The LIGO
Executive Committee meeting for Monday August 16, 2004 is cancelled
The HEPI system at LLO successfully passed a major
milestone, “the one-armed test”.
Congratulations to all.
no report
STATUS OF LSC MOUs
(Petrac)
(LSC Research Updates through February 2005, and Progress Reports
through August 2004)
ACIGA
Balearic
Carleton
Caltech-CEGG
Florida
LSU
Northwestern
Syracuse
Trinity
Columbia / New applicant
Vassar / New applicant
(LSC Research Plans through Aug. 2004 and Prog. Reports
through Feb. 2004)
ACIGA
Florida
LSU
National Astronomical Observatory-China (NAO-C)
Northwestern
Orsay Group (aka Groupe Virgo)
VIRGO
Wisconsin
SITE TELECONFERENCE (Lindquist)
A site teleconference was scheduled for Thursday, August 12,
2004. The following issues were discussed:
The list of assigned actions updated through July 22, 2004
may be found Here.
PROPERTY MANAGEMENT (Chargois)
>From: Ed Chargois <chargois_e@ligo.caltech.edu>
DOCUMENT CONTROL CENTER (Turner, Mak)
>From: Linda Turner - turner@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: Gina Salone <gsalone@ligo.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 (Baldon, Kammerling, Lloyd)
>Irene Baldon
>Sharon Kammerling
>Dorothy Lloyd
ADVANCED LIGO (Cost Schedule Control Systems)
T. Frey
>From: Thomas Frey <tfrey@ligo.caltech.edu>
See Advanced R&D Section.
REPORTS (Lindquist)
Some things have been added to the schedule of proposals and reports
planned for the remainder of the calendar year. It currently looks
something like:
Have made some progress on the Annual Report.
CHANGE CONTROL/CONTINGENCY MANAGEMENT
(Lindquist)
There are no open change requests currently in the system.
HUMAN RESOURCES (Akutagawa)
>From: Cindy Akutagawa <cindy@ligo.caltech.edu>
Quality/Safety (Tyler)
>From: Bill Tyler <tyler@ligo.caltech.edu>
No report this week.
.
Summary of Commissioning Activities at
LIGO Hanford Observatory (Landry)
Roughly 200 people visited the lab on a Wednesday, listening to talks and
sitting outside late into the night, observing the Perseid meteor shower at
peak activity.
4K IFO
There was much commissioning activity this week, and the elog should be
consulted for details. Some highlights are noted below.
Investigations of phase noise coupling continued on the 4k. Although the effect
is not completely understood, the noise coupling was improved with differential
heating on the ITMs.
An initial noise curve for the week is shown here.
The curves for 2W and 4W don't show the expected improvements in noise at high
frequency.
Later in the week, a second, improved noise curve was obtained, this with an
inspiral range of 7.4Mpc.
This followed on fixes to AS4's local oscillator cable.
The gain in sensitivity was anti-correlated with SPOB levels, something that
poeople have been ruminating
on. Efforts to turn the input power up to 7W have begun.
Shot noise limits at high frequency.
A revised 4k noise budget is presented here.
A document describing modelling efforts and the output mode cleaner was posted.
2K IFO
The 2k is still reluctant to go into common mode; work continues.
Later, CM mode was re-established.
Periscope work
was summarzied.
The Big News: HEPI passed the 1-arm test on the X
arm with flying colors. With fully commissioned hydraulic preisolators under
the ETMX and ITMX, and with the X arm locked using a special
"low-frequency adapted" MC/PSL control loop, we measured about a
factor of ten reduction in both microseismic (.15 Hz) and stack-mode (1-2
Hz) motion along the X arm. This was achieved in both a nighttime run Sunday
evening and a daytime repeat Tuesday afternoon. The residual RMS velocity
was comparable to Hanford, and comfortably below the empirical "can't
lock" threshold.
For more details see
http://www.ligo-la.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=08/10/2004&anchor_to_scroll_to=2004:08:10:
19:41:46-giaime
Right now the other 6 HEPI chambers stand as follows; the ETMy is ready to
roll, HAM3 almost, and HAM4 needs a small mechanical adjustment before it its
SysID can be done to establish its control laws. HAM's 1 & 2 and BSC2 are waiting
QA checks and debugging. Problems with bad actuator valves are getting
resolved, but still a cause of some concern as we get to the last few chambers
(spares are gone, more on order but delayed). Other problems with
mechanical tolerancing and adjustments are getting resolved more easily as they
come up.
Work continues in alignment and commissioning of the interferometer proper,
including locking of the Y arm and PRM. We had some more work on the TCS
including hookup of a temporary chiller so we can play with both arms at once
while we're waiting for the final chillers to arrive.
We are putting in a best effort to align get the full IFO locked to give us a
warm and fuzzy feeling before....
The EMC retrofit in the corner station is on track to tear down the full
electronics system and rebuild in the new CDS high bay. So far all the parts
(even the rare 64-conductor cables) are in house or on their way "just in
time." Our game plan is to bring up the digital crates first, followed by
suspensions and PSL and then ISC and finally PEM. Cross-connects and cable
bundles are being prepared in advance as far as possible.
In preparation the 60 Hz and RF baseline measurements are in progress to
provide documentation of the benefit and guide the final tweaks to connections
and grounds.
HEPI and IFO engineering (Abbott)
---------------------------------------------------------------------
1. Working on the HEPI punchlist of items associated with the remaining
chambers in the corner station. ITMY is presently under scrutiny to
determine if there are any bad valves etc.
2. Reviewed MC/CM board for Daniel and provided feedback.
HEPI Engineering (Spjeld)
---------------------------------------------------------------------
- received GS13 Ham Sensor Platforms from the machine shop
- evaluated Parker Servo valve - actuator manifold registration issue
- received all HEPI Documentation Web-page files and software from Katrina
- link to HEPI web-site created on the LIGO Observatory Home Page
- working on publishing additional drawings, procedures, etc. on web-page
L1 Commissioning and HPLF (Amin)
---------------------------------------------------------------------
IPG replied last week indicating it would return our 100 W early this week.
Last Tuesday their scientists pleasantly informed me that IPG had damaged the
laser during a test run and consequently would not be returning the device to
LLO this week. The HPLF crew expects a delivery by two weeks. In terms of
infrastructure, HPLF has bought several two inch optics and their respective
mounts to conform with S. Whitcomb's requests. These optics should arrive
tomorrow.
Safety & security (Riesen)
---------------------------------------------------------------------
SITE SAFETY
1). I found no safety concerns during the weekly site tour.
2). Working with Bill Tyler on safety audit action items.
SAFETY TRAINING
Conducted a short safety briefing at the weekly staff meeting concerning
forklift safety and reminded all present that hunting season is nearly here and
well have hunters around the site. Also touched on the fact that deer are
plentiful in this area and auto/deer collisions can be deadly. Minutes
were taken for this briefing.
PROPERTY CONTROL
Working with Ed Chargois in the disposal process of computers and monitors that
are obsolete, broken and no longer in service.
Livingston Facilities (Sibley)
---------------------------------------------------------------------
The fan in AHU 1 at the south end station has had a failure of the fan actuator
on the outlet fan. We have contacted Howden Buffalo who is the owner of the Joy
Fan line. They don't have a service tech available at this time they can send.
They have requested we disassemble the fan to uncover the parts that will be
required. Star is now on site to see what will be required to do the job.
Data analysis (Franzen)
---------------------------------------------------------------------
Have succeeded in running off-line WaveMon jobs on the LLO Condor cluster.
However, since it was not possible to store the S3 RDS data on disk, the data
was taken from tapes. This slows down the procedure a lot which is bad news.
However, we have shown that it is now possible to run DMT jobs on Condor which
might be useful at a later stage. We will try to install DMT at the Condor
cluster at Caltech and rerun the S3 data if necessary.
General computing (Roddy)
--------------------------------------------------------------------
Successfully upgraded the spare (backup) LDAS VPN to OpenBSD 3.5 from
3.1. This was then tested successfully while Igor had a replication
process running between Livingston and Caltech. I am in the process of
creating documentation so that the other sites can also upgrade their VPNs.
Looking into a strange problem with Gnome 2.0/gnome-terminal on Solaris
9. No matter what the path is set to via bash, etc. when gnome-terminal
is started under gnome, it prepends /usr/bin to the users path. This does
not happen under CDE. Also, if gnome terminal is set to run as a login
process, this does not happen. This is at best a crude workaround.
Received a quote for the Solsoft software maintenance renewal. Solsoft
has modified their product line and there are a fewfeatures that would be
beneficial to LIGO, so I am checking into the cost of these.
Arkeia has been working well for the backups here. I am now satisfied
with the reliability and have started rolling it out to additional machines.
Received a quote for the Arkeia software maintenance and support
contract. I will be purchasing this in the next couple of days once I
receive answers to a couple of questions I had.
Still installing more software in the /apps directory. I now have Mozilla
1.7 running with the java and flash plugins working.
CDS (Parameswariah)
--------------------------------------------------------------------
Fixed the HEPI Watchdog screen to have links to the SUS and HEPI screens.
Installed Solaris 9 on two control room machines.
SEI-RM_SPARE_1 and SEI-OUT_SPARE_1 were found with bit noise. Rich M mentioned
it was working before. This is on the same 110B that had channel hopping few
days ago. Some how the 110B did not read the incoming signal correctly. We have
been seeing 110B problems - may be a bad 110B. Rolf will be here visiting us
next week. We will look into this problem.
l0tidal processor was frozen. Rebooted it and since the hepi watchdogs were
also on the same backplane, they were shutdown too. I made sure the offsets
were ramped down and brought backup.
Fixed database records for the ASC WFS PICO MOTOR coarse right actuation
buttons. Also Rusly, Valera and Peter King moved the REFL half waveplate to
rack in the PSL. Installed the code to run this on the l1iool1 which also runs
the PSL laser power half wave plate.
Made measurements to the applications startup timing at LLO and found it to be
comparable if not better as compared to LHO. Looking into ways of speeding up
the process.
HEPI ITMY POS Sensor inputs were reading garbage. Rebooting the l1hepi2
processor fixed it. A 110B problem again.
hepi1 and hepi2 processors stopped running last night. Shutdown all hepi drives
after Valera called me at home. Restarted them this morning.
CDS software support (Khan)
--------------------------------------------------------------------
1) Developing a script for making generic EPICS database
2) Improving the timing of 1-WireDAQ module
3) Helped Sany in making filters for his mode cleaner locking E2E model.
Data analysis (Yakushin)
---------------------------------------------------------------------
LDAS admin:
1) After 4 visits from SUN, old dataserver is finally fixed and backups are
running.
2) Set up a database replication configuration between LLO_TEST@LLO and
RLLO2@ldas-dbcentral. Data insertion at LLO is done using slope. Replication
have been successfully running since Sunday. I'll let it run while I am at LSC.
3) Installed GSL on the cluster and ldas-grid.
4) Configured nodes60-5 to access /samrds so that Ken Franzen can run WaveMon
(which needs L1 S3 RDS and there is not enough disk space to spray them on the
cluster at LLO). At first /samrds was mounted from gateway and it did not go
well: all the computers that mounts something from gateway (all LDAS computers
do) were slowed down. Remounting /samrds from dataserver instead made a huge
difference. Now only computers that are doing actual job are affected.
5) Started giving user accounts on Condor@LLO in response to
applications done through PSU web site.
LDAS data analysis:
1) Ran the new Condor/DMT version v5 of waveburst on all S2 and MDC data. The sensitivity
is better than it used to be for all the waveforms, especially the improvement
is noticable at low frequencies.
2) Ranv5 of waveburst on S3 playground. The false alarm rate seems to be 10
times higher on S3 in comparison with S2.
3) Generated S3 playground MDC frames with sine-gaussian injections and ran
waveburst on it. The results are puzzling: triple coincidence sensitivity is
much worse for S3 (roughly factor of 6) than for S2. Also, previosly the
maximum of sensitivity was near 235 Hz, now 100 Hz is the best and the
sensitivity gets worse and worse as frequency increases. This needs a lot of
testing and double checking. Might be some error on my part or bad calibration?
e2e IO Suspension Modeling (Dodda, Jamal, Rogillio and Yoshida)
---------------------------------------------------------------------\
In the past three weeks, we made a substantial progress in our e2e modeling of
IO. We continued study on characterization of HAM table motion based on our
model of coupling between translational floor motion and table yaw motion that
we reported previously. Using this model and the DAQ signal of the
accelerometer attached to the cross-beam of HAM1 and HAM2, we can now estimate
realistic table top motion for MC1, MC2 and MC3, respectively, taking into
account the locations of these SOS?s on the HAM tables. Using these table
motions as suspension point input, we computed the optic motion for MC1 - MC3,
respectively, and the resultant osem pendular and yaw signals showed reasonable
agreement with the corresponding osem DAQ signals recorded concurrently with
the accelerometer DAQ signal.
In the 3D suspension box inside the SOS box, we found that previously we had
not set the magnet-to-magnet distance on the mirror surface correctly, and
fixed the problem. After this correction, we re-calibrated the local damping
gain, and found that gain setting of 3 to 5 (as the number given in the OSEM
box inside the controller box which can be assigned in a parameter file as
FilterPos.OSEM.gain=3.0) gives local damping with a quality factor of ~10. This
improvement in the local damping of MC1 - MC3 makes it easier for us to lock
our MC box. Although we still have to make the cavity Q lower than the reality
(we use transmission of 2% as opposed to the designed value of 0.2%), the MC
box outputs stable optical power from MC3, with the above-mentioned realistic
HAM1 and HAM2 table motion as input to the suspension points.
for the 2 weeks ending 8/12/2004
see also the CDS weekly meeting minutes in the commissioning archives:
Rolf Bork reporting
8/5
Have begun work on modifying 40m software to use the latest
version of DAQ used at the sites. Primarily, this involves changing the VME
front end software. In the process, also trying to take care of fix requests
from Rana. Alex has been taking care of some of the sysadmin tasks, such as
loading the requested applications (perl, nedit, etc.).
Right now, I'm working on the 40m suspension code changes. I
wanted to do a couple of things in the process:
1) Make it more modular. The new code is set up such
that it can be passed the arguments of how many optics to control and which
ones they are.
2) Make it more like the sites such that we can test
more here before shipping code to sites. This will involve some hardware
changes:
- Remove the two Linux PC which now
run the realtime control for the 8 vertex optics.
- Replace the VME crate which has
all the vertex optic I/O modules with a split backplane unit.
- Install two VMIC7805 (2.2GHz P4)
VME processors into the VME I/O crate. These will take over the tasks now being
done by the two Linux PC and the 1.26GHz P3 in the VME I/O crate. In testing so
far (new code performs all present functions, just needs more optimization and
testing), each of these new VME processors will easily handle four optics each
(<50usec). This arrangement will also remove a lot of traffic from the
reflected memory network (signals now running back and forth between Linux PCs
and VME CPU).
- Move some I/O cables around in
the VME I/O crate to match modularity of new code. Note that this does not
require rewiring of cross connects. This is only moving a few I/O module front
panel connectors.
3) Generate as much of the suspension control EPICS
as possible using the script already use for some other systems. I've
already completed most of this.
I hope to have this piece wrapped up early next week.
Updating the rest of the systems should be much simpler.
8/12
- Working on new suspension code for 40m lab. I have a VME
crate here put together will the same modules presently at the lab, with the
addition of two Pentium 4 VME processors. The code for 8 optics is running, and
at first glance, seems to be functioning properly in the allotted time. I now
need to go through the testing systematically to ensure everything is fully
functional, including the new DAQ. However, this testing will be delayed for a
week as I need to be at LLO next week to help them out. In the meantime, Alex
will continue the conversion to the latest DAQ by modifying the ETM and IOO
software.
Jay Heefner reporting (8/5)
Fiber Optic Timing Link (Sander)
=========================
Ben and Sander will test the link on the 40 Meter next week.
LLO EMI Retrofit (Mohana)
==========================
- LSC and ASC wiring diagrams are complete, except for
designators. To complete the drawings we are waiting for rack drawing markups
from LLO.
-IO drawings are started, but it is unclear as to what is
actually installed where. Mohaha is talking to Rus.
TNI (Jay)
============================
- The arm cavity servo amp is complete and out for fab. Don
will stuff and test the board next week when it come in.
- EPICS software is being developed to control the boards,
auto lock the ifo and monitor lock times.
LSC PD Redesign (Ben)
============================
- PCB out for fab.
- Boxes on order.
ISS (Flavio)
============================
- Board layout almost complete. It should be done today.
- Front panels are designed.
- Parts will be ordered when layout is complete.
- Boards should be in house to test by end of next week.
40 Meter (Jay, Ben)
===============================
- Ben has issued a drawing for the current rev of the MZ
servo amp.
- Normal clean up and support of operations.
Helena Armandula
The procedure to clean small optics (E990034) has been
revised; the new revision "C" will be placed on the DCC
shortly.
The revision calls for new materials and equipment as well
as a new cleaning procedure.
The clean room wipers have been replaced for more effective
ones in the control of particle generation.
Ionizing bars are now required inside the cleaning bench, to
provide effective static neutralization during the cleaning process.
http://www.terrauniversal.com/products/hoods/ionizingbar.shtml
Implementation of the new cleaning procedure should take
place immediately, the wipers could be replaced when the present stock needs to
be renewed.
PeterKing
The 10-W laser power supply was repaired. Problems
arose when the time came for the power supply to ramp up the pump diode
current, this led to an over current condition in the laser power supply.
The fault was found to be with the internal 24V power supply.
I finished writing the additional on-line documentation
reflecting the slow loop code. The nroff help file can be pulled up from
the MEDM screen.
Ken Mailand
I contacted Keita Kawabe @LHO re. the 4 element mode cleaner
cavity.
I sent the finished main cavity spacer drawing along with
the PZT holder bracket, and sent to Keita to comment on or approve for
manufacturing.
I asked Mike at the CES shop here if he had time to do it, he said yes and that
he has machined invar before.
The optical parts will not be delivered for 4 weeks from CVI according to
Keita.
Mike Smith
I am in the process of using Zemax to model the OMC design
by K. Kawabe. I have developed a single pass, sequential model, and a complete
non-sequential multi-pass model that allows multiple rays to build up in the
cavity. I am trying to analyze the amount of astigmatism present in the design,
using geometric ray tracing. I will probably have to propagate Gaussian beams
in order to evaluate properly the cavity design , and I don't know how to do
that yet.
Lee Cardenas, Liyuan Zhang
OTF Lab. (W. Bridge)
The chamber has two samples, white Ceramabond, and disks of
TRA-BOND #2254
color light brown epoxy. Cavity is locked.
We are taking measurements everyday.
Absorption Test Measurement prototype in standby!
absorption of the 2" dia. x 1.5" annealed sapphire
piece from Stanford has been completed!
The scan of the 2" inches sapphire mirror(cracked in
the middle) annealed from
Standford
has been completed.
It was measured at Zsapphire= -9, -18, -27 mm. and at
Zair= -5, -10 & -15 mm. The results are released by Dr. Zhang's graphs.
We also have completed the scanning of the sapphire mirror
ANCO #2, 3.0" Dia X 1.0" thick.
All the results from the data are being treated by
Dr.Zhang's.
The results are the same as both of these sapphire mirrors
was previously measured.
Scatterometer system
in standby!
Scatterometer head under modification!
We need to modify some more on the scatterometer head,
We need a complete and freely motion that can be easily
moved.
so we are going to open some slots instead of changing
position
by moving the screws.
The head is heavy and bulky.
The Quantronix 60 watt laser
Optics set-up, and alignment completed1
The running of the cooling system on the laser is done
periodically
to maintain the plumbing and the cooling of the mirrors
cleaned.
OTF Lab at Lauritsen ROOM 38 NO CHANGE!
Cavity #3
Chamber is pumping with (6) disks of TRA-BOND #2151 color blue
epoxy and (4) disks of TRA-BOND #2902 color silver
epoxy. The cavity is locked.
We are taking
daily measurements.
Cavity #2
We are taking apart the old chamber and placing orders for
the new parts needed to build the new cleaned and bake chamber.
PSL Laser lab
Laser in standby
I disassembled the old pre-mode cleaner brought from
Standford. (complained a poor transmission from mirrors)
removed the mirrors and cleaned them both. They are ready! For
the new PMC.
Misc.... tasks:
sent a 135lines/mm Co2 grating to LLO.
no report
no report
Simulation and Modeling
(Bhawal)
E2E Weekly Physics meeting
-------------------------------
Sany Yoshida and his students Raghu and Nafis from Southern Louisianna
University discussed recent simulation results related to Mode Cleaner control
and related topics of mode-matching and use of e2e modules. Matt gave an
overview of commissioning activities at Hanford.
BS curvature vs Differential heating
--------------------------------------------
(Hiro) Hiro ran additional test cases to understand the effect of BS curvature.
One clear effect predicted is that ITMx needs to be less converging than ITMy,
i.e., we need to heat ITMx more using the thermal compensation system in order
to make upper and lower sidebands balanced. This is consistent with
experimental measurements at LHO. Hiro is trying to apply the same procedure in
the simulation as the experiment to optimize CO2 laser power and to compare the
observations and simulated result to understand better what is happening.
FFT Locking
-----------------------
(Xiao, Hiro) Hiro and Xiao tried to lock the FFT simulation based on the same
algorithm as the LIGO LSC using Asym, Refl and Pob / Poy. L+ and L- (DARM,CARM)
changes by ~10^-12 m, while el+ and el- (MICH,PARM) change by several x 10^-11.
Observable physics quantities, like gains in Michelson or in arm, both CR and
SBs, change only at the 1 % level. Other quantities, like sensitivity defined
as error signal / L- displacement changes at the same 1 % level.
FFT and WFS
--------------
(Biplab) After calculating WFS matrix and signals, obtained equivalent tilt
offsets representing asymmetries in the mirror phasemaps. The values are
typically a few 10^-8 rad or lower (LIGO I requirement is 10^-8 or less). Tried
to manually null these angles but it did not work well. There are improvements
in some angles but others get worse. So, we tried to take a closer look at the
fields. Analyzed two sets of Hanford 4K FFT simulation data - one set uses raw
mirror maps and the other set uses mirror maps with tilts artificially removed
beforehand. Naively we would expect that for the case using tilt-removed maps
the WFS signals would be less but that did not turn out to be the case for some
degrees of freedom. Decomposed the beams at various ports into Hermite-Gaussian
modes and trying to understand how these two cases could be different.
e2e with BS curvature
-----------------------------
(Hiro) Hiro has modified e2e, based on the modification by Biplab, so that BS
curvature effect is included in SimLIGO simulation. It is impossible to
implement consistently. So, some approximation is implemented. Unfortunately,
SimLIGO would not lock even when BS has very large curvature. The code is being
checked now. The SB imbalance effect was tested by injecting phase noise in EOM
by comparing SimLIGO runs with differentially heated ITMs.
Alfi
----------
(Bruce)
- Completed work on bundles and
bundlers.
- Implemented data port creation
features requested (PR
463).
- Fixed bundler inconsistency
problem (PR 469).
- Designing opaque box
implementation (PR 251).
(Melody) Modifying the ALFI bookkeeping objects to allow displaying the FUNC_X
data variables in a friendlier user interface. The dialog will show the member
variables in the table just as any parameter setting variables (instead of
having to look for them in the C++ code).
Data Analysis Activities (Lazzarini)
Creighton:
Repaired FrComputeFStatistic after it was broken by changes made to modules
common to it and ComputeFStatistic. Started working on input files for
the Condor pulsar search.
Mendell:
The StackSlide code ran on the LHO cluster this week, producing preliminary
upper limits for comparison with the Hough and Power Flux searches. These
will presented at the PULG F2F meeting on Aug 15, 2004 and at the LSC meeting
pulsar session on Aug 16, 2004.
Weinstein:
- Careful read and review of S2 burst untriggered paper.
- In process of careful read and review of S2BNS inspiral paper.
- Looking at galactic distribution of burst sources.
Yakushin:
1) Ran the new Condor/DMT version v5 of waveburst on all S2 and MDC data. The
sensitivity is better than it used to be for all the waveforms, especially the
improvement is noticable at low frequencies.
2) Ran v5 of waveburst on S3 playground. The false alarm rate seems to be 10
times higher on S3 in comparison with S2.
3) Generated S3 playground MDC frames with sine-gaussian injections and ran
waveburst on it. The results are puzzling: triple coincidence sensitivity is
much worse for S3 (roughly factor of 6) than for S2. Also, previosly the
maximum of sensitivity was near 235 Hz, now 100 Hz is the best and the
sensitivity gets worse and worse as frequency increases. This needs a lot of
testing and double checking. Might be some error on my part or bad calibration?
LIGO Data Analysis Software Infrastructure
Software Systems (Blackburn)
Changed FrVerify application used to verify frame files to be based on calls to
read() function instead of mmap() to resolve issues within the SAM/QFS
filesystem.
Two new LSC software developers were added to the GDS CVS repository after
confirming with John Zweizig that this was an approved action.
Modified frameCPP I/O library to throw an exception if either of the
diff_zero_suppress_short or zero_suppress_int_float compression modes were
requested by the application layer. These compression methods are suspect and
can lead to data corruption. LDAS doesn't allow these as a result of filters
for these modes at the TCL layer, but other LIGO/ LSC application that are based
on frameCPP may or may not have the necessary filters to prevent data
corruption so the constraint has been pushed down into frameCPP. NOTE: Normal
processing of LIGO frames do not use either of these methods so no legacy
issues exist. LDAS has had unit tests for quite sometime that identified this
issue in development, not in usage.
The effort to port LDAS to Solaris' c++ compiler has experienced excessive
damping due to particulars of the compiler not meeting all the supported
syntaxes of the ANSI/ISO standard. It is looking unlikely that we will succeed
with this port in time for the release. A new release of the compiler from Sun
was even tried out in hopes that it was more compliant. We will have to resort
to "dumbing down" our use of some C++ language constructs to succeed
which will take more time.
Fixed PR 2651 - zooming too far in the cmonClient results in an endless loop.
Closed PR 2419 - default value of "enable_memory_mapped_io" in
frameAPI to be zero (false).
Placed PR 2685 into feedback - ldas generated emails should have consistent
unambigous subject lines with LDAS system name and originating LDAS API
included.
Placed PR 2151 into feedback - faster cleanup of LAM processes by using SUDO
enabled commands on cluster nodes.
Closed PR 2653 - updated frameAPI's resource files to include defaults for all
createRDS frame inspection flags.
Continued to work on PR 2392 - generalize the definition of an interval in the
diskCacheAPI to improve performance and efficiency.
Worked on improving the syncronization functionality of the LDAS user account
cgi-script and its interactions with the controlMonitorAPI.
Completed all integration and system testing using LDAS version 1.94. Results
are posted on the LDAS-DEV website. Added tests for testing complex data sets
in the FrADCs as well as testing of GEO frames in LDAS in datapipelines. There
are some issues to be sorted out with the GEO frames but they may be specific
to the LALwrapper code and not to LDAS code. This is "to be
determined".
Had to reboot the lightWeightAPI at LLO this week. It was using over 99% of the
boxes CPU resulting in unresponsive ancillary processes. It is suspected that
an XML file with unusual syntax triggered the cascade but this is not confirmed
yet.
The TCL/Globus software was broken by recent code changes to the shared
SWIG/Globus CVS repository by the Python team. After careful review of these
changes on our part, we are once again able to compile TCL/Globus.
Presented report on the lengthening of the RDS frames to the LSC Computer
Committee. Informed the DASWG of the report at the recommendation of the LSC
Computer Committee.
Identified an issue with email sent from the LDAS user account cgi-script on
LDAS-SW which caused users to not get notification emails of account status.
This was traced to spam hammering acrux. Plan to shift over to becrux for the
time being and possible change the code to use sendmail instead of the PERL
module method for emails currently in the code.
Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* Continued the transfer of non-run second trends for both LHO and LLO
(about 1,000,000 files total) from HPSS to SAM-QFS (and their subsequent
deletion from HPSS). We had a brief period of time when we had two tape
drives working in HPSS, but now we're back down to one.
* I started with LLO: of the 458620 LLO files, we've transferred 322562
as of this writing (70%).
* I've also LHO second trends transferring--we've transferred 88429 of
526258 files as of this writing (14%).
* Got Mike Gleicher to make the SFS backups in HPSS to work correctly so
they don't fill up the /opt/sfs_arch disk.
(Hari Pulapaka)
* Working mostly on the 1% data transfer from sites.
* Spent time modifying publishing, so that it can publish a list of files.
* Spent time debugging changes made to the publishing script by Ben.
* Wrote a small utility to calculate the average speeds achieved during 1%
data transfer.
(Al Wilson)
* Node8 Replaced EEPROM. Will burning over the weekend.
* Still have not received the replacement MD forpcdev1 it is shipping via
ground.
* Interfacing BB to tripwire. Work is progressing.
(Stuart Anderson)
* Testing SAM-QFS 4.1.9 and currently pushing ~10 open cases with Sun to help
ensure version 4.2 has what we want for the next Science run.
MIT
---
(Keith Bayer)
* With Stuart's help, mirrored root disk on dataserver and created
mirrored swap partition.
* Upgrading filesystem / firmware on old pcraid#1.
* Received pcraid#7 and mounted into rack.
* Installing o/s on pcraid#7 for benchmarks and burntests.
Livingston
----------
(Igor Yakushin)
* After 4 visits from SUN, old dataserver is finally fixed and backups
are running.
* Set up a database replication configuration between LLO_TEST@LLO and
RLLO2@ldas-dbcentral. Data insertion at LLO is done using slope. Replication
have been successfully running since Sunday. I'll let it run while I am at LSC.
* Installed GSL on the cluster and ldas-grid.
* Configured nodes60-5 to access /samrds so that Ken Franzen can run WaveMon
(which needs L1 S3 RDS and there is not enough disk space to spray them on the
cluster at LLO). At first /samrds was mounted from gateway and it did not go
well: all the computers that mounts something from gateway (all LDAS computers
do) were slowed down. Remounting /samrds from dataserver instead made a huge
difference. Now only computers that are doing actual job are affected.
* Started giving user accounts on Condor@LLO in response to applications done
through PSU web site.
Hanford
-------
(Greg Mendell)
* This week I am working with Ben Johnson to prepare the LHO cluster for the
DMT/Matlab/Grid workshop after the August 2004 LSC meeting.
(Ben Johnson)
* Fixed iced up 5-Ton liebert. This was most likely due to a very cracked fan
belt. The belt has been replaced, and air flow from the unit is much greater.
No freeze-ups again, yet.
* t3-1 (/samrds) controller went offline. The filesystem was quiet at the time;
ran samfsck anyway. The system has since been working for a week.
* Installed matlab7 on the ldas-grid box, the compiler has been
installed today (8/12/2004) on pcdev1 and grid.
* Copied and published some missing S2 L3 LLO frames to cluster/LDR with
Hari's assistance.
* Preparing the local grid install for the grid demo by Keith Thorne.
* Replaced two T3 disks in fault.
General Computing (Wallace)
MIT:
(Keith)
-Starting upgrade of Ultra 10s to Solaris 9
-Deactivating and recycling old Sun monitors
-Working on getting matlab R14 for gc
Livingston:
(Shannon)
-Successfully upgraded the spare (backup) LDAS VPN to OpenBSD 3.5 from
3.1. This was then tested successfully while Igor had a replication
process running between Livingston and Caltech. I am in the process of
creating documentation so that the other sites can also upgrade their VPNs.
-Looking into a strange problem with Gnome 2.0/gnome-terminal on Solaris
9. No matter what the path is set to via bash, etc. when gnome-terminal
is started under gnome, it prepends /usr/bin to the users path. This does
not happen under CDE. Also, if gnome terminal is set to run as a login
process, this does not happen. This is at best a crude workaround.
-Received a quote for the Solsoft software maintenance renewal. Solsoft has
modified their product line and there are a few features that would be
beneficial to LIGO, so I am checking into the cost of these.
-Arkeia has been working well for the backups here. I am now satisfied
with the reliability and have started rolling it out to additional machines.
-Received a quote for the Arkeia software maintenance and support contract.
I will be purchasing this in the next couple of days once I receive answers to
a couple of questions I had.
-Still installing more software in the /apps directory. I now have
Mozilla 1.7 running with the java and flash plugins working.
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
- Most of my time has been spent getting ready for the F2F, LSC and detector
summer camp activities. I've coordinated computer access, created user
accounts, purchased a Matlab compiler license, compiled lists of registrants,
lists of MAC addresses, lists of e-mail addresses, coordinated with John
Zweizig to put DMT environment variables in the global cshrc, and more.
- Notified Fred Raab of some finacial issues.
- Other misc. user support.
CIT:
(Mike)
-Spent a lot of timeworking on PC's that ended up getting infected with
viruses. I was able to clean these PC's. I also had to go around to many user
workstations and manually update their virus scan software due to LiveUpdate
not updating to the current DAT file.
-Worked on loading two-visitor workstations. These were complete rebuilds that
were loaded from scratch. I still need to load additional GC software.
-Worked on all NTSRV's updating an additional service packs and manually
updating anti-virus software due to the latest attack of viruses.
-Larry Jones: Laptop freezing up on him. After troubleshooting his laptop, I
had no choice but to reload this laptop from scratch. I load XP pro and all
GC/engineering software.
-I loaded XP service pack 2 on my VMWARE installations to run tests with our
current engineering/general computing software. I still have much more testing
to do before I start upgrading our XP users. So far everything seems to be ok.
-I worked on Mike Smith's laptop that has a corrupted OS. I ghosted this laptop
for him to recover his data. This laptop needs a rebuild.
-Other misc. onsite/phone user support.
(Veronica)
- LSC website: As the August meeting gets closer, more last-minute
changes/updates to be posted. Posted a bunch of technical papers for the LSC to
review.
- LIGO website: Working with Sharon on bringing her up to speed on web updates.
Helped her on working with Barry's Calcium calendars. We will continue going
through the how-to of the webpage maintenance. Updates to various pages of the
LIGO website. Working on roster updates. Working on scripts for automated
author list generating.
(Lisa)
- Made changes to the permanent archive backup scripts. Verified the
integrity of the changes. Randomly checked the archives for problems.
- Finished a script to reset conflicted dhcp addresses in the dhcp server.
- Chased down a couple of pc's that got infected with the latest varient of the
Bagle virus. We had a delay of a few hours before a virus definition that
could detect it was available for the mailserver. It was many hours after
that before a definition was available for everyone's norton.
- Tried to jumpstart ancha, a blade1000, to configure it for the 40m. After
some hours of testing I determined that it has a problem with the motherboard
that prevents it from recognizing a hard disk.
- Higher than usual amount of user support.
Mail Stats 08/05 - 08/11/04
Messages Accepted:
Spam Rejected:
Viruses Rejected:
False Positives:
Total Mail Thru:
% spam blocked:
(Larry)
-Spent a number of hours clearing up purchase related items. Along the same
line thanks to Ed C. and Mike for tracking down items that had been delivered
to the wrong location. Received a couple of equipment orders and they will be
distributed when I return to CIT.
-Assisted in a number of computer configuration setups with different people.
Went over a number of items with Alan W. for the 40M lab. They will be
installing some new equipment in the near future. Delivered a few systems to
different people needing some specialty items.
-Cleared up a number of accounts and started moving old dead accounts to
another storage area.
-I was able to assist the group at Hanford (mostly working the parking lot) in
their program for the Meteor shower. There was a good crowd and I had a good
time with the group.
From: "Thomas Frey"
<tfrey@ligo.caltech.edu>
Progress Period from 08.06 to
08.12
OUT OF THE OFFICE ON
VACATION AUGUST 6th AND 9th
·
·
See http://www.ligo.caltech.edu/~tfrey/index.html for a
complete listing of all project related cost and schedule data.
Accomplishments:
·
·
Sub-system
PLANNING activities
§
§
Continued work on preparing web space for posting Adv. LIGO reports.
("The Whole Enchilada")
§
§
Posted updated data for FY05 non-labor cost breakdown, based upon
progress data through June 30, and includes an estimate of carry over.
§
§
www.ligo.caltech.edu/~tfrey/rd2005/
§
§
Continued to input of changes / preparing RFIs / responding to RFIs, as a
result of the meetings with Carol and the Subsystem Teams.
§
§
Worked with Janeen and assisted her with preparing a response to David S.
regarding budget data for FY2005.
§
§
Started input of progress data, for progress through
·
·
ROSTER DATABASE:
§
§
Input two authorship list corrections/changes provided by Peter Saulson.
§
§
Updated the source files so that Stan can issue the next draft for
review.
§
§
www.ligo.caltech.edu/~tfrey/authorship/compare.htm
§
§
Met with Stan and agreed on procedural issues related to the control and
collection of authorship list changes.
§
§
Continued to work with Irena to provide Barry with information regarding
FTEs and Institutions.
·
·
COST BOOK DATABASE:
§
§
Executed meeting with Carol, Phil, and
§
§
Posted list / statistics of accounts and activities as requested by
Carol, for
§
§
http://www.ligo.caltech.edu/~tfrey/rd2005/Acct_Stats.htm
From: Larry Jones <ljones@ligo.caltech.edu>
Advanced LIGO Seismic Structure
SEI Structure:
ASI's detailed Replan #2 was received, with a breakdown for the Cost To
Complete (CTC). Significant cost increases were reported in Project Oversight,
Design & Analysis, and Fabrication, and the need for an Integration
Engineer has been identified. ASI presented options for descoping, including a)
deleting the HAM unit design task, b) reducing the documentation for the BSC
unit design & eliminating the final verification model and c) deleting the
assembly, testing and cleaning of the BSC unit prototype. Other options are
also being considered. The Advanced LIGO will be making a proposal to the LIGO
Executive Commmittee on August 23 on how to proceed.
Corrections to the previously-received BSC unit fabrication costs are being
made by ASI to properly account for G&A and Fee; these are expected
shortly.
Delivery at LASTI of the BSC structure prototype is now scheduled for
Actuators:
Kyle Ryan is proceeding with preparations for conducting a Thermal Vacuum test
of the large actuator at LHO.
PSI is proceeding with fabricating the 6 large and 6 small actuators for the
BSC prototype structure. We've agreed on a specification for anodizing winding
areas of the bobbins.
The precleaning of magnet wire for the BSC unit prototype actuators will be
done at Caltech, after all. Proper cleaning solutions for the paraffin in the
wire's dry lube cannot be used in Branford's ultrasonic wire cleaner, and
purchasing wire without the standard dry lube application has a prohibitive
cost. Bob Taylor will be modifying our fixturing used previously to better
maintain wire straightness during the process.
Displacement Sensors:
Nothing new
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: Janeen Romie <romie_j@ligo.caltech.edu>
AdLIGO Suspensions
Answering questions and tracking progress on the blade RFQ effort
with Gina Salone's help.
Provided Primavera updates to Thomas for the month of July. Provided budget and
schedule milestones for suspensions R&D FY05 to Carol and David.
Facilitating a meeting between Mark, Caroline, Norna and I about mirror tilts
based on some modeling and calculations Mark did. It will be tomorrow morning
at
Coordinated with Caroline, Calum and others on appropriate quad noise prototype
tasks for the new RAL engineer to work on after Oct 1st. Established the
paperwork required to kick-off this effort with Carol, Dennis and Justin. Met
the new RAL engineer, Tim Hayler, via phone and coordinated with him on a
proposed visit by him to Caltech for
Started conversations with Calum on a possible change in the interface between
the quad upper structure and the catcher. Russell and Calum worked on this last
week while Calum was at the
Asked Mark Barton to consider starting up work again on the earthquake stop
testing or coordinating with Gregg Harry to find someone else to finish up this
work.
Starting convesations between Gari/COC and SUS team members on clocking
requirements of the sapphire optic.
From: Peter King
<pking@ligo.caltech.edu>
Had a telecon to review the progress and
budget for the PSL.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Advanced LIGO SUS - SF2 glass
A brief experiment was conducted to evaluate the effects of temperature on SF2
glass.
Details can be found in T040159-00-D; the document was placed in the DCC OUT.
From:
Bill Kells <kells@ligo.caltech.edu>
I am down to
the wire on presenting (or at least airing and discussing) ideas I have been
working on concerning the propagation of intensity noise in AdL. Its become a
much more intricate topic that my original questions. For instance I now
believe that the anticipated suppression of Intensity noise ("double
cavity pole") is not what has been assumed. At least the problems are all
delineated so that I will be able to discuss with "experts"
conveniently next week to see if it all holds water.
From: Michael Smith <smith@ligo.caltech.edu>
OUTPUT MODE CLEANER
I am in the process of using Zemax to model the OMC design by K. Kawabe. I have
developed a single pass, sequential model, and a complete non-sequential
multi-pass model that allows multiple rays to build up in the cavity. I am
trying to analyze the amount of astigmatism present in the design, using
geometric ray tracing. I will probably have to propagate Gaussian beams in
order to evaluate properly the cavity design , and I don't know how to do that
yet.
For additional information about this report, contact whitcomb_s@ligo.caltech.edu