LIGO Weekly Report

The list of current actions revised to reflect the status of open actions assigned through July 17, 2003 may be found at ACTION LIST.

Web pages for the DCC give simple how-to's for document numbering, easy access to the latest on-line documents, and search capabilities for the DCC database. Take a look. . .

Advanced LIGO (Frey)

Project Web Site for posting schedule and progress related data continues to be updated with the latest and greatest.

I have requested input by Friday, July 25, 2003 for the Annual Report and Request for FY 2004 Funding. So far, I have received material from the Detector Group, Hanford, and the 40-Meter.

The following change request has been distributed Executive Committee for possible discussion July 28, 2003.


CR-030015	FY 2003 Livingston Observatory Detector Maintenance Expenses (Increment)	R. Wooley	July 14, 2003

Quality/Safety (Tyler)

LIGO Hanford Observatory (LHO) and Interferometer Operations (Raab)

LIGO Livingston Observatory (LLO) and Interferometer Operations (Zucker)

-------------------------------------------------------------------
Interferometer Commissioning
--------------------------------------------------------------------
Last week was mainly spent baselining relevant performance data in
preparation for this week's vent. High-accuracy (order 1% errorbars!)
cavity storage time measurements indicated no significant degradation in
the mode cleaner or arm cavities. Additional measurements of the
anomalous PRC -> AS_Q coupling factor continued to show the anomaly;
according to some theories this will be changed by the correction of the
PRC length, although recent H1 measurements seem to show a similar
effect. RF phases were measured for best signal separation (these will
definitely change with the PRC length change). The WFS2 head was
repaired and its response characterized. An interesting correlation was
found between known seismic stack features and RF sidebands picked up by
a radio receiver tuned to the modulation frequency (!).

The vacuum system incursion was undertaken to correct the recycling cavity lengthtand Schnupp asymmetry, install protective baffles in the input optics section, and re-orient viewports to center them on critical beams. The first part of this week we completed preparations, primarily tuning work procedures, prepping the materials and re-re-re-calculating the planned optic displacements.

We received the vacuum baked baffle components from Caltech and
assembled and checked them on Tuesday, then backfilled the corner
section and Y manifold and loosened 5 doors Wednesday. Thursday morning
three crews began work at 6 am; Team A (led by Gary Traylor, who was
also overall vent coordinator) went into BSC1 to verify the optics
placement and then translate ITMy 34.6 mm closer to the beamsplitter,
Team B (led by Harry Overmier) went into HAMs 1 and 2 to install the new
baffles, and Team C (under Rich Riesen) did rigging, door
removal/replacement, viewport modifications, and QA. All tasks were
completed, preliminary in-air alignment checks and table leveling done
and QA punchlists completed by 7 PM and by 8 PM the doors were back on
and roughing started (about 36 hours vent to pump, 14 hours with doors
off). At this writing (9PM, i.e. Miller time) the system is at half an
atmosphere.

Tomorrow after crossing over to turbopumps we will light up the mode
cleaner and realign the PRM to repeat the RF length measurement.
Saturday we hope to "peek" down the arms to verify global alignment, and
then quckly shut the gate valves to finish degassing the corner section.
Based on the quick turnaround we are hopeful that the water vapor gas load will become acceptable for full-time beam tube illumination within two or three weeks.

Safety and security (R. Riesen and J. Kern)
--------------------------------------------------------------------
A laser safety protocol violation occurred July 17. The
Modulated 980 nm RF Detector Test Laser was operated in the LVEA, while he area was "Laser Safe". The incident is under management review.

General laboratory support (J. Kern)
--------------------------------------------------------------------
Most of my time this week has been spent delving through our lab books
and old e-mails reconstructing the events of Aug '00 when we used the
CAS to move the BS and ITM-y. Crafted a plan to make 4 length
measurements within the recycling cavity when Gary and Andri enter to
move ITM-y. This will be done with a SS tape.

Civil & vacuum facilities (MZ for A. Sibley):
-------------------------------------------------------------------
Lightning protection and new technical ground field installation are
complete and backfill grading is done. The site entry gates are
operational; contractors have walked us through the technical
capabilities. We are developing a protocol and expect to activate the
gates next
week.

LLO seismic retrofit (J. Kern)
------------------------------------------------------------------------
Documentation for the NSF was completed for Gary's approval, and has
been sent to Washington. Quotations have been received for the External
Pre-isolator package. Bellows manufactured from SS 17-7 alloy have been
satisfactorily He leak checked. Correctingerrors on the actuator
drawing package that were noted by Stanford.

E2E Suspension Modeling (T. Findley and S. Yoshida)
------------------------------------------------------------------------
We continued making E2E suspension boxes with local damping control. Both
LOS and SOS box seem to respond properly to changes in the position gain.
In these boxes, the position gain can be given from a .par file (parameter
file) by assigning a number to a variable FilterPos.OSEM.gain (for Pitch
and Yaw, the corresponding variables are FilterPitch.OSEM.gain and
FilterYaw.OSEM.gain). For both LOS and SOS, a gain of 5 - 7 seems to damp
the optic well. Now that the boxes work, we are creating a realistic
suspension point motion that we can give to the box as input. We are
planning to compare the simulated optic motion with measured motion.

General computing (S. Roddy)
-----------------------------------------------------------------------
The vast majority of my time has been spent on LDAP migration issues
again. I have sucessfully authenticated a Solaris 9 machine to LDAP. The
aliases, auto mounts, etc. seem to be working. However, there are a
couple of things that need to be looked into. One issue is if Solaris 8
will work. There seems to be some differences in how Solaris 8 and 9 use
LDAP. However, it seems the best solution to this is to just update the
last few Solaris 7/8 straglers to Solaris 9. This should not be a
problem for the majority of the GC computers. I am still having some
problems with Linux vs. Sun's implementation of LDAP. I have not been
able to successfully authenticate a Linux machine against LDAP. This may
be a problem with the default behavior of Linux and LDAP. I have not
been able to find suitable documentation on changing the default LDAP
account mappings. However, what documentation I have found seems to
indicate that it *should* work. One other issue is that I have not
implemented start/TLS encryption. Linux seems to want this behavior by
default. Also, Solaris should use this, but until I have it implemented,
I will not be able to test this with Solaris. I plan to have this done
within the next couple of days. It seems the road to LDAP is a long one,
and perhaps still not standardized as well as it should be. However,
once it is working, it will solve a large number of problems.

LDAS (Igor Yakushin)
------------------------------------------------------------------------
LDAS admin:
1) Finished integrating the new nodes into LDAS.
2) All the nodes successfully passed the burn test.
3) Net burn test showed that node 31 communicates 10 times slower than
it should.
4) Second IDE RAID with AS_Q only RDS from both sites stopped working on
Sunday (same happened simultaneously at LHO, so there must be some date
sensitive bug in BIOS). ASA recommended resetting CMOS and intalling the
new version of BIOS. It helped. IDE box is up now but it looks like it
needs a disk to be replaced.
5) Helped Partrick Sutton with access to S2 RDS frames for generating
the final SenseMon calibration frames.

LDAS data analysis:
1) Fixing a bug in waveburst output to the database that Mary reported.

Detector/Technical Support (Coyne)

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

This week we opened the vacuum chamber and removed the collimating lenses on the arm-cavities' transmitted beams. We have reason to believe that these lenses are refocusing backscattered light into the arm cavities, and that this noise source may be dominant over some range of frequencies between 100 Hz and 1 kHz. We removed the lenses, realigned the transmitted beams to the output optical table (with diagnostic cameras), and fine-tuned the alignment of the cavities themselves. After we have verified the table balance, we plan to close the chamber and pump out on Friday, July 25.

While the chamber is open, we have also been checking and updating our map of where all of the optical components are. We have taken a number of digital photographs of the inside of the chamber, from several different angles, to use as a reference when the chamber is closed.

We also welcome the addition of a new graduate student to the group, Ivan Grudinin. As an undergraduate, Ivan worked with Vladimir Braginsky at Moscow State University. Please extend to him our warmest LIGO welcome!

LASTI

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)
Weekly Physics Meeting
------------------------
Ken Yoshiki, Sany Yoshida from LLO and Malik from U.Florida discussed
some results (obtained by Ken) on the effect of changing rec cav lengths
on various transfer functions with Biplab and Hiro. After that Virginio
from
Hanford described his and Cella's work on mechanical simulation of
blades.
Virginio and Luca also described current efforts in WFS and initial
alignment
control at LHO. Luca requested for a e2e-generated Table of WFS signals
from
just the recycled Michelson configuration.

FFT study of effects of phase-maps
-----------------------------------
Biplab studied what could be the effect taking zero-value throughout
the
outside region around the circular mirror phase map in the square
pixelized
grid window.
The maximum point in all mirror maps comes to be around 2-3e-9 meter.
The
minimum values are about -2e-9 for ITMs and -5-6e-9 for ETMs. So, the
outside region was filled up with + & - 5e-9 for all mirrors to get a
conservative estimate of the effect.
Results are as follows:
[CARRIER]
No phase-map. As-built H1.
   Contrast = 7.52e-5, Rec gain = 47, Inline Arm power = 3383
With just the circular phase-map. No filling of the outside.
   Contrast = 3.15e-4, Rec gain = 42, Inline Arm power = 3043
Outside region of the window is filled up with +5e-9
   Contrast = 3.21e-4, Rec gain = 41, Inline arm power = 2935
Outside region of the window is filled up with -5e-9
   Contrast = 3.10e-4, Rec gain = 43, Inline arm power = 3104
[SIDEBAND]
    Negligible differences among situations described above

So, the effect of a non-zero outside region would be about ~3% as a
conservative estimate.

Raghu could generate shot-noise curves from FFT results - only a factor
of 2 is hiding somewhere and he is sorting that out. After that he
would
generate shot noise curves of all 3 IFOs with the effect of phasemaps
included.

LIGO Simulation :
--------------------
In SimLIGO, the ASC box has been fixed to reflect the bug fix of the
modeler (the definition of yaw angle of ITM had opposite sign).
Xiao and Hiro studied the effect of the radiation pressure under
different alignment control. When QPDx and QPDy are used to force the
cavity beam axis to point to the ground x-y direction, the system is
stable. But if this control is turned off, i.e., the core optics system
can rotate or tilt as a whole, the system becomes unstable. The effect
of
the input beam tilt is also under investigation.

Source simulation :
---------------------
Jeff completed a calculation to generate a GW source and convert the
strain to the phase change of the laser propagating in a cavity.
He started working to implement two primitive modules in e2e. The first
implementation is good for a period source, i.e., the source has
a fixed frequency. After completing this model, he is going to work on
an implementation for a signal with arbitrary time series.

Dual Recycling model
------------------------
Keiko kept working on writing matlab code for the dual recycling
configuration based on the calculation by Osamu and Hiro.
Now, the model cavity can store fields to get high power in the cavity.
A few known problems exist, including a mirror velocity dependence,
and she is working to fix them and to validate the code.

Code development and maintenance
---------------------------------
(Biplab) Three items in code changes were committed this week.
(surprisingly, all 3 items are mutually exclusive of each other as far
as
files are concerned)
ITEM 1: The way rotation matrices are calculated is now changed.
   Previously, elements of rotation matrices were calculated explicitly
   using algebraic expressions obtained from a Mathematica code.
   This method is now replaced by a function in ModalModel.*
   [ModalModel.h, ModalModel.cc, mirror_operation_matrix.h,
    mirror_operation_matrix.cc, beam_operation_matrix.h,
    beam_operation_matrix.cc, fld_ops.cc]
ITEM 3: The default configuration of FP and Recycling summation cavity
   is that the HR coated surfaces face inside. In case the user
   wants to invert any mirror, the pitch of the mirror as seen from
   the coated side in right hand notation should get a sign flip
   - but that's not true for the yaw. This has been made correct.
   [recycling_cav.h, cav_sum.h]
ITEM 2: We decided to include the longitudinal shift that may
   originate from the transverse shift of the curved mirror.
   Since this is a 2nd order effect (= 0.5*TransShift^2/ROC)
   and in most of the cases mirror ROC is much larger than
   TransShift^2, this effect is very small but in certain situation,
   this could be not negligible.
   [smoptics.cc, cav_sum.cc, tri_mir_cav.cc, recycling_sum.cc]

(Melody)
   - Working on a prototype that dynamically creates C++ source based
     on box files definition. These souce codes will be compiled during
     runtime and combined to form a shared library.

Alfi
-----
(Bruce)
        - Added Show Main Window option in edit windows (PR 386.)
        - Troubleshooting Alfi use for OS X (PR 399.)
        - Fixed incorrect information informing users regarding
          creation of Alfi utility directories (PR 418.)
        - Fixed several minor bugs (PRs 423, 424.)

Software Systems (Blackburn)
The new raid disk systems at LHO exposed an issue in the naming
structure of
files which resulted in db2utils producing errors and unwanted emails
to the
system administrators. A had patch is currently in place and we'll
evaluate
a more permanent solution next week. This bug has been fixed in the
LDAS CVS
repository and will be part of the next release.

An attempt to thread the metaDataAPI's database insertion procedures was
found to be too buggy and was backed out of CVS this week as the C++
developer
is on vacation and the bugs make it impossible to test and other
development
of LDAS components.

The bugs in the dbaccess library used to verify table structure on the
server
to the data structures being inserted on the server have finally been
tracked
down after months of speculation. The cause was traced to the use of
inlined
virtual function used as accessors. These functions did not need to be
virtual
and removing this attribute solved the problem.

Unfortunately, the use of inlined virtual accessor functions have been
identified in a large percentage of the code base. These are
systematically
being removed with each component be corrected and thoroughly tested
before
moving on to fix another component. (Many thanks to Masha Barnes for
finding
this defect in the current GCC compiler and coming up with a
work-around!)
Currently the frameAPI/frameCPP have also been fixed and the ILWD
library
is being fixed.

The controlMonitorAPI performance was significantly improved this week
by
the use of the GNU grep instead of the TCL grep. This has improved
substring
extraction performance by a factor of 3x.

The restoration of the mpiAPI to the 0.7.0 version has resulted in a
more
stable MPI environment, but we still see 0.5% of jobs failing for
unknown
reason. We have determined that simply restarting the mpiAPI will
return the
failure rate to 0% for the first couple of hours. But eventially state
info
somewhere is corrupted and the failures reappear.

The new threaded components in C++ for the diskCacheAPI are integrated
with
the TCL layer. Ongoing testing and debugging continued this week. A
first
attempt at running this code on the LDAS-DEV system identified a cpu
race
issue that required more development in a standalone environment.

The five LDAS standalone LINUX servers have now been upgraded to Redhat
9.
These has now beened upgraded to have the necessary patch to run IBM's
DB2
version 7.2 on this platform. They now all have working version of LDAS
available for the developers.

Current nightly testing of LDAS is showing about 99.5% of the jobs
succeed
as expected with the remaining 0.5% of the jobs unexpectedly failing in
the
mpiAPI. We are also now able to handle over 1200 jobs per hour compared
with
the 480 jobs per hour that the previous (0.7.0) release of LDAS was
able to
sustain. Once the new threaded code is in the diskcacheAPI and
metaDataAPI,
we will probably be close to 3x the performance of LDAS over what was
seen
in the last release. NOTE: Some of this is due to faster network and new
nodes in the cluster but that accounts for only about 20% of the speed
up.

* Showed Greg and Ben how to import/export tapes in L700 and how
to set up archive sets in SAM-QFS.

* Tried to rebuild 3510 (10.0.0.10 at CIT) unsuccessfully. Controller
goes
completely offline from time to time...

* Monitored ldas-archive behavior for Greg's createRDS jobs which have
been running smoothly this week.

* Started writing all RDS data on /frame10 & /frame20 at LHO to 9940B
tape
(via SAM/L700). This is going well, so I added SFT data to the mix.

* Began re-archiving all full frame data on ldas-archive:/archive to
9940B
density tapes. The first phase, E9 data, should be done within 24
hours.

* Troubleshooted various bad/non-working nodes. Have found only two
confirmed bad units.

* M27 is on its last breath. Data from internal hard drive might be
recoverable.

* Started to look at cfdef for MIT. It seem that the files are in a
state
of neglect.

* Finished configuring 3 new servers at Caltech:
    * Dedicated Solaris machine for nightly LDAS builds.
    * New generation Sun machine for Database performance testing.
    * New ldas-sw server.

* Worked with Caltech Engineers to finish off the last few remaining
tasks for the room 215 Synchrotron work to support the new large
cluster.

* Worked with Keith and company setting up the racks and other items
for the
LDAS computer room at MIT.

* Net burn test showed that node 31 communicates 10 times slower than
it should.

* Second IDE RAID with AS_Q only RDS from both sites stopped working on
Sunday (same happened simultaneously at LHO, so there must be some
date
sensitive bug in BIOS). ASA recommended resetting CMOS and intalling
the
new version of BIOS. It helped. IDE box is up now but it looks like it
needs a disk to be replaced.

* Helped Partrick Sutton with access to S2 RDS frames for generating
the final SenseMon calibration frames.

* Continuing to work with Robert Schofield to create an intersite S2
PEM RDS for his analysis.

* Ben Johnson, Dan Kozak, and I completed the initial set up the L700
tape library system and samfs at LHO. Dan solved most of the
technical
problems we ran into, and is running further tests.

* Our old 64 node beowulf cluster was shipped to MIT last week and has
arrived there.

* L700: The L700+SAMFS seems to be operating smoothly now, once the
device and its hosts were set as Fabric devices on the FC switch. Dan
Kozak has been running archive jobs on the robot since Wednesday July
23.

* T3: t3-1 still has the wrong SCSI target ID on fb3. It is listed as
t33 on fb3 (as of Tuesday July 22). /frame10 has been umounted and
removed from fb3's mcf file.

* Nodes: Finished a 4 day burntest on nodes [19,60]. All passed save
node46, which had 1/3rd of the bootstrap files of the other nodes, 1
of
the 250 files it produced also failed its md5sum.

* tekoa: The BIOS on tekoa was updated to 1.0c on Monday. It was found
in a frozen state today (July 24). Though it rebooted successfully,
which is was not able to do on Monday.

* GigE switch: Many, Many difficulties with the Foundry switch this
week.
        a) Multiple port failures
        b) Eventually a whole module went offline
        c) 04:00 Thursday July 24, the switch went bonkers, causing many
           network connections to fail. Removal of cables from module
           7 allowed the switch to return to a normal working state.
        d) I will perform a BIOS update on the switch July24@13:00.

Data Analysis Activities (Lazzarini)
Chatterji:
* Development continues on a library of Matlab algorithms for burst
data conditioning and cross-correlation analysis. Web access to
the CVS repository is available at the following URL.
http://ligo.mit.edu/cgi-bin/cvsweb.cgi/matlabDataCond/
* A Library of ILWD data conditioning filters for burst analaysis are
now available at each LDAS site in the following directory.
ldas_outgoing/jobs/ilwd/filters/S2/
See the README.html file in the same directory for more information.
The Tcl script for launching burst analysis LDAS jobs has been
modified to select filters from this library based on science mode
segment number.

Creighton:
Continued to work with Birmingham group on including second-order
Doppler effects in waveforms from pulsars in binary systems. Started
writing routines for template placement for all-sky pulsar searches.

2) Continued worked with my SURF student Anah Mourant on her pulsar
parameter estimation code for LALApps.

Reilly, K.:
I spent time re-calculating coherence functions and comparing different
methods
of average coherence calculation. Looked at coherence on 600 second
and 60 second
time scales. Made plot of the average PSDs over the entire S2 run
(0-512 Hz range).
This work required some matlab coding. All of this work has been
committed to
the stochastic elog. I have been working with Tania R. a bit on how
best to do the
software injections and how to integrate this with the upcoming grid
demonstration.
I have decided that I will go forward with DSO modifications within
ldas and for
the time being I will abandon the idea of making the DSO working
outside of
LDAS. I am reading through the DSO line by line before making changes.

* Still working with Evan Ochsner on the inspiral search code. We ran
into a nasty problem, which occurs when running within LDAS but not when
running in standalone mode, and are still working out how to debug it.

* Following up on an observation by Bruce Allen, generated a list of
times
during S2 when RDS data isn't available even though the raw data is
available. These are mostly related to gaps in the raw data, which get
extended by 16 seconds at each end in the process of generating RDS
files.

* Investigated a report by John Whelan that incorrect detector geometry
information is being stored in frame files. Sure enough, the S2 frame
files have bad geometry information! The information is correct in the
S1 frames, but the values were not properly updated (to reflect changes
in conventions) when switching over to the version-6 frame format prior
to S2.

* Did another careful read of the final draft of the S1 pulsar paper
before approving it for release to the LSC.

Weinstein:
- Working on S1 burst paper.
- formulated a comparison of our result
with the results from the UG/MPQ coincident run of 1989.

Yakushin:
1) Fixing a bug in waveburst output to the database that Mary reported.

Lazzarini:
Looked that the data Kaice has generated for the 2-month averaged
coherence spectra for the 3 interferometer pairs.
For the two transcontinetnal pairs, I generated hiostograms for the
values to come up with a way of setting a threshold on coherence,
thereby
allowing us to exclude those frequency bins exhibiting coherence above
the statistical level expected for a 2 moth observation.

General Computing (Wallace)
MIT:
(Keith)
-Security audit with Larry
scanned several machines (mail server, nfs server, file server)
-Installed DVD burner in library computer
-Installed replacement RAM in nergis's laptop

(Larry)
-Working with Keith on the local MIT security check. Performed various
scans
and checked a number of machines out. So far things look pretty good.
Most of
the problems are minor and will be addressed in the report when we are
finished.
-Went over a number of configuration issues for different users.

Livingston:
(Shannon)
-The vast majority of my time has been spent on LDAP migration issues
again. I have successfully authenticated a Solaris 9 machine to LDAP.
The aliases, auto mounts, etc. seem to be working. However, there are
a couple of things that need to be looked into. One issue is if
Solaris 8 will work. There seems to be some differences in how Solaris
8 and 9 use LDAP. However, it seems the best solution to this is to
just update the last few Solaris 7/8 stragglers to Solaris 9. This
should not be a problem for the majority of the GC computers. I am
still having some problems with Linux vs. Sun's implementation of LDAP.
I have not been able to successfully authenticate a Linux machine
against LDAP. This may be a problem with the default behavior of Linux
and LDAP. I have not been able to find suitable documentation on
changing the default LDAP account mappings. However, what documentation
I have found seems to indicate that it *should* work. One other issue
is that I have not implemented start/TLS encryption. Linux seems to
want this behavior by default. Also, Solaris should use this, but until
I have it implemented, I will not be able to test this with Solaris. I
plan to have this done within the next couple of days. It seems the
road to LDAP is a long one, and perhaps still not standardized as well
as it should be. However, once it is working, it will solve a large
number of problems.

Lazzarini NOTE: I have been discussing WAN upgrade options with Joel
Tohline (LSU).
He reported of a productive meeting with the Bell South
telecommunications executive
whom he knows. Barry and/or I will plan a visit to Baton Rouge to see
how LSU, LLO, and Bell South can
work together to expand the connection from LLO to LSU to OC3 or DS3.

Hanford:
(Christine)
- Continuing work on the configuration for the gige network. The gige
ports on the router are not usable for LAN connections, it will be 4 to
6 weeks to get an interface card that is. Meanwhile I have decided to
at least upgrade the internal network to fastethernet. I will have
fastethernet from the router to the switch, gige between the buildings
and fastethernet in to each office. I will be installing this network
configuration on Sunday 7/27.
- Purchased new computers for new hires. Set up new accounts for the
new hires. Helped set up offices for the new hires. Set up the new
computers with standard software configuration. Searched the web for a
special keyboard for one new hire.
- Ran more tests and continued communications with esnet and PNNL
concerning how to get increased bandwidth on our WAN network.
- Lots of user support for SURF students, staff and visitors.
- Some time spent doing CDS software for Vacuum Controls and PSL laser
power controls.

Lazzarini NOTE: I contacted the PNNL systems group manager to discuss
the performance
that has been observed between CIT and LHO over the new link. PNNL is
working with the LHO group and ESnet to identify the
performance bottleneck.

CIT:
(Lisa)
- Held down the fort while Mike and Larry were away.
- Did troubleshooting on bsears ultra 10. It was having hardware
problems but
not giving any error messages. I thought replacing the keyboard and
ram had
cleared it up, but it looks like not.
- Tested a blade150 that might have been having hardware problems. It
looks to
be fine.
- More work on the mail servers.
- Ran MU's and recommended patches on becrux, sadal, hamal, spica and
lsc1.
- User support.

(Veronica)
- LIGO website: captured and compressed another batch of the recent DV
tapes of SURF lectures on physics of LIGO. Updated the roster database.
Updated various webpages. Prepared and installed a webpage for the
Amaldi
5 presentations. Working with Cindy on the webforms for
Visitors/Postdocs
data.
- Project Science and CaJAGWR: user support.

(Larry)
-Worked a number of procurement issues. We've received a new SUN server
to be
used for applications and licenses. We will start working on the
install next
week.
Received the CISCO replacement module needed to get on the Caltech GigE
backbone.
Working on the Foundry BigIron swapout.
-Assisted a number of SURF students with various applications.
-Helped DCC out with a couple of file issues on the web server.
-Assisted Cindy with her transition to her new work load.
-Went through a number of computers, checking for spyware/adware
programs. So
far every computer has had some sort of spyware program running on their
machine. Handling of this problem will be addressed when Mike and I
return.

Advanced LIGO Development (Shoemaker)

Advanced LIGO and supporting R&D

Suspension

From: ctorrie <ctorrie@ligo.caltech.edu>

MC
Janeen, Mike-Lloyd and I worked on aligning the MC suspension in pitch and vertical using the adjustment mechanisms that have been added to the suspension. This was a useful exercise for everyone.

SolidWorks
the tools we have been working on in SolidWorks have been submitted to the DCC and we hope to be able to supply references to these and documentation for next week.

ETM
Mike Perreur-Lloyd and I, with help from JaneenRomie and Norna Robertson have been working on bringing together all of the information relating to creating an estimate for the mass budget for the ETM. Several questions remain unanswered and we are planning on meeting with Larry Jones on the 5th of August prior to releasing the document for discussion. A draft copy, with a couple of TBDs, is available for those you are interested.

From: JaneenRomie <romie_j@ligo.caltech.edu>

AdLIGO Suspensions
Working on mass budget calculations with Calum and Mike P-L after defining an earthquake stop assembly concept. Mike's last day of his visit here is tomorrow. His help and support have been quite welcome. I'm giving his Designing in Solidworks a trial run.
Created a talk about Advanced LIGO vacuum requirements for the UK group while I'm in Glasgow next month.
Worked with Calum and Mike on characterizing the suspension alignment features.
Created a notes template for the weekly suspension meeting based on a concept from Caroline.

Gin Gin
Troubleshooting and providing support for Helena while she is assembling the Gin Gin suspensions in the So. Annex.

Auxiliary Optics

From: Michael Smith <smith@ligo.caltech.edu>

PHOTON DRIVE R&D
Tom Essinger-Hileman has re-designed the 40m photon drive optical layout so that we can provide four bounces onto the same spot on the MC2 mirror. He has wired up and tested the 0.500 mw 1047nm laser and the photdetectors. He is in the process of setting up the SR 560 servoamp with Osamu's help to stabilize the laser's intensity.
Aidan Crook received valuable information from working with Peter Fritschel at MIT. He has completed a simulink model of the 40m PSL/mode cleaner system with the addition of a photon actuation loop. He is in the process of determining the suitable frequency range over which the photon actuator loop will be able to perform length control of the mode cleaner.

Other Laboratory R&D

From: David Ottaway <ottaway@ligo.mit.edu>

Infrastructure and General Experiment

We are currently debating whether to switch the arm that is used in LASTI
from the Y arm (Arm with two HAMs) to the Xarm. In the process of
discussing this we realised that we are currently around 200 kg over the
allowed seismic limit for Advanced LIGO seismic. We are doing a trade
study at the moment to determine possible solutions to this problem. There
will be more on this when we know more.
The primary reason for switching arms is to open up greater access to the
LASTI HAM and make it a closer test to what is available at the sites. The
other consideration is that by switching HAMs we willsoemwhat loose the
possibility of using the middle HAM in the experiment. However it will
then be availiable for other experiments that come up as the need arises.

EPI Design and Installation
The EPI pre-isolator solid models have been completed for both the HAM
and the BSC. The models have been sent out on CD's or they can be
downloaded from http://emvogil-3.mit.edu/~kmason/EPI_MODEL/The models
are created in solidworks and can only be opened in solidworks 2003.

MEPI

We have been working on wrapping up the work on the local/modal MEPI-HAM
controller. We are have started work on the documentation (see the LASTI
ilog) and are finishing up the performance measurements.

HEPI

We have implemented the sensor correction coefficient stepper on the
HEPI-BSC system. This is where while the system under control we
automatically change the sensor correction coefficients to find the best
value. This is currently underway.

We have also been examining the issue of pier flexibility, i.e. when you
push with an actuator on the platform what moves. Some results can be
found on the LASTI ilog.

MIT Quadruple Pendulum

We have run into some electronics problems, which have caused a number of
OSEM LEDs to burn out.

Adaptive Thermal Compensation for Initial LIGO
Intensity noise measurements of the CO2 laser we have in house after the
AOM have shown that the intensity noise is less than 2e-5 ie is detector
noise limited. This is already at the level of our specification. Low
frequency intensity fluctuations have been shown to be correlated to
temperature of the cooling water through the AOM. A low frequency servo
loop will be built to suprress this.
RFI measurements have been performed that show that even this laser
operating at full power with the AOM at full power (~30 Watts). The
radiated RF emissions remain small at a field strength of 14 mV/m.

Minutes from COC Telecon – 17/7/03
Compiled by Peter Sneddon for Dave Reitze

In absence of D.R., Jim Hough chaired the meeting, following the tentative schedule
D.R. had sent out previously.

Update on Coating R&D Bids – Helena
All bids have been receiv ed and are being evaluated. Evaluation team is D.R., D avid
Shoemaker, Helena Armandula, Ed Jasnow . The results are due around the 5 th of
August so a decision can be made before the LSC meeting. The proposals were
informally deemed “good”.
Additionally, Pinard gave a talk at Amaldi on coating of Virgo optics. The talk
detailed the fact that they can do more than just coat mirrors. They could measure
scattering, optical losses, etc.
Sapphire Update – Garilynn
They have measured Qs of Rubicon sapphire. Absoprtion was in the 100s to 100s
ppcm range, making it a “reasonable” alternative. The cost of full size optics from
this company is unknown, though their small pieces were reasonably priced. Have
pieces up to 300 mm diameter “off the shelf” so could probably manage full Adv
LIGO requirements. Can grow along multiple axes though 300 mm along c axis is
pushing it. They use a Russian technique that produces rods.
At Caltech, they are going ahead getting their scattering machine into a scanning
mode – laser ordered for this.
It was asked whether we would want another full-size piece of sapphire from Crystal
Systems, as money has been ear-marked for this. Garilynn believes we would like
one as there are several companies (CSIRO, Goodrich, ASML) that are interested in
polishing masses and having 3 full sized optics would allow for parallel comparison.
There was some debate on what the nature of this new piece would be. Does it
already exist at C.S., or would it be grown from scratch. And if the latter, would we
be able to suggest improvements in the growing technique to make the sample better.
However, Garilynn has been given the impression that they cannot control the growth
technique to allow such changes to be made. There was also the caveat that we may
not end up using sapphire at all.
Garilynn also added that the Australians have measured a Q of 2.4 x 10^8 on a piece
of C.S. sapphire.
Stanford Program – Roger
Roger Route was not in attendance, so this section was skipped.

Coating the rmal noise developments – Sheila
Sheila summarised some details from Amaldi:
David Crooks gave a talk at Amaldi that summarised the coating results so far,
including proper interpretation of the results by removing the thermoelastic coating
noise. Still yielded same conclusion that Tantala is lossier than silica. By evaluating
the individual losses of Tantala and Silica in coating, value for Alumina in an
Alumina/Tantala coating could be found. A was similar to S. Thus the A/S coating
was examined and measured values found to match predicated ones well.
Concluded that for a silica substrate, a S/T coating is best – therefore
concentrate either on lowering the loss of the T, or replacing it with another high
reflector with similar thermal properties, so as to not increase the thermoelastic noise
in the coating.
For a sapphire substrate, S/A coating is best. However, in this type of coating,
most of the noise comes from Thermoelastic noise, so a T/A coating would be better,
as it has lower t.e. noise. The same procedure as for a silica substrate would then be
followed to reduce the loss in the tantala, or replace it.
Separately, Garilynn will work on developing a system to investigate the thermal
expansion coefficient of the coatings and substrates. A better understanding of this,
and the Young’s modulus of the coatings is required if the losses are to be minimised
and fully understood.
LSC Planning Meeting
David Shoemaker ran through a list of confirmed talks D.R. had mailed him, as well
as other possible topics. He suggested any comments on this should be sent directly
to D.R., copied to D.S.
Norna again requested no overlap between optics and suspensions sessions.

Weekly Report for Week Ending July 24, 2003

Special Announcements:

LSC Issues (Saulson)

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

Advanced LIGO (Frey)

Quality/Safety (Tyler)

LIGO Hanford Observatory (LHO) and Interferometer Operations (Raab)

LIGO Livingston Observatory (LLO) and Interferometer Operations (Zucker)

Detector/Technical Support (Coyne)