|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday April 28, 2003 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items: furniture for Livingston,
Construction Budgets Update, Operations Budgets Update, DISCUSSION
OF A SCIENCE RUN IN THE FALL
no report
LIGO Weekly Site Telecon (Lindquist)
There was no site teleconference on Thursday, April 24, 2003.
The list of current actions revised to reflect
the status of open actions assigned through April 17, 2003 may be found
at ACTION
LIST.
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
Reviewed and classified contents of the dcc/OUT directory. Analysis showed that only 40% of submittals were able to be processed normally, while 60% required special handling because of inconsistencies, discrepancies, and various other conditions.
ACTIVITY
| WE 04/24/03 | Packages | Faxes |
| In | 37 | 28 |
| Out | 12 | 39 |
From: Esther Cunningham <esther@ligo.caltech.edu>
No report this week (vacation).
Press here for ACCOUNTS PAYABLE HISTORY DATA .
From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
No report this week.
From: Florence Kaufman <fkaufman@ligo.caltech.edu>
From: irena@ligo.caltech.edu (Irena Petrac)
SUPPORT (Baldon, Lloyd, Tischler)
>Irene Baldon
Weekly Advanced LIGO Project Controls meeting was not held this week.
For list of documents that are being used to develop Adv. LIGO Cost and Schedule, see http://www.ligo.caltech.edu/~tfrey/Cost_MTG_082002/
Advanced LIGO MRE Proposal (Highest Priority)
Project Web Site for posting schedule and progress related data continues to be updated with the latest and greatest.
The Construction project quarterly report for the end of February was
scheduled to be submitted at the end of March. Irena has been working
on this, but has been distrtacted by a number of other priority issues:
LSC, contract activities for seismic isolation, core optics, etc.
We have discussed priorities, and the quarterly report is still slipping.
There are no open change requests.
From: Cindy Akutagawa <cindy@ligo.caltech.edu>
No report this week (at Hanford for Safety Review).
Summary of Commissioning Activities at LIGO Hanford Observatory (compiled
by M. Landry)
---------------------------------------------------------------------------------------
Commissioning activities resumed more broadly this week, but with an
emphasis on the 4k.
As a significant portion of S2 lock losses were induced by LVEA-centered
seismic
disturbances, accelerometers were redistributed in the experimental
hall, in order to
better detail these events.
R. Schofield et al. completed a series of measurements of acoustic coupling
on the 4k
AS port table. They argue that the most likely coupling mechanism
arises from clipping
from multiple limiting apertures in the AS chain of the output table.
See this and
experiments in acoustic mitigation (including the elimination of some
discrete peaks,
and table layout recommendations) in the elog suite of:
http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=04/19/2003&anchor_to_scroll_to=2003:04:19:08:25:45-robert
L. Matone and M. Evans measured the full WFS sensing matrix (response
of all WFS and
QPDs to angular dither of all masses), finding reproducible results
(to 10%) on WFS1,
3 and 4, whereas WFS2 results are mixed. A sample sensing matrix
can be found at
http://blue.ligo-wa.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=04/20/2003&anchor_to_scroll_to=2003:04:20:01:22:51-luca
,
along with follow-up elogs by Matt.
R. Savage and D. Gustafson found that the third of three EOMs on the
PSL table had
RFAM levels ~100X larger than previously measured, and that the transmission
was 100X
worse than the other two EOMs. Upon inspection, radiation damage
to the crystal was
observed. The only spare unit available also has physical damage/imperfections;
new
units are on order and the old EOM has temporarily been placed back
on the table.
The elusive 2k POY beam was finally observed again; C. Vorvick found
that it exited HAM10
when an ASC bias was moved ~6 full slider units from nominal (D. Sigg
calculates POY may
well exit the port at nominal alignment if a larger or offset window
is introduced
during the upcoming vent).
Interferometer: The FSS servo is being updated with Rich Abbot. The
cabling
and software for the mode cleaner wave front sensor actuating on the
mode
cleaner mirrors have been installed. Testing is underway and the components
all seem to be functioning. The loops must now be closed. (Joe Kovalik)
I have been working with Andri and Joe K on measuring the parameters
of the
recycling cavity. So far we have measured only the Schnupp asymmetry.
We
working on recycling cavity length now. While doing the RC measurements
we
also characterized and reduced the AM coming from the EOMs. I am also
working with CDS group and Joe K on mode cleaner WFS. We have tested
the
signal continuity and are now moving on to measuring the sensing matrix
and
closing the feed back loops. (Valera).
We looked at coherence between seismic motion of the corner station
floor
(measured near beam splitter chamber) and beam pointing (pitch and
yaw) of
the mode cleaner (MC) transmitted light. Seismic motion in X direction
(perpendicular to the MC transmited light) shows high coherence (~0.9)
in
1.4-1.5 Hz and 2.3 - 2.4 Hz with both pitch and yaw of MC transmitted
light.
These frequencies seem to correspond to the first and second HAM stack
U-U
transfer resonance (at HAM1 U is parallel to x). Seismic motion in
x also
shows several high coherence peaks with pitch and yaw of the MC transmitted
light in 0.5 - 1 Hz where small optics suspension has resonances (pendular,
pitch and yaw). Seismic motion in y direction shows high coherence
(>0.8) in
1.5-1.6 Hz and fairly high coherence (~0.7) in 2.7-2.8 Hz. These frequencies
seem to correspond to the first and second HAM stack V-V resonance.
Seismic
motion in y direction does not show high coherence near the small optics
resonances (0.5 - 1 Hz). These observations indicate that the HAM1
table is
rotated by translational floor motion near the HAM stack's U-U and
V-V
transfer resonances, which is consistent with our previous measurement
of
HAM1 table rotational motion having peaks at 1.45 and 1.6 Hz. Interstingly,
MC length control has very low coherence with either MC transmitted
light's
pitch or yaw, indicating that the beam pointing fluctuation of MC
transmitted light is rather uncontrolled. We are planning to use these
reults in modeling the floor to HAM table motion transfer in e2e simulation.
(Sany)
CDS:
1) Implemented the MEDM screen for the new Watchdog System
2) Help troubleshoot the LOS Coil Driver board for the defective "Fault"
output
3) Worked with Jay on the new Mode Cleaner WFS.
4) Implementing the QPD Pitch and Yaw signals to the MEDM screen
(Ash)
GC: Much of this week has been spent getting the VPN up between the
sites
for LDAS. The VPN is up between MIT, LHO and CIT. LLO will be up tonight.
Also been working with the electricians on some network wiring issues
in the
new building, some of it involving the LDAS room and GC connections.
Worked
with the Matlab people to make all of our licenses terminate in the
same
month to make billing/renewals easier and also transfered the licenses
from
Tom's name to mine. Working on some backup issues. I think that we
have
outgrown our current tape library. The tapes are filling up too quickly,
which is getting expensive in both tapes and time. Starting to move
some of
the stuff at the warehouse in lLvingston into the lab in the new building.
Trying to get most of the stuff moved here and organized so that I
can make
less trips to the warehouse when someone needs something. Received
a quote
from Sun on a 280R with their new matching grant program. Looks like
I can
get a 280R for 50% list price. This is also good on Blade 2000s and
200s in
case anyone has a need for one. (Also 880s, but those are beyond the
scope
of the GC budget). (Shannon)
CDS: Moved all dtt/dvsave and ops users directories to the newly installed
A1000 disk array. Working with Jay to get the Mode Cleaner WFS boards
and
software installed and tested. Chethan.
LLO Seismic retrofit: Work on the Central mass reduction continues this
week
with good results. I have gone through 8 iterations in an effort to
find the
most favorable lightening scheme. I have also entertained changes in
the
boundary conditions and how they affect the mode response. The interesting
affect is that the frequency response is far more sensitive to boundary
condition placement rather than clamped versus pinned condition. The
original analysis was performed on a spring two inches shorter than
what is
installed at LASTI. This at least explains the lower than expected
frequency
in the LASTI HAM setup. I plan to finish this work by this weekend
(mode
shapes), so I can propose a final design and test for static stress
early
next week. (Marcel)
Working with Ken Mason and Ken Mailand to bring our drawings into a
state
appropriate for obtaining bids. Obtaining prices and preparing POs
for long
lead items - maraging steel, Kaman sensors and L4cs. (Jonathan Kern)
for two weeks ending 4/23
Held a design review of the External Pre-Isolation (EPI) effort. The principal aim of the review is to establish whether or not we are ready to commit to production and installation ofone of the EPI approaches at LLO. The review committee, together with the team, continue to debate the relative merits of the Hydraulic EPI (HEPI) and the electro-Magnetic EPI (MEPI); A decision is probably a few weeks away. Documentation for the review is available here
Jonathan Kern has started to organize the engineering effort to final design details and get bids and orders in place for the EPI production.
Last week I spent developing and tagging a version of the dmt softwarethat will compile under the gcc-3.2 compiler. Version 2.6.0 was released tuesday. I have also been looking at the data quality monitor rms band trends for the entire S2 run to determine whether there are any obvious data stretches that should be removed from analysis.
This week I have been working on setting up the automatic generation ofdocumentation for DMT classes. I have also started work on a new base class for use in event data analysis.
ASC code modified for LLO to include IO WFS signals; to be tested at LLO week of 4/21.
1. Working with members of the seismic pre-isolator
group to finalize the order of more capacitance position sensors.
The manufacturer sent us a probe and associated vacuum cable for RGA measurment.
2. Flavio re-sent the files to the board house for the ISS. A problem existed with an internal plane being treated as a negative image instead of a positive image. The result was numerous shorts on internal planes. The corrected boards will be at CIT next Tuesday.
3. Worked with the RFI test equipment recently purchased to gain proficiency. Most of the equipment will suffice, but the handheld probes fall short of what we will need. All items have been shipped to the sites for use there (RF spectrum analyzers, and RFI measurement equipment).
4. Designed, and with the help of Flavio, completed the CAD of an isolation amplifier for use in the hydraulic pump stand. The amplifier will provide separation between the noisy ground of the three phase motor controller and the pump servo.
5. Finished the first cut of the users manual for the pump servo.
6. Working with the RFI measurement team here at LLO.
So far we have completed high frequency testing on the Dawn and Elma crates
as well as the Equipto rack. Mechanical review and low frequency
magnetic testing is next.
7. The changes have been made to the FSS board per instructions from LHO. There are a few small issues to clear up resulting partly from the different optical topology here at LLO. The PMC pole is included in the loop here and should be compensated while we are at it.
8. Fixed a few small issues with the FSS documentation, and once we have consensus agreement, I will post the new revision in the DCC. The DCN will probably have to be re-issued to cover the changes specific to LLO.
9. Flavio has received the boards for the pump control servo from the board house. He will be testing them so they can go to LASTI and allow the pump control servo to be re-connected.
Jay Heefner reporting
Low Noise DAC: Frequency Devices has completed their tests and have shipped the DAC module to us for final evaluation. The prototype has been received at CIT and early tests indicate that the glitch problem has been fixed. Noise tests have been started, but are on hold for EMI tests at LLO. They will resume next week.
LOS Coil DAQ Whitening: Board layout is complete and gerber files will be sent out for fab by 4/ 18. Boards should be delivered by 4/28. All parts are on order or have been delivered.
EMI Retrofit: EMI measurements for the Equipto racks, Elma crates and Dawn crates are being conducted at LLO this week. Early test results are encouraging.
The performance of a modified composite amplifier was measured. Themeasured input referred noise was found to be 1.6 nV/Sqrt[Hz] at 10 Hz, which is about a factor of 2 higher than what was hoped for. The small, nominal, improvements in power supply bypassing did not appear to make any difference to the output noise and in fact may have made things worse out at 100 kHz and beyond. At least the bandwidth was flat to 2 MHz.
I am now debugging this discrete component amplifier. The amplifier does notwork as expected and I think the problem is to do with the biassing of the FETs used. Modeling the circuit in PSpice did not yield any clues, although I do not have a model for the FET. I have tried writing a Spice model for the FET but have had some hiccups deciphering the datasheet.
The warm up beam drift of the 10-W laser appears to be mostly in the vertical direction. A 40 micron variation was observed during the warm-up phase.
Ordered some pipe plug temperature probes to allow a pseudo-inline temperature measurement of the laser cooling water. After some minor re-organisation of the water cooling plumbing these were installed. The
measurements obtained thus far are not all that consistent. This is probably due to the orientation of the probes, as I'm not all that sure both probes are really immersed in the cooling water given the dimensions of the 1/4 in. NPT bushing. The plumbing will have to be done again at a later stage. However when the laser is on standby, indications are that the heat load is approximately half the cooling capacity of the current chiller. This suggests that the long run of hoses at LLO, at least, is quite a heat load.
Larger cooling capacity chillers have been ordered for the observatories. The chillers should be more than ample for cooling the laser, taking into account the long hose lengths. Plus they still fit into the interstitial wall space where the current chillers reside.
The photocurrent versus incident power was measured for a high power photodetector. The output was linear up to the maximum design photocurrent of 175 mA --- corresponding to an incident power of 267 mW, as measured with a calorimeter. However running at this level caused the circuit to thermally cycle after 30 minutes. Letting the circuit cool down, restored its performance. This should be alleviated by an appropriately sized heat sink.
The recent repair of the surface measuring interferometer appears to have made things a bit worse. I have performed a three flat test and get a repeatability of roughly 4 nanometers. Originally it was about .75. This is not sufficient accuracy to measure LIGO optics which have an rms requirement of <.8 nm. I will try some combined calibrations (using old data) but have little hope that the instrument can be made useful in its present configuration.
Ken is working on redesigned baffles with increased
stiffness. A baffle-mounted cavity beam dump for the RM reflected beam
is being designed. Preliminary calculations of back-scattering of an on-axis
beam from the RM onto the MMT3 baffle with re-scattering into the interferometer
cavity through the RM indicate a negligible noise contribution. The scattering
analysis is being extended to assess the effect of the main beam being
offset from the center of the RM.
The reflectivity @ 1064 nm of a sample of the oxidized stainless steel material used for the LIGO beam tube baffles at 5-17 deg incidence angle was measured by Peter King to be <3%. The absorbing surface film was created by baking in air at 850 deg C for 10 minutes. This absorbing surface treatment will be used on all the stainless steel errant beam baffles.
Top level design is complete for all errant beam baffles. The RM reflected beam cavity beam dump design is complete. Working drawings are expected to be completed by 4/25. The upper limit of scattering from the surface of the RM with re-scattering from the MMT3 baffle into the IFO causes a negligible scattered light noise. The errant beam baffle DRD and Conceptual Design document is almost ready for review. West Coast Porcelain, the source for providing the absorbing oxide coating on the SS baffles, was contacted and they can process the baffles in approximately a 1 week.
I am writing a fortran program which manipulates the grids that represent theelectromagnetic fields at the anti-symmetric port in order to compute the ASQ and ASI (D.C. signal). The overall aim is to investigate their changes caused by contrast defect and misalignment.
OTF Lab. (W. Bridge)
Contamination Cavity # 1
The test sample Kapton still pumping DOWN.
We are taking ring down and beat frequency measurements everyday.
Absorption Test Measurement prototype in STANDBY.
Scatterometer
The function of the scatterometer has been fully recovered as mentioned
before.
Measurements of AR, HR reflectance, scattering and transmission
of different mirrors are completed.
The AR reflection for the large mirror ITM S/N 2ITM03 after the first
cleaning has been released.
The reflection of the AR coating is ~2400 ppm.
We have also finished taking measurements for the
AR reflectance of the WITNESS SAMPLE FOR THE
large ITM R= 1300 ppm . This agrees with the previous measurement
for the large mirror ITM08.
Helena will clean the large mirror so we can take new measurement of
the mirror 2ITM03.
We have also finished taking measurements for the
HR transmission for two 1.0 inch wedge mirrors
the results are: 7.4% +/- 0.1% and 7.5% +/- 0.1%
and if we consider the 2nd surface not AR coated
then we'll consider the loss (~3% loss) and the results will be 7.7%
and 7.8% respectively. The manufacture claims
the transmission to be 8% for both mirrors.
As the time progress, we have been updating and improving
our optical train set-up for the entire scatterometer
enclosure. So far, we need more 2" inches mirrors, lens and mounts
in order to catch the reflected beam more
precisely without having to disturb the alignment.
OTF Lab at Lauritsen ROOM 38
Cavity #3
The new test sample SR FR HF from Chem-Sol, Inc. is an Undyed Fire
retardant Hydraulic Fluid.
We are taking ring down and beat frequency measurements everyday.
Cavity #2 Test cavity
The chamber is pumping with new cleaned mirrors ~ 70ppm each.
RGA measurements and optical train set-up is in progress.
no report
MEPI Weekly
Following a concern about the DC range of the magnetic actuator
we
measured the offset as a function of the input current and found about
75
microns/amp in the horizontal direction and 30 microns/amp in the vertical
direction.
In order to measure the low frequency performance of the system
we have
installed a low frequency seismometer (STS-2) on the support table.
It is
mounted on a platform which is connected to the end of the support
tube,
outside of the vacuum chamber.
EPI Design and Installation
Completed a report on structural modifications to the HAM and BSC chambers
with external pre-isolators.
http://www.ligo.caltech.edu/~coyne/IL/EPI/review2/T030073-02.pdf
Also completed assembly drawing with links to detail drawings,
installation procedures and purchasing information for the MEPI actuator.
http://web.mit.edu/kmason/www/mepi_actuator/d020183-02.html
Simulation and Modeling (Bhawal)
E2E Physics meeting
--------------------
There was discussion on current status of the 4K asymmetry problem,
wavefront sensor, new improvements in modeler and problems in dynamical
misalignment studies. Matt, Virginio, Hiro and Biplab attended.
ASC @ LHO
--------------
(Matt) I worked closely with Luca on the H1:ASC system. We developed
and
tested several tools for measuring the 7x7 alignment sensing
matrix, and
for processing the sensing matrix in Matlab to generate a control
matrix
that approximately diagonalizes the ASC plant. Work continues
at LHO in
the effort to control all alignment degrees of freedom.
Outreach effort
-----------------
(Hiro) Worked with Sany and Tiffay to work out a problem of the pendulum
simulation. The mysterious observation, very weak Q dependence
of the
pendulum motion, turned out to be caused by the shift of the
resonance
frequency due to the coupling of the pendulum mode and pitch
mode. This
shift is very small (a few % for a test system they are suing),
but is
large enough to introduce confusion for a high Q (even 30). This
was a
very good case of the model validation.
An updated primitive now prints out the resonance frequencies
of the
coupled system as well.
LLO seismic motion
-------------------------
(Hiro) Sany Yoshida keeps studying the seismic motion at LLO and the
behavior of the mode cleaner. This will be documented in a LIGO
note.
The modeling of the seismic motion of LLO using Sany's measurement
is
still going on.
Dynamical misalignment
------------------------
(Biplab) The transfer function calculated from the outputs of 2 half
photodetectors placed at dark port for differential misalignment
modes
in end mirrors looks incorrect,i.e. it is not giving flat response
at low frequencies. Also for all transfer functions, although
peaks are
appearing at right frequencies, the dips are not coming at frequencies
specified in Daniel-Nergis' paper.
Dual recycling Michelson cavity
------------------------------------
(Hiro) Osamu has been working to formulate the dual recycling Michelson
cavity based on Hiro's summation cavity modeling. There was a
meeting
discussing the progress. A rigorous treatment turns out to be
difficult,
and Hiro suggested Osamu to substitute a realistic values (velocity
of
mirrors, cavity length etc) to see if an approximation keeping
the
leading order of an expansion parameter (mirror motion per simulation
time step / lambda) is practically correct in the current formulation.
If this is justified, a matrix manipulation can be done and the
primitive can be written.
Code development and maintenance
---------------------------------
(Hiro) Several improvements are going on to speed up the simulation
and
to make it easy to setup simulation configuration.
(1) all outputs can be directed to a directory different from
the
default. This makes it possible to run different configurations
at the
same time using same set of boxes.
(2) output data streams can be controlled. For now, it is impossible
to
disable outputs without manually disconnecting output ports.
New scheme
allows to disable outputs using dhr files. E.g., SimLIGO now
has several
hundred outputs and not all are important always and it takes
time to
write out and read in the data file. Now you can selectively
turn on/off
the output channels, like outputs only related to COC and LSC.
(3) Melody has improved the expression parser code. This is being
including in modeler, which can speed up SimLIGO possibly by
up to 30%.
Alfi
-----
(Bruce)
- Fully implemented GenericPort superclasses now used by
several Alfi classes, and to be used by Bundler classes.
- Cleaning all code which can now use these superclasses
instead of having to cast plain Objects to exact classes.
LIGO Data Analysis System
Software Systems (Blackburn)
Began development of new functionality in LDAS to support generation
of RDS
frames using the createRDS command from multiple IFO input frame sources
and
merging the results into single frame files containing multiple IFOs.
This is
complicated by the diversity of frame file structures present in other
projects
such as GEO, VIRGO, TAMA and now the NDAS frames. The requirements
are to support
all of these, but it may be too much to implement for the 0.7.0 release.
The
design is being made general with the possibility of restricting implementation
to LIGO frames if time constraints dictate.
The migration of LDCG, LDAS and LAL to work under redhat 9.0 has been
a
struggle. We now have a working LDCG environment and has shown that
LAL will
compile and pass its unit tests in this environment. But LDAS is proving
much
more difficult to get to work with the newer autoconfig and automake.
We will
continue to pursue this with the possibility of accelerating the migration
to
the newer automake and autoconfig instead of trying to develop a solution
that
works for the older and newer tools at the same time.
Work continue on developing a more compatible LIGO_LW XML document to
be shared
with the DTT. We will be implementing base64 writing capabilities this
week and
will post on the LDAS-SW web site the XML DTD that describes the XML
documents
we generate. However, will will continue to insist that allLIGO_LW
documents
generated by LDAS have this DTD within the document so that network
problems
do not prevent users from reading LIGO_LW documents generated by LDAS.
A bug was found in the managerAPI which cause LDAS account management
to fail
on the LDAS-DEV system. This has now been fixed and it is once again
possible
to add users to this system.
The code base for LDAS tested over the weekend showed reliability levels
at
the highest level seen this year. More than 99.98% of jobs passed.
The struggles
with reliability seen in the LDAS versions used in the second science
run may
finally be coming to an end.
The metaDataAPI improvement (removing the unnecessary slave interpreters)
did
double the performance on the full LDAS development system. In addition
a bug
in retry code for inserting data into the database was discovered and
fixed
this week.
The controlMonitorAPI client had about a dozen enhancements added which
had
been on hold while we froze the CVS repository during the science run
to
assure that we could react quickly to any needed patches for LDAS.
Now that
that phase of our lives is over, the changes are starting to roar into
CVS
in preparation for our next release.
A complete summary for the usage of LDAS at the sites during the science
run
has been completed in preparation for a "lessons learned" meeting this
Friday.
The top level results are the following
LHO LLO
total jobs submitted | 417870 |
228136 |
total jobs succeeded | 409274 |
219206 |
total percent failure | 2.06
| 3.91 |
% failure due to LDAS bugs | 0.65 | 0.28 |
# of rows inserted in DB |24378362|15215169|
These span Feb 14th through Apr 14th, noon to noon Pacific time.
Philip Charlton:
* RDS data production is continuing, although there have been delays
due
to the necessity of backing up some of the raw S2 data on tape. I am
27% complete with LHO and 29% complete with LLO. There still only 2
tape
drives available so the current rates of production is that the ratio
"RDS data produced:wall time elapsed" is about 2.4:1.
* LDAS coding: worked on extending Statistics class to work on complex
arrays where possible and including new sum() action. Code for this
item
has been completed and I'm now working on modifying the test code for
it.
* LDAS testing: retested datacon MDC tests, createRDS, reading/propagating
of calibration data, read/writing of frames with proc channels on latest
build of LDAS after the Swig upgrade.
Hardware Systems (Anderson)
******Stuart is on vacation*******
LHO (B. Johnson reporting)
1) Assisted in the assembly of three computer racks for the upcoming
beowulf cluster.
2) Assisted in the hardware installation of the new StorageTek L700e
tape library.
Installation completed on 4/23/2003.
3) Worked to diagnose continuing problems with an ASA Computers
beowulf
node. Set up self
as LHO contact person for this problem.
4) Assisting Dave Barker with installation of weather station
monitoring/operator alarms
in the LDAS computer room.
From: Dan Kozak <dkozak@cacr.caltech.edu>
Finished archiving all S2 data (copy from AIT2 to SAM-QFS). An
experiment with trying to set up different tape pools in SAM resulted
in a bunch of data only being single copied to 9940 tape, but this
has
been corrected. There were no tape errors found during the staging
and
rearchiving of the single copies and no data was lost. During
this
process, access to SAM-QFS was restricted for efficiency's sake, but
now RDS data is being generated again.
Chased StorageTek vis a vis getting our new 9940B drives installed.
The drive wall is supposed to arrive today and installation of it and
the new drives is scheduled for Monday.
Obtained and installed new, permanent SAM-QFS license on LLO's
dataserver machine.
Data Analysis Activities (Lazzarini)
Charlton:
* FCT -- for the next FCT paper I've written more of the section dealing
with
numerical differences between correlations calculated via matched
filtering vs. the FCT. I have an expression showing that the SNR recovered
by the FCT is slightly less than that recovered by matched filtering,
and
am able to give a bound on the difference. So, given a filter bank
for inspiral chirps which is known to recover at least (say) 95% of
the
SNR of all signals in its domain, it is possible to calculate the
FCT grid density required to recover at least (95-epsilon)% of the
SNR
for any epsilon.
* Chirplets group meeting -- in the past 2 or 3 weeks I've been meeting
with Emmanuel Candes' group in
the Department of Applied & Computational Mathematics. The group
is
working on chirplets and algorithms for fast chirplet transformations
which may have applications in LIGO data analysis.
Creighton:
Last week I presented the pulsar upper limit group's results for the
S1 analysis. Since then I've been working on some finicky final
details of the analysis: most recently, the potential impact of the
200 microsecond time offsets in H2.
Lazzarini:
Completing the systematic error analysis for the stochastic background
paper.
I am looking at the effect of variying calibrations and fluctuation
noise floors
during our 15 minute segments of data that are analyzed as one. No
big surprises expected, however we will be able
to estimate the systematic +/- errors that need to be added to the
statistical errors in our final results.
Mendell:
Mostly I have been preparing for LDAS installations and getting Ben
Johnson started, so not much data analysis work to report from the
last
week. I am continuing to assist the effort to finish the PULG S1 paper.
Shawhan:
* Worked on validating the S1 inspiral upper limit result.
* Reviewed the S1 pulsar upper limit paper.
* Working on validating S2 science-mode segments for offline analysis.
Weinstein:
- Working on bursts paper.
- Working on hardware burst injections.
- Prepared and gave a CaJAGWR seminar on the S1 burst results
http://www.ligo.caltech.edu/~ajw/bursts/burstCaJAGWR_030422.ppt (20
MB)
http://www.ligo.caltech.edu/~ajw/bursts/burstCaJAGWR_030422.pdf (7
MB)
General Computing (Wallace)
MIT:
(Keith)
-Updated modem portmaster to use spare IP addresses
-Brought up console screen on logging server
-Spec'd & ordered new laptop for scientist
-Looking into official certificate for calendar server on emvogil-2
via the DOEGrid Certificate Authority
-Put gc servers (file/mail/NIS+) onto UPS power
-Installed extra RAM into several Sun desktops
-Looking into pc desktop crashes for user
Livingston:
(Shannon)
-Much of this week has been spent getting the VPN up between the sites
for LDAS. The VPN is up between MIT, LHO and CIT. LLO will
be up
tonight.
-Also been working with the electricians on some network wiring issues
in
the new building, some of it involving the LDAS room and GC connections.
-Worked with the Matlab people to make all of our licenses terminate
in
the same month to make billing/renewals easier and also transferred
the
licenses from Tom's name to mine.
-Working on some backup issues. I think that we have outgrown
our
current tape library. The tapes are filling up too quickly, which
is getting expensive in both tapes and time.
-Starting to move some of the stuff from the warehouse in livingston
into
the lab in the new building. Trying to get most of the items
moved here
and organized so that I can make fewer trips to the warehouse when
someone needs something.
-Received a quote from Sun on a 280R with their new matching grant
program. Looks like I can get a 280R for 50% list price.
This is also
good on Blade 2000s and 150's in case anyone has a need for one.
(Also
880s, but those are beyond the scope of the GC budget).
Hanford:
(Christine)
- Trying to get the quote for the Foundry GigE network upgrade going.
- New account set-up and orientation for a new hire.
- Purchases, purchases and more purchases. Some for CDS, some
for GC
and some just consulting for other groups.
- The usual user support.
CIT:
(Lisa)
- Setup unattended rsync between rigel and the new web server.
- Have been aggressively moving people over to using becrux.
This includes
changing those who tunnel and setting up tunneling for those who haven't
been.
Many people forget that when they connect from home they still need
to tunnel or
use their mail client's ssl ability. If anyone is still having
trouble with
their e-mail logging out after 60 minutes, let me know and I will fix
it.
- Added more laptops to the wireless access.
(Veronica)
- LIGO web: most time was spent working on the Advanced LIGO webpages.
Going through the new software on a new machine for video capture and
processing. Captured the first video to a file from a miniDV
tape
archive. Eventually the entire LIGO miniDV and analogue archive will
be
captured and archived in a non-degrading format.
Support for Caltech Media Relations office in tracking a video.
Various updates to the LIGO website.
- LSC web: posting late-arrival talks from the March meeting. posted
updates to other parts of the website.
- CaJAGWR website: captured a video of Tuesday talk and processed it
for
posting. Did usual upkeep of the website.
(Mike)
-Loaded a new laptop for Bill Kells with all GC software an additional
software
packages; plus I transferred his data over from his old laptop to the
new one.
-Swapped out Ben Abbott's computer and brought over his old equipment
over to
Bridge Annex to reload his old computer to swap out one of the older
workstations.
-Flavio Nocera, was having some hardware problems with his PC that
I corrected,
plus I loaded some additional software on his computer.
-Swapped out a new monitor for Ken M. and worked on some software issues
that
he was having.
-Had to do some heavy maintenance on the plotter for this printer went
haywire. After replacing several parts and reloading paper etc. etc
I
finally manage to get this printer backup and running.
-Larry and I worked on optimizing the VRVS software/hardware to
produce a better video/audio output but seem to get know where with
this
system. Larry contacted VRVS support and they have verified that the
output
that we are currently getting, is the best output that we can possibly
get
without the system crashing.
-Worked on the organizational chart for Ed Jasnow.
-Finished up loading a new PC for Veronica and have moved this into
her office.
-Received some new equipment that I had to unpack and test to insure
these
units were not damage and in working condition. This included 3 laptops,
2
monitors, 2 PC's and 1 2U server.
(Larry)
-Spent some time dealing with minor issues concerning the WAN being
installed at
Hanford. So far things look to be on track. The router is to ship in
mid May.
The PO# has been given to Amerion, so they can continue with their
work for the
installation to PNNL. We are now looking into the long-line converters
that will
be needed.
-Worked on a number purchases. Sent out a notice from SUN concerning
their deals
and it looks like a number of people on the project are taking advantage.
-Resolved a number of e-mail issues. Spent time with Thomas going over
how
things need to work so mail aliases can be generated from a database
he is
working on.
-Helped Mike out on a couple of computer installs and a Org. chart
and plots
that he needed to get out.
We also spent time working on the VRVS setup. In addition to the limitation
of
the video resolution of VRVS, we ran tests using Netmeeting but the
audio/video
portion of Netmeeting is point to point and doesn't have multipoint
access as do
other parts of the program. We are working on setting up the sharing
procedure,
which will help in improving the quality of the video of electronic
presentations.
Also, ran a few tests on SUN's Sunforum pkg. so far it looks to be
pretty good.
-Provided some minor support for the DCC.
-Still working with Lisa on the move to the new web server. Presently,
working
out some of the logistics so we can have a smooth transition. Received
a few
good ideas, from the user community, that I am looking into.
Performed a few more security checks on some of the servers and started
making
minor changes to take care of those issues.
Subject: [Aligo_systems] minutes from today's
meeting
Agenda:
1) Preparation for the NSF Review of the Advanced LIGO Proposal
David Shoemaker
6/11-13 at CIT
full days -- committee -- breakout sessions -- 'classic' review
all aspects of proposal (LSC science support, outreach, etc., i.e.
not just the instrument design and astrophysics)
Plenary talks on first day
5 breakout topics:
need someone in each breakout familiar with costing as well as technical
details
Roger Falcone (UC Berkeley) chairs the
review panel; other members include Don Hartel
, Sidney Wolfe, several outside US and LIGO (Giovanni Losurdo, RaffaeleFlaminio,
Francois Bondu), and some new optics/materials
people from NIST (unfamiliar with LIGO)
2 weeks before set of documents posted on web; review panel meets 1 wk before review & will pose Qs
2) Brief updates from each of the Subsystems on key
research and design points and, in particular, systems or interface related
issues or suggestions
All -- please think about what issues are as yet unresolved and for
especially for which there is no action plan. Let's put these issues on
the table for discussion; They may form the
basis of future agenda topics as well.
0) System
layout on tables very tight -- thermal comp combined with ITM quads?,
FM & ITM quads combined?, shorter COS beam reducer or put beam reducing tel
on HAM5?
*** topic for next meeting
*** DC readout decision not formally made yet; affects output MC --
could do in 8-12 wks
draft OO layout by PF et. al. sooner
variable SRM also adds mass -- ACIGA design
support SUS cage from below? last resort
1) SEI
RFP for mechanical design of LASTI prototypes, HAM & BSC
issues from preparation -- weight/mass estimate -- near limits of spring,
crane, envelope
ETF freq OK (fastener stiffness OK)
magnetic field measurements yet to be made
adjustable dovetail --> fixture OK;
*** need a plan for these measurements
2) SUS
4 months concentrate on fleshing out quad designs
concern about spring internal modes being rung up by SEI -- adding
eddy current damping
*** need a better minimum frequency requirement for payloads on the
SEI platform (currently it's <= 150 Hz)
Norna is working on FS quad & sapphire
common design
FS quad FS seems to fit into current envelope
3) COC
SMA virgo sapphire mass absorption patterns
("swish") -- 30 to 130 ppm/cm variation on several cm scale, no symmetry
need to feed this measurement to Dave O. for thermal comp
FS supersil 312, CSIRO polish, no coating
measures 120e6
no other large pieces for LIGO Lab in the pipeline -- GEO has a large
sapphire piece in transit; could measure absorption by Jean Marie -- Gari
to ask
5 vendors -- letter of interest -- 3 expressed interest
4) Laser
measurements 2 weeks before LSC
no laser at Adelaide
80 & 65 W
project to 200W -- no clear winner
rod osc. at LZH chosen for cost &
schedule reasons
goal 200W in 1 yr -- redesigned laser head
PSL ISS -- Jamie Rollins -- ~factor of 10x than L2 goal to ~10 Hz
LZH pursuing in-vacuum PD -- will try ISS, are now about 10x over the
2e-9/rHz
5) Input Optics
2 sci at LLO 5/1 -- high power test
laser purchased & will start at LLO soon
adaptive mode matching experiments started -- need longer, lower doped
materials
design req review -- periscope on PSL
table -- addressing in initial LIGO, or periscope in the vacuum system
RTA EO mod works up to 50W at L2 power densities
6) ISC
DC readout is likely to be the new baseline
acoustic sensitivity --> put *all* sensing readout in the vacuum system?
A dry run of the presentations to be made at the SEI Structure Proposers'
Conference was conducted. Two more of the prospective proposers
have bowed out, leaving five still in the running. All five appear to be
committing to come to the conference, to be held at Stanford on 25 April.
ADVANCED LIGO
1) ROTATIONAL ADJUSTMENT OF UPPER BLADES
The drawings for 2 prototype rotational adjusters that will allow crude
yaw adjustment of the suspension wrt the
structure are in the Physics shop. A similar set up will be used for both
the quad's and RM suspension, therefore it is an important part of the
evaluation of a quadruple pendulum
2) BLADE TESTS
Drawings for a blade testing facility are nearing completion, these
should be available for workshop later this week. This will allow us to
more easily characterise the larger cantilever
blades used in the quadruple and RM suspension.
3) QUAD MATLAB MODEL
I have been working with Norna on the
parameters for the proposed quadruple pendulum suspension for LASTI. Mike
Lloyd and I have created an assembly of an upper mass to aid with this
work
4) VISITS
Mike Lloyd will visit Caltech again from mid-June until mid-July. I
plan to visit Glasgow and GEO around the time of the NSF meeting in August.
Russell Jones another of the young engineers in
From: JaneenRomie
<romie_j@ligo.caltech.edu>
AdLIGO Suspensions
Looking at short and long term schedules and
tasks. Nothing significant to report.
Gin Gin
Mechanical fabrication is almost complete. Travelers for cleaning and
baking have been created. Assembly specifications have been created for
both optics. We will go forward with cleaning and baking the parts
but will delay suspending and balancing the optics until we hear from John
Jacob on a delivery date.
40m
Performing small/various 40m tasks.
One can glean more from the power point original, as the images are
stored with a larger size than the display. E-mail Billingsley_G...
if you would like a copy.
From: Helena Armandula
<ahelena@ligo.caltech.edu>
Advanced LIGO Coating Development Plan
Received a few questions from Dale Ness (REO),
he stated that REO is interested in participating in LIGO's Coating Development
work.
Before the end of the week I will contact the other vendors to confirm
they received LIGO's Letter of Interest package.
Gregg McIvor joined us from Drexel for a 5 month stage.
Alessandro in
Mike Hall just accepted a graduate student position at Cornell. Congratulations.
Francesco C.
Made tests on StefanosMoRuB
test, with again deceiving results. See analysis from Alessandro below.
Charles
Sent in draft of his stage. He chronicled
the brazing and stress strain tests of his stage.
See analysis from Alessandro below.
Alessandro, Riccardo
After a search on literature Alessandro found that, although probably
tough enough, the S-bond has low braze to
metal binding strength. Even brazing to normal metals
the shear yield point of as low as 20MPa (to be compared with the 5GPa
expected tensile strength of MoRuB).
Considering the cross section ratio (10-5/6 10-3) and in absence of shear-lag
(concentration of the shear effort on the edge of the braze joint, see Xaviers
and Stoyans studies) the safety margin
would be only two. In presence of shear-lag, the
braze with S-bond should always fail with a perfect sample. This
is true for thinned (10 micron) flex joints. For 50 micron joints, as used
in Charlestests,the braze
should also always fail in absence of sample defects and misalignments
of the jaws of the stress/strain machine.
Partial failures of the braze also explain
nicely the strain jumps and the large histeresys
observed by Charles and the other results from Stefano.
Finally the low adherence, coupled with the strain concentration at
the edge of the braze point at the base of the diving board, nicely explain
the deceiving and mode dependent Q-factor values found by Francesco F..
For the next few months we will complete some measurements to clarify
a few more points and complete our data base of MoRuB,
but clearly MoRuB, in its present formulation,
cannot be practically used for mirror suspensions, unless a suitable metallurgical
bond is found to connect it to the mirror suspension structures.
The only possible solution is to pass to a material that can be vitrified
in larger thicknesses to allow brazing on larger surfaces and with negligible
amounts of shear lag, or to allow manufacturing non shear braze joints,
or to avoid the shear brazes altogether.
We will search for possible candidates.
Gregg
Great start. Solved the problems of the splatter and brought it back
to close to 100% efficiency. The two key solutions included careful cleaning
of the coil (with HCl and NaOH
alternated) AND (with Bills help) tuning the inductance matching between
the coil and the RF generator. Bill did not make us aware of that requirement
and we never thought of doing it.
35 splats made in a few days.
Eric
Trained Gregg on the splatter. Molten many MoRuB
18 balls for Gregg.
Having problems in reproducing the BCC crystalline
structure of last report.
Analysing the X-ray spectra with fast Fourier
transforms to measure the glass/crystal fraction in samples.
Measuring X-ray cross sections on our samples to
determine the glass/crystal fraction in samples.Investigated
importance of polishing in increasing the overall cross section.
Aso
Simulating a new suspension system for a possible Vertical Suspension
Point Interferometer. Looking into possible practical
designs.
Riccardo
Boxed the old prototype of TAMA tower for shipping to
For additional information about this report, contact sanders@ligo.caltech.edu