|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday June 18, 2001 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items:
no report
WBS 1.2 LIGO Operations--Administration
Due to the convening of the PAC meeting, the site teleconference scheduled for Thursday, June 14, is cancelled. The next teleconference will be held on Thursday, June 21, which is the third Thursday of the month, when the Financial Tracking Report is scheduled to be discussed.
The list of current actions revised to reflect
open actions assigned through May 31, 2001 may be found at
ACTION
LIST.
From: Ed Chargois <chargois_e@ligo.caltech.edu>
>From: Cleveland Mak <mak_c@ligo.caltech.edu>
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. . .
ACCOMPLISHMENTS
From: Esther Cunningham <esther@ligo.caltech.edu>
Press here for ACCOUNTS PAYABLE HISTORY DATA.
From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
From: irena@ligo.caltech.edu (Irena Petrac)
SUPPORT (Wood)
On vacation.
Irene Baldon
Progress Period from 06.08 to 06.14
Accomplishments:
WBS 1.4.1.2 Project Controls (LIGO Construction)
The end-of-May Quarterly Progress Report is due. I have asked for contributions by Friday, June 22, 2001. Again, please keep it brief and report only significant accomplishments (not in progress stuff).
The following Change Request has
been submitted:
| CR-010006 | WBS 1.1.4 (OPS) | Roof Shelter for Concrete Pad Adjacent to Erosion Control Pond (Livingston) | M. Coles |
Press for the latest Contingency Needs Projection.
From: Kris Duncan <kris@ligo.caltech.edu>
From: Ed Jasnow <jasnow@ligo.caltech.edu>
No report this week.
Seismic Systems
---------------------
H.Radkins, G.Moreno, M.Guenther, C.Gray, F.Raab
Coarse Actuation System
While HR was away at MIT, GM & MG, installed the translation stages
on the
piers. We have centered the system and will now exercise the
system.
Fine Actuation System
The PZTs will be put back into service when non disruptive to commissioning
activities except for the Yend which will be interleaved into the COC
installation alignment activities.
Eucon Mining Studies
Working with FR to quantify the future effects from the mining activities
on LIGO if they move closer or change mining equipment. Using
the STS-2
seismometer on site, there is no detectable change in the seismic activity
when the mine is active or shut down. The location surveyed about
10km
closer to the mine(~16km from LIGO) shows definite noise at mine generated
frequencies. Attenuations based on this location suggest LIGO
is OK for now.
ISC
----
H.Radkins, G.Moreno, B.Rivera
The 4k RM Optical Lever receiver has a very sharp angle coming from
the
chamber port and the first attempt to fab a viewport cover pipe from
existing parts came up short. We'll get it shortly. The
receiver of the
4k ITMx should get enclosed today or Friday.
H.Radkins continues to eek out bits of time for E3 & E4 LLO arm
control
comparison to the predicted tides.
We brokedown the IAS Survey Platform and shipped this a few other items
to
LLO in support of their alignment activities.
LASTI
--------
H.Radkins, C.Gray
We traveled to MIT the last couple weeks for buildup of the LASTI system
BSC SEI and the last HAM SEI. We did get all planned accomplished.
The
pump down was underway when we left and there are reports of a leak
or two
but they have tools and the talent now. We had a great time being
with the
folks there and hope to help again.
ISC Table Work
---------------------
C. Gray, H.Radkins)
The optical components for ISCT4 were installed/aligned (minus various
items such as several photodiodes, wavefront sensor, and an optical
spectrum analyzer). The table (and it's enclosure are roughly located
at
HAM4. The next table to be worked on is ISCT1.
Detector Installation: All operators are now in the midst of door and spool removal, optics, and electronics repairs as planned during the vent. The West crane had broken a limit switch that was replaced and the crane is now operational. 2 Met1 particle counters back from repair, all areas requiring a particle counter are now installed and operational. The particle counter air pump calibration stand is now operational and all particle counters will be checked for calibration soon. OSEM replacement is now in progress. (Rich Riesen)
Doug Cook and Betsy Weaver are at LLO this week and will stay into next week. The mode cleaner optics are being reworked by Betsy with the assistance of Ken Watts and Harry Overmeier. Upon removal of MC1-3 from HAM1&2, we observed several features that may be responsible for the unexpectedly low Q values of some of the suspension modes that were measured in vacuum. In addition to the untrimmed wire pig-tail on MC2, we observed excessive glue in wire stand-offs in MC1 and MC3 such that the glue appeared to support some of the mechanical load of the suspension. We also observed discolorations in the suspension wire, and on MC2 we observed the lower left and right PAM magnets were glued onto the beveled area of the optic a "fillet" of glue underneath the magnet was supporting in perpendicular to the optic surface. The clamp plate on MC1 was also installed upside down so that it was not flush with the mating cross member on the suspension tower. Also, one of the OSEMs was constructed upside down so that the pinout orientation was opposite to all of the others, resulting in the LED and detector locations being interchanged in order to place the pinout in the right orientation.
We are concerned about the level of the particulate contamination that we observe within the chambers, in areas that were clean at the time the doors were initially installed. The source has still not been identified.
We have set up a bell jar on the granite assembly table in the vacuum prep lab so that the rehung small optics suspension Q values can be remeasured independently of the reaction losses of the seismic table prior to installation as a way of pedigreeing the work.
As of Thursday afternoon, ITMx has been refit with new OSEMs and realigned by Gary Traylor, Harry Overmeier, and Ken Mason. The beam splitter has been removed. It shows evidence of at least one, and possibly two, additional blemishes (there was one known blemish which occurred during coating). It will be replaced with a spare optic.
GC: Installed a new UPS in the communications room. This replaced a
non rack mount unit with a rack mount.
Installed a 16 port KVM in the communications room for the various
computers that are in there. (9 total) Installing the software for
the blade 100 which will be used as a backup server for the Exabyte library.
LDAS: Finally got the beowulf up and running. I used the system image utility from va linux to install the 16 nodes. (Shannon Roddy)
PEM: Installed the spare seismometer at the Y-end station. Also installed
a microphone in a wall mounted location at each end station. Both the microphones
and the seismometer were installed for student experiments. (Doug Lormand)
Other: The West end forklift had a bad hydraulic leak that has now been repaired. Rebuild and installation of the end station fan motors is nearly complete.
Forty-two teachers from Livingston Parish visited and toured LIGO today
as part of the InTech Teacher In-Service Program. Warren Johnson gave them
a lecture and tour of the facility. This week marked the arrival of our
SURF students and the beginning of our summer lecture series.
| Installation&
Commissioning:
Livingston |
Other Science/EngineeringActivities:
Issues/Concerns |
See also the Installation web page
The mode cleaner will lock stably at all input powers up to the maximum available of about 6.5 Watt. Stable operation requires sufficient servo gain to suppress pendulum frequency oscillations that are driven by radiation pressure. We have not, however, been able to acquire lock with the input power at 5 W or greater. Instead one can lock it at lower power (~1 W) then raise the input power.
As part of the high power operation, we have implemented the electro-optic shutter in the MC reflected beam, which protects the locking photodetector when the MC is not locked. The shutter appears to function well; it switches from low transmission to high transmission after the MC has been locked for ~0.25 sec, then will switch back to low transmission quickly if the MC loses lock.
We measured the change in the MC's length control signal (dc value) as the input power was increased. The control signal should be compensating for the increasing radiation pressure on the three mirrors. The control signal increase is seen to be several times higher (~5x) than we expect based on the radiation pressure -- still scratching our heads over this one.
At higher power we found it more likely that the MC would lock on the small horizontal polarization component (which has a 10x lower finesse). We tweaked the input polarization (by ~1 deg) to reduce this wrong polarization component.
We measured the coupling of the 1 micron laser light into the new osem
sensors, with the technique used on the old sensors of modulating the laser
beam power and measuring how that modulation shows up on the sensors. This
was done on the MC mirrors and the recycling mirror. We did see some pickup
on some of the MC sensors; in the worst case, the pickup was about a factor
of 300 smaller than the corresponing worst case with the old sensors. More
typically the pickup was 2-3 times smaller, or was not even visible at
a ~5x lower level. On the recycling mirror we could not see any pickup,
but this upper limit only says that it is at least ~10x smaller than in
the old sensors.
We then moved on to the 4k system. Here we noticed that the 4k MC locks easily to the laser but the beam was being partially clipped by the EOM on IOT1. We then proceeded in
We then turned the LSC and ASC modulations off and we measured the MC
visibility. We found
visibility = (Pantiresonating
- Presonating) / Pantiresonating = ~ 90%
(see elog 6/13 for details).
The pendulum Q will be measured initially by looking at the losses in the violin modes without the recoil problems of the pendulum. We are in a position to measure the pendulum modes in vacuum on a low recoil table for the small masses. Need to be careful that the vacuum system is not a source of contamination.
The dust is on all horizontal surfaces which collect by gravity driven precipitation. It is easiest to see on the black glass surfaces where the particle count can exceed 10 particles per square inch. At the minimum the density is a particle per square inch. The dielectric surfaces may also have electrostatic driven precipitation near their surfaces. Vertical surfaces are at least 30 times less populated. The dust is much worse (10 to 100) than that seen at Hanford after opening.
Sources of the dust, none of them proved:
The vent lines used in the vent and purge: particle count of the flowing lines is close to zero but can be increased to over 10k counts by shaking some of the bellows.The doors for the chambers are stored in places where there is a good bit of traffic. The doors are placed on plastic pallets, and covered by a cloth gauze to pass no larger than 100 micron particles. Most of the dust seen in the chambers is smaller than 100 microns and could have collected on the doors after the months of storage last summer. The doors were not wiped down before being mounted.
Material of the dust: most likely fluff from the paper and cloth shoes, samples are being gathered at the moment. Prior samples of debris from the chamber floors included aluminium, G10 and Viton. The only material that matters is the dust that could be raised during pump down or venting.
Steps to be taken to avoid redistribution
of dust after cleaning of the optics: We will install witness pieces of
horizontal black glass on easily visible places to be viewed during the
pump down. We will slow the pumpdown by using smaller pumps from 3 hours
to about a day to reduce the chance of lifting dust.
Steps to be taken to clean up the dust: Not clear what to do. We will wipe down accessible horizontal surfaces and clean the floor of the chambers. Thinking about stirring up the air in the chambers and filtering it with HEPA filters or with an electrostatic precipitator. The difficulties are that the system needs to be compatible with our requirement to not insert hydrocarbons.
Difficulties in cleaning the optics: The Carbon dioxide cleaning system is not functioning as expected. There may be a film on the optic or contamination in the gas itself. This is being investigated with witness surfaces and the surface of the damaged beam splitter.
Visitors: Guido Mueller, Stacy Wise, and Michael Hannawald (visitor) will be at LLO for a week, and will likely start OSEM replacement work.
I have been checking the signals coming to/from the temperature sensor box on the PSL table. The EPICS channels need to have their EGUFs and EGULs adjusted as the current values result in readings that are off by a few degrees. The change was probably the result of the EPICS records database being copied from LHO because of the CDS hard disk crash a few months ago.
Currently the 10-W laser is being re-aligned. After approximately 14000 hours on the amplifier, the output power is down to 5.7 W.
Snowmass2001: We are working on the preliminary schedule of the P4.6 (Gravitational Radiation) working group. I prepared the groups page.
The puzzle issue is: why this difference (since the sidebands are not sensing the arm cavity because of the anti-resonance condition)?
In order to find out what is going on in the transition from the "isolated" to the "coupled" situation, different tools have been used.
The Modal Model shows the resonant peak is splitted in several peaks because of the coupling. Those peaks become thinner far from the resonant frequency and the presence of a perturbation affects the two sidebands in a different way. For the ideal case analytical calculations are possible and I was able to indipendently confirm Bill's results.
Moreover there is a shift if the locking is not on the right fringe. I spent some time to understand this result since it was not clear to me but now I have a simple model in my mind: since the cavities have their own resonant frequencies, if I want that a frequency is resonant both for the starting situation "decoupled" and for the situation "coupled", that must be the smallest common frequency for the two systems (otherwise in the transition a shift is shown and I saw that with Mathematica).
The particular Minco process controller used for regulating the temperature of the reference cavity, is apparently no longer made. An alternate part will have to be looked for.
Various optics and hardware items were shipped off to LASTI. I have had a few discussions with David Ottaway about what all the various parts are and what they're for.
Rick Karwoski
40m: PSL Electronics installed and undergoing integration. In
process: verification of signals around the frequency stab loop in preparation
for lock.
Intensity Stabilization Servo: Preparations in process for our trip to Hanford. Objective: Evaluation of the current electronics design with a Mode Cleaner
LASTI: Gathering up electronics, and installation materials for the LASTI site. Objective: Shipment ASAP to MIT in preparation for our upcoming PSL installation trip.
Sander Liu
In the process of putting together a set of PMC Servo and PSL Frequency
Reference modules for LASTI. The followings are our current status:
PMC Servo Module
35.3 MHz RFPD work in progressPSL Freq Ref Module
35.5 MHz Freq Ref completed
PMC Servo completed
21.5 MHz RFPD work in progress
21.5 MHz Freq Ref work in progress
Freq Stab Servo completed
80 MHz VCO completed
I will be soon testing the two more of my new MC servo boards. They have been stuffed by Dave.
I have also written a byte swapper to be used on raw frames in the Data
pusher. The hope was that by swapping the input data bytes in the data
pusher on linux nodes, the task would be offloaded from the monitors resulting
in faster monitor execution times. A preliminary test on alvar shows very
little gain in monitor speed.
This week we succeeded in stretching our arm-cavity
lock time from 10-20
milliseconds to 10-20 seconds. The
lock is reproducible late at night,
when ground noise is low, and typical hold times are about 10 seconds.
The longest hold time we observed was 25 seconds, and acquisition occurs
about once every two minutes. We achieved this by increasing the
low-frequency gain and moving the zero of our lead compensator up to
1kHz, from 100Hz. This preserves the relatively low bandwidth of our
servo, necessary to keep from exciting mechanical resonances, at the
expense of phase margin. There are a number of oscillations in the
error
signal during lock, but none of them appear to be at our estimated
unity-gain frequency of 3kHz, indicating this loss of phase margin
is
not a problem. We are currently investigating these oscillations in
an
attempt to further refine the lock.
LASTI (MacInnis, Gray, Laliberty, Miller, Radkins, Shoemaker, Rollins,
Mittleman, Harry, Ottaway, Mason, Mason, Barker, Zucker):
HAM seismic installation: Installation of the HAM bellows and feedthrus
is complete with no leaks found
BSC seismic installation: Installation of the BSC system is complete.
A
small leak was found in 1 outer flange and will be resolved next week
when people return from travel.
MZ comment: Special thanks to Rich Reisen and
others from Livingston who
supplied all the tools, fixtures, and advice
on this installation.
PSL: Dave O. and Myron began the assembly and leak testing of the
reference cavity vacuum can. Jamie R. has nearly completed the layout
for the LASTI PSL table.
Schedule: David S. went through the overall schedule with members of
the
team and collected some good comments which will be fed back into the
planning process. He has also defined cable and feedthrough requirements
for LASTI so that it can be included in the 40 M purchase order.
Hydraulics: Dan M. has updated the Hydraulic drawings which show the
integration of Stanford's Hydraulic actuators to the HAM seismic
isolation system. These updates were the result of a design review
held
by the active isolation team.
* Simulation and Modeling (Bhawal)
Weekly Physics meeting
----------------------
In this week's Physics meeting Andrea Vicere talked about Burg algorithm
for linear predictor and his related work on issues of (i) fast simulation
of closed loops with high freq poles and zeros (ii) getting a good
PSD
estimator.(Minutes will be available in http://www.ligo.caltech.edu/~e2e).
Mode Mismatch
-------------
(Biplab & Hiro) A problem was encountered while studying the mode
mismatch
problem when the waist was shifted out of a small-length (as compared
to
Rayleigh range) FP or recycling cavity. This was traced out to a convention
problem in between radii of curvature of mirror and phasefront. Biplab
corrected the code and validated several different cases with both
primitive
and summation cavity runs. All studied cases agreed with theory quite
well.
PSD Estimator
-------------
(Andrea) Still working on coding the linear prediction in the E2E for
the
spectral estimation.
Quadrupolar precision
---------------------
(Melody) Downloaded a double-double and quad-double precision C++ math
package from NERSC (but developed in Lawrence Berkeley Lab). Successfully
compiled and ran the sample programs using gcc 2.95.3. Gave some
performance timings to Hiro for double vs. double-double vs. long double
precisions.
Data analysis
--------------
Joe Romano has been contacting with Hiro and Andrea discussing about
the use of e2e simulation for the stochastic UL group data simulation.
Alfi
----
There was a meeting discussing about the new alfi architecture. That
discussion involved Albert, Phil, Ed, Bruce, Melody and Hiro. After the
discussion comparing JAVA and Tck/Tk and some other choices, it was concluded
that JAVA is an appropriate language to use. But before the full development
starts, two issues (1) the long time of JAVA (if it will likely to survive
more than 10 years) and (2) standard, or lacking of (there is no ANSI kind
of standard, SUN dictates), will be pursued and, a design requirement document
needs to be written.
(Bruce)
- Reviewed possible Java libraries for use in new implementation of
Alfi.
- Created class document summarizing the current implementation of
Alfi
in C++.
- Creating similar document as model of new Alfi implementation.
- Discussions with end users about how Alfi is actually being used
at this
point in order to avoid implementing currently unused features
which may
(or may not) decrease the overall stability of the current implementation,
and also to slim down the code by removing obsolete parts of
the Alfi
system.
(Ed Maros) Started work on causing Alfi to refresh the screen when
changes that affect the look of the object happen.
(Melody)
- Learning the code, usage, and the PRs written against Alfi
- Received an evaluation copy of a Java graphics library called JLoox.
This product serves the new Alfi objectives, however, another
library is
less costly and provides the the source code. JLoox costs
$5K for a
single developer license and does not provide the source code.
- Looked at the JGo Java graphics library and figured out how to extend
the \
classes to fit the needs of the new alfi system.
====================================================================
* LIGO Data Analysis System
Software (Blackburn)
Prepared a pre-release tagged version of LDAS in the cvs repository
(with
tag name PAC10_DEMO) which identified the versions of all the LDAS
software
used in the demonstration of LDAS to the PAC.
This week we spent most of the time testing the system. We ran
through the
complete suite of tests at least once (and individual tests multiple
times) on
three sites. The performance numbers are within reasonable limits
compared to
the 0.0.17 release (5/15). The one exception is the numbers for
the socket
tests which consistently show a difference of ~15% slower. The
simulated DMT
test passed. We successfully ran through selected inspiral MDC
tests.
Although, for some reason the 04 section of the InspiralDSO tests fail
with a
Rhosq threshold error (and I don't know if they passed during the MDC).
We also
ran the DC MDC API tests without crashing the dcAPI. The frame
extraction
tests are performing similarly as they did before, with more tests
passing on
ldas-test than the other sites.
Working on Problem report to edit resources on client. Added resource
display
on server for eventmonAPI, metadataAPI. Working with Greg Mendell on
tape
control changes: generic proc getPid fails at hanford to get the tape
control
program id so got a modified version into cntlmonAPI to get pid. Modifiying
code as per discussion with Greg on stop controller and monitoring.
Check out
controlmon functions to make sure they work for demo, especially load,
log
filter, beowulf login and watch. Added support for running client via
/ldas/ldas-0.0/bin as well as local. Added email to notify Ed when
there
serious error on metadata.
In the eventMonitorAPI, fixed documentation for classes methods: need
to prepend
a namespace to all methods within which class is defined in order
for the
documentation to be built. Removed exception specifications from all
class
methods to get rid of possibility to have unexpected_exception being
raised by
API. Working on conversion multiDimData structure to the frameAPI expected
format.
In the wrapperAPI made various fixes such as added SearchParams* argument
to
the conditionData(); Allocate memory for inPut structure on searchSlaves
if
dataDistributor=searchMaster before calling conditionData(); Throw
exception
if more than one 'primary' sequence is detected in inPut data. Kept
track of
how many sequences wrapperMaster distributed in inPut data ===> when
freeing
the memory for the inPut make sure to delete that initial number of
sequences
only. FinalizeSearch() gets called before freeing the memory for the
inPut
structure now.
Did several LDAS build code maintenance and installations. The frameCPP
library was modified to include new frame reader code that allows efficitent
input of multi-second and multi-frame frame files based on the table
of
contents. odified the frame builder code to use this new framecpp reader
class and I am currently testing at Hanford fb1, frame builder test
machine.
Removed most vwaits from the frame, datacond, and mpiAPI's, removing
at
least 75% of the system overhead for a user command. Added support
for the
"Primary" data object to the dataConditionAPI. And fixed a memory lead
in
the dbacess library.
---------------------------------------------------------------------
Hardware (Anderson)
Report on LDAS activities at Caltech:
I. Testing and benchmarking of the LDAS Disk storage is continuing to
investigate disk performance under an I/O load that simulates the
expected real-time data acquisition requirements.
II. A survey of workstation memory configurations has been performed
and a plan to upgrade all development machines to a minimum of 512MB
of memory per processor has been created.
Report on LDAS activities at Hanford:
I. Discussed with Otto plans to enclose the LDAS area in the staging
building, add fire suppression, and provide better temperature and
humidity regulation.
II. Scripts that monitor the ambient temperature measured inside the
Sun
E450 boxes have been running at LHO and LLO for the last two weeks.
The
scripts send email whenever the temperature exceeds 80 F, changes by
more than 10 F in any 60 minute period, or by more than 3 F in any
10
minute period. So far no emails have been received. URLs
for output of
the scripts are:
http://www.ldas.ligo-wa.caltech.edu/ldas_outgoing/tapecontrol/logs/dataservertemps.txt
http://www.ldas.ligo-la.caltech.edu/ldas_outgoing/tapecontrol/logs/dataservertemps.txt
III. Installation of hardware for the upper limits run is complete at
LHO. Testing of the T3s and LDAS on this hardware continues.
The
hardware is:
dataserver = Sun Enterprise 450: Serves ldcg tools and T3 disk
farm to
ldas network. Manages ldas APIs.
metaserver = Sun Enterprise 450: Manages DB2 meta database.
datacon = Dell 6400 quad box: runs data conditioning API.
beowulf server = Dell 2400 dual box: manages beowulf cluster.
admin = Sun Blade 100: manages T3 Storage Area Network (SAN) and system
backups of file systems on to Sun D1000.
controlmon = Sun Ultra 10: Provides graphical interface to monitor ldas
from the control room.
tapecontrol = Sun Ultra 10: Writes data to tape (currently this
is
sunbox1) for transfer to Caltech
Three onsite user machines for visitors to develop/test ldas and search
code:
linuxbox1, sunbox1, plus lynx on the GC network: see item IV.
14.5 TB SUN T3 Disk Farm (Two Hundred Sixteen 73 GB Hard Drives)
Will hold several weeks of online data.
Meta database storage is on a separate T3.
Beowulf Cluster:
16 Dell Precision 330 Pentium 4s.
1.5 TB Tape Library System:
Tape Robot and AIT Tape Drives
Data Rate: 2 tapes 24/7 at 6+ MB/s
Networks:
Gigabit connection to framebuilder, 100BaseT martian network to ldas
and
dmt boxes,
100BaseT beowulf network, 10BaseT SAN network.
UPS:
2 APC 2200s and 4 APC 5000.
IV. Set up a new GC machine, called lynx. It is the computer
just to
the right of the door when you enter the computer users room. (The
box
is a Sun Ultra 10 running Solaris 8.) Over the next few weeks I will
be
setting up LDAS, LAL, and LALWrapper software on it (the software needed
to develop searchcode and maybe monitors). The idea is that preference
for use this box will be given to visitors in need of that software,
but
otherwise anyone in need of a GC machine can make use of it.
V. Most of the work need to perform backups of file systems on the ldas
network are completed. Plans for backing up the live meta database
are
still under discussion.
VI. Significant progress on writing code to search for known pulsars
was
made at last week's ldas camp.
Report on LDAS activities at Livingston:
I. The datacon and beowulf servers as well as 16 Beowulf nodes have
passed the installation stage, i.e., network connection and operating
system installation.
---------------------------------------------------------------------
Data Analysis Activities
[A.Vicere]
* Working with Dhurandhar for an application to the NSF
for network analysis studies.
* Exchanging emails and results with Archana Pai (a collaborator of
Dhurandhar) on issues related to the computation of the factors
for combining different interferometer outputs.
* Prepared a few slides for the PAC meeting.
====================================================================
* General Computing (Wallace)
MIT:
(Keith)
-Setup auto-backup of cvs repository using rsync/cron
-Added summer student user accounts
-Installing os's on new laptop
-Fixed communication problem with file server raid array
-Switched ethernet card from quad to dual ldas machine
-Looking into wireless setup
Livingston:
(Shannon)
-Installed a new UPS in the communications room. This replaced
a
non rack mount unit with a rack mount.
Installed a 16 port KVM in the communications room for the
various computers that are in there. (9 total)
Installing the software for the blade 100 which will be used
as a backup server for the exabyte library.
LDAS: Finally got the beowulf up and running. I used the system
image
utility from va linux to install the 16 nodes.
(Tom)
-Many of our summer students and guests have arrived, and we
have been setting up accounts and workstations for them.
-We have also been getting additional Matlab toolboxes.
Hanford:
(Christine)
- Replaced the batteries on two PCs. Reinstalled all software
on one
PC. Backed up user files on another PC that is transferring between
users.
- Took an inventory of spare monitors. Swapped some monitors
around to
meet user needs.
- Purchased several items for CDS and GC. Put in a PR to buy
a disk
array for GC user accounts. Requested quotes on a new tape library.
- Set up new accounts for SURF students. Installed some new software
on
the application server for use by the SURF students.
- Helped LDAS Sys. Admin. set up a Ultra 10 on GC network that will
have
all the LDAS development software installed on it to be used as an
LDAS
code developer machine on the GC network.
CIT:
(Barbara)
- Worked on packaging and printing issues with the costbook printed
reports.
- Loaded IO data into costbook. Updated PDF reports on costbook
web site.
- Began changes to the web versions of the costbook detail reports
to make
them similar to the Totals web page.
- Added missing fields to production LDAS equipment database.
- Posted talks from April-May NSF review and a number of LSC documents.
Updated publications pages.
- Met with Irena and Rita about LSC issues and Thomas about costbook
issues.
(Lisa)
- Rebuilt alcor, an ultra10
- Working with the 40m to get them setup. This will be ongoing
for the next
few weeks.
- Helped out Riccardo with some user stuff.
- worked with mike on setting up the network for the PAC in the 114
conference
room.
- Created a couple of new accounts.
- Made a ligo T-shirt for Sam.
(Sam)
GRADUATING!!!!!!!!!!!!!!!!!!!!!
Showing family around.
(Mike)
-40Meter, loaded full software installs on two new PC's.
-M15 had to do a full reload win98 and all general computing software
plus
Auto CAD 2000i.
-Loaded some applications on a couple of laptops.
-Ken Mailand, had a problem with dropping a image into word, version
office
y2k. After installing service pack 1 and 2 for office 2000 which did
not
fix the problem; I ended up having to install a service pack for the
registry. This worked 'yes'. This machine had to have WIN98 loaded
to
get Autocad working properly.
-User on site support which included upgrading and replacing hardware
as well as patching OS's.
(Larry)
-Worked with Donna on a number of issues; Laptop repair and disk recovery,
PAC mtg. setup and preparations, a couple of procurement investigations.
-Worked a number of issues for 40M. Mostly procurement related.
-Worked a few network issues. Mostly installation of cables and hubs.
From Mike Smith:
40 METER
WBS
The WBS dictionary and development plan for the AOS, ISC, and IO sections
of the 40m IFO Upgrade were re-organized into two sections--Optical Systems
and Sensing, and AOS--to better represent the division of effort on the
40m project.
Design Requirements
The Design Requirements and Conceptual Design document for the Optical
Systems and Sensing and AOS sections of the 40M upgrade is being developed.
From: Janeen Hazel Romie <romie_j@ligo.caltech.edu>
Suspensions
New osems for 40m, LASTI, TNI etc.: So. California Braiding has
agreed
to make 100 pigtials for TNI and LASTI by the end of June. Final
delivery date has not been finalized.
Advanced LIGO: Helena and I are working with Thomas Frey on schedule
and
cost data for Advanced LIGO Suspensions.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Silicate bonding
Made five 1/2"dia. sapphire/sapphire bonds. These bonds will be tested according to R. De Salvo specifications.
Made 11 (1/2"dia.) silica/silica bonds using a sodium silicate solution.
These substrates were used in previous bonding experiments so the surface
quality was degraded.
Found that sodium silicate solution gives very robust bonds.
All bonds when tested in tension pulled glass.
Pictures showing the effects of tension on the bond with a polariscope
can be seen at Sirius/ahelena/silicabonds/polariscope
5 of the 11 bonds were cured at 120 degrees C for 24 hrs.
The strength of these bonds was somewhat higher and more uniform than
the non-cured bonds. More tests are planned to confirm this fact.
Testing was done when the bonds were 3 and 7 days old.
No difference in strength observed due to age.
The cured bonds were able to withstand from 86 kg to 105 kg of pulling
force. Unbaked bonds from 48 kg to 88 kg.
An interesting fact is that a bond that broke at 48 kg of pulling force
pulled about 3/4 of the bonded glass area; one that broke at 99 kg pulled
very little glass around the edge of the bonded area.
From: Riccardo DeSalvo <desalvo@ligo.caltech.edu>
Florian
Tested DSP, ADC, DACs found one channel with half gain, investigating
with Roberto Taddei of IDED
Virginio
Sent GASF paper to NIM
Riccardo
Submitted IP Actuator paper to internal committee,
please do not forget
to send in your comments, suggestions and corrections
if you have not
done yet.
Szabi
Got printed circuits of 8 channel stepping motor controller.
Populating
and testing.
Last chips due tomorrow. Ship to Japan next week.
Work on IP paper.
Alessandro
Received 30 KHz actuator from Rita. Mounting Aluminum and later
titanium horns on it, testing U.S. cutting power soon.
Yukioshi
Partly occupied with TAMA run.
Studying basic stepper controls.
Suspension transfer function characterisation.
Akiteru,
Given seminar on SAS to TAMA collaboration.
Waiting from parts from machine shop.
Ordered additional feed throughs.
Yukiko
Measured magnetic field in modified actuators.
Hareem
Analysing old Q-measurement data for GASFs.
Found 1 set of data completely disconnected from all the others.
Clearly something happened to Blades. Reconstructed that at that time
the blades had been dismounted, erroneously mounted upside down and
then
redressed. This plasticized the wire hook point.
Starting MGASF analysis
Brett
Characterizing detwinning of existing MM springs.
Built a couple of Memory metal springs dimensioned for actuator for
IP.
For additional information about this report, contact sanders@ligo.caltech.edu