|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday February 26, 2001 will be:
CANCELLED DUE TO NSF REVIEW
no report
WBS 1.2 LIGO Operations--Administration
There was a site teleconference held on Thursday, February 22, 2001. The cost reports as of the end of January 2001 were reviewed.
Copies of a purchasing manual have been provided to the site heads to satisfy a need for guidance concerning the use of Caltech accounts. However, it has been decided that these documents do not provide the information required, and the action to acquire a chart of accounts remains open.
The next site telecon is scheduled for Thursday,
March 8, 2001. The list of current actions revised to reflect open
actions assigned through February 22, 2001 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
| Packages | Faxes | |
| In | 35 | 28 |
| Out | 9 | 22 |
Press here to access the DOCUMENT CONTROL CENTER WEB PAGE.
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)
Accomplishments:
None to report at this time.
WBS 1.4.1.2 Project Controls (LIGO Construction)
Working on material for the NSF Review scheduled at the end of the month in Hanford.
A reminder that I will require contributions for an end-of-February Quarterly Progress Report in March. I would like to set a target date of Friday, March 23rd. Thanks!
The following Change Requests have
been submitted:
| CR-000018 | WBS 1.1.4 | Curbing for Service Roads at Livingston | G. Stapfer |
| CR-000019 | WBS 1.2 | Additional Lab Equipment | D. Coyne |
| CR-000020 | WBS 1.1.4 | Staging Building and Renovations to Existing Building--Livigston | F. Asiri |
| CR-010001 | WBS 1.1.4 | Return of Unused Construction Budget To Contingency | F. Asiri |
Press for the latest Contingency Needs Projection.
From: Kris Duncan <kris@ligo.caltech.edu>
From: Ed Jasnow <jasnow@ligo.caltech.edu>
General Items:
--------------
(F. Raab)
WA4K Optics Installation:
-------------------------
D. Cook
The corner station now has its full compliment of LOS core optics and
most
of the SOS optics installed. However we discovered that the OSEM sensor
voltages vary greatly after the optics are aligned due to a much higher
degree of sensitivity. As PAM magnets are adjusted to fine tune the
optic
positions it takes very little movement to change sensor outputs which
in
turn affects the damping control. To remedy the problem we are having
to
reset the OSEM 50% light level positions in situ, which requires
re-adjusting the PAM screws to maintain to the aligned position. It
takes
several iterations of these moves to complete the alignment. Unfortunately
this was not discovered until we neared completion. We will set up
on at least
three of the LOS optics and repeat this process. The SOS optics have
not
been checked as yet, but it is likely that we will need to make some
adjustments as well. It is our hope to complete the fix in the next
few
days and not greatly impede the COS alignment and installation. The
ETMy-4k
installation will follow on in the near future. We have the next
set of
OSEMs (57) out of the vacuum bake and ready to process. Again it has
been a
large group effort to keep the installation moving ahead and for the
most
part sucessfully.
| Installation&
Commissioning:
Livingston |
Other Science/EngineeringActivities:
Issues/Concerns |
See also the Installation web page
We have started to use a single arm cavity to characterize the frequency noise of the light after the mode cleaner as a function of the gains/filter settings of the new MC servo. The data still need a detailed look, but the general behavior seems to be as expected.
We installed one of the new Watec cameras looking at the beamsplitter. When the full interferometer is locked, some scattering centers can be seen on the face of the optic. The vertical position looks close to center, but horizontally it appears to be about 5 cm left of center.
We found having the alignment equipment from the 2 sites to be a big help in improving the time for a core optic alignment.
Stan Whitcomb, Dennis Coyne
Discovered that although
the shadow sensors in the new OSEM heads have a larger linear range, they
have a much smaller total range. The consequence has been that it is easy
to adjust the shadow sensors out of their range when using the PAM magnets
to align the optics. In addition, the alignment of the two ITMs shifted
significantly (~70 microrad) after the elliptical baffle was installed
instead of the couterweight. It may be that an imperfect weight substitution
in the elliptical baffle replacement causes a slight optics table mis-balance
which is coupled via the magnetic Kovar, on the lens seals for the photodiode
and the LED, and contributing to the misalignment. This is a serious enough
concern that we have stopped assembly of the LED & PD circuit boards
into the OSEM heads until the issue can be investigated further.
We are close to schedule.
Correlation measurements between the motion of the mirrors and the fluctuations
in the transmitted power of the locked cavities, now (convincingly) point
to a strong coupling between the final mirror in the mode matching telescope
(MMT3) and the transmitted power. The large peak at 0.74 HZ seen in the
transmitted power is due to the poor damping of MMT3. The cause is a varying
misalignment of the input beam relative to the cavity optic axis. The reason
for the poor damping is being
investigated.
The optical lever damping servoes have been activated on the recycling mirror, beam splitter and MMT3.
The Nd:YAG oscillator (MISER) in the laser has been losing power continuously for the past several months. This week the output fell below 40 mW (should be 700mW) and the entire laser system no longer functioned reliably. The MISER is being replaced at the moment. (see discussion on the PSL below)
An attempt is being made to bring the vacuum control software at Livingston to the same state as that at Hanford. It is not going easily.
Since the frequency servo loop gain relies on having at least a certain amount of light incident on the reference cavity, it was decided to turn up the master oscillator (NPRO) diode current. Especially since the NPRO power monitor suggested the output power of the NPRO was approximately 50 mW - its usual value is around 500 mW. The diode current was increased from a nominal 2.0 A to 2.2 A. This resulted in the NPRO power increasing to just over 100 mW. The frequency servo was then re-locked without any apparent problems. Once the modecleaner was locked, everything appeared fine out in the PSL enclosure. About 30 minutes later, the modecleaner beam pulsed sporadically and the cause was traced to glitches on the fast actuator.
After deciding that changing the NPRO current one more time would probably not sufficiently extend the life of the laser, it was decided to replace the NPRO. The PSL spare NPRO (SN #238) was installed into the laser. This required a full re-alignment of the laser and phase-correcting Pockels cell located between the NPRO and the power amplifier. The main beam was then aligned into the pre-modecleaner (PMC). The sample beam was re-aligned and modematched into the reference cavity. All of the OptoSigma Ultrastable mounts in the sample beam path were replaced with Ultima U100-AC38s. The frequency servo was engaged and a quick inspection of the in-the-loop noise suggests that the frequency servo is performing as it should.
Some problems are being encountered with the PMC servo in that once the PZT is adjusted to the correct length, engaging the servo results in a 50 V shift in the PZT output. This makes the PMC difficult to lock. The PMC servo does lock when the mixer offset is adjusted. However the setting for the mixer output offset adjust seems to change. The good news is that both the PMC and frequency stabilization servo are functional.
Ben Abbott
I continued to redesign the DC photodiode for the ISC. Modelling
is being done in Intusoft SPICE.
The mount holding the phase-correcting Pockels cell was re-designed to accommodate a number of space constraints imposed once the assembly is placed into the 10-W laser. These are currently being fabricated. The mount permits adjustments to the optical height, pitch, yaw and rotation about the optical axis of the Pockels cell.
A small quantity of precision flats have been ordered to permit practising the technique of optical contacting. Assembly of the first reference cavity will most likely be done later this month.
Rick Karwoski, Paul Russell,Ben Abbott, Sander Liu
Macro Metalics is overnighting their two TiCr samples (one virgin, one acid etched), due here tomorrow. They said they'd quote today for the polishing and acid etching. I already have the coating quote.
The rectangular connectors for the pigtails were incorrect. CES is making 105 rectangular connectors, needed for the 2k IFO, due to be completed by Monday 2/26. This should not affect Oliver's estimated pigtail fabrication delivery date of Friday, 3/2, for 105 pigtails. Mike at CES hopes to have the other 395 rectangular connectors fabricated by Tuesday, 2/27. After we install the pins in the connectors, an outside vendor can resume pigtail fabrication. So. California Braiding quoted for pigtail assembly work. Pal Pilot is due to quote soon; an RFQ went to them last Thursday, 2/15.
Mark Barton
This week I've been working to include the magnetism in the Kovar lens
seals in my model of the optic, optic magnets and PAMs to assess the effect
of any cross coupling.
This week we continued studying different control topologies for the
system consisting of the laser and one of the suspended test cavities,
which include control loops to the laser pzt driver as well as one
of the
mirrors' actuators.
We succeeded in running a configuration with a unity gain frequency
of 400 Hz and a crossover frequency of ~180 Hz. Despite its simplicity
and servo stability, this setup was not robust enough to be satisfactory.
We settled for a control setup consisting of a unity gain frequency
of
~12kHz and a crossover frequency of ~200Hz.
Once settled with a satisfactory control topology, we re-characterized
the system, this time by locking not only on the TEM00 but also on
the
TEM01. In particular, we focused on frequencies below 40 Hz: last week,
we
stated that there was clear agreement between our matlab model and
the
measurements only for frequencies above 40 Hz. We carefully repeated
the
measurement below 40 Hz and realized that the noise level and the overall
gain is too high for a SNR>1.
Once the system characterization was completed, we began investigating
the fundamental noise sources in the instrument. First, we studied
the
readings from the optical sensor head, when the local controls were
on
and when they were off. This gave us a qualitative understanding on
the
mirror motion along the cavity axis. We could clearly observe the
pendulum resonances around 1 Hz, as well as the stack resonances and
pinpoint the OSEM readout noise level. At the moment we suspect that
the control oscillations observed at ~21 Hz are due to this stack
resonance. This, however, needs to be verified.
We then proceeded in measuring the noise level from the photodiode,
the
electronic chain, and the mixer by simply blocking the laser light
onto
the photodiode and measuring the power spectrum density of the error
signal. The laser frequency noise has been measured in the past by
locking the laser onto the reference cavity.
We also measured the power spectrum densities of the error signal and
the control signals once the instrument was locked, in order to estimate
the TNI's first sensitivity curve and pinpoint the noise sources that
limits it.
We also did a series of experiments in preparation for inserting a
broadband Pockels cell into the beam for arm cavity lock acquisition.
Our
current setup bypasses the mode cleaner and uses the laser's PZT to
"stretch" the beam while the mirror catches. Since the mode cleaner
servo
will eventually act on the laser's PZT, our current setup will not
be
practical for arm cavity acquisition. We have been experimenting
with ways
of measuring the transfer functions of individual elements in the system,
both with the loop open and with the system locked, and we have applied
these methods to our current arm-cavity lock and to an auxiliary,
diagnostic lock of the laser to our fixed-length reference cavity.
The most important transfer functions we have to measure are the frequency
discriminant, the mirror response, and the laser's PZT response.
Direct
measurements made in January of this year gave
mirror response:
1.6 microns/volt 2 poles at 1 Hz
frequency discriminant: 20 volts/megahertz
1 pole around 100 kHz
PZT response:
0.7 megahertz/volt 1 pole around 90 kHz
with uncertainties of around 20% for the mirror response and around
10% for
the other two measurements. These measurements were made by sweeping
the
system through resonance and observing the error signal, using the
sidebands for calibration.
This week we made some in-loop measurements with the system locked,
and we
obtained results that were consistent with last month's direct measurements
for the mirror response and the PZT response, within experimental
uncertainties. Our measurement of the frequency discriminant,
however,
appears to differ from the direct result by nearly a factor of ten.
This
discrepancy only occurs for the arm cavity lock; the PZT response and
the
frequency discriminant for the reference cavity lock obtained by both
direct sweep and in-loop transfer function measurements are in good
agreement. We are continuing to investigate this anomaly.
This week we also placed an order for a second broadband Pockels cell
for
the second arm cavity, and started work on integrating the TNI into
the
rest of LIGO's advanced R&D efforts.
LASTI (MacInnis, Harry, Mittleman, Rollins, Shoemaker, Zucker)
---------------------------------------------------------------
Program planning (Shoemaker/Zucker): DHS solicited
more detailed testing plans from Suspension and Seismic groups, for
integrated review at LSC meeting LASTI session.
Held internal planning discussion to identify program breakpoints and
bring new team members up to speed.
Engineering (Ken Mason):
Continuing to create fabrication drawings for a custom BSC clean room
to
allow insertion of new suspensions through the top section.
We recieved the installation fixtures from LLO for the BSC seismic
isolation installation scheduled for 4/30/01.
Recieved quotes for the fine adjust and air bearing adapters for the
BSC
seismic isolation.
* Physics Studies
-----------------
Biplab, Hiro, Rick discussed about pursuing the studies of frequency
noises due to doppler shifts from table motion and motion in
PSL reference
cavity. Cella has been contacted to supply us the MSE code for
the
(mechanical) simulation runs of the reference cavity including
the
fused silica spacers.
* The Code
-----------
(Tavio) Last week was spent examining the PETE expression template
engine
to gauge the effort required to use it's efficiency within the
e2e
simulation. Unfortunately we determined that the documentation
for PETE
is somewhat sketchy on a couple of key areas that are needed
for us to
implement it.
Blast++ was examined as another candidate expression-template
based class,
but it did not directly perform matrix multiplication and is
quite large.
We are still looking for a templated matrix class.
* Client-Server Framework
--------------------------
(Ed Maros) worked on compiling E2E on CACR systems. He wrote
the TCK script,
'Jobber' that allows submission and query of jobs at CACR.
He also worked on getting jobber to run under Linux.
* Alfi
------
(Bruce Sears) - Continuing the work outlined last week:
- Work on connection and port problems, particularly checks
against
sink ports being connected to multiple times via
derived nodes.
- Reorganizing some of the core ALParser code to make sure methods
are
not being made redundant or are unused if they are
not easily found.
Software Systems (Blackburn)
The LDAS MPI team held its first formal telecon with the MPI MDC group this
Thursday. The discussion focused on the close out for the report
conclusions.
This will be completed and submitted to the DCC in roughly one to two weeks.
Other topics discussed included the logistics of a load balance test which
was outline in an email last week, a progress on the eventMonitorAPI, the
documentation for the lalwrapper, development of a dso to test load balance
with an expected delivery date of one week, and plans for the next MDC to
take place May 15 and will test a full pipeline of data from frames through
dataConditionAPI, wrapperAPI, eventMonitorAPI and into the database.The next release of LDAS is almost ready. Testing of modifications to the
ilwd library to support zero length arrays has indicated that the fixes
are complete. Both MDC tests for the dataConditionAPI and wrapperAPI are in
very good shape, thought there are some non deterministic problems reported
with a few performance tests in the dataConditionAPI which may have been
fixed
by this weeks modification to the ilwd library. A few more issues with the
documentation of the build procedure were identified as a result of a new
installation and start up of LDAS on an ANU system. A few last minute bugs
in the manager and controlMonitor APIs were also fixed. Plans are to have
the code tagged and rebuilt later today.Work has already begun on a new release of LDAS which will have limited
additional functionality in support of the next Engineering run. To be
added are support for the new tapeControl script in the controlMonitorAPI
used to record data to tape for transportation to the LIGO archive and
support for user commands to send results of channel analysis performed
by the dataConditionAPI on data read in from frames by the frameAPI to
the LDAS database. An effort is being headed up by Philip Charlton to
be able to use this functionality to contribute to an E3 investigation
using the LDAS system at the sites.John Zweizig profiled the frameCPP library and made several modifications
to the the source code which in his application have doubled the performance
of the library. These changes are being reviewed to determine if they are
universally safe for all code using the frameCPP. It is hoped that this will
be the case and the new code can be used in the E3 run.This week's progress on the Database Mock Data Challenge includes:
- Completing the full set of test scripts for functionality tests.
- Verifying that there are no remaining serious functionality
problems with the version of LDAS being considered for release
soon. There are some issues with error handling which will be
addressed over the next few weeks.
- Improvements to the C library for parsing LIGO_LW table files.
- Some refinements of the utility programs for dumping and comparing
LIGO_LW table files.
Andrea Vicere'.* Keep working on familiarizing myself with the LDAS bibliography
Hardware Systems (Anderson)
Charlton:
Renamed the engineering run directories in the LIGO frame archive at Caltech
from eng.2000.apr to E1 and eng.2000.nov to E2.Initial testing of the copper-to-fibre network media converters at LHO needed
to support the future (post-E3) move of LDAS servers to the Mezzanine area
of the assembly building.Moved the repaired LDAS AIT-2 tape robot to the U10 server that will support
the tape backup of frames during the upcoming E3 run. Cybernetics confirmed
that one of the drives failed their internal diagnostic tests and replaced
that drive.Re-opening of the job search for an LDAS sys-admin/scientist resident at LLO.
The new AMD K7 and Intel P4 test machines at Caltech, and the old linxbox1
at LLO with replaced memory, have passed their LDAS acceptance tests as
functional and stable machines.Started looking into migrating LDAS from ssh1 and ssh2 to openSSH as
a possible more secure, reliable, and less legally encumbered implementation
of the SSH1 and SSH2 protocols.
Data Analysis Activities
* Drafted timetable for development of FCT GW chirp search code for
LAL
and wrapper, aiming for May 15 end date.
* Created "stub" web page for FCT group actions and development info
http://www.ligo.caltech.edu/~charlton/FCT/
Andrea Vicere'
* Started contact with Keith Riles and Rana Adhikari to get involved
in the E3
run.
[Bruce Sears]
* (BS) Hanford Ilog Mirror:
- More Livingston Ilog
maintenance.
Larry was at MIT this week.
MIT:
(Keith)
First week: Began learning MIT setup, investigated mail
trouble on various Sun boxes, investigated NIS+,
reinstalled SSH on several nodes, fixed several inconsistencies
on workstations.
Working with Larry and fixed a number of computers with minor problems.
Working on setting up a new mail server.
Livingston:
(Tom & Shannon)
-Preparing things for the E3 run and the LSC conference.
-More testing on the wireless network. Discovered the 3com cards have
an
extremely flimsy mount for the antenna. Stay away from the liksys
cards, they have windows driver problems. The best one found
so
far is the lucent cards. Also there is an added bonus - they
are
compatible with linux.
As far as access points go, only tested the D-link access point.
They seem to work fine here.
Hanford:
-Preparing for the NSF review and the E3 run.
CIT:
(Barbara)
- Installed DCC changes to support document folders. Fixed problems
with
file size of DCC database; problems were due to new tables supporting
the
text version. Fixed problem with "duplicate key" error message
when
creating new tables for the text version.
- Created and installed web page for January NSF review. Created
and
installed new organization chart web page with pop-up windows describing
the functions of each area. Revised the internal bulletin board
slightly
to include a templates section with Powerpoint templates and the travel
mileage report form. Made several other quick site changes.
- Resumed work on contacts perl script for LDAS.
(Lisa)
- Attempted to rescue data off of Ken Mailand's failing hard disk.
I was able
to rescue some but not all (this is why people should backup their
PC's). Am
currently working on installing a new hard disk and OS in his computer.
- Ordered a second ISDN card for the new modem pool. Am working
with Liz and
the telephone office to move the 800 numbers to ISDN. The ISDN
card won't be
available until mid March.
- The hard disk in Jay Heefner's ultra 10 went belly up. I am
currently waiting
on a warranty replacement from Sun.
- Fixed some exceed/calendar problems for Rita.
- Finished building Liz's new computer. I have a ghost image
and am ready to
start transferring her files.
- The laptop with the nonfunctioning display is back from Dell.
Am working on
getting that set up for Ed Chargois.
- Working on configuring the new laptop for the DCC. The drivers
for this are a
challenge because it is Win 2k.
- Replaced the cdrom drive in hamal.
(Suresh)
-Working on Internet based video conferencing unit.
-Installed Matlab Release 12 in two PCs in bridge.
-Continuing restoration of data from old tape archives.
-Rebuilt Irena Petrac's old PC by replacing it hard drive.
(Larry)
-Finalized a number of issues with SUN. We don't know when the ship
date will be
for the new SUN BLADE system. Too much demand and not enough supply
I guess.
-Worked on some of the documentation Albert needed for the NSF. Also,
arranged
to have a number of PC's and projectors sent to Hanford for the NSF
meeting.
Thanks to Lisa and Ed C. that little project took place in a timely
manner.
-Chased down a number of orders that were late on shipping.
-Spent a good deal of time for P-card reconciliation. Mainly trying
to get
invoices.
-Fixed a few server problems. Most were just processes that needed
to be
restarted.
-Checked on a reported network problem over the weekend but could not
find
anything wrong nor duplicate the problem.
-Continued working with DCC on LSC and other issues.
(Sam)
-Worked on a couple of PC setups.
For additional information about this report, contact sanders@ligo.caltech.edu