would be
whether BLOBs are supported by replication tools.
* Installed more RAM into beowulf nodes.
* Running burntest on them.
Hanford
-------
(Greg Mendell)
* Initiated work with Richard to obtain quotes for the purchase and
installation of 50 micron fiber between the Mass Storage
Room and the
LDAS room in the new building.
* Contacted Ed Chargois about shipping the LHO beowulf cluster to
Caltech, once we are ready to replace it.
* Applied patches in ldas:/export/mirror/sun/solaris8-patches/sf880
to
the LDAS SunFire 880 server, and the patches in
ldas:/export/mirror/sun/solaris8-patches/mu7 to all LDAS
Sun servers at
LHO.
Data Analysis Activities (Lazzarini)
Mendell:
1) I have updated my dso to store SFTs as COMPLEX8 rather than as
COMPLEX16 data.
This reduces their size from 67.1 MB to 33.6 MB per SFT. I have run
my
script to produce new sample SFTs from the H2 LHO E7 data. I have also
tested that the new dso still produces correct out for known input
and
that the sample SFTs produced from E7 data contain the same output
(up
to round-off error) that they contained in the old format. For those
with permission to do anonymous ftp to LHO, the sample SFTs can be
downloaded this URL:
a) anonymous ftp to ldas.ligo-wa.caltech.edu.
b) cd output/sfts/sft_H2_test030702
c) ls will show these SFT files:
H-P_NORMAL129219_KPD_SFT_H2-693633680-2048.gwf
H-P_NORMAL129227_KPD_SFT_H2-693635728-2048.gwf
H-P_NORMAL129233_KPD_SFT_H2-693637776-2048.gwf
H-P_NORMAL129251_KPD_SFT_H2-693639824-2048.gwf
I am still waiting for the best clean_locks to be identified to start
generating SFT again.
2) Discussed with the AEI group modifications to the LALDemod function
that they have proposed to allow shifting the output frequency bins
to
exactly targets known pulsars frequencies.
PS. When things have not been too busy I have been writing an R-mode
paper based on research I did with my SURF student last summer.
Creighton:
Integrated inspiral simuation code into FCT DSO to perform tests of
detection efficiency.
General Computing (Wallace)
MIT:
(Keith)
-Installed 2nd hard-drive mirror and disksuite tool on pendragon.mit.edu
-Installed 2nd hard-drive mirror and disksuite tool on emvogil-2
in preparation for cutover to new e-mail server 'ligo'
this weekend.
-Looking to purchase backup hardware for main mit ligo file server
-Building logging server PC out of spare parts
-MU6 patches on more boxes
Livingston:
(Shannon)
-This week has been spent preparing the new building for the LSC meeting
(networking, audio visual, etc.). Performed some preliminary checks
on
the
Video conferencing and working out some logistical issues involving
that
service.
Hanford:
(Christine)
- The Sun technician was out on Tuesday and is returning today trying
to
fix a brand new Sun Blade 100 with a bad power supply that overheated.
So far we have replaced the power supply but the computer still won't
boot. The technician is due back today to replace the CPU, memory
and
mother board. The technician is also working on an Ultra60 that
has had
multiple problems since it was purchased a little over a year ago.
The
technician has already replaced the mother board on the Ultra60 and
is
now going to try replacing the CPU and memory.
CIT:
(Mike)
-Setup two monitors and added more memory to two computers in the 40M.
-Helping Larry with some testing on foundry switches.
-PC swap for Cindy:
I backed up her old computer then transferred data to a new Dell computer
that I loaded with win2000 and G.C. software.
-Cleaned up a couple of laptops, reloaded another and backed one laptop
up
to get ready to ship to Livingston.
-Performed some maintenance on one of the plotters by changing chemicals,
print heads and print head cleaning kits.
-We were having a lot of problems with the plotter drivers for Autocad
2002
running on a WIN2000 operating system; it turns out that the HP drivers
listed on their website are not the correct drivers for a DesignJet
1055CM.
You must use the drivers on the 2002 AutoCad cdrom.
- Continuing to set-up new Win2K PCs
(Veronica)
- LIGO website: updated MOUs. Reformatted and installed the new LIGO
Charter webpage. Installed February newsletter. Working with Mike and
Larry on security for the Elba 2002 GWADW webserver.
- CaJAGWR: videotaped the talk by Nobuyuki Kanda. Compressed and posted
the video at the CaJAGWR Seminars page. Making organizational
arrangements
for future seminars.
- Edited images for the LDAS website.
- Made a package for staff to post publications lists online: wrote
a
template htm file and provided a finished example.
- Working on the webpages for the Advanced Suspension group.
(Lisa)
-Changed the mount points on acrux: disabled /var/mail automount and
implemented a loopback virtual file system instead.
-Worked on resolving some of the printer configuration stuff that still
points
at acrux.
-Got the rma disk replacement for electra up and running. The
disk was
needed
in LDAS, so I swapped out their bad disk.
-Installed patchdiag (in /opt) on all of the servers.
-Installed emacs on the suns in the 40meter martian network.
-Discovered that my ssh2 agent set up is incompatible with LDAS OpenSSH
configuration. This making it difficult to automate patch updates on
those
machines. Working on resetting the configurations on the GC systems
that
have
come over from LDAS.
(Larry)
-Worked on a number of procurements. There are a lot of hoops to jump
through to
get the 2 for 1 offer from SUN, that adds a couple of days to the
process.
Cleared up some issues with Dell. They are no longer supporting the
60GB
drives
for their laptops and that created some issues with existing orders.
-Debugged a number of problems with the new installs. One problem is
the
powerd
on the Solaris boxes does create a number of problems on the SUN Blade
1000
units. Presently, the daemon is disabled until the patch has been
installed and
checked out. Also, resolved some of the hardware installation
procedures. The
files needed for some hardware is now on two different CD's, no longer
on the
same CD.
-Spending time on working out the kinks for the new network backbone
system. So
far there have been a few problems that have put things behind schedule.
Mostly,
configuration issues. There are a couple of new edge switches that
Foundry is
still getting the bugs out. They don't affect our setup other than
we
have to
change the configuration of the units to avoid the problems they may
encounter.
-Resolved a couple of web issues. Most with the conference servers.
-Worked with the LIGO Sysadmin group and checked out the VRVS
and VNC
again. It
is looking a lot better and more stable.
-Cleaned up a few more e-mail issues. Most were logistics dealing with
mailing
lists.
(Bruce)
Gnats maintenance:
(0.25 days)
- Ilog maintenance.
From Peter King:
AdLIGO PSL
I added some extra wires
for grounding the unused channels of the
3123A and 3113 digital-to-analog cards but this did not remove the
voltage
offset reported last week. Leaving the possibility of a software
calibration not being performed. The version of the device driver
will be
checked.
The power amplifier, power supply and refrigerated bath/chiller ordered
from Lightwave for the high power laser experiment at Stanford arrived
late
Friday afternoon. These were shipped to Shally Saraf at Ginzton
Lab.
From: GariLynn Billingsley <Billingsley_G@ligo.caltech.edu>
Goodrich has submitted a plot of the sapphire transmission after compensating polish. It can be found for now at:DCC/OUT/C020137-00.pdf
CSIRO has completed the compensation feasibility study. They find Ion beam etching to be very promising. Compensating coating is also promising, fluid jet polishing does not appear feasible.
A very brief summary of findings in Ion beam polishing:
5mm diameter etch was demonstrated, the vertical rise from the floor
of the etch was ~ 90nm/mm. This is the smallest diameter etch attempted.
The report draws no conclusions as to the viability of an even smaller
footprint.
The microroughness of sapphire appears to improve consistently with ion beam bombardment. <1 Å microroughness was demonstrated repeatedly, with values down to 0.54 Å.
Removal rate can be varied from 0 to 50 nm/min, the accuracy in achieving the depth they wanted in these trials was within 2 nm.
For now, please find the draft report in: DCC/OUT/C020136-00.pdf
From: Mark Barton <mbarton@ligo.caltech.edu>
This week I concentrated on validating the provisions for wire bending
elasticity and thermal noise in my model. Because I wanted to try many
different combinations of damping I made a number of refinements to
the
calculation procedure to make it easier. In particular I unified the
way
the damping and the other numerical values were specified, so that
it was
possible to override the default damping specification. I also made
the
existing provision for using previously saved results more selective,
so
that it could calculate the damping afresh but use saved results for
the
potential matrices (which are independent of the damping and take most
of
the computation time).
I then did a toy version of the model with only a single mass (<http://
www.ligo.caltech.edu/~mbarton/SUSmodels/toy/>) so that I could compare
the calculated thermal noise curves with analytical values. The curves
have the right general form but I haven't checked the scaling
ctorrie <ctorrie@ligo.caltech.edu>
Prototype quadruple pendulum, MIT
As mentioned last week we planned to discuss the changes required for the quadruple pendulum. Both Rich and Norna seem happy with the analysis and Rich is going ahead, at least, with the first set of revisions.
Suspensions and other..
I have been designing some of the pre-fixtures for the LIGO-MC suspension in SOLIDWORKS.
I now have a 3rd set of drawings for the Physics Machine Shop, checked by Janeen. Parts include clamps and fixtures that we plan to test and use when Norna visits again in April.
On Monday afternoon I had a visitor here from COSMOS to give a demonstration on their non-linear package. I was very impressed. COSMOS runs with SOLIDWORKS so there was NO interface problems. I asked them to run some analysis and expect to here from the application engineer in the next few days.
I have been continuing with my workshop classes in the Physics Machine
Shop.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Advanced LIGO Coatings
I have not received any absorption results on the 27 parts coated by MLD that I sent to Stanford .
However, some progress is being made. They started measuring the MLD
coated optics using a coated substrate I provided as a secondary standard.
The primary standard is a surface-coated fused-silica neutral density
filter from Newport which they can measure using conventional techniques.
The absorption measurement they obtained on this coated part was close
but less than the number I reported to them.
For additional information about this report, contact sanders@ligo.caltech.edu