|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday March 4, 2002 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items:
no report
LIGO Operations--Administration
A Site Teleconference was held on Thursday, Frebruary 28, 2002. The agenda included the status of FY 2002 costs-to-date, of NSF funding and the Cooperative Agreement, the establishment and budgeting of FY 2002 accounts, the FY 2002 Financial Reporting format, as well as the status of the construction at the two sites.
From: Ed Chargois <chargois_e@ligo.caltech.edu>
>From: Linda Turner - turner@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
ACTIVITY
| 02/28/02 | Packages | Faxes |
| In | 44 | 37 |
| Out | 06 | 43 |
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 (Baldon, Torres, Lloyd, Tischler)
Progress Period from 02.22 to 02.28
Accomplishments:
We are still doing a standard Annual Report for the Construction Project as of the end of November. Actually it has been completed and will be ready to be printed to be sent to the NSF shortly.
Completed initial responses to the DCAA Proposal Audit questions. Completed draft of responses to Price/Waterhouse/Coopers (PWC) audit and distributed for comment. Meeting with Finance tomorrow (March 1, 2002) to discuss approach to DCAA Incurred Costs audit.
The following change request has been submitted:
| CR-010012
Revision B |
WBS 1.4.4.1 | Closeout Construction Budgets for Initial Computer Equipment Complement at the Sites | P. Lindquist |
In addition, Gerry Stapfer submitted a change request (CR-020001) for cost increases incurred on the HPD contract. We held a meeting of the LIGO Change Control Board on Tuesday, February 26, 2002. This change request was approved and minutes have been distributed (LIGO-M020117-00-P).
Press for the latest Contingency Needs Projection.
From: Cindy Akutagawa <cindy@ligo.caltech.edu>
No report this week.
no report
CDS: Made a list of software changes/upgrades to be done at LLO. Also
included
were some hardware requirements. Upgrading london to a new ultra 60
on
Friday. Moved all the web page files to the new machine. Configured
it to
run on ATM. Working on fine tuning the MC Auto Locker script. Wrote
test
code for xycom 200 and pentek testing. Created a new medm screen for
weather
stations. Checking on disk to disk backup of LLO1. Planning on making
the
disk backup of london after installing new london. Added additional
licenses
for MATLAB and installed DSP blockset. (Chethan Parameswariah)
LDAS; Reconfigured backups at sites to use cron instead of internal
DB2
scheduler (the advantage is the possibility to use scripts instead
of
GUI). Restored LHO_1 and LHO_2 backups at LLO and LLO_1 and LLO_2
backups at LHO. Finished reading "Fundamentals of interferometric
gravitational wave detectors" by Peter Saulson. Read first 2 chapters
from
DB2 Replication guide. Got quotes from USPS and couple motor freight
companies for shipping Beowulf PCs to MIT. Called SUN to schedule power
supply replacement for SunFire machine. (Igor)
Ordered ram for the beowulf. The first batch that was ordered
was back ordered indefinitely, so the order was canceled and reordered
from a different vendor. Pricing the fiber to run to the new
building. (Shannon)
GC: Ran network cables new building in prep. for the LSC.
Checking
into the fiber & cat5 terminations in the new building. Checking
on
fiber prices for ldas/GC. Wrote a script to manage group IDs
for NIS+.
Finished a web page for tracking IP address allocation. General
preparations for the LSC meeting. (Shannon)
IFO: The transfer function of the mode cleaner servo was measured including
the
crossover frequency between the MCA path and the PSL path.
The unity gain
frequency is about 15 kHz. The crossover frequency varies between
100 Hz
(lowest) to about 350 Hz (at robust locking mode).
The air legs on the PSL table were removed after they apparently sprang
a leak. (Joe Kovalik)
Optics and Installation: The installer reports the Laser Safety
Interlock
complete and ready to enable. We will issue access cards following
a
training session next Monday. Piecing together some DC photodiodes
with an
imaging system to monitor scatter of the cavity optics. Working
with Dr
Weiss and Peter Saulson to design a system to detect shorter wavenlength
harmonic generation on the large optics. Staged material and assisted
Joe
Giame in measuring the transfer functions of BSC piers. This
is ongoing. (Jonathan)
Control Room:
On screen help files are being updated and in some cases generated.
The Mode Cleaner AutoLocker has been tested and needs a little fine
tuning.
A Spectrum Library is being created and should be completed in 2 to
3 weeks.
The pictorial configuration control on the ISC tables is 80% complete.
The MEDM Weather station screens have been reconfigured for better
access.
Other:
The annual crane inspections have been completed only in the corner
station, the end stations will be inspected next week. There were no
major
problems with the cranes inspected. (Rich Riesen)
|
Commissioning:
|
|
Installed a Faraday isolator on the PSL table to test
whether the parasitic interferometer observed in the MC_F signal is due
to feedback to the laser.This time
the FI was installed without any difficulties, and the modecleaner relocked
easily.So far no clear answer about
whether this has suppressed the excess noise.
Continued work on the remaining WFS.The emphasis this week has been on characterizing the beams going to the WFS using the BeamScan to find the waists and adjust the Guoy phase telescopes.
This week
we worked on the DC Calibration procedure and the Commissioning of the
Output Matrix Filters. The two methods in study are the 'Counting of Fringes'
and the 'Sign Toggling' methods. A fair amount of time has been spent on
simulation (e2e), trying to understand the different systematic contributions
of each procedure. We have addressed, so far, the effect of offsets, the
restraints on the gain, the sampling and the drive linearity. Once these
methods are used on the 2k and 4k, both of them yield precise calibrations
(1% level), but their calibrations differ by ~10%. The difference was found
on both 2k and 4k, by about the same amount, and in the same direction.
The systematic contribution listed above could not justify such a difference.
Are the two
methods really calibrating the same thing? The 'Sign Toggling' method calibrates
H1:LSC-MICH_CTRL, whereas the 'Fringe Counting' method calibrates H1:LSC-ITMX.
If the Output Matrix has a 1 for the element relating the lm path with
ITMX, then they would be identical. It turns out that there is a 6% difference:
H1:LSC-MICH_CTRL is always ~6% greater than H1:LSC-ITMX. This is true for
both interferometers, 2k and 4k. It may be true also for Livingston, that
needs to be verified.
With this
systematic in place, the two methods agree to within ~2%. We will analyze
the data carefully in the weeks to come.
The Output
Matrix Filters have been tested on ITMY and their effect is promising (see
2/27/2002 elog). The objective is to minimize the cross-coupling between
the drive and pitch at all frequencies, not just DC. First thing was to
test the filters. An artificial damping term of Q=1000 was introduced to
make them stable but we noticed that once enabled, the mass would start
to oscillate at ~1Hz and the local damping and optical lever damping could
not damp it. We modified the Q from 1000 down to 10 and once engaged, the
mass would not start to oscillate like before.
The second
step was to modify the gains of the upper and lower coils by ~11% (because
of the filters DC gain). We also tweaked the yaw gain by a few percent.
It turns out that the coupling was attenuated by 10x at 2Hz, as seen by
the optical levers (this experiment was done by exciting the mass along
the beam and recording the pitch and yaw signal from the optical levers;
we measured this TF with the standard DC compensation enabled and with
the Output Matrix Filters enabled; by comparing them, we noticed an attenuation
in the coupling of 10x already at 2Hz). More attenuation seems to be present
at higher frequencies. At DC, both compensations yield the same coupling.
The next step is to apply these filters to the other masses and see if we obtain an improvement also there. The ultimate test will then be to lock one arm, engage these filters and verify that the filters are indeed of help and they are not hindering us. Looks promising....
The recycled
interferometer continues to lock at night. Work was done in the following
areas:
·Noise analysis and reduction- The electronics chain from the DAC, Anti-image, de-whitening and differential driver to the LSC input of the coil drivers has been characterized. The noise at the LSC input is still larger than desired (100's nanovolts/sqrt(Hz). The noise in the ASC is being studied. Noise in the optical lever damping system on some of the masses is due to fluctuations in the optical lever laser. The levers are not operating at shot noise.
·Common
mode servo - The transfer function of the mode cleaner servo was carefully
measured to allow calculation of the gains and filters for the common mode
servo. A filter at the frequency of the roll mode of the beam splitter
was inserted in the control loop to avoid ringing up the mode.
·Wavefront
sensor - Some of the inhomogeneity measured between photo detector quadrants
was due to measurement difficulties arising from errors in access to the
digital test points. Earlier work did indicate that the wavefronts at the
wavefront sensor are not uniphase. The phase variations is, however, not
significant enough to keep the sensor from operating.
·Intensity
glitches in the PSL - the pulses noticed in the E7 run have been further
analyzed and cataloged. They are easily seen in the time and frequency
domains. We still do not know their source.
·Coupling of angular motion into longitudinal motion - Work continues in finding the best location of the beam on the test masses to reduce this coupling.
The team is working on documentation to support a requirements and conceptual design review. Delays in the effort call into question timely readiness for an originally proposed 3/15 review date. A decision to delay or proceed with this date will be made in the next day or two.
·Pier
compliance tests are being planned at LLO by Jonathan Kern and Joe Giaime.
Inertial instrumentation will be placed on a pier, the air bearing will
be floated and the pier will be excited with a shaker.
·Differential
inductive displacement sensors (Kaman DT-?) and geophones (L4C) have been
ordered for the LASTI prototypes.
·Seismometer
Signal Conditioning: In the process of designing the printed circuit board
for the remote interface unit.
·It
was hypothesized that the compliance of the narrow cylindrical "shell"
of fluid around the central aluminum plug (into which the L4C geophone
is placed) was responsible for the low bellows resonance. The plug was
removed and the frequency was even lower (dropped from 40 Hz to 30 Hz).
As a consequence we have a mystery that is not understood. However, it
appears that some plug is needed and the Stanford group seem reasonably
certain that its size will accommodate the L4C.
·Plan
to implement and test the new integral hydraulic bypass network in the
new actuator design next week.
·SEI
Hydraulic Pump Station:
oLayout Drawing: I have started a Solid Works layout drawing of the location of the components on the granite base, and configuration of the plumbing. I have an isometric pencil drawing showing all the components. Final word on the accumulator, from the Stanford group, is hoped for this week. Stanford is attempting to reduce the accumulator inductance.
oManifolds: Station noise filter1, and test load 1 &2 have been started, just sketched out dimensions so far, configuration TBD.
oLong
Tube Run: Design of a method to assemble, support and isolate the 360 foot
tubing run is in progress.Test taps
at the 60 foot mark [shortest run] and 180 foot mark.
·Placed
an order for the BEI large electro-magnetic forcers for the LASTI prototype
test on the HAM chamber.
This week’s project-wide weekly CDS meeting and a
follow up meeting on suspension noise and auto-alignment were used to discuss
and set priorities for CDS tasks required for the Science 1 run. The start
date for the S1 run has not yet been defined. Current thinking is 5/31
<= S1 start <= 6/28. We would need to freeze the configuration for
all but minor changes about 4 weeks in advance of the S1 start. This leaves
9 to 13 weeks for all of the tasks listed below:
|
|
CDS
Commissioning Tasks in support of the Science #1 Run (*)
|
|
1
|
Investigate/characterize
SUS noise (implementation of fixes after S1)
|
|
2
|
Remote
control rooms at CIT & MIT
(purpose is to increase participation by campus interferometer experts) |
|
3
|
RFI
cleanup
|
|
4
|
60
Hz ground loop cleanup
|
|
5
|
LLO
ASC code & hardware changes to permit simultaneous use of optical levers
and WFS
(without going to the DSC) |
|
6
|
EPICS
security (enables remote control rooms as well)
|
|
7
|
Common
mode for LHO 4k
|
|
8
|
New
EDCU (EPICS data collection unit)
|
|
9
|
New
filter module for ASC and then LSC code
|
|
10
|
New
MC servo daughter card with variable gain (eliminates transients in gain
stepping which can break lock)
|
|
11
|
Photodiodes:
- 2 PD limit on APS port OK for S1 - ensure that sufficient spares in place, or populate non-APS ports with GDP PDs to recoup EG&G PDs for APS |
|
12
|
New
Timing modules with variable delay (to reduce noise from sample jitter)
|
|
13
|
New
GPS clock driver modules
|
|
14
|
Differential
driver/receiver for Penteks for the LHO 4k
(10dB reduction in MC freq noise) |
|
15
|
CDS
Object version tagging & filter parameter files
|
|
16
|
Automate
initial alignment (without autocentering on QPD, or on ISCT)
|
|
17
|
Make
sure that all the slow channels are written to the frames
(with new EDCU, including all channels will not increase the data rate) |
|
18
|
quiet
RF cooling solution for PA85s in SUS controller
|
|
19
|
PSL
Epics/VME ISS (inner loop) to provide observability (or witness PD at MC
transmitted port?)
|
|
20
|
Microseismic
& tidal control with fine actuator:
|
|
21
|
deferred
|
|
22
|
Framebuilder
capability to read arbitrary trend frames
|
|
23
|
deferred
|
|
24
|
deferred
|
|
25
|
deferred
|
|
26
|
Coordinated
SW/HW switch ramping for the dewhitening filters? (pending further definition)
|
|
27
|
deferred
|
|
28
|
Epics
LSC name cleanup
|
|
29
|
Combined
with #15
|
(*)
Notes:Does not include seismic
retrofit tasks (also high priority) and low noise DAC development oversight
(technical contract management)
Priorities and action item assignments have been made. We are continuing to look at the resource requirements to see if all that is needed/wanted before the S1 run can be accomplished. Completing the hardware tasks (engineering, fabrication, board/unit test and installation) look feasible, though it may impact planned seismic retrofit prototype design and fabrication.. Field testing (commissioning) these hardware changes/additions in time for S1 looks very tight. Likewise, the software changes/additions, and commissioning tests of these software changes, appear to be very tight, especially to meet a late May start.
The requirements
for the summing box were collected and sent out for review (T020029-00).
After the initial review a conceptual design was developed and sent out
for review (T020032-00). Comments should be received by Friday and a design
for the box can be completed in ~1 day. It will then be fab'd on a quick
turn around and should be ready for installation at LLO the week of Mar
18.
The drawings
for putting WFS 3, 4, 5 back to their original wiring and adding the optical
levers to the ASC in a manner similar to the DSC are complete. Rusyl is
reviewing them. Installation will begin the week of March 18.
Preliminary designs for differential driver and receiver modules to be used with the penteks have been completed and sent out for review. Layout of the 4 different modules will begin next week.
Responses have been received from the vendors for development of a higher performance DAC module. One of the responses is very encouraging. The contract should be awarded in the next week.
GPS clock driver board is in manufacture. A total
of about 50 boards will be made available after manufacture.
My variable delay board has been sent for fabrication and will be back end of this week. I have also ordered all the necessary components and kept ready so that we could soon stuff once it arrived, and I can start testing right away. Did some Spice simulation on this board.
Doing the layout for a variable gain amplifier to put on the MC servo board as a drop-in replacement. Board will be sent to manufacture this week and will be ordered on a one week turnaround. There are only a few components on the board to stuff, so the release will be quick.
Lori appears to have the standard dig filter code working; we are now adding to the ASC software. Alex got the GDS AWG and DTT working in our development system, so we hope to start sweeping and verifying the filters in the code by end of this week. Note that this is only the beginning of the ASC code changes.
Found a DW
filter switching problem in LSC code; testing at LHO. Code mod made to
LLO4k software, but waiting for LHO test results prior to shipping.
Added common
mode servo code for LHO4k; testing.
Alex has
begun work on framebuilder code to read trend files generated by sources
other than DAQ.He is also working
on the new frame5 development.
Developed
a means of tagging all CDS object code with version numbers and extracting
that info.We still need to develop
scripts to find the loaded objects and archive/display the info. Plan to
also include filter coefficient files also.
Proceeding with EDCU software; presently adding interrupt capabilities to obtain time marks.
I have been looking at the 80 MHz VCO circuit to see if some improvements can be made. A modified circuit for the tuning voltage to the VCO was stitched together but its performance is not good enough.
Flavio Nocero
Creating the schematics for the VME controlled ISS electronics.Schematics to be complete by the end of the week and will then need review.
We are again studying the "parametric instability", focusing more on the "physical aspects". Although the calculations for the excitation of the other 00 modes are rightwe are trying to address the more general case when a different mode is excited. It turns out that the instability is a second order effect, coming out of equations cut at the first order and in general there are a lot of approximations I would like to understand more.
·Progress on OTF contamination cavities. They lock with some improvements.
Contamination
Cavity # 2: As reported before, It has been locked.New
changes made, replacedthe new 12.3
Mhz resonator EOM (Pockel Cell) and put back to the old broadband EOM (PockellCell).Also,
replaced the old version (blue bulky box)Si diode photodetector for a new
PD255 InGaAs detector for the RF signal.The
new photodiode detector made locking of the cavity easily.
We increased
the RF signal from the 5w RF amplifier allowing us to see the side bands.We
measured the side bands as to be 2.5 millivolts and transmitted laser signal
230 millivolts.We are putting 120
milliwatts of power into the cavity and having 75milliwatts of transmitted
power.The cavity is locked, robust
and stable.
Contamination
Cavity # 1: This cavity is being upgraded. New CVI lens and lens holders
replaced for cavity mode matching. Two lens mode matching calculated and
verified with the Beam scan and fixed the beam waist size and location.Alignment
is in progress.
·Following up with Jordan on archival OTF data for LIGO I COC (with Garilynn)
GariLynn Billingsley, Helena Armandula
Instability in the lab temperature does not compromises or delays phase map measurements. We are still characterizing SPETM01 (a COC spare).
We are pursuing a new approach in taking surface data.In the past I have been reluctant to run the thermal control fans at high volume as it clearly causes vibration.However, we may be able to overcome vibration with a lot of averaging (something we're already doing to beat down camera noise.)The higher fan speed appears to make the temperature of the experiment more stable.Comparison of data with and without higher fan speed has not yet been attempted.
I have also been trying to collect statistics on some noise bursts noticed by Peter Saulson in the L1:PSL-ISS_ACTM_F (this channel is rumored to have been connected to the reference cavity transmitted light signal during E7). My first attempt didn't succeed very well because of a lot of low frequency noise obscured the noise bursts of interest. This should be easily fixable by adding a 100Hz High Pass filter.
no report
LASTI (MacInnis, Mason, Mittleman, Ottaway, Rollins, Shoemaker, Zucker)
=======================================================================
VACUUM ENVELOPE:
Using site and PSI test chamber data we
verified that our water and air signatures are consistent with degassing
of Flourel spring seats and O-rings. Making one last attempt
to
resurrect our RGA electron multiplier before we vent for stack
vibration tests (see below).
EXTERNAL PRE-ISOLATORS:
In situ BSC stack characterization (Mike and Rich and Myron):
We received two shaker setups from LHO to support the test. LLO
is repeating the pier test and will ship back the remaining complement
of shakers and accelerometers end of this week. We currently
plan
to be ready and backfill to install the instrumentation by
Tuesday.
Pre-isolator mechanical design/fab (Ken):
Completed the fabrication drawings for all long lead componants. Visited
several machine shops and fabricators with drawings for quotes. No
one
had a problem with a 4/19 delivery.
CDS/DAQ/GDS:
No problems to report.
PSL:
This week the one meter SOS test cavity inside HAM13 was aligned such
that bright TEM00 mode fringes were observed. Work testing the FSS
servo continues; the problem that arose was amplitude and phase
instabililty in the pockel cell which resulted in an unstable cross
over between the fast path and the PC path. This (we think) prevented
stable locking. Work is in progress to cure this problem.
Simulation and Modeling (Bhawal)
E2E MODELLING:
_______________
Hiro attended a workshop in Kyoto University, invited by TAMA group,
to give a
status report of LIGO. The talk covered LIGO I, E7 run and advanced
LIGO. During
his visit to Japan, he contacted several TAMA people to exchange opinions
about
the collaboration between TAMA and LIGO.
LIGO I simulation system
--------------------------
(Matt) I built the seismic environment model and the ASC sub-system.
System debugging and validation are next on the agenda.
Mode mismatch
---------------
(Biplab) after discussing some results with Hiro and Matt, trying to
understand
numbers for mirror displacements originating from mode-mismatch.
Code maintenance
------------------
(Ed Maros) Worked some on build documentation
Alfi
----
(Bruce)
- Bulletproofing parser and node element consistency.
(Melody)
- Addressed problems reports for the alpha version of alfi5.
Worked on the box files popup menu and the window manager.
LIGO Data Analysis System
Software Systems (Blackburn)
The modifications to the dataConditionAPI and wrapperAPI to support
the
passing of FrDectector geometry data continues. Work in the wrapperAPI
is almost complete. The dataConditionAPI support the geometry information
only for the simple cases of timeseries and frequencyseries which are
not
operated on. Work continues on the complex cases.
Continued to refine a few points in the new resampling library used
in the
frameAPI and dataConditionAPI. More testing has been done at the TCL
layer
and more unit tests have been added to the make check.
The development of a new action called shifter() which is a simplified
interpolation routine has begun. This is the only new action being
added
to the dataConditionAPI in this release, though more work is also still
needed on the Kalman filter action.
Added new user command activeJobSummary which reports the status of
all
jobs currently running or queued in the local LDAS installation.
Added support to mpi API for LDAS e-mail messages to be generated based
on
embedded strings in wrapperAPI shared object status reporting code.
This
allows dso developers to have themselves be notified via e-mail of
notable
conditions arising during data analysis.
Factored out all file type tests required within LDAS into the genericAPI
procedure "fileType". This allows for all necessary file types to be
identified from a consistent interface in all file input routines.
A corresponding procedure for typing the contents of tcl variables
is still
needed for file output routines.
Continued integrating diskcache API into LDAS. Many complex/compound
-framequery options were resulting in erroneous output or error conditions.
A full set of tests are required to verify that all conditions are
now
supported properly.
Added a sanity check to the LDAS generic API socket communication callback
"reattach" so that malformed return values from API's to the manager
generate descriptive messages as opposed to "unexpected socket closure"
errors. Analysis showed that these types of errors are never
fatal to the
LDAS job in which they occur, so they are treated as warnings now.
Began reimplementing diskcache API cache data structure according to
new
design worked out in meeting with Kent Blackburn. The specification
for this
persistent form of this structure will describe the structure.
Added rotation to logging of job success/failure data file generated
by
manager and used by controlmon API for statistical calculations.
This
file grows by one megabyte per 25,000 jobs; it is now rotated when
the
size exceeds one megabyte and the old file is timestamped and moved
to the
log archive.
Began writing per-API statistics generation routine which will maintain
data on a per-api basis of jobs succeeding and failing, how often the
API
is restarted, and each API's uptime. Report generation and persistence
across system restarts via a state file are pending.
Lazzarini: We have discussed with the detector group the timing of making
a transition to the next release
of the frame specification that has been agreed upon by Virgo, LIGO,
and the LSC. It will likely not happen before
S2 due to personnel constraints (too much to do for S1).
Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* Continued HPSS migration from Redwoods to 9940s.
* Still trying to get StorageTek to make good on coming up with a
replacement Ultra 10 (they in turn are trying to get Sun to
do it, with
a little help from me via Linda Feather). Still no concrete
results.
* Mostly worked on setting up the LIGO SAN testbed. Accomplished so far:
SAN admin network set up.
Firmware upgraded on switches, HBAs and T3s.
Relevant Solaris patches applied to SunFire machine.
9940 drives connected to switch via fibre.
9940 drives given basic test (tar files to and from
tape).
* Now working on details of SAN configuration (specifically setting
up a
multi-switch fabric so we can test inter-switch links).
(Al Wilson)
* Prep work for the new network switches. Ran cables.
* Had to make some changes with the cfdef file so the Linux def's will
work
with version 5 of the installer (cfdef)
* Working on tripwire, ran some test with it reporting to big brother.
The
project is coming along fine.
(Stuart Anderson)
* Finished integrating the new SunFire V880 servers in the ldas-archive,
ldas-lho, ldas-llo. Started integration in the completely new
ldas-cit
system.
* Shutdown the ldas-test and ldas-cit system for construction work.
* Created a new top-level /lal directory structure in the ldas-dev system
for the LAL code contributed by the LSC to run in LDAS. If this
works
it will be replicated on the other LDAS systems coincident with
the
next LDAS software release.
* Verified that Gigabit Ethernet works over 450m of the new 50micron
multi-mode fiber installed at Caltech.
MIT
---
(Keith Bayer)
* Contacted sun again regarding SF880 component recall.
* Investigated and installed cmonClient tool on laptop.
* Restarted frameAPI to serve E7 playground data.
* Build concatenated strip across remaining (4) internal
disk drives on SF880 to hold playground data.
Livingston
----------
(Igor Yakushin)
* Reconfigured backups at sites to use cron instead of internal DB2
scheduler (the advantage is the possibility to use scripts instead
of GUI).
* Restored LHO_1 and LHO_2 backups at LLO and LLO_1 and LLO_2
backups at LHO.
* Finished reading "Fundamentals of interferometric gravitational
wave detectors" by Peter Saulson. Read first 2 chapters from
DB2 Replication guide.
* Got quotes from USPS and couple motor freight companies for shipping
Beowulf PCs to MIT.
* Called SUN to schedule power supply replacement for SunFire machine.
(Shannon Roddy)
* Ordered ram for the beowulf. The first batch that was ordered
was back ordered indefinitely, so the order was canceled and
reordered
from a different vendor.
* Pricing the fiber to run to the new building.
Hanford
-------
(Greg Mendell)
* The host dataserver is now gateway on the LDAS network. A new
dataserver, a SunFire 880, was brought on line. As a part of
testing,
the new server is running the ligolwAPI.
* The LDAS network backup script was updated to include the new server,
and reflect the name change of the old dataserver to gateway.
The
ldas_admin_all mailing list was informed of the needed changes
to this
script.
* Rebooted the Sun Blade 100 admin host to complete installation of
kernel patches that were applied by Stuart.
Data Analysis Activities (Lazzarini)
Shawhan:
I have been overhauling the
http://www.ldas-sw.ligo.caltech.edu/ligotools/faq LIGOtools
Frequently Asked Questions page, adding several new notes about
working with data from the LDAS database, etc.
Tracked down an apparent discrepancy in the list of "clean locks"
during E7, finding it to be due to conlog overestimating the length
of locked segments (since it records the time when the LSC servo was
turned Off, which is typically a few seconds after the interferometer
loses lock).
T. Creighton:
This week I submitted the new LAL inspiral simulation interface
function, and have been working with the AEI at validating my pulsar
modulation and their pulsar demodulation code.
Mendell:
1) A memory problem at LLO that was causing the frameAPI to crash when
knownpulsardemod DSO ran has been fixed. (The problem was caused
by
large files stored in /tmp on the server running the frameAPI.)
2) Saving SFTs data as COMPLEX8 instead of as COMPLEX16 will reduce
the
size of SFTs by 2 without loss of precision. (SFTs are computed
using
LAL functions to single precision only, regardless of how they are
stored. However, the LALDemod function currently inputs SFTs
in double
precision.) This will require only a minor change to the
knownpulsardemod DSO.
3) Bugs in the clean_locks reported for E7 are being fixed. SFT
generation will begin again, once new clean_locks are posted.
General Computing (Wallace)
MIT:
-Ordered and installed second ide hdd in Ultra10
-Investigated mirroring with disksuite
-Installed MU6 on compute-server (lancelot)
-Setup new desktop PC for technician
Livingston:
(Shannon)
-Ran network cables new building in prep. for the LSC.
-Checking into the fiber & cat5 termination's in the new building.
-Checking on fiber prices for ldas/GC.
-Wrote a script to manage group IDs for NIS+.
-Finished a web page for tracking IP address allocation.
-General preparations for the LSC meeting.
Hanford:
(Christine)
- Setting up new PCs and laptops.
- Investigating software and/or hardware that would allow management
of
bandwidth usage for the T1 line.
CIT:
(Mike)
-Working on Mohanna's computer troubleshooting some hardware problems
that she was having with her computer. This is looking like a secondary
IDE
controller is going out on the motherboard. I made some adjustments
to keep
her up and running.
-Loaded two new PC's with win2000 with all General Computing Software.
-Loaded a PC to be placed in synchrotron for Riccardo.
-Received two new PC's that I checked out to make sure they were what
we
ordered and I now am going to reload one of them with win2000 for Cindy.
-Assisted Larry and Lisa on configuring gige network.
-Pc72 is hosting Project Science website which is having a slow response
time; looking into stopping some services and a few other changes that
might help increase performance.
-Backing up a Optics Lab PC and getting it ready for a reload this
has win95 and
is to be upgraded to win98.
(Veronica)
- LIGO website: updated Employment, worked with Bonnie Wascom on LSC
March
registration form; submitted the new LIGO roster pointer to Caltech
Reference Desk webpage; updated Student Programs, LIGO Participation.
- CaJAGWR: Posted updates to the seminars page.
- Worked with General Computing group on LIGO website's access issues:
The
website was inaccessible from a subnet, which was blocked due to a
recent
hacker attack.
- Worked with Mike on web-hosting for Project Science, per Gary's
request.
- Planning the webpages for the Advanced Suspension group, per Calum's
request.
- Looking into the digital video setup.
(Lisa)
-With Ed C's help, I did a warranty return of the dead hard disk from
electra.
-Set up ssh keys on most of the unix servers
-Moved pleione up to a visitor space in Millikan
-Spent some time with Larry on the foundry equipment.
-Modified the access permissions on ligo.
-Setup e-mail logging on acrux. The U of Washington imap/ipop puts
a voluminous
amount of log info into the maillog. Am looking at ways to reduce
or eliminate
that.
-Testing out my automated patch script. I'm going to begin running
the patch
clusters tonight or tomorrow night.
-Started monthly backups.
(Larry)
-Finally, was able to spend some time working on the Gige installation.
All of
the edge switches have been tested. All but two have been installed
at their
assigned locations and have been tested. So far things are looking
good. Along
the same line the preliminary tests on the new fiber installs went
well. If all
the testing goes well the move over to the Gige backbone should start
next
Friday.
-Checking out the h/w on a couple of SUN boxes. Appears one will need
a reload
and the other will need a board replaced. These boxes are going to
be used as
servers on the new Gige backbone to help distribute some of the services,
and
ease the migration off of sirius.
-Resolved a couple of e-mail issues. It looks like we have a configuration
that
can be used on all of the SUN boxes that will accommodate the different
mail
tools being used.
-Resolved a couple of web server issues. Working on the logistics with
Mike and
Veronica on a number of other web related issues. Setting up plans
to migrate
some of the web services off of NT and over to Win2000.
-Worked on a number of procurements. Still quite a few to go.
-Updated some of the documentation and working on the updates for the
1-800
policy documentation.
-Worked with Ed C. on a couple of property issues and logistics for
the LSC.
From: Mark Barton <mbarton@ligo.caltech.edu>
This week I returned to debugging my Mathematica
model of the
suspension and with the rederived and corrected version of the Matlab
as
a guide I was able to weed out a couple of minor errors and get the
desired four-figure agreement in the mode frequencies. (As well as
being
an important test for my model, this also greatly increases confidence
in
the Matlab.) I adjusted the default values to better reflect the MIT
quad
prototype and released it as v2.04 (<http://www.ligo.caltech.edu/
~mbarton/SUSmodels/asus4/>).
From: ctorrie <ctorrie@ligo.caltech.edu>
Prototype quadruple pendulum, MIT
(Norna Robertson, Mark Barton and Rich Mittleman)
Rich Mittleman at MIT has measured transfer functions and impulse responses of the quadruple pendulum in the yaw direction. Norna and I have shown that these agree very well with the model.
I now have a list of modifications for the quad at MIT, in order to
solve the problem of the undamped pitch mode. Norna is back on Monday and
Rich, Mark and I plan to discuss the changes then.
As I mentioned last week this will be able to be done in situ.
Suspensions
Janeen and I have been busy to re-organising the North area in the Synchrotron Lab. Janeen took a list of our questions with SOLIDWORKS to a drop in class on Friday. This was very beneficial and we plan to attend other classes in the future.
LASTi
Still working on completing a full set of drawings for the prototype
LIGO-MC suspension. This has included close interaction with Eoin Elliffe
and Russell Jones in Glasgow on parts for the coils.
From: Janeen Hazel Romie <romie_j@ligo.caltech.edu>
AdLIGO Suspensions
Working with Thomas on GEO proposal numbers.
Receiving quotes for maraging steel for RM and ETM blade prototypes.
Hope to place the PO by the end of this week.
Working on RM suspension parameters.
Working on CAD and FEA software translation capabilities.
40m
Bartie's baking schedule has fallen behind due to a potentially faulty
RGA but we are still on schedule.
Bill Kells
· Following up on tasks
from Gingin oversight telecon.
40 Meter
Jay Heefner
- Preliminary design for the 40 Meter timing system has been completed
and
drawings were issued for review. The small VME crate and the IRIG B
distribution
amplifier for the system were ordered and due in the next few weeks.
- Drawings for the mode cleaner length and alignment controls are complete
and
ready for cross connect list generation.
Sander Liu
- EO Shutter Controller - In the process of building three sets of
circuit boards for the 40M. We are currently waiting for delivery of the
high voltage power supplies from UltraVolt to finish repairing and testing
two exciting units.
AdLIGO PSL
Peter King
Some minor teething problems have emerged with the wiring harness.
All the DAC function fine, along with the binary input and output channels.
However a problem exists with all the ADC channels, with the presence
of a
~60mV offset on all channels, suggesting a possible grounding problem
of
the cross connect.
Only one problem was encountered with the EPICS records database and
that
was due to a typing mistake. The MEDM screens were modified to
reflect the
changes in the records.
For additional information about this report, contact sanders@ligo.caltech.edu