LIGO Weekly Report

to: LSC executive committee
from: R. Weiss January 30, 2002
concerning: Notes from the LSC Executive Committee meeting January 25, 2002

Plans for on coming LSC meetings
--------------------------------
Feb 20, 2002 @ 11AM - 12:15 EST the LSC will hold a telephone conference open
to the LIGO I collaborators. The upper limit chairs, detector characterization
chairman and leaders of the LDAS development program will give presentation on
the work in progress and issues that need to be addressed at the March LSC
meeting concerning the analysis of E7 data. The phone call will be made with
some but limited capability to interact.

The plans for the March LSC meeting are being made. A symposium honoring
Bill Hamilton will be held at LSU on March 19. Two sessions will be held
on March 20. The LIGO I collaborators will meet in closed session while
a group of numerical relativists with interest in coupling to LIGO and LISA
will meet separately at LSU to formulate plans for activities that will
be useful for the observational programs and guide the theoretical research.
The formal LSC meeting begins on March 21. The numerical relativists will make
presentations in the afternoon plenary session and are welcome to stay
throughout the LSC meeting.

Upper Limit group activities
----------------------------
Alan Wiseman reported on the activities of the upper limit groups.
Some groups were able to carry out limited on line analysis
during the E7 run. Calibration signals and simulated astrophysical
signals were injected into the data stream for extraction by the
various analysis programs that have been developed. The groups are
grappling with:

* Generation of reduced data sets - a contraction of the full data set is
necessary to be able to carry out analysis at locations without the full
data storage capability in both the Laboratory and other LSC member
institutions. Univeristy of Oregon has taken on the task of proposing
reduced data sets. Weiss has taken on coordination of the detector
commissioning group with this effort.

* Need for faster lines (copper to fiber) providing a gain of 100 in network
bandwidth is being called for. The LSC will ask for this improvement in
letters to the Laboratory and to some member institutions.

* Data selection - the groups want to carry out practice analysis of
sections of data (or in domains of the full data) that will not be used
to establish the scientific results. There is here some confusion
about the meaning of the E7 run. Data anlaysis for the detector
characterization needs to use all the data and the first look at the
data also should not be inhibited by artificial rules (RW opinion
not necessarily shared by all).

* Need to develop a plan for the allocation of computing resources available
to the LSC

Report on the activities of the LSC/Lab computing committee
------------------------------------------------------------
Al Lazzarini reported on the first meetings of the committee. The initial
focus of the committee's activities will be on:

* Establish guidelines for the use of computing grids in the analysis
of LIGO data

Discussion of K.Riles proposal to couple shifts at the sites with authorship
----------------------------------------------------------------------------
Jim Brau and Peter Saulson reported on their interviews with LSC members on
the proposal to couple taking shifts at the sites with authorship on the
LIGO scientific papers. There are arguments pro: distributing the load,
training of the scientists ... and against: contributions by people not
on shift nor at the sites, contributions by engineers... No simple
resolution was formulated. An unnattractive part of the concept to several
committee members is the aspect of a conscripted army without a well understood
job to do. What became clear is that when there is scientific
work to be done at the sites such as online analysis or significant
scientist impact on the operation by helping to decide on operational mode, the
roles of the shift scientist will be much better defined. We have agreed to
come up with a first order plan to increase the scientific participation at the
sites during runs. Barish and Weiss are to write a proposed plan for discussion
at the next LSC Executive Committee meeting.

Next LSC Executive Committee meeting
-------------------------------------
Friday Feb 22, 2002 at 11:00AM ET

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

A "lessons learned" session for the E7 run is scheduled for the time slot normally held by the Executive Committee Meeting on Monday, February 4. The 24/7 Operations Session is moved from February 7 to February 14 during the site teleconference.

The list of current actions revised to reflect the status of open actions assigned through January 24, 2001 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)

Weekly Advanced LIGO Project Controls meeting was not held this week. Next meeting date has not been scheduled.

Followed-up on incident with Primavera regarding actual finish dates for finish milestones being dropped from the update files through email. (No Change or Solution yet).

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

We are doing a standard Annual Report for the Construction Project as of the end of November. Started working on the report again.

A number of issues have been raised regarding the FY 2002 Operations budgets and the support needed for Advanced R&D Equipment. We have scheduled a meeting Tuesday morning with Gary and Dennis.

Quality/Safety (Tyler)

Met (25 January 2002) with representatives of LIGO, PMA and Physical Plant to review the proposed LDAS room 215-Synchrotron modifications.
At this meeting, it was agreed that the modifications to extend the raised floor and the installation of a electrical fire suppression system should be done. It was also agreed to do the modifications in two phases. The first phase will consist of extending the raised floor, closing off the old elevator doorway and some of the electrical and HVAC preps for the second phase work. The first phase work will start as soon as a revised cost estimate is presented by Physical Plant and PMA provides a POETA. The second phase will be the installation of the fire suppression system and will be done later in the year.
PMA agreed to fund the first phase work now with the funding for the second phase later.

The installation of the new LDAS dual fiber optic cable network started this week and should be completed in a week or so. Stuart Anderson did the job walk with Caltech Electrical Shop personnel.

Met with Caltech Safety, insurance risk assessment representative and Steve Vass at the 40M laboratory to discuss the need for and ways to install an electrical fire suppression system. A risk assessment report will be submitted with recommendations and system design and vendor contact information wthin two weeks.

Working with the Gang of Five to update reference/contact information/employee checklists for meeting with Ryan Tischler and Gary Sanders.

LIGO Hanford Observatory (LHO) Operations (Raab)

LIGO Livingston Observatory (LLO) Operations (Coles)

Interferometer: Optical tables ISCT1 and ISCT4 were examined and some of the optics removed or changed to make them safe or to have better functionality. The optical shutters were investigated and some work needs to be done on the electronics to have them work properly. The beam on the interferometer reflected port was found to be clipping on the Faraday isolator in HAM1. There needs to be some re-alignment work done on the input beam to the interferometer. We have not been keeping track of the power levels in the interferometer. They have changed by almost 80% and this has affected the gain of various servo loops. We will undertake an audit of laser power levels and keep them as part of our standard parameters. (Joe Kovalik)

Used the earlier part of the week to finish final touch ups on PSL acoustic enclosure. Replaced a dead optical lever laser on MMT3. Continuing the Auto-Cad drawings for the IOT and ISCT table layout drawings and built a fixture for taking digital photos of the table layout for our own documentation. (Gary Traylor and Harry Overmier)

CDS: Completed work required in response gnatsweb problem reports 8 & 9 (new power supplies for optical lever lasers & LOS controller rework). Reinstalled EO shutter controllers on ISCT1 & ISCT4. Both are currently being tested. Completed building breakout boxes for testing Penteks. Completed fabrication of the new 1X7 digital suspension rack. (Rus Wooley)

Upgrading the OS to Solaris 8 ongoing on the CDS workstations. Also upgrading RAM on these machines along with the OS upgrade. Found slow network problems with the CDS ATM switch. Will look into it when Rolf is here next week. Writing an epics interface for displaying the PEM signal diagnostics on medm screen.
Setting up an Ultra 60 to swap with the existing london. Added two new hard disks on control room machines for disk to disk backup. Modified the LLO1 backup scripts to accommodate the new disks. (Chethan)

Doug Lormand has completed a nice (and very simple to use) web-based retrieval tool to export realtime data. This tool is linked to the main LLO website under Installation & Commissioning. Currently, the only channels available are PEM channels, but I will be adding new channels over the next few days. Also, the web tool is password protected. Doug will send the password to LSC members who request access to this utility.

GC: We have been working with LSU on the configuration of our router, and now have it in a more flexible configuration. (Tom Evans) Continued to do some work on the development web server here. Planning the temporary network in the new building for the LSC conference. Working with Bonnie on setting up a secure web page for
registration & payment for the LSC. Helped Doug out with password protection on a data access web page.
LDAS: Installed the tape robot from Cybernetics. I believe that everything was working OK. Gregg will be writing data to it today, so I will know if there are any problems shortly. Continuing to look into prices for the planned beowulf system. Getting prices for more ram for the 16 nodes currently here. (Shannon Roddy)

PEM: The weather stations are being upgraded to make them operate more robustly - operate off site power instead of solar (which discharges the batteries during extended cloudy periods), sensor re-wire, and investigate problems with data conflicts with dust monitors in DCU. (Rich Riesen)

LDAS:
* Made an offline backup of the database on metaserver at LLO.
* Installed DB2 Administration Server and Client on metaserver at LLO.
* Changed some Solaris kernel parameters on metaserver at LLO.
* Configured DB2 to take online backups every day at 3am.
* Investigating the possibility to restore a database from backups
to a different computer.
(Igor Yakushin)

Detector/Technical Support (Whitcomb, Coyne)

Commissioning:

Hanford
Livingston

Science & Engineering Support

See also the daily electronic logs for the installation and commissioning activities:

Hanford Detector Log

Livingston Detector Log

LHO COMMISSIONING

2km Commissioning

Dennis Coyne Reporting

Nergis and Peter made a survey of the noise on the coil monitor outputs of the PRC optics, as part of the search for the limiting noise seen in the 2km interferometer. While the coil current noise on these optics currently does not explain the interferometer noise, some of them are not too far below (in particular, the MICH and PRC LSC control inputs to the FMs from the Pentek DACs). The Michelson loop (limited by sensing/input noise) brings in noise that is about a factor of 3 below the level of differential arm noise. So before long (hopefully), the Michelson loop will need to be improved, by, for example, getting more light on the pick-off detector; putting a (digital) low-pass filter in the loop, with cut-off frequency somewhere above the MICH unity gain frequency.

The improved sensitivity of the 2km interferometer is shown in a calibrated spectrum of our present displacement sensitivity in this elog entry (username: reader, password: readonly) Compared to the E7 noise level, the present spectrum is about a factor of 5 lower at 100 Hz to 1 kHz band and about 7x lower at higher frequencies. For amusement, the displacement sensitivity goal for the 2km interferometer is included on the same plot. At low frequencies, below 1 kHz, the main gains were made by engaging the common mode servo, while above 1 kHz sensor noise was reduced by gain reallocation (increasing the gain at the input to the ADCs). Presently, the extremely high noise levels measured on the ETM coil monitors is most likely limiting the sensitivity.

Rana and Peter found that the ETM coil noise was dominated by enabling the ASC input. There is a lot of high frequency noise coming out of the ASC dewhitening filter (perhaps Pentek DAC glitches). Modifications to the dewhitening filter stages made a significant reduction in the coil driver noise. However, this did not result in improved interferometer sensitivity.

After switching from 1W to 6W, the MC was having trouble locking until the reflected power was set to original levels with optical filtering, the cross-over gain was raised and a gain scheduling sequence (pre- and post lock) were established. Full interferometer lock at 6W has been achieved, but for short duration (a few minutes) and the common mode servo fails to engage. There is some evidence that there may be light coupling to the OSEMs.

4-k Commissioning

Work has focused principally on attempts to get longer, more robust locking. Currently lock durations are a most a few minutes and generally shorter.

LLO COMMISSIONING

Rai Weiss reporting

The results of having installed the acoustic isolation chamber around the prestabilized laser at Livingston are a general reduction of the acoustically coupled frequency noise by a factor of 5. Significant is the clean up of many of acoustically driven lines in the 100 to 1000 Hz band, some by more than a factor of 20. The broad noise peak at 100 Hz is unaffected and due either to electronics or mechanical motions. The prominent peak near 200 Hz, due to the cantilever motion of the periscope, is reduced by about 3, so still primarily driven by direct mechanical vibrations.

The installation and repair of the electroptic shutters, in preparation for recycling, is being more aggressively pursued.

The antisymmetric wavefront sensor (WFS1) has been reinstalled with a new combination of lenses in the Guoy phase telescope. The discrimination of cavity input and output mirror motions is currently being measured.

The innovations made at Hanford during the past weeks are being incorporated at Livingston. The sequencing script developed for the common mode servo is being installed and modified to be appropriate for Livingston. The low frequency servo to reduce the tidal excursions of the arm cavities will be installed in the near future.

A survey of the noise of the Pentex A/D, D/A converters in the length sensing system was finished. It showed that several D/A units were excessively noisy (these have subsequently been replaced) and indicated that we may want to reconsider the design of the whitening/de-whitening filter functions given the realities.

After further work on the noise in the recombined system in the coming week, we expect to return to recycling of the interferometer since some of the noise reduction strategies are beginning to depend on the interferometer configuration.

SCIENCE & ENGINEERING SUPPORT

Seismic Isolation Upgrade

Dennis Coyne reporting

The performance testing for the first generation of prototype hydraulic actuator was completed at Stanford. 1 DOF isolation of a factor of ~10, limited by the displacement sensing in this prototype, was achieved over a bandwidth > 10 Hz. A second generation of hydraulic actuator has now been installed on the Stanford test stand and is undergoing testing. This unit incorporates a modified bellows for higher breathing mode frequency and will (eventually) incorporate a more effective bypass network.

An alternate (less expensive) eddy current position transducer has been obtained for evaluation. The displacement noise spectrum is comparable to the (baselined) Kaman sensors. Further evaluation will be done before selecting.

Characterization of the differential, bridge valve (DYP2S) continues at Stanford. The transition to turbulent flow was found to occur at flow rates less than required for our application. Initial modifications to the nozzle geometry are promising, but do not yet yield an acceptable margin.

The detailed design and procurement for the prototype hydraulic pump station to be used in the LASTI testing has begun at CIT, based on the earlier prototypes at Stanford and MIT. This unit will be tested for pressure noise spectrum with a simulated load before shipping to LASTI for integrated testing.

The assembly structure atop each BSC pier for integrating the sensors/hydraulic actuator assemblies and providing the spring support for the crossbeams, is in detailed design. We expect to release the drawings for prototype fabrication soon. A similar unit will be designed for a LASTI HAM prototype, with the alternate actuator (EM linear motor or piezo)

Fabrication of the prototype springs has nearly been completed. We were only able to find one company capable and interested in the spring fabrication. A spring tester is also being fabricated and will be available soon for measuring the spring rate.

In addition to the external pre-isolation approach (with either hydraulic or EM actuators), we are beginning to explore the use of in-vacuum sensing and actuation (across the stack from the support structure to the table). This later approach has the potential to actively damp stack modes, in combination with active isolation from the external pre-isolator. The internal sensing & actuation might be based on the units developed for TAMA, or other commercial sensors (e.g. geophones and the EM linear motors being developed for advanced LIGO).

A working meeting of the external pre-isolation team is being held at MIT this week. Key is to get a statement of the requirements defined, status/revision to the plan and schedule and establish a date for a project requirements & design review.

CDS Software

Rolf Bork reporting

Finished up design document for the CDS standard digital filter software; Lori is in process of coding it now.We will begin testing next week.

Worked with Richard at LHO tracking down a problem with ASC hardware filter switching; this is now working properly.

Hongyu finished up changes to dataviewer to read 16sec frames; she is now back to working on the multi-CPU Epics data collection unit.

Now that at LHO we have a couple of DAQ data collection units directly on the FrameBuilder reflected memory network, we encountered problems whereby if a DCU was reset, the controller would stop. Alex found it was a DMA timeout problem due to the network interruption and has fixed the problem.

Began discussions with Alex on how we could have all the real-time processers report the code presently running along with its version number (as requested by Sigg). After we sort out a few ideas, we'll submit a proposal.

Alex has begun work on having the FrameBuilder read and export data from trend frames produced by other systems, such as the DMT and eventually the Facilities systems.

I'm at LLO next week. Present items on list are track down ATM problem, upgrade primary framebuilder to 16Hz frames and check into the LSC dewhite filter switching noise.

RF Photodiode

Rich Abbott

I have finally completed the rework and testing on 6 LSC RFPDs.They are tuned to 24.495 and 29.508 MHz.The travelers still need to be filled out and then they will be shipped out to Nathan at LHO for final performance testing.

Variable Timing Delay Board

Mohana Mageswaran

Added VME interrupt capability to the variable delay timing board and is testing the prototype and working in close contact with Jay and Rolf.

RF Distribution

Rich Abbott

An investigation into the RF distribution system has revealed that some of the inputs and outputs are not well impedance matched. Although not a significant factor in the radiation problems we have seen at the sites. I prepared a two element single frequency matching circuit which seems to function well. It can be retrofitted into the existing design without much effort if needed.

EO Shutter Controller

Sander Liu

EO Shutter Controller - In the process of testing two new units. One will be sent to LHO and the other to LLO for spares when complete.

Digital Suspension Controller

Jay Heefner

Continued to study the LOS Coil driver PA85 cooling problem. A larger heat sink is on order. In addition we are experimenting with ways of locating the fans to reduce fan noise on the signals.

Pentek Noise

Jay Heefner

Pentek has modified their tests to poll the module and write data to the module on a sample to sample basis and they now see the noise that we see. Hopefully they will be able to come up with a fix soon.

Optical Metrology

GariLynn Billingsley

We have received a large shipment of Mode Cleaner mirrors from Wave Precision as follows:

1 MCCM, (40M)

3 MCFMs (40M)Results are posted at http://www.ligo.caltech.edu/~gari/40MAsBuilt.htm

3 MCFMs (L1)

3 MCCMs, (L1)

All of the flat mirrors have been measured and are within spec.I am working on a process to measure the curved mirrors.

Bill Kells, Liyuan Zhang, Lee Cardenas

Continuing work to get two contamination cavities properly resonating in OTF. Good resonating modes so far, but not locked yet due to lack of demod signals. We are tracking down the possibilities one by one.

Optical modeling

Bill Kells, Erika D’Ambrosio

Continuing analysis of SB behaviour in LIGO-like interferometers. We have made a lot of progress understanding the SB behaviour, and are trying to put it all together for a paper. This last study turns out to overlap considerably with work the e2e group is now doing to model necessary HTMs in their simulations, particularly of SBs. have been discussing the various issues with Biplab, Hiro, Matt.

We have reviewed some Ligo literature on the mode expansion of fields when a mirror it tilted. Some approximations have been puzzling us because results beyond the linear approximations are kept from calculations started with a linearization of the perturbation. Keeping the second order terms has introduced some issues that must be tackled. An example is the lost of the orthonormality property for the perturbed modes.

PSL

Peter King

The output of NPRO #259 was circularized and mode-matched into the pre-modecleaner. The NPRO power supply displays "CABLE ?" on the LED display which makes controlling the NPRO a little difficult.

I dug out some information about a shutter suitable for use as a laser safety shutter for the 10-W laser.

An 80 MHz VCO was put through its paces, prior to being shipped to LASTI for replacement.

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

LASTI (Zucker)

LASTI (Bayer, Fritschel, Goda, Harry, Laliberty, MacInnis, Mason,
Mason, Miller, Mittleman, Ottaway, Phinney, Rollins, Shoemaker, Zucker)
=======================================================================

VACUUM ENVELOPE:
We've been under vacuum about 300 hours since the installation of the
two SOS suspensions, wiring and viewports in HAM13 and repair of the
bellows leak on BSC0. Calibrated RGA spectra show the system is still
extremely clean, with no trace of hydrocarbons up to AMU100 (at
sensitivity level of about 1e-11 torr partial pressure, turbopumped).
We are still
dominated by water outgassing at a pressure of 2e-6 torr (falling as
~1/time), and there is also a substantial air signature. The former is
consistent with degassing of the Viton, the latter is still open to
debate (we've He leak-checked and certified every joint, but
nitrogen, oxygen and argon seem higher than normal).

HYDRAULIC ACTUATORS:
The analysis of the Housing assembly has been completed. First mode is
at 147 Hz and the maximum loaded stress is 5000 psi. A left and a right
hand version has been designed for the BSC chamber.

Dennis Coyne, Brian Lantz, and Joe Giaime will be at LASTI this week for
a week long discussion on the project design and status.

PSL:
This week the modified
FSS card was tested. From the preliminary results it appears
that this card should provide improved bandwidth for the FSS
servo. Unfortunately before this could be tested on the PSL, it was found
that excessive power from the VCO had damaged the AOM, hence
preventing light from reaching the reference cavity. A replacement
VCO and AOM are en route from Peter King at Caltech. This week we have
also commenced what we hope to be the final tweaking of the modematching
including the insertion of a cylindrical lens to correct for the
astigmatism in the input beam. Results of this will be included in next
weeks report.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Bhawal)
Weekly Physics meeting
-----------------------
First Hiro presented equations describing the effects of spatial modal mismatches and
explained why implementation of the Modal model upto 4th order was necessary. After that
Biplab presented new results from studies of in-lock states of LIGO as it gets hotter.
A lively and useful discussion with Bill Kells, Erika, Luca, Matt and Stan followed.
Details available at http://www.ligo.caltech.edu/~e2e

Collaboration starting with SLU
----------------------------------
Sany Yoshida is now a faculty of South-Western Louisianna University, and he is going to
initiate an activity using e2e for LIGO research and for the education of students.
Hiro is preparing an MOU attachment for the distribution and use of e2e at SLU. Hiro had
a first telephone conference with Sany and his students on Jan.31.

Locked hot LIGO
----------------
(Biplab) In-lock states of LIGO as it goes from cold start to final hot state could be
obtained. Understanding and analysis of several issues are going on.

LIGO I simulation system
--------------------------
(Matt) I transplanted the Han2k simplified PSL and IOO systems into the new simulation.
I am currently working on the simulated ISC electronics.

Modal model
-------------
(Hiro) Calculated various field interactions to understand the modal model.

Code development and maintenance
---------------------------------
(Ed Maros)
1. Built and installed latest beta version of e2e code. This version has only
been installed at Caltech.
2. Worked on build-e2e script to simplify building of e2e software.

(Biplab) After discussing with Hiro and Matt, started working on a change in mirror
module to account for a phase that might originate from changing modal parameters of
the input beam to a cavity.

Alfi
----
(Bruce)
- Port reordering implemented.
- Junction repositioning implemented.
- Working on Rotation operation for member nodes.

(Melody)
- Finishing up the alfi5 implementation. Working on the node user modifications.

Software Systems (Blackburn)
The new diskCacheAPI has been integrated into the LDAS CVS repository
and is being configured and built into each of the nightly builds on
the LDAS development system. It has taken over the frame disk cache
management from the frameAPI. The frameAPI is no longer handling the
disk cache management and has more CPU cycles available for frame I/O
requests. Open issues are evolving on an hour by hour bases as we are
carrying out our first suite of tests or this new configuration. It is
also visible on our development webpages.

The configuration modifications for migration from alpha version
numbering to beta version numbering have disclosed half a dozen issues
with how LDAS is configured and built this effect everything from makefiles
to webpages. We are slowly sorting the all our and hope to have stable
nightly builds on a regular basis soon.

Changes to the metadataAPI to separate out the handling of getMetaData
and putMetaData have increased our database performance (the modivation
for the change) by 20%.

A custom compilation of TCL to attempt to improve possible thread issues
resulted in a "no-go" for these configuration options. We will stay with
our current configuration options.

We have now captured two dataConditionAPI core files and problem reports
have been filed. Hopefully they will provide clues into the thread issues
limiting the reliability of this API to about a 98% duty cycle.

We tested sending multiple channels to datacondAPI. All the
combinations that were tried (1-second frames, 16-second frames, fast and
slow channels) worked without noticeable problems. I am still looking
into the channel ordering issue discussed in Wednesdays meeting.

We received a request from the LSC to modify the information passed
along with the data to search codes running on the LDAS beowulf.
The specific request was to add a list of channel names which were
used to construct the input data, and a user-specified comment.
After discussing some alternative approaches, we have started to
implement the requested changes.

This week we've also been working on porting the classes needed for
resampling into a new "filters" library. These are:

and some incidental source files, some of which will be eventually moved
to libgeneral. The porting involves recasting all the interfaces in terms
of valarrays and/or pointers to arrays, and removing all references to UDT
and UDT-derived classes.

UWM has been running single jobs on their LDAS system which produce over
38000 events in a single job. Attempts to insert this many events into the
DB2 database failed for the default DB2 configuration. UWM requested support
in understanding this problem. We were able to determine that this was not
an LDAS limitation and after some research found the configuration options
needed to allow this large a request to the database. We hope that this is
just a transitional need on the database server and as data understanding
increases single jobs will not generate this large a insertion request on
the database (otherwise we under estimated our database needs by several
orders of magnitude). UWM is now able to run the large DB2 request jobs.

The controlMonitorAPI is being enhanced to include monitoring and control
of core files produced by LDAS APIs at runtime. The new interface will allow
movement of core files into a staging area, email notification, tabulated
graphical summary information and the ability to load core files into a
debugger.

Support for frequency bandlimiting of data stored in frames has been added
to the frameAPI shared object. More functionality is still needed in the
TCL layer to fully exploit this new functionality.

Thread functionality in the genericAPI was both fixed and enhanced to
include a debugging tool.

The LSC has requested additional formating for channel names in the
interface to the wrapperAPI. This has been added to the wrapperAPI and
will soon be available for LAL and LALwrapper coders. Still needed is
a parser to pull out names from actions.

* Tuned purge policy for SC=32 (the disk cache used for COS=31/32/310/320/4320)
to better handle LIGO E7 data ingestion. It seems to have helped.

* Fixed various breakages in HPSS (deleted core files, restarted servers,
added tapes to storage classes, another stuck 9940 tape resulting in
STK service swapping out one of the drives)

* Last week's problem with the Ultra 10 that controls the STK9310 silo
reoccured: memory was swapped out (with memory that LIGO had available)
and the problem has not reoccurred since (still doesn't mean that it's
solved). STK is having bureaucratic problems with supporting the Sun
box, but they're trying to resolve this so that they can get memory,
motherboard, etc.

* Started implementing revised plan for splitting up the large directories
in the LIGO HPSS archive by moving files around.

* I have delayed developing a simple kickstart script for redhat 7.2. losing
focus on it. I have completed and testing the rpmsync and cfdef setups for
RH7.2. This works just like it did for 7.1.

* Getting up to speed on BigBrother for ldas. I am testing some configurations
on ldas-sw. So time from time sysadms might get some warnings
that are not true.

* I have updated the sendmail files on Linux systems (cit). The fix was for the
senders return path. This fix could be applies to other sites as well. See
/ldcg_admin/cfdef/Linux/etc/mail/sendmail.mc/cit and change the return line
for your site.

* Renamed all LDAS gateway machines from "dataserver" to "gateway" in
preparation for bringing on-line new internal "dataserver" servers.

* Started discussing and planning the Caltech LDAS network topology to
take advantage of the currently being installed 50um fiber.

* Investigating the possibility to restore a database from backups
to a different computer.

* Installed the repaired LDAS tape robot from Cybernetics. I believe that
everything is working OK. Gregg will be writing data to it today, so I
will know if there are any problems shortly.

* Upgraded my tapecontrol script to read in the time intervals from
text files, and to process data only within these intervals. Currently
the script is running at LHO and LLO, writing to tape the playground E7
data identified in the files PlayH1noH2L1.txt, PlayH2noL1.txt,
PlayL1noH1H2.txt, as explained at this URL:
http://www.phys.lsu.edu/faculty/gonzalez/E7Segments/.

* Installation of a crane is taking this place in the staging building
this week at LHO. Discussed with Richard and Otto the implications for
LDAS. It was decided this should not affect LDAS operations at LHO.
However, several temperature alerts were generated in the ldas area due
to the doors in the staging building being open, causing the building
heat to come on. This was handled by lowering the heat set point on the
building thermostat.

* Work on the knownpulsar DSO has focused on handling data quality and
drop out, getting SFT output to go to a designated directory, and naming
issues associated with SFTs.

Data Analysis Activities (Lazzarini)
* (Mendell) The final version of the November 2001 Known Pulsar Demod MDC was
written and submitted to the DCC with the name T020014-00.pdf.

* Modified conlog web interface to correct for time discrepancies
in the raw log files during E7; times should now be accurate at
the 1-2 second level.

* Fixed the LIGOtools 'getFrames' utility to be able to retrieve
E7 data from the LDAS systems at the observatories.

(Lazzarini)
I received a request from I. Leonor, a UofO postdoctoral scholar, for O[0.5TB] disk space at LHO in order to store a reduced data set (RDS) that she proposes to produce use her own filtering code. We are in the process of understanding what filtering is being proposed, which are the 40 channels that would be kept in the RDS, and how the body of code she has created single handedly fits into the larger scheme of things. Rai is looking into this from the detector side.

General Computing (Wallace)
MIT:
(Keith)
-Configuring new pc for video in conference room
-Started updating gc (and LASTI) Solaris nodes with MU6
-Testing out networking equipment for NW22
-Moved LASTI gateway machine onto 100Mb/s hub
-Installed matlab 12.1
-Clearing out lab pc for postdoc
-Removed most SSH v1 daemons

Livingston:
(Tom)
-We have been working with LSU on the configuration of our
router, and now have it in a more flexible configuration.
-We have ordered two PCs for upgrades for users, and received
one of them.

(Shannon)
-Continued to do some work on the development web server here.
-Planning the temporary network in the new building for the LSC
conference.
-Working with Bonnie on setting up a secure web page for
registration & payment for the LSC.
-Helped Doug out with password protection on a data access web page.

The network traffic analysis for the Observatory can be found at the following
location: (There are two T1 lines)
http://kahuna.net.lsu.edu/mrtg/remrouters/ligo-t1-1.html
http://kahuna.net.lsu.edu/mrtg/remrouters/ligo-t1-2.html

Hanford:
(Christine)
- Wrote comments to the GC Computing Policy document.
- Continued setting up new PCs.
- Bought more PCs as upgrades.
- Created several user accounts.
- Provided user support for various problems.

CIT:
(Ed)
-Working with Veronica on updating web pages for the LDAS web servers.

(Mike)
-Maintenance on the HP LaserJet 4500 color I had to swap out the transfer belt.
-I upgraded Janeen Romie & Calum Torrie to solid works plus.
-I ran a cable drop in the bake oven room for Robert Taylor.
-HP DeskJet 1600CM I swapped out this printer for a serviced identical
printer that is in better working condition.
-Worked with Larry on setting up a New employee account, program settings and
work environment.
-Backed-up Rick K's personal files and created a ghost image of his hard drive.
-Transferred equipment to and from Millikan.
-Started reloading two PC's with win98 and General Computing software.
- On a couple of PC's the OS is corrupted and installed programs are not
responding. I am backing up all personal files then reinstalling to try and
repair OS this is a 2000 Pro installation.

(Veronica)
- LIGO website: posted various updates (missing talks at the last LSC
meeting; Fellowships/postdoc openings; employment listings; LSC talks to
be presented; new accounts numbers). Finalized and posted Elba GWADW
webpages. Made some minor changes to the LIGO home page.
- LDAS: met with Albert and others to discuss updates to the LDAS
website. Working with the graphics for it.
- Project Science Jan 18-19 workshop: posted presentations webpage.
- CaJAGWR: posted transparencies of the last talk and the schedule for the
next seminars.

(Lisa)
- Figured out how to run mailman, a tool for web based mailing lists, in our
configuration. There was a problem with the GID resolving correctly between the
mail server and the nis+ server. Mailman will be available once the new mail
server is up
- Compiled an accurate list of unix computers in active use to submit to ITS.
- Worked with Larry on putting the new mail server on line. There were some
problems so we backed down to the old one until we got things figured out.
- Began monthly backups. So far I am getting chronic write errors from the
cybernetics using ufsdump. It's less than pretty and I need to call
cybernetics.
- Rebuilt an ultra 10
- Worked with Larry on network problems in wilson house. Ran patches on polaris
to see if that helped clear things up.

(Larry)
-Went through a number of procurement issues. Have to reorder a number of units
since the deal through CatechWired fell through.
-Received the new licenses for Matlab and IDEAS master series 7. The contract
with SDRC (IDEAS) and Caltech has not been renewed and I have not received any
word on when the new contract will be in place. However, SDRC informed me since
this is a license for and older version we should not have a problem with the
license they sent. For the latest version of IDEAS we would have to wait for the
contract to go through.
-Met with the Gang of 5+1 to resurrect some guidelines we worked on a number of
years ago.
-Started working on the GigE setup. We will need to order three more edge
switches, to replace the FORE edge units we were planning on using. Foundry will
be out next week to help with some of the installation.
-Installed another printer and worked on a couple of PC fixes.
-Spent time getting the DCC Web server working correctly again. Hopefully I
fixed more things than I broke.
-Worked on the General Usage Policy document. It has been sent out again for
comments and I have received a number of good replies for changes.
-Met with the Caltech computing audit group and Albert. They will start running
a security audit for us next week. We will notify everyone when they start
testing the network and server systems.

LIGO II/Advanced R&D (Sanders)

This week I made a new release (1.03) of the triple version of my model
and started work on the export to Matlab. I reverse-engineered Peter
Frischel's Simulink wrapper for Calum's Matlab model and modified it to
take one huge state-space model (60 DOF for the quad) instead of 4
independent models for longitudinal/pitch, transverse/roll, vertical and
yaw. I've also nearly finished the code for export of the state-space
from Mathematica.

I have started drawings and an assembly for a planned single pendulum experiment in SolidWorks. This will include crossed blades, angled clamp and 2 wires off 1 blade. Information gathered from this will be very useful for adding various effects to the pendulum models.

Norna and I have measured the deflection on several cantilever blades. For comparison we now plan to model the blade with its pre-stressed radius of curvature and load it in ALGOR to make it flat.

We have been looking again in detail at the pendulum model of the LIGO_MC design, following the changes noted above. We now have a working design which can be taken forward in order to create a full set of engineering and assembly drawings.

Mechanical loss - Coating studies
Shipped the last coated substrates received from Lyon to MIT
Placed an order for 10 more substrate blanks of each type and to polish 2 substrates, 3"dia. x 1" thick, made from Suprasil 311 SV (low water content)
Received a revised quote from MLD to coat 2 thin and 2 thick substrates. The new quote includes the special masking fixture required to coat the thin samples.
There are "Q" measured substrates available for this task.

Materials
Identified another company that may cast and form high density glass masses. A company representative will probably stop by Caltech next week.
Also, located a polishing facility to polish 4 samples made of glass # 24935 to start bonding tests.

(ERGO) mechanical arm
Discussed with M. Gerfen from CES the specs and requirements of the device. He is working on a cost estimate.

1. Testing is complete on the +/- 10A voice coil driver that came back from the board house. Here are some of the interesting numbers.
a. Current drive capability - +/- 10 amps into about 4 ohms
b. Output short circuit and over voltage protection circuitry built in and functional
c. Worst case analysis of power dissipation indicates about 52 degrees C. case temperature, which is well within the recommended operating spec.
d. On-board ambient and heat-sink temperature read-back
e. On-board output current monitor read-back
f. Modular design that allows rapid access to all components for maintenance or repair. There are no solder connections to the unit. Six drivers will live in a specially designed 2u high chassis with an external 2500 watt power supply. The total construction time for one unit can be as low as about 4 hours.
2. Preliminary testing of the capacitance position sensors revealed that there is a construction error on the cabling. The center pin of a SMA connector is not properly inserted. I contacted the manufacturer and worked out a way of fixing this discrepancy. The three units still to be sent from the manufacturer will be corrected prior to shipment.

40m electronics
Jay Heefner
- Continued work on the mode cleaner alignment and length controls. Drawings are ~75% complete.
- Began drawings for the frequency distribution system.
- Selected a video switch for the lab. It is the same model as the sites (for s/w compatibilty) and has 64 inputs and 32 outputs. Quotes will be obtained this week.

40m. South Annex
Lee Cardenas
Fluorel pieces have been baked by Bob Taylor and are still inside the oven. We are waiting on Larry Jones for final decision on if & how to remove free flourine with the "Walker" process. The total pressure for the bake oven is 4.2E-10 Torr.
The sum of outgassing rate for selected AMU (41,43,53,55 & 57) exceeds the requirement for Fluorel (in LIGO-E960022-03-E ). We may consider baking the parts again but I believe that the parts are clean.
In one of the new bake oven chambers, heating pads have been fully attached. It is now in preparation to have them covered with insulation foam.

40m. Ref. Cavity chamber.
Lee Cardenas
I have installed the heating jacket and am waiting on the thermistors.
I have cut the insulation foam (Polyimide fiber temp. good up to 287C) to make the insulation jackets for three Ref. Cavity chambers. The insulation jackets will be covered with Nomex fabric(temp. good up to 315C)

The draft wirelist, EPICS records database and cross-connect
diagrams have been generated for connecting the existing GEO 10-W to a LIGO
style VME-based control system. Currently the documentation is being
cross-checked, prior to fabrication of the wiring harness.

Hareem, Riccardo @ Pasadena
Setup the new cryo-lab in West Bridge, working on measurement software.
Decreased creep oven temperature from 80 to 60 centigrades.

Jan, Bill, Riccardo @ Pasadena
Nothing new yet for brazes or spin cast from Carnegie Mellon

Riccardo @ Pasadena, Gianni @ Bientina
Updated preliminary design of auxiliary bench for Hongo low frequency
TNI

Szabi, Virginio, Alessandro, Akitero, Virginio, Luca, et al.
Following an idea from Szabi we are evaluating the possibility to apply
viscous damping using the LVDTs and constant force actuators directly
across the LIGO I stacks to eliminate the ill effects of their
resonances.
The main problem is that this is violating our own basic tenet of
isolation, which is to make attenuation in steps and relegate the
damping to the initial step. The danger is to effectively short circuit
the stack attenuation properties at high frequency.
The actuators seem to be adequate, we still have doubts on the sensors.
Then we have to study the viscous damping loop feasibility.
This study is getting first priority, over everything else that we are
doing.
Given the long lead time of some components we have already ordered some
peek to make UHV compatible LVDTs and some horizontal accelerometers
which in any case can be recycled for the TAMA interferometer or the
thermal noise facility at Hongo.

Papers
Also the LVDT paper was accepted by the referees (with minor comments).
The paper is sent back to NIM after fixing the comments and will be
published back to back with the actuator paper.

Minutes of the Core Optics Subgroup, 1/10/02
9 am PST US/Europe meeting
UF: David R.
CIT: Norna, GariLynn, Helena
MIT: Gregg, David S., Peter F.
Glasgow: Jim, Sheila, Peter S.
Stanford: Roger, Marty, Vlad
1) Coating Status
- Gary S., Helena and Gregg visited MLD in early January. Since the last visit (~ 1 year ago),
MLD has made many improvements to their capability. In particular, they are setting up a
characterization lab to look at coating parameters. MLD is willing to work with LIGO on
absorption and also Q studies if needed. Gregg noted that one interesting technology they
employ eliminates the sharp boundary between coating layers such that they can control how
much blending (gradients) occurs between coating layers. We’ll start them with a classic 30
layer coating to compare Q results between the different vendors.
- Achieving AdL goals for low absorption in the coatings is appearing more feasible. Mackowski
(JMM) reports 0.2ppm coating absorption (AdL target is 0.1 ppm).
- We have just received 30 layer lambda /4 wave Ta2O5 and SiO2 parts from SMA-Virgo: 2 thin (for
MIT) and 2 thick (for Glasgow/Stanford). We have 4 more of each type at SMA for 30 layer
3lambda /8 and lambda /8 Ta2O5 and SiO2 coatings to look at material thickness dependence. Following that is
a 60 layer run of lambda /8 Ta2O5 and SiO2. After these are in process, it was agreed that we should
send MLD and REO substrates for 30 layer coatings to see if there is a difference between
vendors. Two suprasil substrates are ready for coating.
- LIGO is in the process of ordering 10 more of each type of substrate. Substrates are the
limiting resource right now, so we have to plan our coating runs carefully.
- Some discussion of altering the ordering and nature of the Q runs was discussed. One
possibility suggested was to use Aluminum Oxide instead of SiO2 as the low index material.
Dave R. advocates waiting until we have results for this first run before we change the plan.
- Jim requested a small delay in shipping substrates to SMA-Virgo to complete characterization.
It was agreed to delay the next shipment of substrates from Glasgow to Lyon until the 18 Feb.
We will re-examine the coating plan at the next LSC
- Helena reports that MLD is trying Nb with different anneal temperatures for absorption studies;
will await absorption results.
- CIT is working on the ergonomic arm for lifting large optics. We have decided to adopt the
vacuum cap for lifting because the process is less sensitive to tooling tolerance. This work is
being done with the CIT machine shop.
2) Polishing Status (Gari)
- Goodrich notes that they see two different patterns in the substrate homogeneity depending on
polarization. This was known and documented in the CSIRO report on such measurements, and
Goodrich has a copy of this report but apparently forgot. The homogeneity depends on which
polarization the bulk is probed with. Unfortunately, Goodrich has not been careful about using
only one polarization and been compensating for one then the other. CIT informed them of the
problem and it has been corrected. As of now either polarization will work for testing purposes,
however, the IFO design may specify a particular polarization.
3) Sapphire Absorption (Roger)
- The absorption measurement system is now reproducible and reliable. Vlad is working through
measuring 36 1/2 windows which will then go to CSI for annealing, hopefully by 1/11. They
will then work on standards for fused silica absorption.
- The CIT furnace has arrived at Stanford, with some damage due to shipping. We are working
with the vendor to determine if Stanford should fix it or get new one with 4 week delivery.
Consensus to have a new unit shipped out.
- The first set of measurements on sapphire before and after heat treating showed a reduction
from 65ppm/cm to 55ppm/cm in the bulk.
- The 10W laser provides sufficient resolution measuring sapphire and shouldn't have a problem
measuring FS. We will get to work on the samples from MLD beginning 1/14
4) Q measurements (Gregg, Sheila)
Sheila reports working on characterizing blanks for the coating experiments followed by
measurements of the coated samples that came back from JMM.
Post-telecon news: The first of the samples coated at SMA-Virgo have been characterized; Peter
Snelling at Glasgow has preliminary results for both the 30 layer and 2 layer SiO2/ Ta2O5
coated samples. The results suggest that the loss in the coating in the 30 layer sample is slightly
greater than 2 x 10^-4, although more work is needed to be precise about the number. (This is
close to both the Syracuse coated silica numbers and the Stanford/Glasgow coated sapphire
numbers both of which were for the same coating materials). The 2 layer coating has had a
much smaller effect on the measured Q's of the modes of the silica sample, basically at the
resolution limit of the initial measurements.
MIT results on the thin sample with the 30 layer coating that are in good agreement with the
Glasgow results, in the range of 1 to 4 x 10^-4 for the coating loss. MIT is performing more
measurements and modeling to better determine the energy distribution in the sample.
This suggests that the loss is not associated with the first interface of the coating with the
substrate material, but instead is related to the body of the coating (i.e. material or
multiplayer interfaces).
5) Wrap up, upcoming deadlines
- Peter F. and Gari requested inputs for their visit to CSI taking place the week of 1/15.
- The Glasgow group is getting a 25 X 12 a-axis piece of sapphire (polished in the UK) for
bonding and suspension trials. The current plan is to demonstrate hanging from a heavy glass
mass, looking for stresses, response to temperature change, Q. We would like to have it at
Christmas next year for the hanging test. This piece may be available for other testing.
GariLynn B noted that Glasgow could use the large piece of sapphire (currently at Goodrich) for
their testing, with Glasgow buying a piece that has more general testing value for the AdL effort.
5:30 pm PST Telecon
CIT: Gari
UWA: David B.,
Adelaide: Peter V., Jesper M.
- David Blair notes the difference between theory and measurement of Rayleigh scattering, a
factor of ½ in Sapphire where the theoretical value was 20ppm and the measured value was
10ppm. Fused Silica theory yields a few ppm, the measured value is 10ppm. For further info
see: Setah Benabit thesis at UWA
- UWA is now getting vacuum system and suspensions in place. We are also getting preliminary
optical cavities going in order to get instrumentation in place for the high power experiments.
- Peter V. presented a strawman design and set of experiments for the Gingin High Power test
facility. ACIGA wants to present the Gingin plan at next LSC

From Gerry Stapfer quoting a communication from HPD, the new seismidolation prototype builder:

"The basic Stage 2 assembly is complete! The assembly went very smoothly, with all the parts fitting together easily. We are now at work on assembling to Stage 2 the forcer and sensor mounts, flexure mounts, etc.

I have included a few pics; we will be posting these and other pics on the HPD website within a couple of days. Please circulate these pictures to whomever would be interested in the progress.

After the Stage 2 final assembly is complete (minus the optics table) we will mount the pillars onto the support table ("the yellow table") and then lower Stage 2 in place. Stage 1 will then be installed, and then the springs and flexures. Finally, the optics table will be mounted."

bh installing pins in ribs 2.jpg

dan installing table on s2.jpg

s2 without table 2.jpg

Weekly Report for Week Ending January 31, 2002

Special Announcements:

LSC Issues (Weiss)