|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday January 14, 2002 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items: E7 REPRISE
A good sign
is that sometimes it actually gets boring... The E7 run
status...
: Notes from the LSC Executive Committee meeting December 21, 2001
Collaboration with gamma ray astronomers
----------------------------------------
Encouraged Sam Finn to explore correlative data analysis of gravitational
wave bursts and gamma ray bursts with the gamma ray astronomers on
SWIFT and
at Goddard Space Flight Center. Sam is requested to report to the Executive
committee on the arrangements planned once they begin to be formulated.
LSC Computing Oversight Committee
----------------------------------
Al Lazzarini reported briefly on the efforts to bring this new committee
together. Several committee members suggested that we look at the way
high energy physics collaborations have set up comparable committees.
Initial Discussion of Data Analysis White Paper
-----------------------------------------------
It was generally agreed that the new version of the White Paper is
a
significant improvement over the first version.
The discussion centered on two fundamental issues raised in the
White Paper.
1) the way data analysis groups evolve in the Collaboration
2) the mechanisms for making the entire Collaboration current and aware
of the
work in the data analysis groups.
1) The discussion was quite general and not specific. I will assert
some
executive privilege as the correspondent. There is support and need
for
a "permanent" framework within the LSC that deals with continuing issues
such as strategies for detector characterization, the coordination
of software
and software libraries, methods for maintaining software standards.
On the
detector hardware the technical development groups have and are serving
the functions the Collaboration needs to coordinate and guide the development.
There is no need to change many of the elements of the framework that
has
been established.
The actual data analysis leading to the scientific results
has proved more volatile. ASIS has accomplished its initial task of
getting
the ball rolling in developing analysis software but it was necessary
(so I felt)
to organize the initial analysis more broadly than the Collaboration
was
doing, in particular, to include experimenters and theorists
in groups organized around initial searches for specific types of sources,
the
so called upper limit groups. These were encouraged to use the proposal
process
in which they outlined their demands on resources and gave a prospectus
of what
they were going to accomplish and a schedule. The proposals also gave
the members
of the Collaboration an insight into the work and allowed a more informed
way for
members to decide whether they wanted to join. The upper limit groups
are designed
to accomplish a limited goal - the analysis of the initial data with
the
instrument still not at design sensitivity. From these groups will
arise more
sharply focused efforts through the proposal process. The upper limit
groups
are intended to be transient and not permanent.
Similarly a strategy had to be developed to fullfill the need to test
the LDAS
and the Collaboration developed software. There had been a group designed
to
do this in the LSC but at the time it was premature. The need was met
by the
Mock Data Challenge groups, formed ad hoc, and these have been successful
in getting the tasks done.
The lesson learned is that we need both a "permanent" structure and
ad hoc
groups to fullfill our needs. As long as we keep our senses and do
not get
bound up in rigor (mortis), things will continue to work.
2) The more central problem is that with the accelerating pace of the
data
analysis effort and the reality of a large amount of data going into
the
archives, we need, as a Collaboration, to become more tightly aware
of what
is going on in the various analysis groups and to be able to really
discuss
and understand the scientific case for the various investigations.
This is
the second and more substantive issue. It is clear that two meetings
per
year is insufficient to deal with what is expected to become a fast
moving
part of the Collaboration's activities. We decided that a web page
be
dedicated to the activities and posting of the results of the upper
limit
groups. After preparation of an agenda, sent to the entire Collaboration,
the Collaboration would meet on the telephone monthly for a presentation
and dicussion of results by the various groups. Working
papers for discussion would be posted on the web site.
TAMA Coincidence run
--------------------
TAMA is hoping to be ready with a power recycled interferometer in
the Fall
2002 and has approached the LIGO Laboratory to arrange for an exchange
of
data to be analysed in coincidence. Professor Kanda (?spelling?) is
visiting
LIGO Caltech during the spring and one of his functions will be to
help arrange
this data exchange.
The committee is in favor of the laboratory exploring the possibility
but urged
caution that a protocol be established for publications and the role
of the LSC
in this exchange be defined.
Numerical relativists coupling to LIGO
--------------------------------------
Barry raised an issue brought to him by Kip Thorne about the role of
the numerical
relativity community in LIGO and LISA. A way to give relevance, increase
the
funding and the activity in numerical relativity at both NSF and NASA
is to bring
the source simulation work closer to the needs of both LIGO and LISA.
Kip is asking
us to invent a way for the numerical relativists to be involved with
the LSC but not
to be tied (at least initially) to deliverables. Several ideas were
talked about
at the meeting such as associate membership in the LSC or observer
status or full
status but with a different type of MOU. In any case, all were in favor
of trying
to figure out a mechanism to bring numerical relativists into closer
contact with
the LSC.
Susequently, Barry, Gary , Kip and I had a meeting in which we decided
to initiate
coupling to the numerical relativity community at the March LSC meeting.
We will
ask the new numerical relativity group at LSU to help organize a meeting
of the
numerical relativists on the first day of the LSC meeting. The first
day of
the LSC meeting is closed for LIGO I data analysis groups. On the second
and open
attendance day of the LSC meeting, we will dedicate part of the meeting
to reports
from the numerical relativists on source simulation and open discussions
on topics
of mutual interest. The numerical relativists will be welcome to join
the rest of
the LSC meeting. Based on these initial meetings we will learn better
what is a
good way to make the coupling.
Next LSC Executive Committee meeting
-------------------------------------
Friday Jan 25, 2002 at 11:00AM ET (Please note change of day)
LIGO Operations--Administration
A Site Teleconference was held on Thursday, January 10, 2002. Discussion centered on the status 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.
Gary requested that a review of the status of the Joy Fan issue and spares be placed on the agenda sometime in the future.
The list of current actions revised to reflect
the status of open actions assigned through January 10, 2001 may be found
at ACTION
LIST.
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
| 01/10/02 | Packages | Faxes |
| In | 33 | 33 |
| Out | 12 | 23 |
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)
CONSTRUCTION:
SUPPORT
Progress Period from 01.04 to 01.10
Accomplishments:
We are doing a standard Annual Report for the
Construction Project as of the end of November. I have received some
contributions and am working on the report. Still waiting for material
from some groups. (This effort has been interrupted to respond to
DCAA Audit issues.)
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.
No report
E7: Day 13 of the run and the interferometer continues to operate in a stable configuration. As of noon CST on Jan. 10, we have accumulated 205 hours of recombined lock with durations > 15 minutes for a duty cycle of approximately 66%. We have been unusually lucky with seismic noise -probably because the rain last Saturday prevented logging and other outdoor operations from going on the following Monday, so the seismic environment was as quiet on Monday as it is on a normal Sunday. The New Years holiday period, and Jan. 2, were also very quiet seismically. More details have been gathered by Gaby Gonzalez in the llo elog. See: http://www.ligo-la.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=01/10/2002&anchor_to_scroll_to=2002:01:10:14:52:31-gaby
CDS: Found the disks used for disk to disk backup in the control room are getting full. Changed the backup script and ordered two hard disks for the control room machines. Updating the CDS Red Book for LLO. Working on setting up an Ultra 60 as a replacement for london. Wrote a script to monitor the temperature of the mass storage room and the CPU's, and another script to monitor the Status of the Frame Builder 2. These scripts produce web pages that are updated every ten minutes and the links can be found on the CDS web page. (Chethan)
LDAS: * Tapes: shipped 20 tapes to Caltech, 15 more tapes are
ready for
shipment.
* SunFire 880: performed a diagnostics boot, installed the graphics
card,installed Solaris 8 and configured metadevice.
* Built a standalone LDAS system to play with.
* Talked to Chethan about an fb2 status indicator for operators:he
just added it to the webpage that operators are checking.
* Wrote a perl script to extract data rate time series from log files
in a format that can be used , for example, in gnuplot. Wrote corresponding
gnuplot script to generate a postscript plot. (Igor Yakushin)
Ordered the UPS for the SF880. The Cybernetics tape library came back from repairs and is in the Mass Storage Room. After the run is over, I will reinstall the unit, return the GC unit to decatur, and test the LDAS unit. (Shannon Roddy)
GC - Set up a web page for Rai Weiss and all others concerned
where they can post ascii data files (or any file for that matter - .jpg,
.pdf,
etc) for later reference. A search form will be implemented later this
week or this weekend so that users can search in the description field
for each file and/or by date, name, etc. The web page is on a temporary
server currently awaiting a recompile of php on the main LLO web server.
Snort (IDS) is now up and running at the gateway between the LLO network
and the T1s. I have already gotten some useful information from it.
(Shannon Roddy)
Facilities: The new construction is making good progress in the auditorium and other areas that are renovations of the existing staging building. We are watching the activity closely in this area to ensure that it is available for the LSC meeting March 20. Auditorium seats were delivered this week. HVAC for the existing building will be turned on next week following completion of the fire alarm system and then much of the final finish out in this area is expected to occur very rapidly in the last two weeks of this month. The contractor has agreed to obtain a temporary use permit, at no cost to LIGO, from the state Fire Marshal so that the building may be used for the LSC meeting even if something delays the work past the present contractual completion date of March 1. Ed Jasnow is preparing a bid package for the A/V equipment and there will be a bidders conference and job walk Jan. 15.
The patio cover kit was delivered today by Polygon. The erection contractor
walked the site this afternoon in anticipation of erection. They will first
complete an engineering calculation of the footing dimension required for
the shelter anchor bolts and then mobilize. Erection and painting is expected
to require two weeks.
|
Commissioning:
Livingston |
|
During the year rollover a major glitch occurred in the two network time servers which jumped to the year 2021 which in turn affected the clocks of the computers in the control room. This did not affect the time stamp of the data but, since the LLO frame builder went down, a few hours of data have been lost. Otherwise, the only significant (non seismic) down time was due a software/hardware failure at LHO/MX.
A first look at the sensitivity revealed no surprise and it is approximately the same as before the run.
Rai Weiss
Szabi Marka
LLO 4K:
i. in the simple Michelson configuration, let the mirrors freely swing,
drive one of them at 100mHz and count fringes, thus calibrating the drive;
ii.lock the Michelson, toggle the sign of the servo so the fringe changes
from bright to dark. This change corresponds to lambda/4 and so you calibrate
the control signal;
Both methods have been automated and both methods yield very precise results, down to the 1% level (statistical). Their results, however, differ by ~4%.
The two methods differ in many ways:
1. the "sign toggling" method gives you a pure DC calibration and the
calibration range is within 1-2 lambda; the counting of fringes is performed
at 100mHz and the drive excursion is of ~>10lambda (more "vulnerable" to
the drive non-linearities);
2. offsets in the control signl need to be taken into account in the
"sign toggling" method, not the case for the fringe counting method;
3. in the counting of fringes method, you need to identify precisely
the peak position, in order to decrease this systematic errors;
There are ways to get around some of the systematic contributions. The control signal offsets in the sign toggling method can be cleverly eliminated by summing the lambda/4 calibration factors (obtaining a lambda/2 calibration, not sensitive to the offsets). Fitting the peaks, in the fringe counting method, with parabolas enables to decrease by >100x the systematic due to peak identification. All of this has been included in the analysis, making a simple calibration procedure a little more complicated.
There is, however, another systematic contribution which I was made aware of only by using e2e. A few week ago, I had decided to reproduce the fringe counting method with e2e. It's not a complicated problem, not many degrees of freedom, and it could easily be mapped in the simulation world. Perhaps I was thinking to myself what could I possibly learn from this since the problem is not very difficult (besides I know what I'm putting in the simulation). I knew, however, that I could at least test all of my analysis scripts with the simulated data.
Indeed, I did find and fix a few bugs here and there in the analysis scripts. But I was mostly disturbed with the result from the simulated counting of fringes method. E2e had been setup in a simple Michelson configuration, with a digital filter modeling the suspension actuator and no seimic noise injection (yet). The calibration value obtained by fringe counting at 100mHz was 2% off of the value I had put in the simulation. My chase for a bug began, not realizing that the simulation was correct: I always assumed that the value of the mechanical tf (modeling of the mechanical response of the suspension, described as a simple harmonic oscillator, resonance at 740mHz and effective Q~10) at 100mHz was the same at DC. Not so. The tf magnitude at 1mHz differs by 2% from the magnitude at 100mHz.
Experimentally, we have observed a ~4% difference between the two methods, fringe counting at 100mHz and sign toggling at DC. The results from e2e show that there is already a 2% difference between the calibration at 100mHz and 1mHz due to the mechanical (pendulum) tf. If an accuracy of 1% is desired, this systematic contribution also needs to be addressed when fringe counting at 100mHz.
The interferometer has been operating since Thursday afternoon Dec
27 for the E7 run. No changes in the configuration have been made since
then. The configuration is as a recombined Fabry-Perot/Michelson interferometer
without power recycling.
During low seismic activity the system is operating at a sensitivity of 1.5 x 10^-19 strain/sqrt(Hz) between 300 to 2000 Hz and has a best performance of 8 x 10^-20 strain/sqrt(Hz) between 500 to 1000 Hz. The duty cycle at night has been over 80%. During the day the system is operated in a robust mode able to hold lock against most of the seismic noise but at a cost of a factor of about 10 to 20 in sensitivity. The duty cycle is about 50%. The numbers quoted will be refined when the data is analyzed.
The data acquisition system has been working with a duty cycle over 90% and we believe that the data is being acquired and stored without significant errors. Again, the final assessment will be made when the data is analyzed.
We have been unusually lucky with the low level of seismic activity during the day, that is until this morning (Jan 3,2002) when logging activities resumed.
Immediately after the E7 run work will resume on improving the interferometer performance. In the coming week the acoustic enclosure to reduce the sound induced frequency fluctuations of the laser will be installed. Work will continue in chasing the sources of the noise in the A/D and D/A system and the revised optics for the antisymmetric wavefront sensor will be installed.
Peter King
I have been going over the reference cavity temperature control software. Previously I thought that perhaps the gains were set too high. After reducing the gains by a factor of 2, it was reported that it was now too low. A software modification was made to make the gains adjustable but this resulted in a temperature jump of a degree. Whilst it is not clear what the problem is, I think there might be some problems associated with the event handling in the state code.
This operation doesn't come straightforward. I have been thinking of what goes on in other fields where the time reversal brings up a dual representation of SU(N). The formalism I started to study exactly applied to a simple example of a tilted mirror that Bill pointed me out and this made the problem very clear and interesting to discuss.
Bill Kells
Continued work on modeling (with Erika) to get a final write up on
the work we have been doing on the general properties of SB balancing in
Michelson like interferometers.
In the meanwhile, I am preparing the other contamination chamber. The new NPRO S/N 316 laser installed and power output measured to be 740 mW . (The laser power supply shows only 403mW of power and i couldn't change the reading of the digital readout as i turn the knob to match the power measured. I have contacted Lightwave for technical support) The Optical height is 141 mm. Beam Scan of the npro laser, gathering the optics needed to complete the path and complete alignment efforts in progress.
Scatterometer.
Steve bell and I got together and discussed the possibility to modify
the cage to hold Helena's new sapphire test mass (33 cm in dia. ~ 5 inches
thick and ~ 70 kg in weight). We decided to design a tool to
carry the new test mass that will be suitable to carry on and off the scatterometer
as well as for cleaning without taking off the tool. We will be looking
into the motor drives to see if it will hold the weight. Steve Bell will
be talking to Helena and Garilynn to accommodate their needs in terms to
design the new tool.
Bill Kells
Worked more in the OTF (with Liyuan Zhang and Lee) to get the primary
contamination rig locked.THere now seems to be a major problem with its
behavior. A suspicion is that the cavity left inside the Vac chamber is
hopelessly contaminated. We have decided to go ahead and examine a different
cavity.
Some spare 10-W laser shutters were requested from Lightwave, since Doug Cook reported that the one on the LHO 4k PSL failed. These should be shipped from Lightwave later this week.
Lightwave have accepted that the 40m 10-W laser (S/N #118) comes under the warranty due to the falling output power of the master oscillator. The NPRO will be replaced with a spare 700-mW NPRO and the 1-W NPRO will be sent back for repair.
The 10-W laser (S/N #104) on loan to Stanford will be shipped back to Caltech for re-tuning and re-alignment. This should take place towards the end of the month. Both the NPRO and power amplifier diode currents will be increased to their maximum, in an effort to get around 20 W of output power.
Rick Karwoski
Mohana has been coordinating the final packaging for the photon recoil calibration system, which is now mounted in a large box along with miscellaneous optics.
Mike Zucker
Continued efforts to resolve manufacturing and implementation problems
with LSC photodiodes. Contactedprincipal supplier (Perkin Elmer)
and also backup vendors; no conclusive solution yet. Revisited SNR
models to determine if re-optimization of parameters can render previously
rejected lower-performance alternates acceptable.
Lori is moving from EDCU code to updating the ASC code. What we want to do is provide the ASC with the same capability to download filter coeffs as in the LSC system, rather than a code recompile. Some other changes have been requested in the ASC code and will be incorporated as well.
no report
LASTI (Bayer, Fritschel, Goda, Harry, Laliberty, MacInnis, Mason,
Mason, Miller, Mittleman, Ottaway, Phinney, Rollins, Shoemaker, Zucker)
=======================================================================
VACUUM ENVELOPE:
We resolved our viewport supply problem by cannibalizing ports off
the
old PNI reference cavity chamber, and pumped down for leak checking
Monday. All ports and electrical feedthroughs passed. Unfortunately,
the repair of a small leak in the big-end seal of one of the BSC
bellows inadvertently introduced a much bigger leak in the
corresponding small-end seal. Backfilling today to take another
shot
at the repair.
HYDRAULIC ACTUATORS:
The design of the BSC hydraulic isolator is just about completed. A
spherical joint has been added to the actuator attachment point. The
actuator mounting brackets and outer housing has been completed. The
design work that remains is to provide mounting areas for Stanford's
manifold and control valve. Some additional finite element analysis
of
the housing assembly needs to be completed.
All spring attachment hardware has been designed and ordered. They will
arrive in two weeks when the first springs are also scheduled to be
delivered to Livingston.
Dan* has completed fabrication drawings for 8 parts. There are
six
remaining and an assembly drawing to create.
*Thanks to Dan Mason for
donating his winter semester break!
CDS/DAQ/GDS:
No problems to report. We ordered a bigger ethernet
hub to support more user terminals and a web-ready scope. Also
we
are experimenting with a wireless hub in the high bay.
PSL:
This week we locked the Reference cavity, a unity gain frequency of
90
kHz was achieved. We are currently looking at the FSS board to try
to
improve this. The PMC was locked last week with a throughput of 7.5
W;
we hope to improve this by improving the modematching. Beam enclosure
tubes and viewport adapters are being fabricated to prepare for locking
the beam to the suspended 1m cavity in HAM13. Due to the acute
shortage of site LSC photodiodes, we're ordering a commercial low-power
RF photodetector to get us started.
Simulation and Modeling (Bhawal)
_______________
Lock acquisition in presence of mode mismatch
-----------------------------------------------
(Biplab) After sorting out several difficulties encountered, continued
lock acquisition for the cold state starting from a not-actual LIGO
configuration that is well-matched (except small mismatch due to Schnupp
asym) with the input beam modes ..then trying to do the same including
real
parameters of LIGO which is not perfectly matched with the same input
beam
(although, when locked, coupling could remain to be high ~99%). The
procedure followed is this:
(i) lock for matched parameters.. this is almost equivalent to locking
in plane-wave approx. (ii) change parameters little bit which makes
it a
little closer to actual LIGO parameters, (iii) lock this. Note mirror
positions at locked state and use those as initial mirror positions
for the
next run with changed (again) parameter values,
(iv) continue above procedure a few times and then try to extrapolate
mirror positions at various locked states with mode-mismatch, so that
parameters can be changed with bigger steps.
It would take 1-3 more days to complete this procedure.
LSC characterization
------------------------
(Luca) Recently, I've been focusing on the LHO 4k ITMs calibration
procedure. Two methods have been used so far:
i. in the simple Michelson configuration, let the mirrors
freely swing, drive one of them at 100mHz and count
fringes, thus calibrating the drive;
ii.lock the Michelson, toggle the sign of the servo so the
fringe changes from bright to dark. This change corresponds
to lambda/4 and so you calibrate the control signal;
Both methods have been automated and both methods yield very
precise results, down to the 1% level (statistical). Their
results, however, differ by ~4%.
The two methods differ in many ways:
1. the "sign toggling" method gives you a pure DC calibration
and the calibration range is within 1-2 lambda; the counting
of fringes is performed at 100mHz and the drive excursion
is
of ~>10lambda (more "vulnerable" to the drive non-linearities);
2. offsets in the control signl need to be taken into account in
the "sign toggling" method, not the case for the fringe
counting method;
3. in the counting of fringes method, you need to identify
precisely the peak position, in order to decrease this
systematic errors;
There are ways to get around some of the systematic contributions.
The control signal offsets in the sign toggling method can be
cleverly eliminated by summing the lambda/4 calibration factors
(obtaining a lambda/2 calibration, not sensitive to the offsets).
Fitting the peaks, in the fringe counting method, with parabolas
enables to decrease by >100x the systematic due to peak
identification. All of this has been included in the analysis,
making a simple calibration procedure a little more complicated.
There is, however, another systematic contribution which I was
made aware of only by using e2e. A few week ago, I had decided
to reproduce the fringe counting method with e2e. It's not a
complicated problem, not many degrees of freedom, and it could
easily be mapped in the simulation world. Perhaps I was thinking
to myself what could I possibly learn from this since the problem
is not very difficult (besides I know what I'm putting in the
simulation). I knew, however, that I could at least test all of my
analysis scripts with the simulated data.
Indeed, I did find and fix a few bugs here and there in the
analysis scripts. But I was mostly disturbed with the result
from the simulated counting of fringes method. E2e had been
setup in a simple Michelson configuration, with a digital filter
modeling the suspension actuator and no seimic noise injection
(yet). The calibration value obtained by fringe counting at 100mHz
was 2% off of the value I had put in the simulation. My chase for
a bug began, not realizing that the simulation was correct:
I always assumed that the value of the mechanical tf (modeling
of the mechanical response of the suspension, described
as a simple harmonic oscillator, resonance at 740mHz and effective
Q~10) at 100mHz was the same at DC. Not so. The tf magnitude at
1mHz differs by 2% from the magnitude at 100mHz.
Experimentally, we have observed a ~4% difference between the
two methods, fringe counting at 100mHz and sign toggling at DC.
The results from e2e show that there is already a 2% difference
between the calibration at 100mHz and 1mHz due to the mechanical
(pendulum) tf. If an accuracy of 1% is desired, this systematic
contribution also needs to be addressed when fringe counting at
100mHz.
Alfi
----
(Bruce)
- Save, Save As, and (safe) Quit fully implemented.
- Working on edit window resizing.
LIGO Data Analysis System
Software Systems (Blackburn)
LDAS continues to execute well during the E7 run with slightly better
job loss rates than were reported last week. There was a new problem
with the metaDataAPI which appeared this week. A bug was causing some
database request threads to be miss-managed. This lead to a state for
which all request to the metaDataAPI failed. This was simple fixed
by
restarting the metaDataAPI at Livingston. A problem report was filed
on this bug and it is being reviewed by staff programmers for a fix.
Work continued on separation of the disk cache functionality out of
the frameAPI and into its own "new" diskCacheAPI. It is expected that
with no major distractions, testing of this new LDAS configuration
will
begin next week.
The managerAPI has new functionality to log timing information on a
per job basis. This information is parsed by the controlMonitorAPI
where statistical reports can be custom generated.
The db2utils packages and resource files have been updated to display
E7 dates and the 0.0.23 insertion rates at all Lab LDAS sites.
A bug was discovered in the LSTRING data type which allowed binary data
to creep into the database. This lead to the identification of several
exception handling issues in the LDAS code which are being cleaned
up.
Work continued this week on the migration of LDAS C++ source code to
conform to the stricter ANSI/ISO compliant GCC 3.0.3. This is still
in a discovery mode. LDAS has not yet adopted this compiler platform.
Hardware Systems (Anderson)
TO BE PROVIDED
Data Analysis Activities (Lazzarini)
Lazzarini attended a 3-day GriPhyN and iVDGL meeting at the University
of Florida (Gainesville).
It was the semiannual GriPhyN meeting, with the external advisory board
(the equivlaent of LIGO's PAC).
LIGO presented the status of its work on the SC01 prototype demonstration
and its Year 2 GriPhyN Plans.
Lazzarini discussed with Reagan Moore (SDSC/Teragrid Project) how the
DTF program can be used to replicate the LIGO datasets at SDSC. This was
origianally proposed as a way to provide offsite redundant storage for
LIGO data and SDSC is interested in doing this. They recently produced
a metadata catalog for the entire BaBar dataset as a 2 day exercise.
General Computing (Wallace)
MIT:
(Keith)
-Researched AP1000 point to point connection mode
-Experimented with access point security
-Configured/cabled/installed AP1000 in highbay
-Troubleshot Linux penguin computer problem
-Relocated a number of computers
Livingston:
-Support of the E7 run continues.
-More testing on the VRVS system.
(Shannon)
-Set up a web page for Rai Weiss and all others concerned where they
can post ascii data files (or any file for that matter - .jpg, .pdf,
etc) for later reference. A search form will be implemented later
this
week or this weekend so that users can search in the description field
for each file and/or by date, name, etc. The web page is on a
temporary
server currently awaiting a recompile of php on the main LLO web
server.
-Snort (IDS) is now up and running at the gateway between the
LLO network and the T1s. I have already gotten some useful information
from it.
-Had Dell come in and repair a broken pcmcia slot on my
laptop.
Hanford:
Nothing to report.
CIT:
(Mike)
-Had to update Matlab expired licenses for a few LIGO users.
-SONY Laptop, OS was corrupted; I reinstalled OS then loaded General
Computing
Software, this laptop was put back into the loaner pool.
-Repaired a PC with a hardware problem it keeps blue screening with
.VDX errors.
This turned out to be bad memory.
-Another PC with a hardware problems with IDE 1; this turned out to
be a bad
data cable. Reloaded OS and General Computing software.
-Backed up and removed data that was left behind from previous users
on Bob T.'s machine in the 40M plus uninstalled a series of unnecessary
applications such as games and things in that category.
-I helped Stuart assemble two computer racks.
-Multiple user onsite and phone support,
that included software problems, printing problems and networking issues.
(Lisa)
-Made ufsdump tapes of the root filesystems of most of the servers
-Setup some new accounts.
-Downloaded the new site license version of norton anti virus. I will
put it
on pcapps.
-Helped Larry resolve some problems with the main web server.
-More pcard updates.
(Veronica)
- Made several updates of the LIGO website (Undergraduate Programs
for
2002; Fellowships/Postdoc Openings; LIGO Publications List). The webpage
that lists postdoc openings needs to be periodically revised to check
which positions have been filled. Set up automated e-mail messaging
to
remind those responsible to check the job's status. Contacted with
Marie
Woods of MIT regarding LIGO policy on access/posting of updates to
the
LIGO web server.
- Project Science website and workshop: updated the website; made minor
changes to the Registrants database; met with Gary and others to discuss
the final details of the workshop preparation.
- CaJAGWR: posted the seminars for the Spring term.
- Met with members of the Advanced Suspensions Group to discuss a webpage
that the group needs.
(Larry)
-Worked on the documentation for the User Policy.
-Had a few discussions with Noanet concerning WAN connectivity at Hanford.
More information should be received next week.
-Worked on some tweaks with the VRVS system. That will be a continual
process
but we do have the VRVS system working at all four locations. We are
also
working on getting the procedures down for the VNC part of the video
conferencing.
-Worked a few security issues. Most were with a DOS attack on our WEB
server.
Evidently, there were a number of servers at CIT that were targeted
but so
far no major problems.
-The regular support with user issues. More support this week was given
on the
SDRC program. There will be more future work with it and the other
CAD programs
as we start switching over to SolidWorks.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Advanced LIGO coatings
"Q" tests
Received 4 thin substrates coated from Lyon, 4 thick coated substrates
were sent to Glasgow.
Samples had been coated with 30 quarter wave layers of Ta2O5/SiO2 and
2 quarter wave layers of the same materials.
These coated samples, when re-tested for "Q", will provide information
on the effects of number of layers on loss.
Visited MLD facilities.
They have expanded and added some coating characterization capabilities.
They are willing to work with LIGO.
Some coating runs with different materials are taking place at this
time. We need to expedite our absorption measurements to provide them with
timely feedback.
From: ctorrie <ctorrie@ligo.caltech.edu>
Suspensions
Norna and I have started work on some pre-prototpyes in the lab.
Janeen and I have been investigating transferring files between
different software. I have been transferring a file from AutoCAD to
Algor and she has been doing the same to I-DEAS and we hope to compare
these results.
Janeen, Norna and I have been looking in detail at the design for the
LIGO mode-cleaner suspension in particular the design of the upper
blades interfacing with Janeen's cage structure and we have been
looking again at the damping of the pendulum in the various degrees
of
freedom.
We have submitted 2 sets of prototype blades to the Physics workshop.
This is so we can compare the material from GEO to the material Janeen
got
from a company in the US.
Next week Janeen and I are at a SOLIDWORKS class all week.
Calum Torrie (and Norna Robertson)
From: Rich Abbott <abbott@ligo.caltech.edu>
1. The recently received capacitance position sensors have been
mounted in their Eurocard cages after solving an issue that required some
on-the-fly machining due to the electronics being longer than advertised.
We will power them up and begin testing which should be done in about 1
day once begun.
From: Janeen Hazel Romie <romie_j@ligo.caltech.edu>
Working on AdLIGO MC and RM suspension and structure.
Working with Calum on blade design and testing.
40m
Cleaned and baked SOS parts are due here from LHO today. Bartie will
be cleaning and baking a small alum. load this week and starting to clean
and bake a large load of stainless clamps after that. I'd misunderstood
CES last week, many of the TM suspension machined parts are done but not
stamped with part numbers. This work, along with the other machined parts
should be complete in a couple of weeks.
From: Riccardo DeSalvo <desalvo@ligo.caltech.edu>
Happy New Year everybody.
Akiteru, et al.@ Hongo
Bad News (he says)
“Hello,
As you know, we have been measuring SAS transfer function with a small
Michelson interferometer. I have a bad (partially good) news about
it.
The idea was to excite the IP with the coil magnet actuators already
implemented for local control, and detect mirror displacement with
the
interferometer, however, in the first result the measurement was
reasonable just below a few Hz. Apparently, the actuation was not
sufficient to excite enough mirror motion comparing with the
interferometer sensitivity. This was due to misevaluation of their
capability (we had reduced the actuator efficiency 300 times smaller
than that in the original configuration to meet noise requirement).
Good
side of this is that the passive chain seems to have attenuation factor
at the expected level.
We have already spent enough time to prepare the measurement, so I
am
very reluctant to abandon it immediately and hope to make it successful.
Now we are implementing a voice coil actuator to increase the IP
displacement (and that of the mirror) and will restart the measurement.
This may cause a few weeks of delay, so I would like to ask for your
understanding and patience.
In any case, the electronics for the second tower is not here and would
take a few weeks to arrive. So at earliest we should start the FP cavity
measurement at the end of this month. Best regards
Akiteru
Please send us more of these bad news!!
The actuator’s efficiency
referred in the report is reported in DCC Number
LIGO-P010026-01-D, just
accepted for publication on NIM.
Hareem @ Pasadena
Studying the cryostat programming to restart the thermal conductivity
measurements.
Bill, Riccardo @ Pasadena
Waiting for spin cast of MoRuB sample, found FeMoCB glass much lighter,
similar strength (5 GPa) and much wider temperature working range.
Have
sample to be sent to spin cast.
Hareem, Riccardo @ Pasadena
Following creep measurement. Still ongoing with the well known
idiosyncrasies.
Preparing fast and dirty ribbon Q-measurement setup.
Riccardo @ Pasadena, Gianni @ Bientina
Updating drawings of TAMA-SAS as built.
Preliminary design of composite suspension wire to increase the SAS
yaw
frequencies and bring them under the inertial damping frequency range
and eliminate the fifth degree of freedom of inertial damping.
Calculations being done in Hongo if new design is sufficiently stiff.
Preliminary design of auxiliary bench for bottom of existing 3 m towers
to be used as TAMA 300 injection bench prototype and isolated bench
for
low frequency TNI. Kenji evaluating kinematic performance of
design.
Alessandro @ Pisa
Working on brazing problems for glassy metals.
More work on vertical accelerometer.
Following Ultra Sound milling machine construction.
Franco, Alessandro @ S. Piero
Restarting stress-strain machine to test glassy metals. Waiting
for
parts from G&M.
Charlotte @ Lyon
Sent in final report for her stage.
Akiteru @ Hongo, Riccardo @ Pasadena
Contacting prof. Mio to see if his funny metal could make high loss
springs to replace LIGO I springs in stacks. These funny metals
are
very similar to Szabi’s memory metals. They switch between Austenitic
and Martensitic crystal under stress variation instead than under
thermal variations. There are substantial technical problems
and needs
R&D but it may work.
For additional information about this report, contact sanders@ligo.caltech.edu