LIGO Weekly Report

There was a site teleconference held on Thursday, October 10, 2002. The following issues were among those discussed:

Budget/Costs--It was noted that the carry-forward from FY 2002 is expected to be $3.6 million. Still no word on continuing resolution funding from congress.

Livingston Staging Building--October 21 Ed Jasnow will be traveling to Livingston in October to negotiate a conclusion for the building contract with Brunt. Estimates are being sought for costs to repair the auditorium stage flooring and the HVAC in support of those negotiations.

Hanford Laboratory Building--The contractor is working the punch list. The fire marshall tested the fire detection and suppression systems yesterday, and an occupancy certificate is expected shortly. There is a list of additional items to be resolved.

The landscaping contract will be issued today. A change control board is scheduled for Monday, October 10, 2002.

All Terrain Vehicle for Livingston--Expecting delivery tomorrow or Monday, October 14.

The list of current actions revised to reflect the status of open actions assigned through October 10, 2002 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.

Irena is still gathering information for the Quarterly Report for the end of August. We have contributions from most, but one or two are still outstanding (not in the good sense).

CR-020016	WBS 1.1.4	Landscaping for Hanford Laboratory Building	O. Matherny

I have distributed (via .pdf attachment to email) Change Request CR-020016 submitted by Otto Matherny for consideration during the Executive Committee Meeting on Monday, October 14, 2002. The request is for $60,000 for irrigation, erosion protection, and landscaping around the new Laboratory building at Hanford. The amount requested is less than the amount that was held for this purpose on the liens list for LIGO Construction.

Quality/Safety (Tyler)

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

Summary of Commissioning Activities at LIGO Hanford Observatory (compiled by F. Raab)
-------------------------------------------------------------------------------------
(see ilog for details - if it's not in the ilog, it didn't happen...)

2K IFO (H2)
-----------
The installation and checkout of the new digitial controls is almost done and should be completed by end of this week. Then commissioning of the new control system can begin in earnest. Some issues are already on the radar screen. The day after relieving the "horror of static cling" on SM2, Betsy and I were able to align the input optics by applying biases that brought the sensors on small optics to near their "S1" values (to within a sign flip) and biasing the RM and MMT3 to restore the optical levers to their "S1" values. With a slight extra tweak, we were able to thread the reflection from the RM back through the Faraday Isolator and onto the REFL photodiode. However the bias module for the RM could not supply enough current to restore the RM alignment and we needed to use the additional bias from the DSC. At first glance, it appears that we are drawing anomalously high bias current, but we are checking it out more carefully. Our attempt to restore alignments of other large optics have been somewhat hindered by a problem in one of the pentium controllers. A new loaner unit is in the mail.

4K IFO (H1)
-----------
In the process of adjusting the 4k MC xover, Peter and Luca discovered that the output of the MCL path (out of the MC board) had an attenuator (1:20) at the output that limited its dynamic range to the Pentek. Fixed the problem by replacing a resistor. They also removed the notch filter daughter board (not used) in the MC board and we moved to ~6kHz 2 poles in the MC length Whitening board. This change gave us more phase to push the MC xover to the ~100Hz region. The new recommended settings for the MC in acquisition and run mode have been recorded. Justin Garofoli has become quite practiced at the new FOTON feature in DTT, and has digitally notched out the 19-Hz bounce mode in the MC loop, among other projects. After these fixes we obtained good locking of the full interferometer, although we cannot say for sure whether the changes or the calendar have caused a good effect. Nathan has cleaned up a number of automation scripts that needed fixing due to various name changes. I have started updating calibration files for the new channel names, but testing was slowed by a problem with test points that should be cleared up by software fix today. We are also cleaning up templates for various tests to accomodate new channel names and take advantage of new channels that should make loop gain measurements a snap.

Moving soon!
------------
At long last, we have gotten the fire inspection we needed for occupancy of the new building and a move date of October 18 has been set.

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

Wavefront sensing work continues apace. Andri, Gaby, Valera, Vijay and Brian have all had a hand in checking and troubleshooting the RF chain (head, cables, demod boards). The results are mixed (no pun intended). The WFS 3 & 4 heads were found to oscillate at about 150 MHz. The oscillation was sufficiently large that it was saturating components on the board and reducing the sensitivity to signals at the sideband frequency of 61 MHz. After some work, these oscillations were quenched by adding capacitor between the quad photodiode -100V bias and the cathodes. However, if the head RF gains are raised by more than about 6 dB without the oscillation reappears. Since the signals going into the demod board are now considerably smaller than anticipated, this puts stricter constraints on the demod board noise than we otherwise would have had. The frequency of the peak response of the quadrants of WF S3&4 were found to have significant variations even before adding the capacitors described above. After the oscillation removal above, we retuned all the peaks and traps of these sensors. WFS1 and WFS 2 tuning was found to be adequate, however, over the course of a week, one of the quadrants of WFS2 "spontaneously" became detuned by several MHz. Of coarse that's worrisome and we don't know why that happened. The WFS2 demod board was found to add different amounts of phase to the quadrant signals. The maximum differential phase shift added by the demod board was about 18 degrees. The same measurement needs to be made for the 61 MHz WFS. Presumably the differential phase added will be greater. This will probably need to be accounted for in software since it can't easily be tuned out in the hardware. We also measured the transimpedance of the symmetric port heads in their current configuration. This was done at 50 MHz and 61 MHz for WFS3&4 and 25 MHz for WFS2. (See http://www.ligo- la.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=10/09/2002&anchor_to_scroll_to=2002:10:10:01:54:53-andri). (Andri Gretarsson)

Overall, the interferometer is slowly coming into operation following installation of the digital suspensions. Numerous small software incompatibilities have required resolution, and some significant rework of the ETM coil drivers (fabricated nearly two years ago and stored until now) was also needed to get them working. The mode cleaner now locks, and an arm was locked for the first time Wednesday evening.

Daniel was visiting LLO Sunday - Tuesday. Worked with him to fix some of the medm screens and understand the epigen - the new way to generate the medm screens. Downloaded and installed the new wachdog database. Setup new sun blades and now installing the Os. Experimenting with Solaris 9 on one machine. Ordered DVD backup drives. Trouble shooting bugs in code. Ash is working on the Mode Cleaner Auto Locker. Tidal software was installed on a processor and tried. A few more bugs need to be worked out. Ash and Joe Kovalik worked on the testing the end station software and hardware functionality. (Chethan/Ash)

Fixed some driver issues on a new laptop. The internal Dell wireless card can be a pain to get it working. Worked a Pcard issue. After many emails back and forth with CalTech's network people the issue was resolved. Somehow there was an update somewhere on their network that was blocking observatory access. The problem is fixed now. Installed the Solsoft software and have begun setting up a configuration for LLO. Did some benchmark testing for Marcel's machine. His machine runs extremely slow on Solid Works models. I will be replacing it soon with a faster machine. (Shannon)

We completed the welded assembly of the first hydraulic actuator and brought it to LLO for leak checking. You can find some images of the procedure at: http://www.ligo-la.caltech.edu/~jkern/ActuatorAssemblyOct8/ We found three leaks. Our welder cautioned if it leaked anywhere, he was concerned with the 1/8" tube weld between the actuator and manifold. Sure enough we found a leak at the manifold. No big deal and that was easily remedied. Of greater concern are two sizable leaks somewhere inside the bellows weld assist band. This is a weld provided by Hyspan where they attached the band to the bellows. You can see the point of leakage best on image PA080631. Allen Sibley is confident that Hyspan delivered good bellows, based on his previous experience with the firm furthermore each bellows was He leak tested at HYSPAN at 1/10^9 We've come to the opinion that our welding of the bellows into the assembly is re-melting the small bellows-to-weld-band attachment weld done at HYSPAN. There is sufficient heat available from our TIG weld process to do this. We are going to use different sequencing of our welding to minimize the heat buildup, and permit more time for cooling prior to initiating the next pass. Not likely, but we will try to repair the leaky bellows by soft soldering. The leaks are small and will not prevent Stanford from operating the unit in their test stand for characterization. Our machine shop in Houston is completing the few remaining parts this week and shipping them as completed. Assuming no problems with these last few parts we will have the actuator at Stanford next week. (Kern)

Created a script to generate the VPN configuration files. It will evolve a little more in the near future, but it generates accurate output at this time. (Shannon)

1) Replaced failed hard drive on gateway.
2) Fixed the network cards ordering problem on RAID using Stuart's
suggestions.
3) Helped Phil fix the problem with LDAS reported by Alan (some frames
were unavailable through LDAS).
4) Together with Chethan mounted fb0 directories to
dataserver:/fb0/frames
5) Replaced failed memory on dataserver.
6) Studied DB2 license for scholar program. I do not see any problem
that might prevent us from using DB2 the way we are currently using it:
if we are not using it for administration or commercial purposes but for
research or teaching only, I think, we are OK.
7) Wrote and distributed draft version of DB replication/federation
document.
8) Checked tapes containing frames missing from CIT archive. One tape
was OK, one seemed to be damaged. With Greg's help copied corresponding
frames from disk to tapes and shipped them to Caltech.
9) LDAS was shut down for hurricane Lili last Wednesday and turned back
on last Friday. Everything came up OK.
10) Upgraded DB2 from 7.1 to 7.2 on LDAS-TEST and LDAS-DEV systems at
Caltech.

First Baptist Church of Denham Springs' Senior Citizen Group toured LIGO today. Some were return visitors from the group tour from last year and were excited to come and see what was new. First timers were anxious to see what was out here. All had a good time even though it was raining and they couldn't go to the observation deck on the roof. (Bonnie)

Detector/Technical Support (Coyne)

DETECTOR SCIENCE & ENGINEERING

Seismic Upgrade Project

Hydraulic External Pre-Isolation (HEPI)

Electro-Magnetic External Pre-Isolator (MEPI)

CDS Software

Rolf Bork reporting

- Finished requested changes to Digital Suspension Control (DSC) software and

have been testing on LHO2k.Code will be compiled and installed on LHO4k and

LLO4k in next day or two.

- The GDS router software has been modified to run on the LSC system and appears

more stable in testing done on LHO2k. This code will also move to other IFO in

the next few days. This code no longer places limits of 1 GDS EXC/TP per optic

at a time.Limit is now LSC limit of up to 9 GDS signals.

- Work ongoing to replace xmgr with 'grace' software in dataviewer.Expect

completion in 1 to 2 weeks.

- DAQ framebuilder reflected memory network at LHO has pretty much hit its limit

at ~11MByte/sec (max rated network capacity is 13.8MB/sec).We are starting to

see some network overload delays. So, we are looking at upgrading this link to a

new 2GBit/sec network.I sent out a request for a price quote for this upgraded

equipment today. Earlier this year, we had received a few of these new boards on

loan and verified compatibility with our software drivers.

DMT

John Zweizig

This week I have continued an investigation of the S1 data quality. I have

completed trends of AS_Q and REFL_I glitch rates in all three

interferometers, and have rms trends for narrow frequency bands for four

channels (as_q, refl_i, pob_q and mc_f) of the LHO interferometers. I am

working on plotting these trends for the LLO interferometer. These plots

are available for perusal here

CDS Hardware

Intensity Stabilization Servo (ISS)

Flavio Nocera, Rich Abbott, Peter Fritschel, Mike Zucker

Worked on requirements for the next generation laser ISS (Intensity Stabilization Servo) control system electronics. These are almost ready for review. Requirements grew from 1/2 page to about 3 in two iterations; writing this stuff down is apparently a very good idea.

Anti-Image Board

Jay Heefner

Anti-image board design for the Adv. LIGO Suspension Prototype incorporates some of the changes that might be needed for the LIGO EMI retrofit. It is plug compatible with the existing anti-image board and could be used for testing on the IFOs. A separate requirements & design review for the initial LIGO anti-image boards is pending.

Timing Board

Jay Heefner, Mohana Mageswaran

FPGA based Timing Board: FPGA trial “breadboards” are expected back from the board house 10/10. We will stuff one to test the functionality of the FPGA. Requirements for the board are being collected and should be available for review next week.

Frequency Stabilization Servo (FSS)

Mike Zucker

Looked over results of testing on prototype next-generation laser FSS (Frequency Stabilization Servo) with Kirk McKenzie and Dave Ottaway.

Coil Readout – DAQ Interface

Mike Zucker

Began modeling to define updated electronics requirements for back-end DAQ readout of coil actuation currents. This system will likely take a modular form such that refinements can be added as we approach SRD sensitivity.

EMC

Mike Zucker, Jay Heefner

Completed draft detector EMC (ElectroMagnetic Compatibility) requirements and site EM interference remediation plan, slated for laboratory review next month.

In connection with this, began dissecting the DoD MIL-STD-461E EMC specification book in order to customize tests to LIGO applications and concerns. Looks pretty straightforward so far.

A prototype of a Eurocard to VXI adapter board has been designed and should be in house by 10/14. This board will allow our present eurocard form factor modules to use a fully enclosed and shielded VXI form factor.

Note by D. Coyne: Review of EMI requirements and proposed design changes is scheduled for 11/22, 8:30 – 11:30 am PT.

Sideband Camera

Keisuke Goda

Here is a brief explanation of what the sideband camera is:

The sideband camera measures the amplitude and phase of a sideband in presence of the carrier. The fact that sidebands superimposed onto the carrier by phase modulators are much weaker than the carrier has not so far allowed us to see them in presence of the carrier. However, the sideband camera can be tuned to a particular sideband among many sidebands and its monitor independently displays the images of spatial variation in its amplitude and phase which are as clear as those of the strong carrier. Therefore, the sideband camera is useful for seeing one particular sideband in presence of the carrier and other sidebands and can be installed everywhere in LIGO.

Here is my weekly report:

We successfully demonstrated the sideband camera and obtained the images of the first and second order sidebands which are 25dB and 50dB respectively weaker than the carrier.

SUS Earthquake Stop Redesign

Janeen Romie

Working on alternative earthquake stop designs. Held a meeting on Friday and got some good feedback and suggestions from Betsy, Doug Cook, Mark Barton, Mike Smith, Calum and others.

Note by D. Coyne: Janeen is gathering requirements for a design review. Also looking into getting a small optic suspension setup in a small vacuum chamber for testing the designs. Basic problem is to minimize charge transfer while proving a compliant and damped limit to motion that prevents magnet and wire breakage in an earthquake event and can be easily adjusted.

PSL

PeterKing

The optically contacted pre-modecleaner was locked to an NPRO

-- previously this was done the other way around -- and the length noise

was measured.The results obtained again confirm that at low frequencies,

the length noise was reduced by just over a factor of 2.The first

resonance observed was 13 kHz and this limited the bandwidth of the

SR560-based servo.The power throughput was consistent with the cavity

visibility of 90%, which is a good sign.No serious attempts at optimizing

the cavity visibility have been made, so some work needs to be done on

that.

Optics Analysis

Erika D’Ambrosio

I have been studying the alignment control system

reproducing the formulas of the articles and I have been

applying the corresponding correction, simulating

in the FFT-code what the control system would do

if the reflecting surface of the mirrors show a large

odd distortion compared to the ideal profile.

The aim is to fix some requirements on the fabrication

of the flat topped mirrors for Garilynn to investigate about.

Optical Contamination Cavities

Lee Cardenas, Liyuan Zhang

OTF Lab. (Bridge)

Contamination Cavity # 1

Cavity is baking as of right now.This cavity is being prepared for the new test sample Glycerin -Alcohol (plus perhaps Biocide and rust inhibitor).

Scatterometer

Efforts to fully recover the Scatterometer are in progress.

I took the drawings to the shop for the newbase plate to hold different ETM diameters.

New OTF Lab at Lauritsen ROOM 38

Cavity #3 Reference Cavity still pumping down. Optical train ready.

Continue taking RGA measurements.

Cavity #2 Test cavity

This cavity still pumping, optical alignment and installation in progress.

Continue taking RGA measurements.

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

LASTI (Zucker)

LASTI (Coyne, Harry, Kern, MacInnis, McKenzie, Mason, Mittleman,
Ottaway, Rollins, Shoemaker, Zucker)
=============================================================

PSL (Jamie R, Kirk M, Mike Z and Dave O)
Kirk continues his final characterization of the PSL board to finalise the optimum design and increase the phase margin of our FSS board at 1MHz. Also an evaluation of the headroom at all frequencies has started. Jamie has sent of his new high power low noise photo-diode board off to be manufactured. Characterization of the large area EOS photodiodes has
commenced.

Thermal Compensation
Ryan continues to write his thesis and gave us a tutorial on the latest
changes to Melody.

The installation procedures for the MEPI pre-isolator have been completed and a preliminary copy
sent to several people for comments;

We've come to the opinion that our welding of the bellows into the
assembly is re-melting the small bellows-to-weld-band attachment weld
done at HYSPAN. There is sufficient heat available from our TIG weld
process to do this. We are going to use different sequencing of our
welding to minimize the heat buildup, and permit more time for cooling
prior to initiating the next pass.

Not likely, but we will try to repair the leaky bellows by soft
soldering. The leaks are small and will not prevent Stanford from
operating the unit in their test stand for characterization. Our
machine shop in Houston is completing the few remaining parts this week
and shipping them as completed. Assuming no problems with these last
few parts we will have the actuator at Stanford next week.

We have been taking single actuator to all 16 sensor transfer functions.
Some plots and the transfer function data can be found at:
http://emvogil-3.mit.edu/~richard/HAM/.

Professor Nayfehs group has spent almost two full days gathering more data to construct their model.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)
E2E Physics meeting
-------------------------
Luca (at LHO) described his work on frequency noise in various parts of LIGO
and comparison of measurements with results from Simulink's linear model.
He would be interested in studying a separate Mode-cleaner model (time-domain
optics model) of e2e to investigate, for example, the extent to which the
nonlinear effects might exist. He would start giving his inputs for SimLIGO
model to Matt next week.

Dynamics of FP tilt
-----------------------
(Hiro) Analytic calculation of the multi mode fields in a FP cavity with
time dependent mirror tilt is completed. The effect in the reflected field
is proportional to L/c theta d(theta)/d(t), where L/c is the one trip time
and theta is the time dependent misalignment angle. The power in the cavity
has two time dependent components, one is theta^2 and the other is
L/c theta d(theta)/d(t). The time dependence based on the adiabatic
calculation is theta^2, so if the mirror is rotating in a sinusoidal form,
then the reflected field has 90 degree phase shift. This is observed in the
e2e calculation. Observation in LIGO data by Bill K. and the simulated
result by e2e is being compared with this analytic result in hand to have
a better understanding of the LIGO arm behavior.

SimLIGO
---------
(Biplab & Hiro) After fixing another bug in summation cavity's misalignment
calculation, the WFS signals for pitch and yaw modes match very well.

Code development and maintenance
----------------------------------
(Hiro) Adlib code: Continued code modification to support the new alfi
output format (merged boxes file) and for the better support of SimLIGO.

Alfi
-----
(Bruce)
- Fixed a serious problem involving the movement of grouped elements.
- Continued implementation of cut and paste.

(Melody)
- Finished implementing the revised format of the exported modeler
file from alfi5.
- Continued working on the automated tester for alfi5.

LDAS environment. Testing turned up several holes in the code which have
been patched. There has been a noticeable slowdown in the performance of
this new frameCPP which is now being looked into. It was also discovered
that the behavior had changed enough in this new library to cause the creation
of RDS frames to fail. This has been traced to the management of the FrDetector
structure. All other APIs which have been impacted by this change now seem
to be in good working order. There is still plenty of testing to be done with this
new library and also the issue of backwards compatibility to be demonstrated.

A comparison test of the data products from the various DSO codes run in
testing LDAS has shown that the results are the same for all search codes
tested with the exception of the tfCluster codes. A further investigation is
needed, but it is thought to be a change in the behavior of tfCluster.

The diskCacheAPI mechanism for learning the frame files available to an LDAS
system has been speeded up and tested on the LDAS-CIT system with the full
dataset of S1 frames from both sites.

Several issues with cleanup of lamd processes managed by the mpiAPI have
been cleaned up, improving the performance and reliability of these underlying
processes used by out MPI jobs.

We worked on making the individual connections to the database managed in
a separate thread within the metaDataAPI. This has not fully been solved and
will require more C++ code development after the release.

We have begun the upgrade to a newer version of IBM’s DB2 database (7.2).
Preliminary tests indicate that the new database server is about 30% faster.

A new action for low-pass filtering has been added to the dataConditionAPI. The
default behavior is the same as that used for downsampling.

Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* HPSS: Got Redwood migration scripts to run continuously (by hardcoding
dce_logins with passwords). Worked on setting up new V880.
* Got quote for SAM-QFS pricing.
* Worked on getting 2Gb FC test hooked up. Having problems with Qlogic drivers.

(Al Wilson)
* Some additional tweaking of the cfdef system for the LDAS-DEV system. Should
be cut and dry for the other systems now.
* We still have a problem with the DEV system Beowulf nodes. They will not
reboot under 7.3 without a keyboard attached.
* Node15 is being put back into service CND.
* Completed getting all of the ldasbox(s) on-line they are ready for use.
* Please note that the mirrored version of RH7.3 on our site seems to be a
bit buggy, contact me for details [awilson@ligo.caltech.edu] so use
/export/mirror/linux/iso-rpm-7.3/en/os/i368 for installs

(Stuart Anderson)
* Finished the SAM-QFS vs HPSS evaluation and started serious price
negotiations with Sun over the cost of SAM-QFS.
* Rewiring the 6th floor Millikan network closet to avoid using any ITS
fibers as they will be disconnected Oct 11-13 as part of the Millikan
Library construction work during a critical time in the current LDAS
software release cycle.
* Started planning the final LDAS construction procurement of Sun storage
and server hardware now that a decision on the mass storage architecture
has been made.
* Finished integrating all of the S1 frames into the LDAS-CIT system for
the upper limits groups to work on.
MIT
---
(Keith Bayer)
* Sent S1 LLO RDS tapes to Greg.
* Ordering IP addresses for Shannon's network boxes.
* PC raid box stable after 1 week of uptime with new firmware.
* Received extra 3ware 36" ribbon cable.
* Met with MIT electrician this morning regarding new circuits for lab
job is more complex due to location and existing electrical equip.
Fairly firm estimate will have job completed by end of next week.

Livingston
----------
(Igor Yakushin)
* Replaced failed hard drive on gateway.
* Fixed the network cards ordering problem on RAID using Stuart's suggestions.
* Helped Phil fix the problem with LDAS reported by Alan (some frames
were unavailable through LDAS).
* Together with Chethan mounted fb0 directories to
dataserver:/fb0/frames
so that LDAS can see current post-S1 frames again.
* Replaced failed memory on dataserver.
* Studied DB2 license for scholar program. I do not see any problem
that might prevent us from using DB2 the way we are currently using it:
if we are not using it for administration or commercial purposes but for
research or teaching only, I think, we are OK.
* Wrote and distributed draft version of DB replication/federation document.
* Checked tapes containing frames missing from CIT archive. One tape
was OK, one seemed to be damaged. With Greg's help copied corresponding
frames from disk to tapes and shipped them to Caltech.
* LDAS was shut down for hurricane Lili last Wednesday and turned back
on last Friday. Everything came up OK.
* Upgraded DB2 from 7.1 to 7.2 on LDAS-TEST and LDAS-DEV systems at Caltech.

(Shannon Roddy)
* Created a script to generate the VPN configuration files. It will
evolve a little more in the near future, but it generates accurate
output at this time.

Hanford
-------
(Greg Mendell)
* LDAS at LHO can now see raw frames written in real time by the fb0
framebuilder. This means real-time frames are available for LDAS jobs
at LHO, as well as S1 data. (Note that this has no effect on S1 data
which was written by the fb3 framebuilder; it is still available at
LHO.)
* Assisted Igor in regenerating two tapes with LLO S1 data that was
incomplete or missing from HPSS. Once these tapes are ingested into
HPSS all the raw S1 data from the sites will be archived at Caltech.
* I'm currently running a script to read LLO S1 RDS data from tape to
disk at LHO. I am assisting Robert Scofield in creating an RDS from
this for PEM investigations.

Data Analysis Activities (Lazzarini)
Weinstein:
- Continuing to generate lots of burst simulations (gaussians,
sine-gaussians, Zwerger-Muller SN waveforms) through burst
event trigger generators (tfcluster, power) with S1 data.
- Continuing to develop flexibility of burst simulations
(random burst injection times, sine-gaussians with constant
Q instead of constant duration, choosing more representative
ZM waveforms, etc).
- Studying different methods for applying response function
to simulated bursts.
- Studying effect of response function filtering and whitening
filtering on short-duration simulated bursts.

Shawhan:
* Continued to work with Vladimir Dergachev (at Michigan) on his prototype
for a multi-platform program for viewing LIGO trend data.
* Constructed a list of positions and frequencies of known pulsars at the time
of the S1 run, for use by the LIGO-GriPhyN demo analysis utilizing the
period source search code.
* Wrote my article for the proceedings of the ACAT'02 conference.
Laura:
My principal activity for this week is the analysis of software injected
signals (gaussians and sine-gaussians) in the S1 playground in order to
tune thresholds, pipeline cuts and efficiency vs strenght curves.
Focusing on Slope and Tfcluster so far. Also worked on dataCond issues
as well as handing data to GEO.

Shourov:
I investigated sixteen seconds of H1:LSC-REFL_Q data from the middle
of the third S1 playground segment. Confronted full frame and version 4
RDS frame data that involved no downsampling and found no difference.
Looked into the the recently discovered anomaly on REFL_Q and prepared
plots and scripts for investigations.

Stefan:
Worked on vetoes. Came across the REFL_Q problem and investigated
with Shourov/Laura.

Erik:
Run S1 playground pipeline analysis for TFClusters/Slope
as part of BAM#2. Compiled an approach to look at DSO coincidences
and prepared first cut analysis of those that appears on MIT's
Data Analysis web page.

G. Mendell:
1) I have identified a problem that can occur in SFT generation due to
bad locked segments or round-off error in my code when it finds the
start/end of a locked segment. I have only seen the problem in one SFT
so far, but once the problem is fully debugged it will be fixed and S1
SFTs will be regenerated. The plan was to do this anyway, once the new
release of LDAS and v6 frames is released. Other issues such as
additional filtering and windowing of the time series before producing
SFTs is under investigation by the PULG group, and may be implemented
before the SFTs are regenerated.

2) I'm currently working to debug LALDemod and enhance my code to
calibrate SFT data and normalize it with the inverse of the noise power
spectral density.

3) In general, I am working on how to actually implement in code our
plans for setting upper limits on the gravitational wave strain from
known pulsars.

General Computing (Wallace)
MIT:
(Keith)
-Created accounts for new urop/grad students
-Fixed/deleted several dead accounts where e-mail had been accruing
-Reinstalled win2k on DSpace computer (after bad kernel tweak from
-Matlab install made box unusable)
-Installed LINUX 7.3 on grad student laptop
-Troubleshot windows networking issue on laptop
-Continued to investigate MIT bandwidth troubles
more sampling and utilities
-Coordinating with CSR sysadmin
-VRVS testing / setup

Livingston:
(Shannon)
-Fixed some driver issues on a new laptop. The internal Dell wireless
card can be a pain to get it working.
-Worked a Pcard issue. After many emails back and forth with CalTech's
network people the issue was resolved. Somehow there was an update
somewhere on their network that was blocking observatory access. The
problem is fixed now.
-Installed the Solsoft software and have begun setting up a configuration
for LLO.
-Did some benchmark testing for Marcel's machine. His machine runs
extremely slow on Solid Works models. I will be replacing it soon with
a faster machine.

(Tom)
-Installed Win2K on a Laptop.
-Resolved a printer issue.
-Cleanup from Lili.

Hanford:
(Christine)
- Out sick most of the week. Returned to work to discover that the
Cisco router was reporting out of memory errors. A reboot appears to
have fixed it for now. Cisco troubleshooting guide suggests the memory
errors were due to IOS bugs that can be fixed by upgrading the IOS. I
am looking into this and into getting more memory.
- Once the Cisco was working again, I added another IP address to the
access list for the TRINET computer.
- Changed the login shell for one user. Dealt with an IP number
conflict for another user. Ordered some PC upgrades for another user.
Helped another user with Acrobat version problems. Upgraded Windows on
two PCs.

CIT:
(Lisa)
- Reprogrammed the wireless network. The growth of the network exceeded
original plans and some adjustments were due.
- Setup some new accounts
- Updated Inventory information
- Worked with our Sun Technical Rep on getting more information about
integrating the sun crypto accelerator card for secure email.
- Spent a sizable chunk of time doing user support for Syd.
- Spent some time helping out the PMA group with their new installation of a SUN
box.

(Mike)
-Worked on troubleshooting problems with DCC; this is regarding the
electronic links not automatically updating. I am still running tests; it
looks like I may have fixed this problem.
-Have quite a few PC's to upgrade to 2000 Pro this has been consuming a
lot of time but I am almost there. This week I upgraded:
Win98 to 2000 Pro:
Lori Hines
Bob Taylor
Started:
Laptop
Steve Vass
Dennis Coyne's engineering workstation.
-Larry, Keith and I ran some tests on VRVS sharing option, which seems to be a
great tool for future conferences.
-Larry, Lisa and I swapped Gary, Ryan and Cindy to the 115 Subnet.

(Veronica)
- LIGO website: modified one of the DCC databases to handle a wider range
of document dates.
Posted several updates to various parts of the LIGO website.
Per Albert's request, reworking a webpage for interferometer sensitivity
curves.
Posting updates to the Project Science website as they arrive.
- LDAS website: installed the cgi script with updated ilog links.
- CaJAGWR website: videotaped a lecture. Processing the video file in
preparation for web posting.
- LSC website: met with Rita and Irena to discuss a problem with a
database for submitting semiannual progress reports. Working on the
problem.

(Larry)
-Ordered a couple of Laptops to replace some of the older ones. So, we will be
playing the laptop shuffle in a couple of weeks.
I am now going through our third rep. for the Matlab licensing. He assures me
that he will be the one to get the job done.
-Actually was able to do something fun this past week. Worked/working on a Ansys
issue, trying to get the pkg. to work better on a 64 bit machine. Also, helped
out the PMA group on a V880 install and going over a potential new computer room
for their group.
-Worked on the VRVS items. The broadcast from MIT was very good on the units I
connected to, however, I did receive word that others had some problem with the
audio. The group as a whole will continue working on the VRVS issues for the
next few weeks. Keith appears to have the setup in good shape at MIT. However,
we still want to test out the audio system over the phone unit.
-Spent time resolving a variety of PC issues for different users. There will be
a number of boxes getting rebuilt over the next few weeks because of the
problems they are having.
-Worked on the DCC problems with Linda and Mike. Once we got the procedure down
to handle things manually Mike has been able to spend some time repairing the
automation procedure.
-Worked with Shannon and spent some time with others over the Livingston P-card
issue.
-Continued go over some documentation concerning Computer policy and security.
-Worked a small network issue that deals with ITS having to move its network
closets in Millikan over the weekend. It is a good thing we have some spare
fiber to use during that time.
-Continually blocking more e-mail addresses. We have been adding at least 3-5
addresses a day for the past few weeks. We've also put a few more specific
addresses on the allow list.
-Working with ITS computer security to resolve a couple of issues. Just as a
side note, they have sent another request to a couple of ISP's asking them to
check on some machines, generating e-mail that has been forged with LIGO names.

LIGO II/Advanced R&D (Sanders)

SMA/Lyon
One substrate have been coated with doped tantalum and will be shipped to Caltech at the end of this week.
They are having some trouble with the coating systems. The Virgo one needs a new large liquid feedthrough rotary seal.(Ferrofluidics).
The small DIBS also has a rotary seal problem.
Jean-Marie made some special arrangement to coat this part - just one thin substrate.

MLD
Has substrates to be coated with SiO2 / Al2O3. They should be finished during the next 2 weeks.

We have measured most of the modes of the niobia/silica
coated sample. We are still trying to get definitive values
for niobia's index of refraction and Young's modulus, but using
approximate values (n=2.16 and E = 8.5 10^10 Pa) the niobia phi
is about 5.0 10^-4. This is a little worse than tantala's phi
of 3.7 10^-4.

We also have done some preliminary experiments on improving
silica Q by annealing. We annealed, in air, a silica disk
to 1200 C for 6 hours. The Q was a disappointing 13 million.
We then annealed it for 12 hours. We are still measuring
Q's from this, but they seem to have gone down some. From
a glass-half-full perspective, it is encouraging that the
Q did noticeably change after the second anneal. We will
try a 12 hour anneal again but bring the temperature back
down to room temperature more slowly. Depending on the
result of this, we will probably anneal again for 24 hours.
If none of this works, we will be forced to try a vacuum
anneal.

Writing a draft Statement of Work and Design Requirements for the contract to design and build prototypes of the HAM and BSC SEI structures.

The optically contacted pre-modecleaner was locked to an NPRO
-- previously this was done the other way around -- and the length noise
was measured. The results obtained again confirm that at low frequencies,
the length noise was reduced by just over a factor of 2. The first
resonance observed was 13 kHz and this limited the bandwidth of the
SR560-based servo. The power throughput was consistent with the cavity
visibility of 90%, which is a good sign. No serious attempts at optimizing
the cavity visibility have been made, so some work needs to be done on
that.

I finally got the Motorola input/output controller (IOC) to boot without
timing out. For reasons I don't fully understand the old boot up files no
longer worked when the IOC was carted down to the PSL Lab. Thanks to Lori
who pointed me to a set of functional boot files.

Ed Chargois is organizing getting the current generation PSL frequency
stabilization electronics shipped to the University of Hannover.
Installation of testing of which is scheduled to take place early
November.

An R&D program for the Photon Actuator, with schedule and cost, was presented at the LIGO Lab R&D meeting on 10/8/02.

AdLIGO Suspensions
Worked on budgets and viewgraphs for the Tech. Review 10/8.
Working on Norna's questions for her PPARC meeting.
Working on mode cleaner tasks and setting up the lab with Calum.
LIGO I
Working on alternative earthquake stop designs. Held a meeting on Friday and got some good feedback and suggestions from Betsy, Doug Cook, Mark Barton, Mike Smith, Calum and others.

The pendulum suspension is currently dis-assembled in order to allow us to incorporate a number of optimisations. This was after assembling it several times to allow us to get comfortable with the sub-assemblies, wire jigs and procedures.

We have been working on some optimizations for the mode cleaner prototype these week. These include a winch adjustment similar to that being currently used in LIGO I and pitch adjustment similar to the design used on the upmost mass in the MIT quadruple pendulum. These are now ready to be sent to be machined and will be added to a package of work that will be started by the Physics machine shop on Monday.
(Calum, Janeen and Russell Jones, Alastair Grant, Mike Lloyd from U. of Glasgow)

Virginio
Finally returned from Japan, despite the visa and hurricane problems.
He reports that the system still needs work, but the low frequency IP isolation is almost too good. The sophisticated inertial damping is necessary at the beginning when acquiring lock, but then it can be ramped down and the interferometer stays in lock for hours with only controls at the mirror level. Despite the downtown Tokyo location, there is not enough seismic energy trickling down through the IP barrier to significantly excite the chain internal modes. Virginio now thinks that, working on it, it should be possible to operate from the marionetta and run an uncontrolled mirror interferometer. This was and still is also the thought of Diego Passuello.

Akiteru
Found not one but three leaks in the welds of the two new vacuum tower extension rings. Sending back for repair. Writing thesis.

Alessandro, Charles
Alessandro training Charles to work on MoRuB joint manufacturing.
Performed the micro analysis of electrodeless NiP films on MoRuB of many thicknsses.
Composition is only visible for layers thicker than 0.5 microns. It is a uniform 87%Ni 13%P. The deposition rate has been calibrated and is 2~2.5 micron/hour at 70oC.
Some cracks in coating observed, need a proper bakeout at 160oC for 3 hours to anneal stresses if thick films are necessary and in any case before depositing the braze.
Tested many different braze alloys in crucible. S-Bond is not capillary , therefore is not oven reflow braze compatible, and is eliminated.
AuSn20% sticks well on NiP at 280oC. Found that using a new aggressive flux we can get AuSn to stick directly to MoRuB. May possibly eliminate NiP electrodeless.
The problem with in-crucible dipping technique with AuSn is that the capillarity is too high and a drop is formed instead of a nice film. Should use vapour phase deposition, this would allow the creation of the flex joint 300 micron wide strip by simple shadowing. May or may not need the intermediate NiP electrodeless deposition.
Need to test bond strenght in stress-strain machine.
Still remains to try AgSn.

Stoyan
X-ray imaging. Working on calculations of resolution.
Splatting. Running out of splats, will soon need to resume production.
Installing a new (old, recovered from mothballed lab) oven to obviate the black body radiation driven overheating of the sample.
Ansys. Test bending on Al ribbons (for cross check of a real system) and on thin MoRuB joint models.
After many months and effort (including the effort from Charlotte last year) finally managed to correctly model a thin flex joint. It is not clear what caused the success. The entire system geometry was just re-entered from scratch, with no new or different techniques, and now the flex joint behaves just as expected. It is VooDoo F.E. analysis.
But it finally works!

Hareem, Xavier
May have (after a few months of good behavior) new problems of pressure in the cryostat. It is possible that at this stage we simply need a legitimate cryo pump regeneration.

Allyson
Trained herself to make precision indentation measurements, now producing indentation with same or less spread than Maddalena could do. Now she starts real measurements repeating Maddas curve on new samples and making 50 indentation per sample. Stoyan will have to organise splatting programme to provide her the needed samples. Since the Knee of the distribution is between 16 and 17% of B, we will need to make MoRuB16.5%.

Xavier, Valerie
Analysing heat conductivity data.
On old data found an unwanted large variance of measurement length. Switched off the software automatism that generated that problem. Launched another measurement session repeating the conductivity measurement several times at each temperature to get the measurement stability, and the run measurements ad decreasing power levels to interpolate out the b.b. radiation corrections.
Ansys. Found that we were underusing the Sargas power by at least a factor of 4. Working with Larry to make full use of the potentialities.
Working on connecting the stress matrix of different simulated segments of the system to optimise the simulation performance.

Riccardo
Receiving and reviewing reports from all SURF students.
Posted in DCC Erics report and all SURF presentations.

For Week Ending 10/10/02	Packages	Faxes
In	37	29
Out	11	35

Weekly Report for Week Ending October 10, 2002

Special Announcements:

LSC Issues (Weiss)

LIGO I Construction/LIGO Laboratory Administration (Lindquist)