The LIGO Executive Committee Agenda for Monday July 19, 2004 will be:
(Meeting time: 10:30 am Pacific Time)
Special Items:
Sharon Kammerling has joined LIGO in the Project Office as Assistant to the Director.
no report
STATUS OF LSC MOUs
(Petrac)
(LSC Research Updates through August 2004, and Progress Reports
through February 2004)
Balearic:
GEO 600:
National Astronomical Observatory-China (NAO-C):
Rochester:
Groupe Virgo:
Virgo:
AURIGA:
Penn State:
SITE TELECONFERENCE (Lindquist)
A site teleconference was scheduled for Thursday, July 15,
2004. The following issues were discussed:
The list of assigned actions updated through July 15, 2004
may be found Here.
PROPERTY MANAGEMENT (Chargois)
>From: Ed Chargois <chargois_e@ligo.caltech.edu>
> From: Cleveland Mak <mak_c@ligo.caltech.edu>
COST SCHEDULE CONTROL SYSTEMS (Cunningham,
Brambila, Kaufman, Salone)
>From: Esther Cunningham <esther@ligo.caltech.edu>
>From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
>From: Florence Kaufman <fkaufman@ligo.caltech.edu>
Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport.
(For passwords contact Florence)
SUBCONTRACTS MANAGEMENT (Petrac, Jasnow)
>From: irena@ligo.caltech.edu (Irena Petrac)
>From: Ed Jasnow <jasnow@ligo.caltech.edu>
SUPPORT (Kammerling, Baldon, Lloyd)
>Irene Baldon
>Dorothy Lloyd
ADVANCED LIGO (Cost Schedule Control
Systems) T. Frey
>From: Thomas Frey <tfrey@ligo.caltech.edu>
See Advanced R&D Section.
REPORTS (Lindquist)
CHANGE CONTROL/CONTINGENCY MANAGEMENT
(Lindquist)
Change request CR040013 was submitted by Albert Lazzarini.
A copy will be found at:
http://www.ligo.caltech.edu/~phil/ChangeBoard/CR040013.pdf
The request is for additional budget for Hanford LDAS hardware and
maintenance. The change request was discussed during the meeting of the
LIGO Executive Committee held on July 12, 2004. The change request was
approved.
HUMAN RESOURCES (Akutagawa)
>From: Cindy Akutagawa <cindy@ligo.caltech.edu>
Quality/Safety (Tyler)
>From: Bill Tyler <tyler@ligo.caltech.edu>
No report this week.
.
no report
Commissioning:
Now that the gate valves are all open again, after HEPI mechanical
installation, interferometer commissioning has resumed. Last Friday
night, Brian and Andri locked both arms for the first time in a while.
R Weiss has been working on repairing the QPD units, so that they run reliably
in the RFI-fitted end stations. Brian and William have been improving the
optical lever servo compensation. HEPI work in the corner station seems
to have caused some slow payload drift among the corner station tanks.
This is because the hydraulic valves have some hysteresis, which will soon be
tamed by servo action. Right now, we are still knee-deep in electronics
commissioning, so there is some tension between the needs of interferometry,
which requires unmoving payloads, and that of HEPI. This should resolve
itself in a week or so. It may be necessary to move HAM4's payload to
unclip the AS beam.
HEPI commissioning: Chethan has installed new watchdog code. Rich
Mittleman has supplied new compensation filters, and I have been implementing
W. Hua's FIR highpass block in the HEPI processors. All of these should
be tested soon in ETMX.
(Rupal Amin)
Phase noise measurements of the Wenzel crystal oscillators
continue. I have reduced a 4 pole high pass filter to a single pole RC
filter in order to improve data reliablity at low frequencies.
All WFS (1,2,3,4, and the spare) now have the correct power
conditioning capacitors installed. They also each have a electronics
traveller associated with the repair.
HEPI Installation. (Rich Abbott)
1. The HEPI installation phase is complete. All hydraulic actuators
are wet and have been bled of bubbles. The actuators are in the
"run" mode ready for service.
2. We have encountered difficulty in booting the corner station HEPI
crate. There is considerable confusion as to what the correct jumper
settings are for the VME crates. Work will continue on this until it is
solved.
3. The next task for the hydraulics is to put the three hydraulic pumps
in closed loop pressure control mode. This will be done as soon as the
VME booting issue is done.
4. Documentation is being finalized for the HEPI system to ensure all
documents are in the DCC. The HEPI website is coming along well under
Katrina's direction.
5. The Excel workers used to do the piping and installation work have
completed their task and have been released. They successfully welded
1200 joints and probably about 5000 piping bends without a single leak.
It's a very impressive body of work on their part.
6. We are preparing to ship the items for the LASTI HEPI
installation. Jon Allen will be providing local coordination at MIT.
7. I have started to examine the TCS schematics, but will not have
completed the task until Monday (Friday if I am lucky)
LDAS (Igor Yakushin)
1) Upgraded GSI enabled sshd server on ldas-grid;
2) Had to powercycle fiber channel switch: it got into some strange state in
which fb1 could not mount /frames and the ethernet connection to the switch was
lost, however, LDAS had no trouble accessing the disks. Powercycling the
switch, suggested by Dan Kozak, fixed all the problems.
3) Restarted LDAS with the new passphrase.
Data analysis (IY):
1) Processing production and simulation triggers obtained by running the corrected
version of waveburst on S2.
2) Processing production waveburst triggers on S3 playground.
(Ken Franzen)
Have tried to help Alessandra Di Credico to get started with running off-line
glitchMon jobs at alvar. She also wants to be able to create artificial glitches
in some channels using DMTGen. A problem is that glitchMon gives a
"configuration file error" when we try to run. Trying to solve this.
Have also been doing some minor alvar system administration for SURF student
Marc Cenac who requested installation of some code necessary for his project.
CDS Operations. (Chethan Parameswariah)
Worked on the l1dscl2 skew problem. Turned out to be a bad reflective memory
node again. This time it was l1iscepics.
Fixed the MC Trans readout - cable was disconnected.
Installed new watchdog code and database for HEPI in the corner station. Found
that I will have to move the whole watchdog stuff to a baja.
Working on getting the corner hepi processors up and running.
Found that after the frame builder fb1 reboot, the ldas disk could not be
mounted. LDAS traced it to a bad port on their T3 switch. Dan Kozak rebooted
the switch and this fixed the problem.
(Ash Khan)
Developing a generic Data Acquisition card, design to fit on the back of the
Euro-Card chassis. The board supports 16 Digital Inputs, 16 Digital Outputs, 12
A/D and 12 D/A converters all on one card. The board requires only one wire to
interface with and in a multi-drop mode since all modules are connected with
only 1 wire the long cable runs and Cross-Connects are eliminated. The
Optically Isolated RS-232 interface eliminates the ground loops between the
Controller Power Supply and the Euro-Card Chassis. One card (4 and 1/2" X
5" size) replaces 1/2 XYCom220, 1/2 XYCOM 212, 1/2 VME3113 and 1-1/2 VME 4116
cards. The cost is less then $200 per board. Directly supoports EPICS Sequencer
and MEDM interface.
Computing (Shannon Roddy)
* Current bandwidth usage can be seen at http://teche.ligo-la.caltech.edu/mrtg/LLO-Router/130.39.245.1_1.html
* Archived bandwidth usage can be seen at http://teche.ligo-la.caltech.edu/mrtg/archive/
under the dated folder for the week of -- Shannon Roddy LIGO Livingston
Observatory California Institute of Technology 225.686.3106 Work 225.933.7821
Cell interest.
* The main item of any importance this week is that the application server
started corrupting several binaries. When the problem was noticed I
started checking on the server and it seems that it was suffering from file
system corruption. I tried to salvage what was on the server and repair
the file system late last night, but at about 2 am it finally barfed during the
reiserfsck process and the entire file system was lost. It is unknown at this
time whether this is a hardware or software problem on the server. So,
the old applications directory was salvaged, but all of the new applications
that were added in the last two months or so are lost. I was
not backing up this server yet since we just changed software and
tape libraries. The most critical items were being backed up (in other
words the non-replaceable files).
So, at about 2 am I started rebuilding the applications. This is still
ongoing, but I am going to break for sleep. I have replacement hardware coming
in tomorrow which should end this problem permanently. Right now I am
using spare space on the home directories disk (which is backed up nightly) to
store the applications. I have a fair amount of the software rebuilt, but there
is still a lot to go. It should be enough to keep the staff and students
going since Matlab, etc. is back at this point, but some of the support apps
are not there yet, such as acrobat, etc. Much of this software is still
available to users, but in older obsolete versions.
High-power laser lab / U of F (Rupal Amin)
HPLF:
IPG Photonics, supplier of the 100 W laser, has indicated that a
"fiber fuse" occurred two weeks ago. This failure resulted in
internal components being damaged. However, LLO has not been informed which
internal components are irreparable. The supplier did request a
description of the table setup to determine the reason for failure. I sent
photos of the table, beam dump, and sensor which were catching the beam at the
time.
Modeling (Dodda, Jamal, Rogillio and Yoshida)
We continued characterization of HAM1 table yaw motion using e2e HAM stack box
and MC1 box. Previously, we used the rotational-rotational transfer function of
the HAM stack box to calculate the table top yaw motion. When this table yaw
motion was used as suspension point input to MC1 box, the power spectrum of the
resultant osem yaw was an order of magnitude higher than the corresponding DAQ
osem yaw signal in the frequency range of 1 1.8 Hz. This time, we first
calculated table top translational motions U and V using the corresponding
translational-translational transfer functions of the HAM stack box, and
subsequently, calculated the table yaw motion as 1/2 (dV/dx-dU/dy). Here we
assumed that dV/dx (dU/dY) is proportional to kV (kU) where k represents the
phase delay of V motion along the x-axis (U motion along the y-axis). For k, we
used the phase delay of floor U motion that we measured previously using a portable
seismometer. This new modeling of HAM table yaw motion results in the
power spectrum of the calculated MC1 osem yaw much closer to the corresponding
osem yaw DAQ signal (the discrepancy in the 1 - 1.8 Hz range has been improved
to a factor of 1.5 from an order of magnitude).
We continued the construction of Mode Cleaner box as well. For the purpose of
testing the locking mechanism, we set the MCs Q factor about a factor of ten
lower than the reality (by increasing the optics transmissivity to 2%). Under
this condition, the MC box can now be locked stable for realistic HAM table
motion.
John Zweizig
On the DMT software development side, I have made the GDS
software
compatible with the gcc-3.4 compiler. This new compiler has
much improved
syntax checking an revealed a few lurking errors in the code
which might
have show up in the future. I also released a table
conversion and
manipulation program that can be used to convert xml trigger
lists to and
from flat ascii text which several people prefer for trigger
analysis
studies.
I also reviewed a set of triggers I generated from the power
line monitor
channels. To make these triggers, I notched out the 60Hz and
300Hz
components in the power lines. The signals resulting from
this filtering
were then searched for glitches using the PSLmon glitch
tool. The triggers
produced in this way were of three basic classes. These are
1) single sample "glitches" resulting from
the LSB DAQ error. These are
are obviously artifact of the
readout rather than something that
might be used as a veto and can
therefore be ignored.
2) Brown-outs - These are cases where the line
voltage sagged, usually
for a few tenths of a second. These
are probably a result of
large loads switching on the power
line.
3) Ugly nasty transients.
I looked for coincidences between these triggers and
triggers in AS_Q and
found a few (~10 in first month of S3 playground) cases
where the triggers
were coincident. The next step is to look to see if there
are any obvious
subclasses in types (2, 3) that are more like to correlate
with AS_Q
disturbances.
see also the CDS weekly meeting minutes in the commissioning archives:
no report
Jay Heefner reporting
Anti-Image Board for Low Noise DACs
============================
10 modules are being stuffed. The first should be tested and
ready for shipment by 7/16/04. The final 9 should be ready for shipment by
7/23/04.
LLO EMI Retrofit
=============================
Designed and ordered all required rack and crate panels
necessary for LVEA SUS installation.
40 M
==============================
- Updated and installed the autoburt and saverestore
functions for c1psl and c1iool0. Still pending are the save/restore functions
for c1epics, c1vac1 and c1vac2.
- Added high power amplifier to LO drive for MZ servo card
to bring LO level to +23dBm. The amplifier is a Mini-Circuits block with SMA
connectors. A more permanent solution using a board mount amplifier is being
investigated.
AdL SUS
==============================
- BNC to Dspace ADC adapter modules are complete and will be
shipped to LASTI on 7/16/04.
PeterKing
A problem with the PSL Lab's pre-modecleaner servo caused by
an over
zealous cleaner and mop was fixed.
Another problem was that last week the lab had a strong
smell of something
burning that lasted in the lab for quite some time.
This was common to all
the 3 labs at the east end of Lauritsen. The cause was
apparently an over
microwaved lunch in the machine shop downstairs
Lee Cardenas, Liyuan Zhang
OTF Lab. (W. Bridge)
We have introduced a new sample of epoxy. the white Ceramabond,
along with the disks of TRA-BOND #2254 color light brown epoxy.
Cavity is locked and we are taking measurements everyday.
The "good news" is that the ring down (loss) and the beat frequency
(Absorption)
are the same as before, therefore we can say that these epoxy samples are clean
and there is no contamination. We'll continue
taking measurements.
Absorption Test Measurement prototype in progress
Scatterometer system in progress!
We have installed the scatterometer head after a severe modification.
Now the scatterometer has the ability to move in the X-Y-Z direction.
We can adjust the sensitivity more precisely as the up and down movement
is controled by a micrometer.
We have encounter with a small situation as the working space between the
scatterometer base and the top of the mirror is not large enough.
We are
going to take some measurements as to find exactly how much more space
is needed, then we are going to have the base holder posts for the mirror to be
shortened as required.
The Quantronix 60 watt laser is in progress.
OTF Lab at Lauritsen ROOM 38
Cavity #3
The cavity is locked with (6) disks of TRA-BOND #2151 color blue epoxy and
(4) disks of TRA-BOND #2902 color silver epoxy. Both are in.
as stated before these samples show NO CONTAMINATION!
Our ring down (loss) and Beat frequency (absorption) are the same as our
base
measurement and we continuing taking daily measurements.
Cavity #2 Test cavity optical set up is still in standby
no report
Lasti Weekly (Allen, Mittleman, Pfau, Mason, McInnes and Ottaway)
Site Support
David Pfau has finished the first round of testing the new cylindrical
Periscope. His found that when hanging from a wire its lowest resonance is
~600Hz. On the optics table there is a lower resonance ~250Hz. His report is on
the 7/015/04 Lasti ilog.
Jon Allen is close to completing the layout for the TCS interface board completed.
Research on the adaptive feedforward scheme for HEPI is progressing. We have
had some good success with Leis feed forward scheme. We switched over to using
the platform geophones and this seems to work better. Results will appear on
the ilog in the near future.
Advanced LIGO Installations
Myron has installed four of the eight actuators onto the X-end HAM. He has also
commenced assembling the six in-vacuum geophones for the Xend HAM optical
table. The cabling for the in vacuum geophones is in the bake oven.
Ken has updated the MC triple suspension installation and alignment procedures.
The installation of an optical lever for the MC triple suspension has begun
using parts removed from the HAMs during the triple installation. We found that
three old style HAM piers and a scissor table dummy gave us good stiff piers
for both the optical lever sender and reciever. The optical lever breakout box
is being built by Jon Allen
Simulation and Modeling
(Bhawal)
Hiro and Biplab gave a 2-part presentation on the "FFT study of the
realistic optics of LIGO I" at the weekly LIGO commissioning meeting
(viewgraphs at http://emvogil-3.mit.edu/commissioning/)
Weekly Physics Meeting
-----------------------------
Monica Vervella from LAL, Orsay reported her progress in studying the control
of Mach-Zehnder interferometer using e2e. Sany yoshida of Southern Louisiana
University presented results of his comparison of the measured pendular and yaw
motion at Mode-cleaner mirror with results from e2e model in which accelerometer
(at HAM) outputs are injected as noise. He calculated Table yaw motion after
passing the seismic translational motion through the HAM transfer function
instead of using translational-to-yaw transfer function. That way he got good
agreement with the measured spectrum.
FFT physics
-------------------------
(Hiro) The effects of the beam splitter curvature and the thermal lensing
effects were studied and the results were presented at the commissioning
meeting (http://emvogil-3.mit.edu/commissioning/FFT040712-Hiro.pdf).
My calculation of the thermal lensing effect showed different results
from Erica's previous calculation. The cause was that Erica's input beam was
not set properly. I asked Erica to confirm this point by checking the beam
curvature on the HR side of the recycling mirror which is kept in the FFT run
output file.
FFT Mirror Phasemap
-----------------------
(Hiro, Xiao) The mirror reflection phase map has been extended to outside of
the measured region of 15cm diameter to the entire coated region of 24cm
diameter based on the suggestion by Matt of using outer region as a regulator
(require to be 0 outside of 30cm diameter). This generates a reasonably smooth
behavior. The noise is used to test if the systematic behavior is properly handled
by basic Zernike terms. Basic work is done, and, a new sets of maps, with three
different variations of the extrapolated regions for systematic error
estimations, will be produced and will be available from e2e home page.
Mirror Maps and Beam profiles
------------------------------
(Biplab) The effects of mirror aberration on Beam intensity profile and
contrast defect have been studied and reported in the commissioning meeting.
The contrast defect calculated from FFT for Hanford 4Km interferometer has reasonable
agreement with the measured value at LHO (several 10^-4) but the measured value
of LLO (~ 3E-5) is surprisingly better than even an FFT simulation without
mirror aberration and beam-splitter curvature.
FFT does not have any Angular control built into it. Although tilt has been
artificially removed from the mirror maps, it's not obvious that the intensity
profiles obtained so far truly represents the effects in a tilt-controlled
interferometer. For that reason study of WFS signals for the interferometer
with mirror aberration but no pitch/yaw and then the manual correction of
equivalent tilt are currently going on.
Mach-Zehnder Simulation using E2E
-------------------------------------
(Monica) The Mach-Zehnder package has been completed both for the configuration
with 33MHz and for the configuration with 29MHz: open and closed loop has been
evaluated giving a motion (a simple ramp) to the steering mirror. PSD have been
done for the error signal and the transmitted power. Seismic motion on the steering
mirror with a resonance peak at 800 Hz (steering mirror resonance) has been
implemented. Next step is to calculate transfer functions using the 29MHz
configuration as this is the one actually mounted on 40m PSL.
MiniLIGO
-------------
(Matt) Worked on new MiniLIGO simulation. This is a striped down version
of SimLIGO (no digital, no ASC, no DSC, etc.) which can be used as a learning
tool and for investigations that do not require the full complexity of
SimLIGO. Currently, MiniLIGO is working but needs lock acquisition code.
FFT code modification
------------------------------------
(Hiro) FFT code was revised to include the beam splitter curvature effect and
to simulate the thermal lensing effect easily and consistently.
E2E code Modal Model class
------------------------------
(Hiro) ModalModel class is modified so that it matches with the formula with
the e2e modal model document. The interface to the existing e2e code is being
modified to generate a correct answer. Codes and primitives needed to study the
OMC are tested.
Code development and maintenance
---------------------------------
(Melody) Working on revamping the modeler's parser to make it more efficient
and maintainable.
Alfi
-----------
(Bruce) Continuing work on re-institution of bundle and bundler code.
(Melody) Working on PR 468 to prompt the user to save all affected box
Data Analysis Activities
(Lazzarini)
Creighton:
This week I committed the frame-reading version of ComputeFStatistic, tracked
down an error in the ATNF pulsar catalogue that was causing an overestimate of
pulsar timing uncertainties, and rewrote sections of the PULGroup known-pulsar
search paper.
Mendell:
I am working with the CW group on the S2 paper, and on a new LAL function,
LALFastGeneratePulsarSFTs, that hopefully will significantly speed up the Monte
Carlo simulations performed by the CW group. The function should be ready
to checkin to LAL by next week's report.
Shawhan:
* Erik Katsavounidis and I have been working out how to organize the
astrophysical analyses to be carried out by the Burst Group using S3 data. We
have made a proposal to the group to follow a more loosely-managed model, with
people taking on various topics and tasks according to their interests.
We still need to work out the details of providing some structure to this
process and coordinating the release of results and papers.
*Helped Teviet Creighton and Rejean Dupuis track down a discrepancy in
systematic error estimates for the S2 known pulsar search.
* Worked with SURF student Matt Wroten to run the lalapps inspiral search code
on sample data with simulated signals, so that he can continue to explore
possible statistical tests to distinguish real signals from glitches.
Sutton:
This week I've been dividing my time between a number of projects. I put a few
hours into my calibration class for the DMT, specifically handling input and
output of reference calibration data. Progress has been slow - relevant
existing codes are not well-documented, but I want to avoid writing un-necessary
new code. For LIGO-TAMA I've been post-processing WaveBurst triggers,
doing sanity checks on the distributions of various trigger characteristics,
and computing the WaveBurst 3X efficiency. The next step is to set a
coincidence window for WaveBurst-TAMA and do 4X coincidence. I'm working
on the TFClusters review report, due July 19. Finally, I spent a
significant amount of time on characterizing the LIGO burst signal space, and
presented some results at the July 13 bursts telecon.
Yakushin:
1) Processing production and simulation triggers obtained by running the
corrected version of waveburst on S2.
2) Processing production waveburst triggers on S3 playground.
Lazzarini:
Working on first draft of the stochastic S2/S3 paper and associated figures.
Working with the Virgo-LIGO joint analysis team on the first draft of the white
paper/proposal for joint analysis projects.
LIGO Data Analysis System
Software Systems (Blackburn)
Made code changes to support to both the C++ and TCL parts of the diskCacheAPI
and the controlMonitorAPI (including the client) improve the performance of
graphing the available frame files under LDAS's diskCache by promoting the
previous sorting and concatenation steps in the time interval lists into the
C++ layer. This did however turn up an issue with the TCL layer of the
diskCacheAPI using multiple threaded calls per job to generate the results. It
was determined (as was the case with the framequeries in the frameAPI over a
year ago) that attempting to use multiple threaded calls between the TCL and
C++ layers is very dangerous and leads to unpredictable behavior. The code is
now being reworked to have a single interface per job.
Reran the system and integration tests on our standard weekly cycle. This was
tested against the Thursday night build (1.1.64). The test completed without
issue and the results have been checked into CVS and posted on the website.
Began validating the Solaris compiler code changes made to LDAS against our
current GCC compiler (3.3.3). John Zweizig has provided some changes to
frameCPP which support GCC 3.4.1. We'll look into the compatibility of these
changes against both the Solaris and the GCC 3.3.3 compiler before providing a
frameCPP that works with the latest GCC compiler.
Added suppression of support in user commands for the compression mode
"ZERO_SUPPRESS_INT_FLOAT". This was an oversight in the filing of the
problem report (1828) that was discovered this week.
Continued to work on the mpiAPI issue of the master compute node being used by
multiple jobs. This is PR 2451.
Began working on PR 2380 - controlMonitorAPI not properly notified with the
diskCacheAPI fails to rebuild cache. The first attempt to fix this problem
introduced (or uncovered) a race condition. Continued analysis of the new
problem is needed to fix the PR.
Organized a group to review the size of frames and the best strategy for making
RDS frames longer. This is being formulated under the guidance of the LSC
Computer Committee. Technical experts from CDS, Control Room, LDAS, LAL,
Archives & file servers, RDS generation and DMT will be on board for this
study. The first meeting of the group is expected next week with a report to
the LSC Computing Committee roughly two weeks later.
With the inclusion of GEO data in our archive the missing support for multiple
frames per file in the creation of RDS frames is becoming more of an issue.
Fixing this should overlap well with the expected outcome of the above
mentioned group reviewing how to increase the size of the RDS frames.
Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* Continued the transfer of files from HPSS to SAM-QFS. Also
continued deleting files in HPSS after confirmation that they were
successfully transferred.
* Restarted various parts of HPSS that crashed so that the above could
continue.
* Received replacement T3 disk drive from Sun, put it in spare parts T3
(t3-11) in Synchrotron, put drive from t3-11 into t3-4 (in Booth).
* Got policy clarification on whether rare events (e.g. Vela pulsar
glitch) should get their own directories (they shouldn't) and moved the
files out of the LLO VelaPulsar directory so that it would conform.
* Worked with Chethan and Igor at LLO to resolve the problem where fb1
could no longer see any of the devices attached to the FC switch. It was a
problem with the switch port that fb1 was connected to as best we
can tell. The switch had also quit responding on it's Ethernet
port, even though all the FC traffic (other then to fb1) was working fine.
Taking everything down and power cycling the switch fixed the problem.
(Hari Pulapaka)
* Installed and tested Gridcvs on Solaris.
* Testing the pyGlobus ftp server.
* Working on getting the Matlab Frame channel software working on
ldas-gridmon.
* Upgraded ssh on ldas-grid and ldas-grid.mit.edu.
* Created Indexes on the databases located on ldas-cit and ldas-gridmon so that LSCdatafind queries will run
faster.
(Stuart Anderson)
* Continued testing of SAM-QFS 4.1 and verified read-only mount option
works.
* Identified 5 marginal node disks in the LDAS-CIT cluster.
* Started setting up tandem system to be able to mount SAM-QFS for
additional LDAS software testing.
Livingston
----------
(Igor Yakushin)
* Upgraded GSI enabled sshd server on ldas-grid.
* Had to powercycle fiber channel switch: it got into some strange state in
which fb1 could not mount /frames and the ethernet connection to the switch was
lost, however, LDAS had no trouble accessing the disks. Powercycling the
switch, suggested by Dan Kozak, fixed all the problems.
* Restarted LDAS with the new passphrase.
Hanford
-------
(Greg Mendell)
* Contractors have installed water and the electrical circuit for the new 5 ton
chilled water AC unit in the LDAS room. John Worden is working to get the
unit going today. We will then verify that we now have sufficient cooling
to run the cluster with all 278 CPUs in the Condor pool fully loaded.
(Ben Johnson)
*Re-Installed LDR onto gateway@LHO. It is now an official 0.5.5. version of
LDR. It should be ready for publishing of the one-percent data.
* Replaced HDD in node89. See PR#2609.
*Electrical power to new 5-ton chiller installed in LDAS room here. John Worden
is getting it operational. We'll test the heat load this afternoon.
* Worked on getting locked segments published into the LDAS database
again.
General Computing (Wallace)
MIT:
(Keith)
Nothing to report.
Livingston:
(Shannon)
* Current bandwidth usage can be seen at
http://teche.ligo-la.caltech.edu/mrtg/LLO-Router/130.39.245.1_1.html
* Archived bandwidth usage can be seen at
http://teche.ligo-la.caltech.edu/mrtg/archive/
under the dated folder for the week of interest.
*The main item of any importance this week is that the application server
started corrupting several binaries. When the problem was noticed I
started checking on the server and it seems that it was suffering from file
system corruption. I tried to salvage what was on the server and repair
the file system late last night, but at about 2 am it finally barfed during the
reiserfsck process and the entire file system was lost. It is unknown at this
time whether this is a hardware or software problem on the server. So,
the old applications directory was salvaged, but all of the new applications
that were added in the last two months or so are lost. I was
not backing up this server yet since we just changed software and
tape libraries. The most critical items were being backed up (in other
words the non-replaceable files). So, at about 2 am I started rebuilding
the applications. This is still ongoing, but I am going to break for
sleep. I have replacement hardware coming in tomorrow which should end this
problem permanently. Right now I am using spare space on the home
directories disk (which is backed up nightly) to store the applications. I have
a fair amount of the software rebuilt, but there is still a lot to go. It
should be enough to keep the staff and students going since Matlab, etc. is
back at this point, but some of the support apps are not there yet, such as
acrobat, etc. Much of this software is still available to users, but in
older obsolete versions.
Hanford:
(Christine)
- Network usage can be seen at
http://www.ligo-wa.caltech.edu/~christin/mrtg/
198.129.208.1_198.129.78.122.html
- Last week, on July 3, the tape robot used to backup all the user
directories died. I discovered the problem on July 7 and started creating
tar files of the directories and backing them up to other disks. On July
10 all three disks that I was using for the backups suffered damaged sectors,
unreadable blocks and finally were totally unreadable. On July 12 one of
the disks containing the current user directories started to go bad, also
reporting multiple bad sectors. I ordered more disks from Sun, but they
have a 5 day lead time before they can even ship. Meanwhile, it turned
out that the problem with the tape robot was a bad connection with the power
cord, so I was able to get the tape backups working again on July 14. I
had also just received 5 SunBlade 1500 computers which I had not had time to
set up. Each SunBlade 1500 has an 80 GB disk. I quickly started
setting up the SunBlade computers and restoring user directories from tape to
the disks of the SunBlade computers so that I now have user's files on the
SunBlade disks and can mount those for user's to access their directories if
any more of the original disks fail before my replacement disks from Sun
arrive.
- Un-installed Norton AV on one laptop and reinstalled it. When the user
tried to run Liveupdate, it reported a corrupt system file and wouldn't
update. The user had used the Windows remove software tool and then
reinstalled Norton, but still had the same error. When I uninstalled I
also cleaned the registry of any references to Symantec and Liveupdate before
reinstalling. That seems to have fixed the problem.
- Several users have reported a problem with Eudora 5.2 and virus emails.
When Norton catches an e-mail with a virus in it, it quarantines the virus
file, but the e-mail is not deleted from the server, and Eudora can't download
any further messages from the server. Deleting the virus e-mail from the
server solves the problem temporarily. Netscape e-mail doesn't have this
problem. I'm testing the more recent versions of Eudora to see if they
have the same problem.
- Other Misc. user support including; setting up a new modem account, helping
users with Matlab, fixing a font problem with some new software on the
application server, fixing a problem with the compilers on the application
server, and deleting some pop3 lock files after the server reboot.
CIT:
(Mike)
-Ken Mailand: I reloaded his workstation from scratch. This is a corrupted win
2000 installation that I reloaded to XP professional. I loaded all GC software
and many engineering packages. I also work with Ken restoring his preferences,
desktop settings, printing and mapping drives. I also installed a dual monitor
video card that is compatible with Solid Works. I also swapped out his CTR
monitor with a LCD for compatibility issues I came across with the new video
card.
-Sharon Kammerling: I setup her up with Donna's old workstation & scanner.
This project required a lot of equipment swapping in order to set her up in
Ryan's office, located in E/B.
-Barry Barish: I worked on backing up Barry's laptop and troubleshooting a
problem with his old laptop. I ended up replacing the keyboard.
-Work on a visitor's workstation due to a virus problem. I was able to clean
this workstation. This computer got infected due to the anti virus software
being out of date. This computer was also out of date with OS and security
settings. Everything is updated.
-I worked on Linda Turner's workstation in DCC due to a virus issue. I also
cleaned this computer and updated the security and disable unnecessary
services.
-Setup anadditional workstation over in Synchrotron. This required a lot of
time running a network cable and updating OS and security patches.
-Larry Jones: Working on setting up a laptop for Larry that was previously used
by Jonathan Kern. I still have a lot of work to do on this computer.
-Other misc. onsite/phone user support.
(Lisa)
- Spent time helping Larry chase down a couple of compromised pc's.
- Ran into more problems with the X11 based compiles on the new application
server. I'm still trying to work those out.
- More work on the wireless APs.
Mail Stats 07/08 - 07/14/04
Messages Accepted:
Spam Rejected:
Viruses Rejected:
False Positives:
Total Mail Thru:
(Veronica)
- LIGO website: Per arequest from a publisher, provide a high-resolution image
for an upcoming article about LIGO. Training Sharon Kammerling on updates to
html documents. Posted updates to various pages.
- LSC website: Updates to Observational Results page and other postings.
Working on a project for Albert on streamlining the process of generating an
LSC author list. Updates to the layout of the August meeting website; posting
new information as it arrives.
(Larry)
-Worked with Mike and Lisa chasing down some PC's that were infected with a
virus. It was noted that all of the machines did not have their virus definitions
up- to-date. Mike will be making some changes on the machines not assigned to
anyone in particular to where the definition files and patch files will be done
automatically along with a reboot. Those machines assigned to users we will
continue sending out the notifications for keeping the files up-to-date.
-Assisted in getting a number of items transitioned over to Sharon and still
have a number of items to go.
-Worked a number of mail alias issues. Thanks to Stan there are number of
aliases that have been cleaned up. A new alias has been setup for local
visitors. Just as a reminder when someone leaves the project, the sysadmin
group does need to be notified so accounts and related issues can be addressed.
-Made a number of purchases including a new printer for the 3rd floor of
WestBridge. Also, a quad CPU unit to be used as a sandbox has been ordered.
Worked a couple of maintenance contracts.
-Still working with the group making modifications to the computer room. The
fire suppression unit is suppose to be installed next week. The power
installation should be finished by the first of next week.
-Working with Christine, Ed and others on the logistics for the August LSC
meeting.
-Assisted a couple of the students with a couple of configuration issues.
-Rebuilding a laptop to replace my failing laptop.
-Regular user account support.
From: Larry Jones <ljones@ligo.caltech.edu>
Advanced LIGO Seismic Structure
SEI Structure:
Brian Lantz has agreed to relaxing the vertical positional tolerance of the
Stage 1 CG to as much as +/- 10 cm of the Stage 0-1 horizontal actuator plane.
ASI has combined an 8 cm CG offset with practical structural stiffening and
achieved 151.4 Hz as the lowest resonant frequency for Stage 1. They are
hesitant to go all the way to 10 cm, which would only add a few more Hz. We are
in the process of approving the model producing 151.4 Hz, even though it does
not have the required 10% margin over the 150 Hz to be demonstrated in the
hardware modal testing.
ASI has completed a "bottom-up" cost estimate for fabricating the BSC
structure prototype, and will be sending it to LIGO soon. A replan to determine
the contract Cost-To-Complete (CTC) is in process.
ASI has completed a preliminary sag analysis and finalized the lock/locator
design.
ASI stress engineer Eric Scholten has left the company to pursue an advanced
degree; this will cause some ramp-up inefficiencies with bringing his
replacement up to speed.
The HAM structure Design Review is now scheduled for
Actuators:
The prototype actuators were received at Caltech on July 7. Vacuum bake and RGA
scanning was completed on the large prototype on July 15, but we suspect that
the final cleaning was not fully effective; this may be repeated. Mechanical
inspection was performed on the small actuator, with no significant problems
found.
The early phases of optical contamination cavity testing of the Tra-Bond 2254
alone has shown it to not be the source of the contamination that caused bad
results immediately when three epoxies were tested together. Ceramabond 835M
has been added to the 2254 test chamber for simultaneous evaluation of
acceptability; both look good, so far.
Displacement Sensors:
The Tra-Bond 2902 (silver) coupons were added to the Tra-Bond 2151 (blue)
optical cavity contamination test and both are performing well. Additional
epoxies were purchased from Aremco and will be tested as possible replacements
for the Tra-Bond epoxies.
Seismometers:
Nothing new
Galling/Dusting Test:
Nothing new.
Other:
Planning is proceeding on the design and sourcing of a larger vacuum bake
oven for the quad structure.
From: Jay Heefner <jay@ligo.caltech.edu>
AdL SUS
==============================
- BNC to Dspace ADC adapter modules are complete and will be shipped to LASTI
on
From: "Mark Barton" <maaku@attglobal.net>
* I worked with Bram Slagmolen of AIGO to get a new e2e model of his FP cavity
commissioned from Hiro and Matt Evans set up on his computer.
* I advised Conor Mow-Lowry on suspension design for an optical experiment on
ANU. It appears that the triple pendulum lower blade springs would be suitable
so we explored the possibility of LIGO lending him some with Janeen and Dennis.
* I finished testing and released a new version v1.3 of my Mathematica pendulum
modeling toolkit with significant new features contributed by
Ben Lee of UWA.
* I calculated some more cases of asymmetrical quad pendulums to extend the
survey of cross-coupling reported in T040143-00.
From: Janeen Romie <romie_j@ligo.caltech.edu>
AdLIGO Suspensions
Circulated a 1st draft of the ETM Upper Structure Product Design
Specification, T040141. Also, circulated a 1st draft of a possible
failures and failure mitigation paper, E040329, to a select group.
Having many discussions on ear bonding techniques and fixturing. Participated
in a discussion this morning, hosted by Caroline, on fiber specification. It
included the alignment requirements as we believe them now and some cross
coupling issues.
Completed Primavera updates, with input from many, for
Coordinating an installation fixture design meeting for this coming tuesday.
Worked with Calum and Mike P-L on releasing the mode cleaner drawings. Almost finished
with the DCN to release the hybrid osem drawings, materials list and drawing
tree.
Working on a new design for the magnet holder for an eddy current damper.
From: Helena Armandula <ahelena@ligo.caltech.edu>
From: ctorrie
<ctorrie@ligo.caltech.edu>
Advanced LIGO
SUSPENSIONS
1. MC
The MC files have been released. Here is an extract of the set: -
DCN
Materials List
Drawing Tree
Example of Drawing with embedded file
Overall Assembly as built in tank (incl BOM)
2. PDMWORKS VAULT
The vault is now up and running and hosts all of the MC and QUAD files.
MPL, Mike and I have drafted a users guide. It can be found on the DCC as
follows: -
Drawing & Data Templates and an into to PDMWorks Vault
3. SolidWorks 2004
Upgrade to happen from Monday
4. Design Meeting
At this weeks design meeting we discussed issues with regard to the
interaction with RAL on the cantilever blades and the blade test facility.
Noters can be found at : -
http://www.ligo.caltech.edu/~ctorrie/QUAD_ETM/quad_etm%2520setup%20page2.html
5. Structure requirements
A draft document has been created highlighting the requirements. Sketches
highlighting the lengths and numbers associated with T040028-00 &
M040099-00 have also been included
Notes from the Monthly Core Optics group meeting –
Gregg Harry MIT Q Lab
This isn't new data, but a new interpretation. Both S/N 33 and 32 were
coated by CSIRO with silica/tantala using an ion assist gun. The grid for the
ion assist gun was changed between the two coating runs to try to improve
uniformity. The indices of refraction for the tantala were different,
n_33 = 2.140, n_32 = 2.087, while the index for silica was constant at n_sio2
= 1.474. Here are the coating phi's deduced from each modal Q of
the two thin samples;
phi_coat_||
Sample
S/N 33 S/N 32
butterfly low 4.1 +/- 0.1 X
10^-4 3.2 +/- 0.1 X 10^-4
butterfly high 4.2 +/- 0.1 X
10^-4 3.1 +/- 0.1 X 10^-4
drumhead 5.0 +/-
0.1 X 10^-4 4.0 +/- 0.1 X 10^-4
2nd butterfly low 4.1 +/- 0.1 X 10^-4 3.5 +/- 0.1
X 10^-4
Mode 12 4.4
+/- 0.1 X 10^-4 3.1 +/- 0.1 X 10^-4
Mode 13 4.4
+/- 0.1 X 10^-4 2.3 +/- 0.1 X 10^-4
You can see that the coating with the higher index tantala had consistently
higher phi's than the low index one. Those error bars are preliminary, I
need to check where they came from, but I am confident that most of the
differences between the two samples
are significant.
We should think about whether having CSIRO do further work on tantala
stoichiometry or do more with changing the ion assist gun's parameters is the
best next step.
David Crooks –
At Glasgow, we have mainly been working on the new coating
paper incorporating alumina/tantala and silica/alumina coatings, the new
coating material properties measured at Stanford and also frequency dependent
effects; the paper draft is now approaching completion.
Additionally, work on measuring the control annealed sample for the second
doped tantala/silica sample has been completed. When this control is used with
our data for the second doped tantala/silica sample a loss of (1.30.3)ױ0^-4 at 0 Hz is obtained, compared to (1.80.9)ױ0^-4 for the first doped tantala/silica sample.
Report from Bram Slagmonlen and Jerome
The ACIGA high power test:
David
Shoemaker -- Short update on the Substrate Downselect --
We are now planning to pull together the information on a downselect in July
(rather than June), with permission from the suspension folk and their
schedule. Recent input has come from Garilynn, who made estimates for the risks
in the production of the two substrate types, and also given an update of the
document summarizing our knowledge; and from Eric Black who has written up the
Sapphire Thermoelastic noise measurements done in the TNI. Steve Penn's
ripening summary of the fused silica loss data will also help us gain
confidence in predictions for loss at GW frequencies.
Helena Armandula – Caltech
Adv. LIGO Coatings:
CSIRO
sent a report on the work
performed to test the effect of ion bombardment assisted deposition on Young's
Modulus of Ta2O5 films.
The films were bombarded by an oxygen/argon ion beam from an assist ion gun
during deposition.
The Young's Modulus of the films was measured using a CSIRO manufactured
ultra-micro indentation system (UMIS).
Changes on Young's Modulus were very small, within the sensitivity of the
instrument.
Also, according to the report, " X-ray diffraction measurements of the
above films have shown no evidence of possible ion-bombardment induced
microcrystallinity, the as-deposited films being amorphous according to this
test. The annealing of the films at 600C for 2 hours did not change this
result, however, at 700C, a strong evidence of crystallization of the film was
obtained. This is in agreement with other sources which have shown the
threshold of microcrystallization of an initially amorphous tantala film to be
somewhere between 600C and 700C."
Garilynn Billingsley – Caltech
You can find the most recent version of the downselect document here
http://ligo.caltech.edu/~gari/LIGOII/Downselect/DownselectV7.doc
Updated Advanced LIGO COC pages here
http://ligo.caltech.edu/~gari/LIGOII
Eric Black – TNI
At the TNI, we have finished analyzing our Sapphire data and have written a
short report, including implications for AdLIGO. See the link below for a copy.
We are ready to test a new coating material on fused-silica substrates as soon
as it is available. In the meantime, we are setting up to do a non-Gaussian
noise measurement.
TNI Sapphire Report
http://www.ligo.caltech.edu/~blacke/T040147-00.pdf
Other Laboratory R&D
From: "Anamaria Effler"
<felfre@its.caltech.edu>
Juri:
I have continued working on my presentation and poster, which are basically done
now.
Yanyi:
I have been making the necessary improvements to the 3 circuits- the LVDT, the
Amplifier and the Voltage Saturation Detector.
The LVDT circuit: I am trying to build a Jumpers of the LVDT driver and the
connector JP102, that can change the sign of the LVDT driver.
The Amplifier: I am trying to stabilize the voltage sources using voltage
regulators as well as making some neccessary changes on the connectors in the
circuit.
The Voltage Saturation Detector: I've build most parts of the circuit on the
prototype except for the transistors. Paul's lab has most of the parts in this
circuit, but not the transistors. I still need to purchase the transistors in
order to build a complete circuit on the prototype and simulate the circuit.
Barbara:
I'm back from Italy and after I checked the status of the interferometer I'm
working on the piezo's circuit because channel 1 and channel 4 don't work properly.The
next days I have to retune the GAS springs and I have to prepared the
cabling.
Anamaria:
I've continued with the data acuisition which is stilll causing some beats problems
that I'm trying to solve. Also, I have done a presentation for my group of what
I have done so far.
Nikki:
Last week i continued assembling Barbara's interferometer and cleaned the lab.
With the help of Dr. DeSavlo and Anamaria, my oven is near completion and will
be finished when I return from being ill.
Randal and Jessica:
Our first week here in
For additional information about this report, contact whitcomb_s@ligo.caltech.edu