LIGO Weekly Report

CONGRATULATIONS TO ALL FOR A VERY SUCCESSFUL REVIEW BY THE NSF. THE REPORT DESCRIBED AT THE OUTBRIEFING ON THURSDAY AT HANFORD GAVE LIGO THE "HIGHEST POSSIBLE " RATING. MORE TO COME AS THE REPORT BECOMES AVAILABLE.

LSC Issues (Weiss)

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

There was no site teleconference held on Thursday, March 1, 2001 due to the NSF Review at Hanford.

The list of current actions revised to reflect open actions assigned through February 22, 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. . .

From: Ed Jasnow <jasnow@ligo.caltech.edu>

Irene Baldon

Twenty-Nine (29) new trips were started (including Advance Check written and hotel/car rental authorizations filled out and FAXed to appropriate vendor in various locations). Sixteen (16) trips have been ticketed and are awaiting completion of necessary paper work. I have twelve (12) trips pending final approval before tickets can be issued.
Completed twenty-Six (26) Expense Reports and have seventeen (17) Expense Reports to work on and I'm holding two (2) reports which need a check from the traveler before processing. Jim Covington has seven (7) expense reports that he is working on. I continue to train and supervise him on expense reports.
I have twenty-two (22) items on my P-Card to reconcile this week. Assisted a few travelers with their reconciling and/or data entry. Continue to following up on Al Lazzarini's stolen card and new card needs.
Dealt with several emergency problems on what I call "Doom’s Day for Travel" and continued to work on reissuing upcoming tickets from E-Ticket to Hard-Ticket in case of airline strikes.
Prepared the Travel/Vacation Itinerary for the Week of February 26, 2001.

Rita Torres

Obtained incoming LSC reports from IAP and ACIGA, these are in review. Attachment No. 1 with Shanghai, had to re-input as FrameMaker file would not cooperate. Did change order No. 11 with U of FL. Obtained several copies of responses to RFP IPGS-511, SEI isolation System Mechanical Structure Prototype at LLO.
Obtained Oracle requisition numbers for Change order No. 3 to Crystal Systems, and regular POs: Synergy Microwave, Mindrum Precision, and Sager Electronics.
Chased down some information regarding eye wear from our new vendor, Lase-R Shield. Spent some time purging old email. Larry Wallace, did I do good?
Ongoing activity: Placed Pcard orders, chased documentation to reconcile some.

Dorothy Lloyd

Processed the usual requisitions, invoices and receiving on-line. For more detail, see "Cost Schedule Control Systems" report by Esther Cunningham.Tracked and followed up on invoice problems.
Reviewed and recorded payments processed by Esther for the period February 5-23, on contract summary sheets and LIGO database.
Continued to work with Florence and Ruth on transferring encumbrances to new fiscal period and closing out old POs.
Continue to monitor contract and blanket-order funding levels and notify task managers when supplements are needed.
Jim continues to do data entry in the LIGO database and help out in both the DCC and travel areas.

Elizabeth K. Wood

Prepared a binder for the NSF review as a sample for Jill Berry to use in creating the binders up at Hanford. On Thursday last week, we sent it via FedEx overnight delivery by 10:30 a.m. It arrived on Monday. We will not be charged for the delivery. It’s a good thing I didn’t send all of the binders by FedEx up to Hanford.
Communication with the immigration department has hit an all-time low, and as a result, their response rate is equivalently slow.

Advanced LIGO (Frey, Petrac)

A reminder that I will require contributions for an end-of-February Quarterly Progress Report in March. I would like to set a target date of Friday, March 23rd. Thanks!

Quality/Safety (Tyler)

LIGO Hanford Observatory (LHO) Operations (Raab)

Everyone knows by now about the magnitude 6.8 earthquake that occurred the morning
of Feb 28 near the WA state capitol of Olympia. Prior to this earthquake, several
issues concerning alignment of corner station 4-K optics were coming to resolution.
The earthquake undid all this work. A crew of people were able to restore alignments
for the two ITMs and the Beam Splitter. This allowed the COS crew to get back in
and restart the alignment of the pick-off mirrors and telescope. Currently (2pm,
March 1) this re-alignment is in progress and planning for repair of the 2-K
interferometer is underway.

On the 2-K, it appears that up to four core optics and MMT3 need magnet regluing and
other work. The small input optics are in various states: 1 OK, 5 will need to be
rehung and 1 needs regluing and rehanging. The3 picture here is changing rapidly
as more testing reveals more problems and people here are planning and mobilizing
resources for the repair of the 2-K.

Work continues on the 4-k PSL installation. The refereence cavity is now aligned with 90% visibility and efforts to lock the frequency should begin early next week. The STACIS system is due to arrive this week and Dennis Ugolini will help with installation next week. The system will be tested on the 2-K PSL.

LIGO Livingston Observatory (LLO) Operations (Coles)

OPTICS: We replaced the Opto Sigma Ultra Stable mounts with Newport Ultima mounts on a specially designed pyramidal shaped base. We re-established alignment and locking of the mode cleaner. We found that the low frequency resonant gain for the MCA actuator was not working so we disconnected it until we can understand what is wrong with it. We have increased the focal lengths of the video camera lenses at ETM and ITMx. We've also centered the optical levers for ITM and ETMx. The ETMx optlev spot at the quad cell was ~2mm dia., so we defocused the beam increasing it to ~5mm. Continued x-arm locking (2/27) to get to a point where the ETM may be used as the actuator. We continued checking the mode cleaner reflected beam. We replaced the half periscope at the bottom of the main periscope with a fixed block mount on the main periscope. We then realigned the input beam to the mode cleaner. We find the beam is clipping coming out of HAM1. We do not know where and this is taking place in vacuum and it should be checked once we open up to air. We also put back in the resonant gain boost on the MCA path of the MC servo. It worked, but now the switch on the amplifier in the MCA path is on normal rather than invert. We do not understand why this changed. (Kovalik/King/Traylor/Kern)

Detector/Technical Support (Whitcomb, Coyne)

Installation& Commissioning: Hanford Livingston	Other Science/EngineeringActivities: Design/Analysis/Fab Issues/Concerns

1.1 LHOINSTALLATION & COMMISSIONING

2km Commissioning

A preliminary assessment of damage to the 2 km interferometer indicates that 4 of the large optics and 2-5 of the input optics have had magnets knocked off. Most of the remaining optics have been misaligned enough to require adjustment. A detailed plan has not yet been formulated, but this process is expected to take about 8 weeks before we can again pump down and resume commissioning activities. We will use this opportunity to install new OSEMs in all the 2 km interferometer.

We are still planning to hold our next engineering run (E3) on March 9-12. The E3 run will precede as planned at the LIGO Livingston Observatory, but it will not be possible to operate any interferometer subsystems at the Hanford Observatory. We will take PEM data for the E3 run for correlation with the LLO data.

Nergis Mavalavala, Rich Abbott, Flavio Nocera
With the goal of reducing the frequency noise in the gw signal, we have begun work to approach the final servo configuration for frequency noise suppression. To optimize the MC length/frequency crossover, we recalibrated the mode cleaner control signals to account for the new servo board
and signal conditioning. We were able to push this crossover down to ~14 Hz and made some changes on the board to go even lower (need to be at ~5 Hz for the ultimate frequency noise suppression). The interferometer continued to lock well with this lowered xover, which is encouraging since we are imposing some low frequency MC noise on the laser frequency when we do this. The PSL frequency stabilization loop, however, tended to oscillate at ~12 kHz more readily with the lower xover. We were unable to test our board changes since the MC was irrecoverably misaligned by the Olympia earthquake.

Bill Kells
Some time spent considereing the issue of possible high coating losses on at least some (ETMx ?) LHO 2k coatings. The recent inference that ETMx central region may be associated with extraordinary high loss is perhaps at odds with the (~year old !) X arm visibility measurement, consistant with ~100ppm/mirror loss. However was the ETMx spot centered for that visibility observation ??
##SEW## as one of the few people who claimed to be able to see the spot, I believe that it was reasonably well centered.

Core Optics

COS

The reticle pattern was reflected from the BS and projected onto the BS pickoff mirror in BSC1; a double reticle pattern was observed, which precluded an accurate determination of the center of the projected image. Subsequently, the COS autocollimator was exchanged with the one with AR coated internal lens elements; a test projection pattern in front of the BS displayed a single reticle pattern. The earthquake incident has necessitated that the three COC mirrors be re-aligned before the COS installation can proceed anew.

1.2LLO INSTALLATION & COMMISSIONING

Commissioning

A large component of the power instability in the cavities has been traced to motions of the final telescope mirror (MMT3) which causes the input beam to jitter relative to the long arm cavity axis. The cause of MMT3's motion is under investigation. One problem is an interaction between the side damping and the position damping.

The NPRO in the laser was replaced and the system realigned. The laser and mode cleaner have been operating reliably since this was done.

Mike Zucker
At LLO this week assisting with achieving stable single cavity locking and the transition to recombined operation. So far we have managed to fix or sidestep all the problems we've found, but there's still much work to do.

E3 Preparations

Shannon will take care of the backup of critical disks periodically during the run to eliminate problems due to disk failure. We also have two extra external disks for prompt replacement. We plan to exercise the RDS software/computer every day until the run to endure durability and working channel list. Ed Daw agreed to supervise the RDS system during the run. Comprehensive documentation were produced for the RDSWriter, which is accessible at http://london.ligo-la.caltech.edu/engrun/RDS_UsersGuide.html

A variation of the E2 LHO channel list serves as the base for the E3 LLO list and it will be finalized by the beginning of next week. The evolved versions of E2 checklists are being adapted to use at LLO. Two training sessions are planned before the run to provide up to date information about LIGO software tools and procedures for visitors and interested on site personnel:
http://blue.ligo-wa.caltech.edu/engrun/E3/info/SoftwareTutor.html

Chethan and Doug is working on the installation of the voltage monitors and fluxgate magnetometers, which are necessary for the E3 power line monitoring project. Calibration activities are planned for before, during and after the run. Detailed information is available about the planned investigations at: http://blue.ligo-wa.caltech.edu/engrun/E3/info/tasks_e3.html

PSL

I have been working with Joe Kovalik and Jonathan Kern on re-aligning the beam into the modecleaner. After swapping out the mirror mounts, it was found that the input beam was partially clipped by the short periscope at the base of the large periscope. This was replaced by a mirror mount mounted to the large periscope. Now the input beam into the modecleaner is clean but the reflected beam from the modecleaner appears to be clipped.

2.0 OtherEngineering and Scientific Activities

2.1 Design/Analysis/Fab

Optics Modelling

1. Investigated why FFT sometimes does not run fast on Mulder (HP V2500). CACR is puzzled too: not promising.
2.Further simulations of the "driven" (by input beam) degenerate cavity. Have some nice comparisons with the simple 2D Modal model (which might be expected to completely break down as degeneracy is approached).
3.The immediate aim of the FFT work is to come to complete (note earlier partial lower bound) description on ifo contrast defect on sensitivity for LIGO II. Goal in sight !
4.Still battling with the MELODY validation. We seem, however, to be in a mode of debugging MELODY. This is useful and R. B. is very thankfull, however we are not converging on OUR original goal of validating the thermal lens effects, and ilucidating the hinted at phenomenon of characteristic SB+/- imbalance in the strongly lensed ifo.

LSC

PSL

A number of 1.00 in. diameter super polished substrates have been ordered from Mindrum Precision.

New OSEM Heads

AstroPak is removing the ZrN coating off of about 50 heads. They should be delivered here 3/7. MacroMetalics has quoted on their TiCr coating and sent samples. One of the two samples does not appear to have the required surface resistivity. This coating is a back-up process to Surmet's ZrN coating technique.

Oliver McCullough is still on track to deliver 105 2k IFO pigtails here by next monday, 3/5.

I will order more osem circuit boards that will position the sensor in a more optimum position relative to the coil. They will have a long lead time; 8-10 weeks or more. As is, we should have enough for the LLO 4k osems.

Peter Fritschel
We are continuing to assemble new osems for the LHO 2km interferometer. A batch of 35 long osems will be complete this week, and the short heads will be started next week. We will be modifying slightly the assembly of the long osems used on the side of the optic to provide greater mechanical clearance.

This mechanical clearance issue has also brought to light the fact that the mechanical design of the osems is less than optimal: the nominal axial position of the magnet with respect to the coil is approximately 1 mm away from the point where the force gradient (due to coil current) is zero. Fortunately, the coupling to the optics platform due to this offset turns out to be very weak compared to the coupling through the pendulum (3-4 orders of magnitude of margin), thus there is no significant performance penalty due to this error.

Modulated OSEMs

Digital Suspensions

GDS

Data Acquisition

40 Meter Interferometer (Weinstein)

Thermal Noise Interferometer (Libbrecht)

Since we are planning to break vacuum soon, this week we summarized the
results of some of our recent experimental investigations in a LIGO
technical note, which can be accessed at

For the simple laser/test cavity configuration, which uses an essentially
free-running laser without the presence of the mode-cleaner nor the
reference cavity for the frequency stabilization, the displacement
sensitivity is 2 x 10^-15 m/rHz @ 1 kHz, and it is naturally limited by
laser frequency noise from 100 Hz (fig. 9 of the LIGO note gives the
sensitivity curve). This modest result reassures us that we are indeed
limited by frequency noise and not by other noise mechanisms at this stage.
The detector noise, including the noise from the electronics, the
photodiode and the mixer, is also shown in this figure and it is only 1
order of magnitude below the sensitivity level. Once the laser source is
replaced (at present the free running laser frequency noise is 2kHz/rHz @
100 Hz) and the mode-cleaner is in operation, the detector noise will have
to be addressed.

The performance of the mirror local controls has also been explored by
simply monitoring the motion of each mirror through the z output of the
OSEM controllers. We learned a few things:
1- the front mirror OSEM is a factor ~20 noisier than the back mirror OSEM;
2- the local controls for the back mirror damp well the suspension
resonances at ~1Hz, clearly visible once the local controls are off;
3- a first estimate of the residual mirror motion is 0.4 microns RMS
integrated up to 200 mHz (which is quite low... perhaps a calibration error)
4- that the motion of the two mirrors is highly correlated, at least up to
4 Hz; then we are limited by the OSEM noise of the front mirror.

The high correlation between the motion of the two masses does not come as
a surprise since they share a common suspension. This 'common mode' motion,
then, could justify the control system's short acquisition times.

Before attacking our mode-cleaner, we began to investigate, for just a
brief moment, the properties of a Pockels cell that was placed in the
chamber. According to the original TNI design, this Pockels cell will be
located on the mode-cleaner output beam and it will be used for the lock of
one of the test cavities once the mode-cleaner is in operation. At present,
it is placed on the beam path to the test cavity and a few measurements
have been done in order to understand its frequency dependence.

We have also been in contact with Klaus Witte, at Resonant Optics Corp. to
investigate an alternate source of NPRO lasers. Resonant's turnaround time
is 3-4 weeks, which is much faster than Lightwave's current estimate of 17
weeks. However, since Lightwave holds patents on some key aspects of the
NPRO design, the Resonant Optics laser does not come with some of the
features of the Lightwave model.

On Thursday, Eric gave a talk about LIGO to students at John Muir High
School in Pasadena.

Finally, we (mainly Eric) have been working on a detailed schedule for
measuring thermal noise in sapphire by the LIGO II material downselect
date, June 4, 2002.

LASTI (Zucker)

Data Analysis and Computing (Lazzarini)

The controlMonitorAPI was integrated with the TapeControl script used to
record LIGO frames written at the sites onto tape. This functionality has
been tested at Hanford without any problems being seen which could result
in loss of data. Testing will continue in preparation for the engineering
run.

In preparation for using LDAS (and in particular the functionality found in
the dataConditionAPI) in the next Engineering run, a list of functionality
required for E3 and submitted it to datacon development group. Also started
writing the E3 data conditioning scripts and driver. The dataConditionAPI
scripts are ~70% complete. Driver script ~30% complete - it is
intended that this will be run at regular time intervals (or looped) and
kick off a number of dataConditionAPI scripts, which will also be spaced to
spread the load out on the site datacon boxes and the LIGO database.

Correspondences were made the LSC members Keith Riles, Peter Fritschel and
Nelson Christensen regarding science goals for E3. I plan to store the PSD
of the IFO, the PSD's of several combinations (differences) of seismic
channels,
and the CSD of the IFO and seismic channels, and summary statistics for the
raw
channels and the PSD's. That way we can calculate coherence functions,
transfer functions, and also identify/correlate some transients.

A modular library was developed this week for use in LDAS APIs which need
to insert data into the LIGO database. This library now has support for the
Process, ProcessParams, SummStatistics and SummSpectrum tables. It has been
used successfully to insert statistics of Frame channels generated by the
dataConditionAPI into the LIGO database in preparation for using LDAS to
carry out investigations in the E3 run.

Minor modifications were made to the frameAPI and dataConditionAPI in
support
for being able to statistically characterize frame channel data and send it
to the metaDataAPI. Also, the code changes supplied by John Zweizig to speed
up the frameCPP library were added. These do not however improve LDAS's
performance. There were several minor bugs were identified in the light-
WeightAPI by on going testing in preparation for the database MDC. These
have
been fixed.

A version of a dynamically loaded shared object was provided by the UWM
group
this week which will be used to test out load balancing in the mpiAPI and
the
wrapperAPI. There is some concern on the UWM side that this dso will need
some
modifications to allow the load balancing to be tested.

Verbal approval from the NSF to proceed with Phase I of the major LDAS
hardware procurements has been received. The winning bid vendors for the
initial order of 28TB of disk and 6000 slots of tape storage have been
notified, and the necessary paperwork started through the Caltech
procurement system.

The new LDAS AMD K7 and Intel P4 test machines at Caltech are now open for
general use (and abuse) by those interested in testing LDAS software on
these relative new computer architectures. They are accessible via LIGO CIT
general computing users accounts as ldas-pctest1.ligo.caltech.edu (K7) and
ldas-pctest3.ligo.caltech.edu (P4).

Tests of the copper-to-fibre network media converters at LHO needed to
support the move of LDAS servers to the assembly building have all passed.

openssh installed and passed initial testing as a candidate replacement
for ssh within LDAS for improved reliability, performance, and security.

Data Analysis Activities (Vicere)
---------------------------------
* Resumed contacts with F.Ricci to set up a "use case" for Grid technologies.
* Keep working on getting acquainted with the LDAS system

* Contacts with M.Barsuglia about the E3 run: unfortunately because of
the earthquake it is likely that he shall have to cancel his participation.

    * Some help requested by Virgo for mechanical simulations.
        I keep supporting the simulation activities in Virgo,
        while training a substitute.

MIT:
(Keith & Larry)
-Larry and I rebuilt sendmail (8.9.3) on Solaris2.8
and did some preliminary testing.
-Went over NIS+, patching, general FAQ and security stuff.
-Had a meeting to discuss long range projects with David
Shoemaker.
-Also, spent some time troubleshooting email irregularities over
NFS mounts.
-Investigated Openssh for Solaris 2.8.
-Resolved a network setup problem on the box to be the new mail server.

Livingston:
-Preparing for the E3 run and the LSC meeting.
-Finally, received a message from Bellsouth and they are still
looking into the LAnet charges by the state for the Observatory
connection to the LIGO WAN.

Hanford:
(Christine)
-Helping out with the NSF.
-Working out issues with the backup system.

CIT:
(Lisa)
- Setup a laptop for the DCC. It is running Win2k. Win2k has some really nice
features in the user interface and the software load went well. I have had some
problems with the USB drivers, but it is on a device-by-device basis.
- Rebuilt a PC for the 3d floor of Bridge.
- Installed a hard disk and Sol 8 on Jay's Ultra 10.
- Finished rebuilding a laptop for Ed Chargois.
- Swapped out Liz' old computer and transferred her files.

(Barbara)
- Finished the perl/cgi script for a Data and Computing Group contacts web
page. I just have to do some polishing.
- Installed a revised web page for the upcoming LSC meeting that includes a
registration form. Added directions to the conference site and payment
information.
- Installed a news article about the earthquake. Created a graphic showing
the lab and the epicenter using one of Ray Sterner's maps.
- Designed and installed a revised LDAS home page for review.
- Updated PubDocs and PAC transparencies page to pick up document folders.

(Suresh)
-This week I mostly spent on testing and establishing Internet Based (MBONE)
video conferencing system. A production version has been set up at the
Millikan's conference room. This technology so adopted has been developed by
Virtual Room Video Conferencing System (VRVS) team at CERN in the sponsorship of
Caltech. A reflector (video server) encapsulating the multicast packet in
unicast packet and tunneling over the Internet has been set up at Lauritsen. For
our system we have used remotely controllable SONY EVI-D30 video camera,
Polycom's SoundStation and WinTV PCI card installed in Dell computer. Presently,
we have video/audio duplex transmission, Desktop sharing and chat services.
Thanks to Philippe Galvez (CERN) and Julian Bunn (HEP) for their cooperation in
resolving some technical issues.
This system will also be presented for use at the Observatories and MIT after a
few more items are checked out on it.
-Enabled a ethernet port in 115 subnet for Virgini and also given him access
to local disk space in his system vega.
-Restored a number of lost directories and lost mail files.
-Set up a new user account for a undergrad student at 40 meter lab.
-Installed MS Office 2000 in Irena's PC.

(Larry)
-Installed the P-card s/w on the general usage PC located on the third floor
of Bridge. Being to install the s/w ourselves and not depending on the CIT group
to install it has helped in a couple of situations.
-Updated more of the e-mail aliases. We still are not getting enough information
when new people arrive so they can be put in the correct groups.
-Worked a number of licensing issues. Suresh should be installing the new Matlab
pkg. and licenses next week.
-Resolving some maintenance contract problems with Marconi. The third party
vendor they used has not been passing on the correct information on the
equipment we have purchased so that is being worked out.
-Still working out some of the logistics for the LSC with various people.
-Working on a few purchases for new equipment.

* Physics Studies
-----------------
Alan Weinstein, Biplab, Hiro and Julien (MIT) had a discussion meeting
on how to use E2E for studies of the upper limit (data analysis) group
for burst sources. Alan and his visiting graduate student started
some preliminary investigations using the Han2k model. A new version of
the model including all thermal and high freq seismic noise will be
supplied to them in about a month (current version of Han2k generates
low freq seismic and shot noises).

Hiro went to LHO to attend the NSF review.
He discussed with Rick about the effect of the motion of the photo
diode detecting the reflected light from the reference cavity on the
frequency noise. Biplab has written the first draft of the paper on
this issue, and we discussed about the content as well.
He also worked with Malik building a simplified IOO model. There is one
version of PSL and IOO model, but they include too many elements and a
very small time step had to be used to make them work. Malik is
building with the minimum elements to make it work with longer time
step. This will be used for the in-lock state noise simulation.

Biplab, Hiro, Matt discussed about future developments related to
LIGO I and II simulation software and Physics issues, ways to make e2e
faster etc..

* The Code
----------
(Tavio) Spent last week reviewing several public-domain mathematics
API's looking for a toolkit that used expression template methodology
to implement clean mathematical expressions without the use of
temporary variables.
The only one available was Blitz++, which I compiled on both our
Solaris machines and on CACR's exemplar V. Unfortunately, Blitz++
consumed almost 150MB of memory during the compilation of a simple test
program, and since Blitz++ is mainly billed as a smart array class, it
was decided that, after studying the Expression Template methodology's
seminal papers, I will attempt to build an expression-template based
Matrix class.

* Alfi
------
(Bruce Sears) Alfi 4 Software Development:
- Wrapped up work on ALGNode reorganization for connections.
- Wrote and conducted tests on Alfi stability regarding ports and
connections. It highlighted a few old problems which I am now
in the process of fixing.

LIGO II/Advanced R&D (Sanders)

Received and set-up a polariscope to be able to detect stress on silicate bonded pieces under load.
Bonded some prisms to 1" fused silica substrates for this purpose.
Made 5 silica/silica bonds with potassium hydroxide solution to be tested for tension.
Made 5 silica/silica bonds with sodium silicate solution also to be tested for tension, and, compare the strength of the bonds with the different solutions.

3.1 Advanced LIGO PSL
I spoke with a sales rep. from Coherent Laser about the pricing of pump
laser diodes and the quotation that was provided to me late last year. I
had enquired as to why the price of each diode was so expensive. The
quotation will be re-worked.

3.2 40m Lab & LASTI
All the optics and mirror mounts for the 40m and LASTI PSLs have been
ordered. A number of adapters have been put into the workshop in the
basement of Lauritsen for fabrication.

Date: Fri, 18 Dec 2037 12:41:19 +0900
From: Riccardo DeSalvo <desalvo@ligo.caltech.edu>

SAS towers safely arrived in Tokyo, started installation details below.
Arrival photos will be updated into:
http://www.ligo.caltech.edu/~takamori/images/tamasas/ as soon as
possible.

Pasadena activities:
Virginio
Got new baby called Henrique (Sunday February 18 at 5:02pm (3.30+-.03)kg
(53.3+-.5)cm.). Virginio is clearly the most productive SAS member.

Edwin
Searched for hysteris into the old GAS filters, found none at slightly
below 500 mHz.
Checked hardness of old wedges, could not identify which one of the
spares was used in the GASF filter (we will do at our return from Japan)
but found three batches one at Rockwell 15, one at 21 and the third at
30 to be compared with the 4 Rockwell of the TAMA SAS wedges
(confirmed). Clearly we need to pay attention to this. It is a pity
that soft wedges went into TAMA SAS. Will change them at first
occasion, will need to check on blade housing hardness in future
models. Writing report.

Lyon activities.
Fred working with Sonic to get some Sapphire sample demonstrators
machined. Some will be done to be tested in his sapphire quality
measurement machine, some as flex joints for suspensions.
INSA students simulating mini-MGAS blades resonances and behavior.

Pisa Activities.
Alessandro.
Finalising thesis. Accelerometers slowed down by machinist sickness
(did not dare replace him with replacement half way into machining),
machining should be ready by Tuesday and EDM in another week, then ready
for shipment.
Preliminary work on low frequency vertical accelerometers for Newtonian
noise diagnostics.

Tokyo Activities.
Monday 19th start working. Original manpower: Riccardo, Akiteru, Kimio,
Yamamoto, Tatsuo, Asso, Kobayashi.
War bulletin: Tatsuo immediately got into bike accident, out of action
for a week. Tuesday Akiteru and Asso out of action for a day and a half
due to intestinal flu, Riccardo followed next day. All three still weak
at end of week.
Kobayashi occasionally available.
Took reception of TAMA SAS towers.
Took reception of vacuum tanks.
Both SAS towers arrived in good order, except for a little of cosmetic
damage to one of the double pendulum top masses that had not been
secured properly for shipment. No functional damage, but if suspensions
had been attached to it, it would have been damaged. Towers entered
standing up into the laboratory with one inch to spare.
Vacuum tanks filthy with machining oil, spent long time in cleaning
them. Note only mechanical vacuum needed in this test. Positioned and
installed vacuum tanks on ground and bolted down tower feet and bellows.

Vacuum tested tower both tanks, checked OK, below 10 mTorr rapidly,
identified leak outside the tank, will take it away easily.
Opened and plugged in TAMA-SAS, needed some shimming to level in tower
two, may put some shimming on tower one but maybe not even necessary.
IP operating very well.
Found that one of the magnetic actuators had been mis-mounted, coil was
touching yoke and under compression instead of floating, may need
rewiring at worst. Filters untested for lack of dummy payload, but look
perfect.
Close both towers and connected link pipe, pumped down reached <50 mTorr
overnight, ready to apen up again.
Started assemblying setup for mirror Suspension assembly. Next week
will start Suspension assembly.
Too bad that we did not have one additional month in Pasadena to
pre-assemble the suspensions and ship complete functional package
otherwise next week we would be aligning mirrors.
Colin Taylor (from Kashiwa) and Tomaru (from KEK) came to see SAS.
Drafted preliminary plan on how to implement SAS in the KEK cryogenic
interferometer. They will come periodically to Hongo to assist Akiteru
and familiarise with system.

Installed two TAMA towers in the 3 m interferometer, precision levelled,
and coarsely aligned and tested under vacuum within 4 days. The
cartridge installation works.
Started assemblying the mirror suspensions from scratch, found some
minor problems, will iron out.
After suspension mechanics debugging the Tokyo team with support from
visitors will install, cable, and align the interferometer. Many weeks
of work still ahead.
During the final testing before shipping, found hysteresis in the main
MGASFs, then traced to soft wedge materials (hysteresis that has been
cross checked not being present in the preceding prototypes) We will
continue testing.
Embarassing show, but the problem should not be harmful because present
only in the main GASFs above the final miniMGAS springs at the head of
the double pendulum.
Preparing design for sapphire flex joint cut demonstrator, hopefully
testing by next meeting.
Visit to KEK and inspection of site in view of installation of seismic
attenuation system for the cryogenic facility.
No major problems found. Will proceed after approvals and certified
results from the 3 m prototypes.
The seismic isolation cryogenic facility will serve to test various
heat extraction schemes and as a testbed for advanced sapphire
suspensions.

Fiber/ribbon research:
We are continuing the study of the nonlinear thermoelastic damping in
the violin modes of our fused silica double suspension. By elevating
the temperature of the suspension to about 120 Centigrade we see Q's of
all five modes at about 1e8. This may be due to the lower loss of SiO2
at that temperature or may be a change in the viscosity of the pork fat
on the upper wire suspension. Either way the apparent cancellation of
the linear thermoelastic damping is even more apparent for the
fundamental since the measured Q went up while linear thermoelasticity
predicts higher losses at higher temperatures.
Silicate bonding research:
We have purchased a polariscope to allow us to monitor the stress
distribution in loaded bonds to check the stress levels and (hopefully)
directly observe incomplete bonding when it occurs. Tensile strengths
of KOH-bonded silica/silica samples are clustering around 1e7Pa now that
we are getting smarter about our tester. This is a factor of three
lower than our measured shear strengths for similar bonds. It is now
apparent that should a fiber break in an advanced LIGO suspension, the
next fiber over and its bonds will have to bear 3x the design static
load for short times while the weight redistributes itself. This eats
up a lot of safety margin and will probably force us to redesign our
ears for larger surface area of bonding.
A first tension test of a silica/sapphire bond gave 8e6Pa tensile
strength before the glue holding the bond in the tester failed, so this
figure is a lower limit. This part was analyzed through the polariscope
during pulling and showed uniform stress, so the bond was uniform.

Analysis of the results obtained for the Q measurements of coated and uncoated samples of sapphire and fused silica, in collaboration with Univ. Glasgow, suggest that we see measurable levels of coating loss in each case. FE modeling has allowed the spatial energy distribution in the modes to be determined and thus the loss of the coating material to be estimated. The loss of the coating applied to Corning 7940 silica by General Optics is ~ 1.7 x 10-4 and the coating applied by REO to the sapphire mass ~ 4 x 10-4. This is in reasonable agreement with the results of parallel experiments carried out by the Syracuse group.
Nakagawa et al (Iowa/Stanford/Syracuse collaboration), have extended their work on modeling the effects of coating loss on the thermal noise of coated test masses to include the case of the coating and substrate having different materials properties.
Interpretation of our measurements using this work suggests that such levels of coating loss will affect the sensitivity of advanced LIGO, and so this needs more study.

Helena visited Stanford to help with installation of an uncoated LIGO I size fused silica blank in the Q measuring apparatus here in preparation for measurements of the loss of coated and uncoated LIGO I size masses.

Jordan visited Stanford to discuss proposed plans for combined R & D on reducing both optical absorption and mechanical dissipation associated with dielectric coatings.

… Milestone for assessing initial impact of coating on test mass Q by 4Q00 is met
… Next milestone is that we should know final Q measurements we expect of coated sapphire/silica substrates by 4Q01.

(B) Mechanical losses due to silica/sapphire bonding
The Q of the bonded sapphire sample, which was previously measured when suspended using a wire sling, was re-measured with the sample suspended from a silica suspension. Using this type of suspension, loss factors for the sapphire of around 10-7 were measured, slightly higher than when the sample was suspended on wires.
Further work in this area is being undertaken; in particular a second sapphire sample has been purchased by GEO and transferred to Stanford where it will have an attachment bonded to it, so that the previous experiment can be repeated in an attempt to gain further understanding of the sources of the excess loss.

(C) An attempt was made to measure the thickness of a silicate bond between sapphire and silica substrates using an interferometric technique. The results suggested the bond thickness was
< approx. 2 microns.

… (C) and (D) come under "test mass mechanical integration", which has a milestone of 4Q02 for characterization of specific techniques

We have been making slow progress with the tests of cross couplings in
our suspension prototype, nothing to report there yet (Amber Stuver,
Aran Glancy, GG).
We have been analyzing E2 data for angular fluctuations: they are LARGE,
up to a few microrad, even 0.5 urad with a few Hz bandwidth. (We had
expected 0.1 urad rms total). it is not clear how much is due to x-cplg
and how much to natural fluctuations, but x-cplgs are large, especially
from length drive into angle. When taking this away with controller
"tuning", the noise at higher frequencies (stack resonances) is
increased. We have been taking data with new OSEM monitors (insensitive
to laser light) installed at LHO, they are considerably better than
previous ones, but small range si posing a problem for initial optics
alignment. (Tiffany Summerscales, GG). There maybe some lessons to be
learned for Advanced LIGO...

SWG monthly progress report for the Syracuse group
submitted 27 Feb 2001 by Peter Saulson
Steve Penn has made some measurements related to the question of excess
loss from superpolished surfaces. He suspended a thin disk of fused silica,
which had been superpolished by Zygo. (These are nominally identical to
the "thin" samples ordered by Gregg Harry for future coating experiments.)
He found the Q of three modes: the two polarizations of the "butterfly"
mode (around 3.0 kHz) and the drumhead mode near 4.5 kHz. The butterfly
modes showed Q's in the range of 3.7 to 4.1 million. The Q of the
drumhead mode was in the range of 5.9 to 6.1 million. These have been
checked in two separate suspensions. Because the Q is so much poorer than
the bulk Q for fused silica, we assume that our losses are due to the
polished surface. Suspension losses appear to be small, both because
changing the suspension does not give significantly different numbers, and
also because the two polarizations of the butterfly mode agree well, even
though they have different alignment with respect to the suspension.
These Q values are up to a factor of 3 higher than the best value
obtained by Gregg Harry on supposedly identical samples. The suspension
technique and measuring technique were similar. We wonder if it is possible
for surface losses to vary so much among nominally identical polishes, and
will ask Zygo for any information they may have that could shed light on
the question.
This higher Q shows that at least some superpolished surfaces can be
substantially better than had previously appeared to be the case. On the
one hand, this is certainly good news. Still, we have to understand what
would be necessary to achieve good surface loss routinely.
Josh Smith and Steve Penn successfully rang down a special sample
designed to test the loss in silicate bonds. The sample, supplied by the
Glasgow group, consists of two half cylinders, bonded together along their
long rectangular faces (with a slight lengthwise offset.) Both polarizations
of the fundamental flexural mode (2.3 kHz and 2.45 kHz) were measured.
The higher mode showed Q consistently at 200,000. The lower mode gave
several measurements spanning the range from 220,000 to 320,000. (Josh is
now investigating whether any amplitude dependence might be responsible for
the spread of values in the latter case.) It appears that loss from the
bond must dominate the Q, and we will check this with an unbonded control
sample of the same material. This should make possible the best determination
yet of the loss in silicate bonds.
In January, Gregg Harry measured the excess loss caused by applying
an REO coating to one of his "thick" samples of fused silica. The result
was consistent with his previous measurement of coating loss on a thin s
lide, and with the measurements made by Glasgow of coating losses on a
full-sized mirror. We are working out with Glasgow and Stanford the best
way to write up the accumulated set of coating results; writing will start
shortly. Gregg is now finishing his move to MIT.
Andri Gretarsson has made his first measurements of the statistics of
thermal noise in a violin mode of his thin tungsten fiber. He has a number
of modes that are visible above the readout noise of his sensor. For the
mode that he studied, the histogram of mode energies shows a noticeable
spread, in excess of what one would expect simply from the statistics of
thermal noise. It is too soon to be sure of the interpretation, although
it looks qualitatively like the results that Braginsky's group saw. If this
is really excess mechanical noise in the fiber, we have to learn its cause,
and especially to study the results in systems more like real LIGO
suspensions.
Before making production measurements, Andri plans to build an extra
layer of feedback into his sensor, to make it less sensitive to laser
power fluctuations.
Scott Kittelberger is testing a new set of readout electronics for
the anelastic aftereffect experiment. The 10 x larger bandwidth should help
with making short-time measurements. The circuit is also being carefully
checked to ensure that it is free of any systematic responses to step inputs
that could mask our real signal.
We were joined by our new postdoc, Alexander Ageev. Sasha got his
degree in Braginsky's group on the "excess violin mode noise" experiment.
He has given us a seminar on that work.
We took delivery of our new large bell jar. The next steps are
installation and vacuum checkout. Once the bell jar is commissioned, we
plan to restart our measurements on Q vs. diameter for fused silica rods.

1 State of the art
Mechanics: The assembly and tuning of the last superattenuator (SA)
has been completed in December 2000. All towers are under vacuum.

Inertial damping: implemented on all towers in a robust and simple
conguration (g.1). The implementation of such a control from scratch
does not require more than a day thanks to the automatic diagonalization
procedure set by Lee Holloway.

Diagnostics: A complete slow monitoring of all the suspension sensors
(including the vertical LVDT mounted on each filter) is being imple-
mented, meant to be a complete diagnostics setup (G.Ballardin, S.Braccini,
V.Dattilo, F.Frasconi, R.Passaquieti).

Non linear alignment: filters for the mirror alignment from the
marionetta using a CCD camera as a sensor have been designed and
implemented by M.Barsuglia, E.Majorana and P.Puppo. The control works
nicely. An rms angular displacement of the order of 1 microrad has been mea-
sured in both d.o.f.. There is no evidence of interaction with the inertial
damping loops.

Wideband mirror control: A small table interferometer is being used
to measure the force-to-displacement angular and translational TF from
the marionetta and the recoil mass. The interferometer signal has been
used to control the mirror in translation from the marionetta (L.Di Fiore
and D.Passuello). Results are preliminary but very promising.

2 Comments and remarks
After the long and tough assembly and integration phase we are finally moving
towards a systematic study of the suspensions behavior. Several questions are
still pending and the final answers will come from the work on the central
interferometer (starting right now). But the general feeling is that the system
is good. The hottest point is perhaps the controllability of the mirror from
the marionetta. The results obtained on the SA prototype in air [1] and the
preliminary results on site are quite satisfactory.

References
[1] A.Di Virgilio, G.Losurdo, D.Passuello, B.Slagmolen, R.Cecchi, Two d.o.f.
control of the SA test mass, VIR-NOT-PIS-1390-167, VIRGO intenal re-
port.

· Isolation and local control of the 6 double pendulum modecleaner
suspensions in GEO 600 (3 mirrors per modecleaner, each of round trip
length 8 m) are such that peak-to-peak motion in a mode-cleaner is ~200
nm. Local control systems for the two triple pendulum suspensions (with
test optics on wires) used in the recent 1200 m single arm test
interferometer experiments are well behaved with good damping. The
electrostatic feedback system, operating between the test mass of one
triple pendulum and the test mass of its reaction chain, was tested in
situ and the force produced agreed well with prediction for magnitude of
force as a function of frequency.
· Modelling work on suitable suspension designs for advanced LIGO
continues. In particular the design of a first quadruple suspension (all
metal), with masses (from top to bottom) of 15, 15, 15, 30 kg has been
essentially carried out, and Autocad drawings for the various parts are
underway and almost complete. Procurement of parts has started.
The intention is that initial assembly of this quad. will take place in
Glasgow over the next few months, and then in summer the parts will be
sent to MIT for assembly and further testing. This is in line with the
suspensions development schedule as presented to the NSF review in
January.
· A fibre pulling machine has been installed in Hannover in preparation
for assembly of two full scale monolithic silica suspensions in GEO.
Also a suitable jig to hold the parts for assembly of the fused silica
intermediate mass and test mass has been constructed. The fibre pulling
machine and jig have been successfully used to make suitable fibres and
then assemble and suspend for the first time a GEO-sized silica mass and
upper mass. The target vertical frequency for the test mass on its
silica suspensions is 18.6 Hz. The fibres used yielded frequencies when
measured separately before assembly of 19.8, 19.4, 19.4, 19.8 Hz, a very
encouraging set of results - the frequencies giving information on
consistency of effective diameter of fibres. Unfortunately assembly of
actual GEO monolithic suspensions with mirrors has not yet taken place
due to delay in coating our mirrors by REO. Delivery is still awaited.
Note that the timing of the first fused silica GEO 600 mirror suspension
has slipped from the milestone (1Q01) as presented in NSF review paper
due to the delay in coating.
b) Materials research relevant to Advanced LIGO and beyond.
· Further measurements on coated and uncoated silica and sapphire masses
have been carried out in collaboration with Stanford and are
demonstrating measurable coating losses. FE modelling has allowed the
spatial energy distribution in the modes to be determined and thus the
loss of the coating material to be estimated. The losses of the coatings
applied to Corning 7940 silica by General Optics are ~ 1.7 x 10-4 and
those applied by REO to the sapphire mass are ~ 4 x 10-4. Interpretation
of these results for advanced LIGO, using the analysis of Nakagawa et al
(Iowa/Stanford/Syracuse collaboration), suggest that such levels of
coating loss will affect the sensitivity of advanced LIGO.
The milestone for finalising measurements fused silica and sapphire Q
with coatings as presented at NSF review is 4Q01. It is not clear
whether this target can be met, given that procurement of samples and
coatings has quite a long lead time, and reliable results require
repeated measurements.
· Further measurements of mechanical loss associated with
sapphire/silica bonding have been carried out in collaboration with
Stanford. Previously a bonded sapphire sample was suspended using a wire
sling, recent experiments have used a monolithic silica suspension
welded to the bonded attachment. Using this type of suspension loss
factors for the sapphire of around 10-7 were found, slightly higher than
when the sample was suspended on wires. Further work in this area is
being undertaken.
· Ears have been bonded to three intermediate masses for GEO 600 and
de-bonding and re-bonding of an ear on one of these masses has been
successfully demonstrated.
· An improved cleaning technique for silica before bonding has been
developed based on the use of cerium oxide followed by sodium
bicarbonate before washing
· In collaboration with Stanford an attempt was made to measure the
thickness of a silicate bond between sapphire and silica substrates
using an interferometric technique. The results suggested the bond
thickness was < approx 2 microns.
All of the above come under the "test mass mechanical integration",
which has a milestone of 4Q02 for characterisation of specific
techniques

· A second new fibre/ribbon pulling machine has been completed for use
in Glasgow, the first new machine having been exported to Hannover.
Downselection of ribbon vs. fibre was to be at 1Q01. Clearly this has
slipped. The end of 4Q01 is a more realistic goal.

Most parts for the special-purpose Cleanroom to allow the cartridge nstallation of the BSC-testmass isolation/suspension system have been received. Preparations for the BSC support tube installation are under way with large pieces of equipment shipped up from Livingston; folks from there will be up in the Spring to help with the actual installation.

The laser table is in place; kits of parts for the optics and electronics for the Pre-Stabilized Laser (of LIGO I design) will arrive in March. Small Optic Suspension parts arrived and are being assembled for the initial 'toy' cavity on the LIGO I HAM stack.

Planning for the visit of Calum Torrie and Norna Robertston in July is underway; an early prototype of a quad Advanced LIGO Pendulum from Glasgow will be installed in the vacuum tank now holding the SEI Rapid Prototype and tested for functionality and then studied for an optimization of the actuator hierarchy.

Gregg Harry is just getting installed and gathering equipment to set up a vacuum chamber to permit Coating Loss measurements (continuation of work started in Syracuse, in parallel with the Glasgow/Stanford work).

Starting to look at the optical design for the LASTI Test Mass cavity with Nergis Mavalvala: how large a beam spot is practical?

Stanford activities (Brian, Corwin, Hua, Graham, Jamie N., Dan):
Hydraulics:
Two actuators are now in place on the hydraulic test stand. We are
measuring performance of the vertical system. We have closed loops and
reduced the transmission of ground motion to the test mass by a factor of
10 almost down to 0.1 Hz (limited at low frequencies by the inability of
the witness seismometer to measure the results) We are using feedback to a
small geophone on the test mass to improve the performance from 5 - 30 Hz,
and see isolation factors of a few around 10 Hz.
Active Platform:
At 10 Hz we now measure transmissions of 1.0e-2 vertically and 1.5e-2
horizontally, measured between the ground and witness seismometers on the
platform.
Modeling:
Hua has duplicated the mechanical cross coupling seen at the rapid
prototype by implementing more realistic blade springs and horizontal
seismometers. These tilt-horizontal couplings lead to zeros in the measured
open-loop plant response at frequencies around 0.15 Hz.
We are designing controllers at Stanford based on the model to control the
system at MIT. This is an interesting test of modeling and long distance
teamwork.
MIT activities:
Rana, Jamie R., and Richard continue work on two-stage pre-prototype, to gather final data before mechanical upgrade, described below.
JILA activities:
Giles finalizing design for inter-stage flexure upgrade, which we expect will greatly reduce the coupling of horizontal motion to pitch and yaw, allowing beneficial flexibility in the controller design. This will be fabricated this month, and installed by JILA at MIT.

Prototype preparations:
The bids are back from venders on mechanical large structures, and are being evaluated (Gerry, Joe, et al); the range of prices is not crazy. The forcers and large geophones have been ordered (Marcel, Giles). Systematic survey of all parts needed is underway (Brian, Joe). Modifications to seismometers to allow remote locking in design/prototyping stage (Szabi, Marcel).

CR-000018	WBS 1.1.4	Curbing for Service Roads at Livingston	G. Stapfer
CR-000019	WBS 1.2	Additional Lab Equipment	D. Coyne
CR-000020	WBS 1.1.4	Staging Building and Renovations to Existing Building--Livigston	F. Asiri
CR-010001	WBS 1.1.4	Return of Unused Construction Budget To Contingency	F. Asiri

Weekly Report for Week Ending March 1, 2001

Special Announcements: