LIGO Weekly Report

The voting by the LSC Council was completed yesterday morning.
Peter Saulson was elected the new LSC Spokesperson. He and I
will coordinate the transition between the March and August
LSC meetings.

I am personally delighted that Peter will be the next Spokesperson.
The Collaboration is indeed fortunate to have had three really excellent
candidates for this position. My thanks go to Peter, Keith and Stan
for having taken the risk of standing for election.

to: LSC executive committee
from: R. Weiss March 4, 2003
concerning: Notes from the LSC Executive Committee meeting February 28, 2003

New Membership in ACIGA and Southern University role in S2
----------------------------------------------------------
Neither proposal proved controversial. I will write Edith Cowan University
that they are welcome to come to the LSC meeting.

Status and schedule for S1 result papers
-----------------------------------------
It is clear that the "upper limit" groups are behind in their analysis. A revised
schedule was made to give some time prior to the LSC meeting for comment by the
Collaboration and critical review by the review committees. It seems now that
all groups will emphasize methods and use the results as examples. The burst
group is furthest from having a paper written. In subsequent conversation with
Peter Saulson I get the impression that it will be a stretch to have a paper
to review by the time of the LSC meeting. They will try their best and have asked
to be the last group to present results at the LSC meeting.

Plans for the LSC meeting
-------------------------
The agenda has been through two drafts and is close to final form. There will be
strong emphasis on signing off on the S1 papers at the LSC meeting. The first day
of the meeting is dedicated to "upper limit" group meetings and presentations by the
groups and reviewers to the LIGO 1 Collaboration. Further reports are scheduled at
the end of the second and third day of the meeting with a plenary session for the
LIGO 1 Collaboration on the afternoon of the last day of the meeting. We have
asked the "upper limit" groups to discuss their plans for the analysis of S2 data
when they present their results of the S1 analysis.

Deciding on whether to publish
------------------------------
We discussed the process the Collaboration will use to decide on whether the
papers are of sufficient quality to be published. The proposed steps are to
have the reviewers report their findings to the LSC Executive committee and
a vote will be taken by the committee.

Shifts and other service functions
----------------------------------
Keith Riles reported that after some initial fluctuations, the Science monitoring
of S1 is going smoothly. The system he has put in place seems to be functioning
well.

Discussion of LSC membership dynamics
-------------------------------------
Gary Sanders prepared a set of discussion points for how we might deal with new
groups and new members to existing groups and their relation to LIGO 1 authorship
and service functions. The issue has been around since the beginning of the
Collaboration but has come to a head with the proposal at Goddard to increase the
group size and functions as well as the continuing desire in the outside scientific
community to join in the LIGO scientific program. The substance of the discussion
was how to preserve recognition of the work of Collaboration members while also
allowing room and flexibility for new ideas and new members who are interested in
working in the Collaboration. Gary reviewed the way some of the high energy
collaborations have dealt with this, these range from rigid to flexible. The
procedure we will discuss with the Collaboration is to NOT freeze the membership
of LIGO 1 but rather to allow new groups to join the
Collaboration author list after a trial year (or less if a strong case can be made by
the group PI to the LSC Executive committee). The year of authorship will
extend to people who have left the Collaboration as well. A distinction will be
made between new senior and junior members of the existing groups, with senior
membership requiring a vote by the LSC Executive committee.

Advanced LIGO proposal
-----------------------
The advanced LIGO proposal has been submitted to the NSF for a new cost of $122M.
It is still not clear whether MRE or direct is the best path and we will
be advised by the NSF as to the best route. A review of the proposal is expected
late in April or mid May. Ken Strain reports that the PPARC actions on the
British part of the advanced LIGO proposal are positive and a final decision is
expected in a few weeks. Barry warns that the backlog of big projects at the NSF
is significant and it will take a concerted effort to
have advanced LIGO move on our schedule.

Sapphire Decision
-----------------
David Reitze reported that the down select decision between fused silica and
sapphire for advanced LIGO test masses has been delayed to July or August.

Next LSC Executive Committee meeting
-------------------------------------
March 19 at the Executive Committee dinner in Baton Rouge

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

There was a site teleconferences held on Thursday, March 13, 2003. The following issues were among those discussed:

No site teleconference or executive committee meeting will be scheduled next week due to the LSC meeting.

Budget Concerns: A list has been published of items that will be purchased using Construction funds. One concern is the list includes a number of small items and many of these items are still in the process of design and reviews. It will be difficult to process these before the cutoff. We have processed several items recently, e.g., $50K for RFI measurement equipment. Florence will be requested to update the list to reflect additions and corrections and to provide this list to all concerned.

Regarding the new laboratory building in Hanford, Chevenell is revising their cost estimates for the modifications. We will probably be able to proceed once we get the quotes.

Alan is preparing package for a storage building at Livingston. This information will be added to the list. Schedule is a concern.

It appears that no one thought of the need for a laser security system in the new Staging building at Livingston. Cards, a card reader, etc., are required for the high power laser lab. This will be procured in the account for the building. The estimated cost needs to be put on the list.

Ed Jasnow expressed concern about how we should handle budgets for the remainder of this year in light of the fact that NSF has indicated that we will get the full $33 million as requested. The end of March will represent the completion of the first half of the fiscal year. We will submit a revised change request (CR-030002) as of end of March reflecting expenses to date and updated staffing levels. There are a number of large procurements that have been delayed pending the actual receipt of funds.

The list of current actions revised to reflect the status of open actions assigned through March 13, 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. . .

WE 03/13/03	Packages	Faxes
In	29	29
Out	19	35

Advanced LIGO (Frey)

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

We are starting work on the Quarterly Progress Report for the Construction Cooperative Agreement. The fiscal quarter for this cooperative agreement ended at the end of February. Irena has taken the lead for assembling as much material as she can from the weekly status reports. Some additional input may be required. Since this is the second to last report, we do not plan to make it too complicated (basically everything is done except for buying some computer and detector equipment).

Change Request CR-030002 to adjust the FY 2003 operating budgets to reflect actual staffing during the first quarter has been prepared. We will update this request at the end of March to reflect a full six months of expenditures. Meanwhile, we are reviewing the costs anticipated for the remainder of the Construction Project (before June 30, 2003), and Change Requests will be prepared to close out the Cooperative Agreement.

Quality/Safety (Tyler)

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

Summary of S2 Science Run Activities at LIGO Hanford Observatory (compiled by M. Landry)
---------------------------------------------------------------------------------------

This week showed decidedly better performance out of H1 (72% duty cycle) than H2 (37%).
Last weekend saw the 4k on an excellent stretch, 66h in continuous science mode operation.
On Tuesday, high winds hampered both machines' locking, delaying the mid-run
calibration study. At the end of the week, H2 had drifted into a pathological alignment state
that took D. Sigg's considerable attention to restore locking (alignment and new oplev
filters, see http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=03/13/2003&anchor_to_scroll_to=2003:03:13:21:51:27-daniel ).

Occasional very broadband noise continues to affect H1:
http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=03/09/2003&anchor_to_scroll_to=2003:03:09:15:59:09-stefan

R. Schofield showed that one of our worst noise offenders in H1, couples acoustically into
the ifo at the AS port:
http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=03/11/2003&anchor_to_scroll_to=2003:03:11:23:30:41-robert

L. Matone has been studying the Sensemon integral to determine how to extend our range:
http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=03/12/2003&anchor_to_scroll_to=2003:03:12:05:25:20-luca

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

S2:
The interferometer has been running less reliably this week. A site-wide
power interruption last Thursday caused a disruption in the electrical
service provided by the local utility to LIGO. Almost all of the systems
were recovered right away, but several systems did not.
(1) The timing distribution failed. A board and cable were replaced, but
there have been intermittent failures in the IRIG-B distribution since that
time that are not yet fully understood
(2) A VME controller was not able to talk with the 110B slow ADC modules
for the SOS following bootup. The object code on hand to restore operation
had addresses that were incompatible with the 110B addresses on the card.
The software was rebuilt. Some pilot error in diagnosing the problem further
delayed the restoration of operation.
(3) The AC unit in the LDAS room tripped, although the LDAS computers
recovered automatically, rapidly increasing the ambient temperature. Igor
quickly intervened to shut things down, but this indicates the need for a
thermal breaker to safeguard the LDAS system.

Almost all of my time this week has been spent on LSC meeting prep. and
additions to the auditorium and new building.

LDAS:
Did some limited VPN testing with two nodes of the planned six. The new
configuration file for >2 nodes brought to light a minor bug which I was
able to fix with an email to an individual very familliar with OpenBSD and
it's IPsec implementation. This will be fixed soon in my script which
generates the configuration file for the VPN. The bug was not clear at all
in the man pages or any documentation I have been able to find.
(Shannon)

LDAS admin.
1) LDAS recovered easily from the power outage last week. However
the air conditioner in the LDAS room stopped working and within minutes
it became 80F. I had to shutdown most of the computers till the air
conditioner was restored. Investigating options of automatic LDAS
computers and power shutdown in case of temperature increase and
temperature alarms in the control room.
2) Ran out of i-nodes on /frame10 that holds RDS data although the disk
was only 91% full. Dan had to reconfigure /frame10 and /frame21 to
increase the number of available i-nodes.
3) LDAS@LLO experienced a several hour period of failing mpiAPI. LDAS
programmers, who investigated the incident, believe that the culprit was
tfcluster stuck on some unusual stretch of data and trying to run on it
over and over again. Alan changed the DSO parameters to prevent it from
happening again. However, today the same happened at LHO. So there still
might be other problems.

LDAS data analysis.
1) Modified waveburst DSO to be able to deal with clusters of size 1 and
2.
2) Changed the way pixel mixing is done that would hopefully prevent it
from generating similar random sequences in two channels.
3) Working on fixing the way coincidence is done with nonzero time shift
between channels.
(Igor)

Active Seismic Isolation
Work on 3D digital content pre-isolation is complete. My intentions are to
ask Jonathan Kern to check the digital files to ensure quality. At that
point, the configuration for PEPI,HEPI, and MEPI should be in sufficient
order that they can be utilized by a greater population. On the same note, I
have been in correspondence with Janeen Romie and Calum Torie about
utilizing PDMWorks for our configuration control mechanism. They are on a
parallel course looking for ways to control configuration that will work
with the DCC.

I have been helping with the advanced LIGO proposal by supplying Larry
Jones (or making the preparations to supply) the sensor manuals in digital
format (PDF).
(Marcel)

Detector/Technical Support (Coyne)

DETECTOR SCIENCE & ENGINEERING

Seismic Upgrade Project

See the LASTI report.

CDS Software

Rolf Bork reporting

- Received more details from Peter on the LSC after S2. I have now modified the

LSC code to handle I&Q signals for up to nine RF photodiodes. This includes the

pentek I/O and the additional 10 filter banks. Cycle time for the code is now

about 40usec. In the process, I have added a generic EPICS sequencer such that

any xycom bit can be set in accordance with the switching of any filter, either

OR or XOR. This allows to user to set up such things as hardware

whitening/dewhitening filter switching without having to change the software.

This code is generic so we can use it in other systems as well. I believe the

last part to be done is to finalize the DAQ output routine and then run the

system thru as full a test as can be done here.

- Continued work on the PC/VME version of the digital suspension controllers for

the 40m lab. We have a problem in lack of reflected memory boards. We need the

faster ones, either the 5588 or new 5565 models. We have some on order, but they

will not arrive until May. The sites presently only have single spares, so we

can't use them (until at least after S2). We are trying to figure out some other

work around so we can continue.

- Received more info from Rana on the digital I/O WFS requirements. We are now

trying to match up what is being done for the 40m lab such that the same code

can be used here and at the sites.

CDS Hardware

Rich Abbott reporting

1. Finished the work on the QPD that will allow us to analyze jitter coupling to residual intensity noise. The board was manfactured and stuffed and will be installed on the Lauritson testbed today.

2. Flavio has finished the ISS schematics for submission to the technical review process. We are planning on starting the final CAD process while waiting for approval on the design.

3. Got two quotes for RFI measurement systems. In the process of analyzing the data and itterating with the vendors on missing items.

4. RFPD redesign changes were submitted and are in the review process. As soon as a final decision is made, the RFPD can be re-done and submitted for manufacture. This is on the fast track as the boards will be needed soon in April

Jay Heefner reporting

Freq Devices DAC: Preliminary tests for the prototype module shipped to CIT were unsuccessful. After several telecons and many tries we shipped the module back to Freq Devices for rework. The problem appears to be a communication error in the serial connection used to pass data across the galvanic isolation barrier.

Pentek 6102: The temporary fix made by Pentek on the 6102 does appear to eliminate the glitches on the DAC outputs. It turns out that a 20 MHz oscillator on the board used to generate timing signals was marginal. In the temporary fix this oscillator was changed to 10MHz, but this increases the latency through the board. Pentek thinks that they can get the same result by changing the logic in a CPLD on the board, without increasing the latency. It was decided that we will wait until this option has been tested. Once they have the final solution, we will begin to upgrade our boards as the operations schedule permits.

LOS Coil DAQ Whitening: Final designs have been started. A request has been made to reduce the number of gain choices for each channel from 7 to 4. This will greatly reduce the number of relays necessary in the design.

Earthquake Stops

Mark Barton

Resumed work on

the earthquake stop tests. I've got the dSpace connected up as a

datalogger, and have tested the bounce performance of some Viton stops as

a control. As one might hope they damp quite well - the sharp turnarounds

from hitting a stop are visible in the traces, and even with a rough

shake to the tank there are never more than 3 or 4 of them. I have sets

of a number of other designs ready to be tested in the next day or so.

PSL

PeterKing

I have been helping Flavio with his testing of the intensity

stabilization servo. An extra EPICS channel was added to allow for removal

of an offset on the board. This was previously removed by a voltage

calibrator.

Errant Beams

Mike Smith, Ken Mailand

re. errant beam baffles ref doc. T030006

The AR coated aluminum approach requires a multilayer coating with excessively tight layer thickness control. An alternate approach using a dielectric absorbing layer on the polished aluminum is being investigated. A coated sample will be available next week for testing with a PSL laser. Sources for obtaining the polished aluminum material, polishers, laser cutters, and coaters have been established. A preliminary layout of the baffles on the 2K and 4K ACAD integrated layout drawing has been completed. Detailed baffle designs are in process.

I’ve been in contact with Doug @LHO on locations and other information,
and Dave Reitze in Fl. Re. RM beam dump.

The goal is to mount a #8 mirror polished absorptive-coated AR over coated 2024 T-3
aluminum metal baffle to the tables in front of the exposed suspension wires and Osem control cables,
to protect them from errant beam damage due to overheating by beam.
Gathering information on:
1. Material [Alcoa, Wichita, may have polished sheet in stock]
2024 T-3, sheet .125” thk. 48”x 144” final answer tomorrow.
a. Kaiser Aluminum says may have unpolished available [checking stock]
2. Polisher [MPC Irvine, Ca. can do this] one week
2. Optical coater [Newport Thin Films Chino Ca. is working on this] ~one week
3. Cutting process vendor [Superior metal Fabricators Anaheim, Ca. can do this] one week
they have laser, abrasive jet and shear press equipment, suggest laser cut.
4. Machining of base bracket [s][ASCO can do these as soon as the drawings are finished]

The parts count is: 11 SOS baffles, 3 LOS baffles, 3 IO baffles [2 parts ea.] 4 RM beam
Dump glass replacements, and a foot mount for each baffle

Optics Analysis

Erika D’Ambrosio

The new numerical results for thermal lensing simulations are posted in my home

web page and I added a description of the model I used to produce those, that is

I only considered a difference in the radius of curvature, caused by the thermal

heating of the mirrors.

Optical Metrology

GariLynn Billingsley

The Fizeau surface measurement interferometer was realigned last week by the manufacturer. The instrument had exhibited a measured change in astigmatism of roughly 10nm that could not be attributed to changes in any reference flat. The interferometer intensity was also clearly mis-aligned.The technician could not explain the changes nor deterministically correct them. The interferometer was simply re-aligned, three times. On the third try the astigmatism went away, but left a larger spherical aberration than before. This means that the calibration we have for our "c" test flat is no longer valid, the calibration will have to be redone. I estimate ~2 weeks of dedicated work to perform and check the calibration.

Lee Cardenas, Liyuan Zhang

OTF Lab. (W. Bridge)
Contamination Cavity # 1
With a new test sample. CHEM-SOL Hydraulic fluid
We are taking ring down and beat frequency measurements everyday.
Please see Dr. Zhang graphs for absorption, ring down and thermal lensing.

Absorption Test Measurement prototype is in STANDBY.

Scatterometer
The function of the scatterometer has been fully recovered.
The software for the scattering has been fixed and it is working.
We have run a simple test to map the scattered beam from the coated surface of the existing ETM.

We got the reflected beams from the AR and HR coating(It was tough to get the AR reflected beam, very weak!)
But in order to catch the AR reflected beam, bigger mirrors are needed, then we need this beam to bring it out of the scatterometer enclosure so we can take measurements.
All these efforts will take much time because the modifications it needs to be done on the plate ( drilling tapped holes, adapting optic mounts etc..) We are putting more time on the scatterometer in order to complete the task.

OTF Lab at Lauritsen ROOM 38

Cavity #3
The test sample from Leander Lubricants
CHEM DRAW’ # HSF-75-2, Batch 12302001. (a synthetic, hydraulic fluid.)
with different additives. is completed as of this week.
We have completed taking ring down and beat frequency measurements every day and concluded that there is no contamination from this fluid. We'll stop pumping this chamber as soon as we get a new sample to test it.
Please see Mr. Zhang's latest graphs for absorption, ringdown and cavity thermal lensing measurements.

Cavity #2 Test cavity
The chamber is pumping with new cleaned mirrors ~ 70ppm each.
RGA measurements, optical train set-up in progress.
New NPRO laser installed (M126-1064-700 S/N 414) and it is in operation.
Alignment in progress.

The fabrication of the new contamination chamber is completed and it will be taken to the 40m to be baked.

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

LASTI (Zucker)

LASTI (Allen, Coyne, Lantz, Mason, MacInnis,
Mittleman, Ottaway, Rollins, Shoemaker, Zucker)
---------------------------------------------------------------------
# MZ NOTE: Due to MZ's confusion about revised reporting instructions,
last week's report was lost; relevant parts are included in this week's submission below.

MEPI Progress (Ottaway, Mittleman, Coyne)
----------------------------------------------------------------------
(3/6 report)
We have started to implement the sensor correction that is necessary for
us to evaluate the seismic isolation offered by our control loops.
Locking only to position loops and using sensor correction in the
vertical direction we have seen a seismic isolation of a factor of 3 at
1 Hz. This will be significantly improved with tweaking of the gain
which we are in the process of doing.
(3/13 report)
While trying to make a minor matlab upgrade the Dspace connection
stopped working, leading us into an ever deepening morass of compatibility issues. Meanwhile the new Dspace processor board arrived. The new board has now been installed and a successful installation of the software has occurred. Control efforts have been resumed. We hope to
rapidly improve on the factor of three improvement that sensor correction gave us in the preliminary experiments.

(late 3/13 addendum)
It works. Sensor correction, using a floor-mounted Streckeisen STS-2 to "correct" the local position sensors, is now giving about a decade suppression at 1 Hz in the vertical modes and measurable suppression out to 8 Hz. See

HEPI Progress (Lantz, Mittleman, Hardham, MacInnis, Coyne)
----------------------------------------------------------------------
(3/6 report)
Guralp witness seismometers have been installed to monitor the support
table and the optics table to accurately measure the seismic isolation.
(3/13 report)
When Brian left last week we had discovered an oscillation in the system
at about 550Hz which was driven by the fluid flow (it was present with no control signals). It was isolated to one horizontal actuator. It is
likely that oscillation is in the flapper of the control valve for this
actuator. In preparation for changing out this valve the BSC has been
sealed. Meanwhile we have been able to work on the vertical modes (Z and rotation about the X and Y axis). Controllers for these degrees of
freedom have been designed, but not full tested since the witness
seismometers have been disconnected in order to seal the BSC.
LATE NEWS: The faulty control valve was successfully swapped out in situ according to the reviewed procedure, and no hydraulic fluid was spilled. Corwin is characterizing the replacement valve and we plan to reopen the BSC to reconnect the sensors Friday morning to resume HEPI characterization.

External Pre-Isolator (Ken, Myron)
----------------------------------------------------------------------
(3/6) The BSC chamber was opened and 3-axis Guralp seismometers were placed
on the optics table and on the support table.
An additional layer of isodamp damping material with additional clamping
has been added to the
HAM crossbeam.
Mode Cleaner Suspension (Ken, Myron)
----------------------------------------------------------------------
(3/6) We have recieved the spacers cleaned and baked from Caltech. Additional
hardware has been
designed and ordered to build up the solid HAM optics table.

Electronics (Jon Allen)
------------------------------------------------------------------------
Recabled the voicecoil actuators on MEPI to reduce noise and
crosstalk. Also set up preamplifiers for payload GURALP seismometers
on HEPI testbed and pulled a 30A AC power service for the MEPI/CDS rack
cluster.
LASTI PSL (Ottaway, Rollins, Zucker)
---------------------------------------------------------------------
(3/6) Jamie has achieved intensity stabilization at the few time 10^-8 level at 300 Hz as observed by an out of loop photodetector. This measurement is currently limited by the electronic noise in the thorlabs out of loop photodetector, significantly lower noise will be achieved when Jamie utilises his second high signal to noise photo-diode in the out of loop measurement.
(3/13)
Jamie has now achieved an _out of loop_ noise performance of 3e-8/rHz at 10 Hz for his intensity noise servo.
#MZ NOTE: WAY TO GO DUDE
We do not believe that this is limited by beam
jitter as this is part of a flat plateau of noise around this level that
extends from 6 Hz to 300 Hz. It is certainly not loop gain limited as the in loop noise drops to below 2.5e-9.

Thermal Compensation (Ottaway, Zucker, Mason, Kells)
---------------------------------------------------------------------
(3/6 report)
GEO has implemented a ring compensation heater to correct for a
polishing error in one of the test masses. We have contacted them and
will hopefully learn some valuable lessons from their in-situ test. We
will also use our modelling expertise to help them see if their solution
if optimal.
(3/13 report)
We find a ring heater would involve extensive rework inside our chambers and is hard to arrange for "positive" lensing (case of insufficient collateral heating by the IFO main beam), as might be required. We're developing a less invasive idea for possible pre-S3 LIGO retrofit. The favored scheme employs a staring mid-IR laser beam with a fixed intensity profile played on the ITM's directly. Feasibility study and cost/schedule estimates are in progress.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)
SimLIGO noise curve
---------------------
(Matt) Finished first stage of understanding the SimLIGO noise curve:
Observed the spectra of many noise sources, reduced total noise to
near SRD, then increased it somewhat by re-enabling some of the
non-fundamental noises.

Code development
-------------------
(Melody) Continued working on optimizing the modeler's functions
analyzer.

Alfi
-------
(Bruce)
- Completely rewriting junction/connection standardization methods
due to continued unforseen problems with the prior implementation.

Software Systems (Blackburn)
A new issue with involving the mpiAPI at Hanford has been identified
this week.
The mpiAPI locks up while accessing file descriptor 3. This causes all
logging
and all processing in that API to stop until the mpiAPI is cleanly
restarted
This has not been seen at LLO or CIT, so it is not clear yet whether it
is a
software or hardware related issue. However, it has now happened three
separate
times at LHO. We are trying to catch the mpiAPI in this state so that
we can
attach debugging tools to get a better understanding of the problem.

The version of the frameAPI running at MIT seems to be about 20 to 30
times
less likely to core dump than the original frameAPI in the 0.6.0
release and
the 0.6.20 running at the sites. It does still core dump about once
every two
days (was about once every couple of hours). But the core files
identify that
the failure is in a different location in the code.

The database has on occasion been in a deadlock condition where
metadata from
a datapipeline can not be inserted. This has been seen every couple of
days
at LLO and LHO. On the LDAS development system we have experimented
with a
couple of DB2 configuration parameters which seem to have greatly
reduced the
deadlocks. Correcting this at the sites would require taking LDAS down
for an
hour or two so unless this issue becomes more frequent, we will
postpone any
modifications until after the science run.

While working to move the validation of a frame file from TCL code into
C++
code, we discovered that several frames being used in the science run
for
calibration which were generated with the VIRGO C framelib library were
not
conforming to the specification. We have contacted the author with our
concerns
over this and have met some resistance to fixing the framelib library.
So in
the meantime, we will lift some of the comformancy issues in our
validation
until this can be fixed in the calibration frame - either by a new
framelib
from VIRGO or using a different tool to build these frames.

One of the search code running at LLO sent the mpiAPI into a tight spin
cycle
of trying to clean up after each job using this search code (tfcluster)
failed.
The team here spent the better part of the day tracing the failure
through
LDAS and into the DSO. Once the source (line of code) where this DSO
was failing
was identified, the LSC proposed changing some of the runtime
parameters so
that the jobs could succeed. One of the steps we undertook in tracing
this
bug in the DSO was to move the data and the job onto our LDAS-DEV
system were
we were successful in duplicating the problem with every job.

* Continued to archive S2 data and monitor performance of SAM-QFS while
archiving and staging.

* Reconfigured QFS filesystems (/frame10 & /frame21) at LLO so that they
have enough inode space for storing RDS frame files. Dealt with
fallout from this upgrade due to a QFS bug.

* I have added a new addition to the Big Brother Program. On the
www.ldas-dev/bb site you will now see a column marked Graph. Click on
any of the green bullets below the heading. You will now see a history
graph displayed for the default stats. Click again and you can get up
to
a 2 year history for that parameter.

* Distributed all of the DMT SenseMonitor trend frames to all of the Lab
LDAS systems to supported jobs that request dynamic calibration.

* Tested various Linux drivers and obtained 950Mbit/s between copper
GigE
ports on an Ethernet switch making it a prime candidate for the large
LDAS compute clusters.

* Applied patches for Sol 9 machines and updated LDAS to version 0.6.25
(frameAPI much more stable now).

* LDAS recovered easily from the power outage last week. However
the air conditioner in the LDAS room stopped working and within
minutes
it became 80F. I had to shutdown most of the computers till the air
conditioner was restored. Investigating options of automatic LDAS
computers and power shutdown in case of temperature increase and
temperature alarms in the control room.

* Ran out of i-nodes on /frame10 that holds RDS data although the disk
was only 91% full. Dan had to reconfigure /frame10 and /frame21 to
increase the number of available i-nodes.

* LDAS@LLO experienced a several hour period of failing mpiAPI. LDAS
programmers, who investigated the incident, believe that the culprit
was
tfcluster stuck on some unusual stretch of data and trying to run on
it
over and over again. Alan changed the DSO parameters to prevent it
from
happening again.

* Did some limited VPN testing with two nodes of the planned six. The
new
configuration file for >2 nodes brought to light a minor bug which I
was
able to fix with an email to an individual very familiar with OpenBSD
and it's IPsec implementation. This will be fixed soon in my script
which generates the configuration file for the VPN. The bug was not
clear at all in the man pages or any documentation I have been able
to find.

Data Analysis Activities (Lazzarini)
T. Creighton:
Finished preparing an updated catalogue of pulsar positions and
spindowns (isolated pulsars only), for the pulsar upper limit group.

Weinstein:
- Continue to monitor online burst dso running,
and recover from several LDAS problems.
- Modified the tfcluster filterparams, and the
filtering done in datacond, to reduce these problems.
- Working on quantifying systematic errors for burst analysis.
- Preparing talks for LSC on hardware injections and on
the use of calibration in the burst group.

Reilly:
I have spent most of my time this week trying to understand
and trying to create correct comparison plots of the expected integrand
of the
cross-correlation statistic and the measured integrand. I am
still not certain on the factors so I am checking with the equations
given in the stochastic LAL documentation.

I have also been working on the pre_S2 hardware injection
analysis. I repeated results, I got previously using ldas at lho,
here at Caltech. These results were obtained using the E7
calibration information. Now I am working to put the calibration
info for pre_S2 in frame format here at Caltech. Jolien Creighton
has sent me instructions and a script for doing this.

Shawhan:
* Worked on inspiral upper limit paper
* Prepared a talk about the inspiral upper limit analysis

Yakushin:
1) Modified waveburst DSO to be able to deal with clusters of size 1
and 2.
2) Changed the way pixel mixing is done that would hopefully prevent it
from generating similar random sequences in two channels.
3) Working on fixing the way coincidence is done with nonzero time shift
between channels.

General Computing (Wallace)
MIT:
(Keith)
-Working on PPT to other formats with Marie
(installing OpenOffice on her pc as well)
-Investigating the SSH windows server for lab PCs
-Went over ssh-agent and ssh-keygen utilities with developers
-Gave lunch talk on Unix/gc utilities
-Purchased more matlab toolboxes

Livingston:
(Shannon)
-Worked on backing up the building control computer files to CD for
Excel.
-Almost all of my time this week has been spent on LSC meeting prep. and
additions to the auditorium and new building.

Hanford:
(Christine)
- Purchased some new laptops for staff.
- Renewed a couple of maintenance contracts.
- Working on a complete re-build of a laptop.
- Misc. user support.

CIT:
(Veronica)
- LIGO website: working on several issues regarding the DCC database.
Extensive work on the causes of its most recent glitch was the focus of
the last few days.
Continued he digitizing of LIGO VHS tape archive. Making a trial run of
capturing the footage on a hard disk and compressing it for web
delivery.
Updated various pages throughout the website.
- LSC website: posting the last-minute updates to March meeting pages as
they arrive.
- CaJAGWR website: taped the video of the last talk. Processing it for
web
posting. Rewrote a script for the smil video/slide combo of Braginsky's
talk. Posted updates at the website.

(Mike)
-Updated and added additional software and wireless cards on six laptops
for shipping to Livingston for LSC.
-Had to restore a laptop from a ghost image due to an OS problem. I was
able to recover the old data also. It is a good thing that I had a
current
image of the laptop.
-Updated firewall software on all NT servers.
-Spent a lot of time troubleshooting the DCC database trying to find out
what the problem is with the access database. I cannot give a report on
this matter because the database some how fixed itself. This server
needs
some more investigating to find out why the database flakes out once in
awhile. (Unknown)

(Lisa)
- Spent quite a lot of time trying to find a compromised machine that
was making
DoS attacks on the world. The problem went away but the offending
machine was
not identified.
- Resolved some nis+ problems on mouser.
- Rebuilt the commissioning list mail archive.
- Looked into a suspicious process report on capella. It turned out to
be a sun
service that needed to be turned off.
- Fixed a problem with the Ansys license server.
- Implemented subject filtering on email. There has already been a
case in
which a legitimate e-mail was refused because it included one of the
blocked
patterns. Basically, I'm trying to block as much pornographic spam as
possible.
This means that if I filter on the string "sex" any word that includes
that
string, such as "sextant," will cause the entire message to be
discarded. I'm
trying folks.

(Larry)
-Spent the last 8 days attending the SANS security conference. I
attended 60+
sessions. Of interest are the legal changes taking place. The new
Homeland
Security group is just getting started and will be making changes.
Various
industries are all playing a major roll in getting laws implemented to
protect
copywrited material and recover financial losses not just from stolen
material
but from network outages caused by DOS attacks or similar problems
caused by
other companies or organizations. This is one area we are keeping an
eye on.
-Worked with Lisa on resolving a number of issues including the DOS
issue. We
are still looking into finding the cause of the problem.
-The RAV software package has been installed. It is a package that does
a
preliminary virus check on e-mail. However, everyone still needs to
keep their
local anti-virus s/w up to date. The RAV s/w doesn't catch everything.
Lisa has
also set it up to start filtering by subject. Subjects containing
obscene
language and a few others such as viagra will be filtered. There are
still some
kinks being worked out. Just FYI over 1900 messages were filtered out
over the
last weekend. This doesn't include the number of messages rejected by
source.
There will be an announcement with more details in the near future.

Lazzarini:
There is progress being made on the upgrade in BW on the LIGO WAN at
LHO. I am working through the fiscal
numbers, but it appears that we may be able to migrate to an OC3 link
for LHO while remaining part of the PNNL subnet.

I have been in contact with Joel Tohline (LSU) regarding upgrade paths
at LLO. Our LSU affiliation may help provide a cost effective solution.
Nothing definite yet.

Advanced LIGO Development (Shoemaker)

AdLIGO Suspensions We're still progressing with the validation of Dennis' mass budgets for the suspension designs. We feel confident with the MC and RM numbers. It is our plan to move forward with the ETM suspension design in the next month or so based on the 22, 22, 40, 40kg mass concept. We hope to have the design, based on FEA work that Dennis has provided, far enough along by the end of June, to feedback to the seismic group on the validity of the quad mass budget. This is dependent on personnel availability.

Micheal Perreur-Lloyd he has been progressing with the details of the recycling mirror. He is working with a footprint of 300mm (beam direction) x 480mm. Please see Calum's report for more details.

The mode cleaner RFQ to support the proposal costs has been sidelined. Procurement is researching alternative ways of asking for and receiving this information.

On March 25th at 8am in the ECR, our Solidworks reseller, GoEngineer, will give us a presentation on their data management software, PDMWorks. Please attend if you're interested. Send me an e-mail if you need call-in instructions.

Gin Gin Suspensions The electronics have arrived from Hanford. Jay is making cables. We plan to damp the metal ITM optic tomorrow morning in the So. Annex. Helena is almost done with the suspension assembly document, which will accompany the suspension and its electronics to Gin Gin.

Mike Perreur Lloyd and I plan to write up a full report on his visit for next weeks report.
for now we have new pictures and data at the following links: -

Please see Janeen's report for other information regarding suspensions.
On the 25th of March Go Engineer are coming to give us a demonstration of PDM Works, a data management software package. Please contact me if you would like to be included on this as they are planning to broadcast it on the web.

Gin Gin
============================
- Three controllers were received from LHO. We are now testing them with the rest of the electronics and the dummy mass in the south annex of the 40 meter lab.

Adv LIGO RM Prototype
=============================
- System drawings and wiring diagrams are 75% complete.

---
The LZH high power laser was put through the test suite last week. This week the Stanford laser is being tested. The technology downselect is still expected to take place at the LSC Meeting.

--
The paperwork for ordering the 100-W 1030 nm ELS VersaDisk laser was submitted. The laser should arrive early July.

The ERGO arm, our Advanced LIGO mirror handler, is finished.
We'll be testing it for some time under different circumstances to see if improvements need to be made.

Preliminary estimates of the size, weight, and mounting configurations for the PO telescopes and PO mirrors are in process.

Allyson
Presenting measurements of SEM photos of indentations data.
She shows that:
The old optical measurements were skewed by the squirted material at the edges of each indentation (a squirted halo of the order of 1-2 microns and then practically undistinguishable with the old microscope).
Different concentrations of B might have produced different squirts and change the data trend.
The coarse surface conditions of the samples make the measurement difficult and fuzzy even with the electron microscope.
With the electron microscope the indentation and the squirted halo are clearly separated.
The optimal choice of MoRuB is shifted more towards 19% of Boron (t.b.c.).
The entire Vickers procedure is unreliable with hard materials (which because of their hardness produce indentations of less than 20 microns to start with) and that with these materials careful surface polish and SEM analysis are a necessity.
Bill suggest that, once cleaned up and enriched of statistics, this data is of publication interest, especially the new Vickers procedure. After all publishing before 18 is not illegal, and it is good outreach, and I bet that Westridge High School will not complain if their name appears on a publication. Work to be done though.

It is also noticed that the indentations of MoRuB-14 show a splitting in two classes of wider and smaller indents. Perhaps at these low B concentrations areas intertwined with dendritic crystals start appearing. To be confirmed.

Stefano
AgSnTi eutectic seems to work correctly and stick to MoRuB even in high stress conditions ( a sandwitch of Braze and MoRuB can be bent plastically without separation of the wetted surfaces. Prepared eutectic AuSnTi-0.5% and 2% alloys to check if the good wetting derives from the Ti activation and AgSnTi enriched with Ag too make a stiffer AgSn bronze precipitation hardened braze.

Electropolishing to clean the sample edges. Now close to be crack initiators free. Mapped MoRuB samples before and after polishing. Confirmed that the edge is eaten almost 50 times faster than the surface by electropolishing (10 micron polishing results in 450 micron erosion at the edges). To obtain predetermined width samples must polish the surface, the EDM cut the desired shape, then re-electro-polish the edges to clear away any possible crack initiator.

The parts of the new stress strain machine have just been finished, now need to make assembly and get it to work.

Eric
Prepared in the vertical coil and the mini arc melter a whole series of new recipe MoRuB melts and X-ray scanned many to explore their structure. Cast samples have more structure of arc melted samples.

Alessandro & co
Ground and mounted one of the diamond loaded vitrelloy samples on a Ultrasound milling machine compatible chuck. Eaten up a few grounding wheels in the process. Testing of the cutting performance of the new samples earlier next week.
Produced kapton coated samples for Calun.

Weekly Report for Week Ending March 13, 2003

Special Announcements:

LSC Issues (Weiss)