|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee Agenda for Monday December 11, 2000 will be:
CANCELLED DUE TO PAC/GWDAW MEETINGS
No report.
WBS 1.2 LIGO Operations--Administration
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
From: Cleveland Mak <mak_c@ligo.caltech.edu>
- File naming should be based on the root part of the number plus the revision and without the Group ID. Additionally, the file name should be in capitals, and the format extension in small letters. Thus a document numbered as
- LIGO-G990025-00-D would have a file name of G990025-00.pdf
- All pertinent information needs to be on every document. This includes:
- DCC number
- Author's name
- Date of document
- Title
- Duplicate submittals can be confusing as well as time consuming. If you can't remember whether a document has been submitted to the DCC, you might check the docspublic directory to see if it resides there. Or, feel free to give us a call so we may check for you.
| Packages | Faxes | |
| In | 49 | 34 |
| Out | 19 | 43 |
Press here to access the DOCUMENT CONTROL CENTER WEB PAGE.
From: Esther Cunningham <esther@ligo.caltech.edu>
Press here for ACCOUNTS
PAYABLE HISTORY DATA.
|
Date
|
Incoming Invoices
And Receivers
|
Wires
|
Large Contract
Invoices
|
| December 1, 2000 | 7 | 0 | 0 |
| December 4, 2000 | 15 | 0 | 1 |
| December 5, 2000 | 9 | 0 | 0 |
| December 6, 2000 | 8 | 0 | 0 |
| December 7, 2000 | 10 | 0 | 0 |
| Totals | 49 | 0 | 1 |
From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
From: irena@ligo.caltech.edu (Irena Petrac)
Irene Baldon
Progress Period from 12.1 to 12.7
Accomplishments:
Corrective Action:
None to report
at this time.
From: irena@ligo.caltech.edu (Irena Petrac)
WBS 1.4.1.2 Project
Controls (LIGO Construction)
Annual Report: There will be an Annual Report due as of the end of November. I will be requesting inputs from the usual bunch mid-December after the dust from the proposal has a chance to settle a bit. I propose to keep this report moderate in size, about the size and complexity of a Quarterly Report, to minimize the burden.
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 |
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)
Work continues on locking the full interferometer in the "undamped"
cavity mode. With some help from Hugh Radkins, I have made some progress
using the Morganson code (SURF 99) to "predict" the earth tides for
E2.
The preliminary plots comparing the data to the prediction are at
http://www.ligo-wa.caltech.edu/~fjr/earth_tides/E2_prelim_tides_comm.pdf
http://www.ligo-wa.caltech.edu/~fjr/earth_tides/E2_prelim_tides_diff.pdf
http://www.ligo-wa.caltech.edu/~fjr/earth_tides/E2_prelim_tides_diff.pdf
for your viewing pleasure. The parameters chosen for the tide prediction
were "typical" numbers adopted by Morganson and me back in 1999. We
will
now move on to fitting the tides to the data.
PreStabilized Lasers:
--------------------
M. Guenther, T. Mahood, R. Savage
- The 4K racks were successfully relocated an additional 6' from the
4K
PSL enclosure and rotated so the doors now face the PSL enclosure (with
lots of help from J. Meyers and R. McCarthy). This provides additional
clearance around the PSL enclosure should a future wall
or secondary acoustical enclosure be installed. The relocation
also
cuts down on the rack's acoustical impacts, and with the rack doors
now
facing the enclosure, adjustments to the laser are greatly facilitated.
Following the rack relocation, cables to the laser were run overhead
to
keep the floor free and clear. Finally, the bulk of the electronics
for
the laser were installed in the rack, the cross-connect put in place,
and
cables run to tie in the cross-connect to the rest of the equipment
in
the rack. The laser is back in operation, and now feeding data into
the
EPICS system.
- The broadband EOM was installed within the laser, but a reduction
in
the laser's output power suggests some additional tweaking of the EOM
alignment is necessary.
- Parts for the 4K reference cavity are being staged for assembly next
week (while Lee Cardenas is here to advise). It's planned to assemble
the
cavity in its ultimate location on the 4K table.
The rigid mirror mount bases have now been anodized and are hand for
imminent installation. These bases include rigid mounts for Newport
U100 and "Klinger" mirror holders, a rigid base for supporting polarizing
beamsplitter cubes, and a reference cavity periscope with the lower
mirror fixed, and the upper mirror a steerable U100 mirror holder.
Computing
-----------
(C. Patton)
Reminder - Starting 12/27/00 until 12/29/00, the LHO file and NIS+
server, rainier, and the LHO email/web server, apex, will be undergoing
hardware upgrades. These machines will be shutdown for several
hours on
the 27th and may require periodic reboots over the 28th and 29th.
I am adding more disk space and more memory to rainier. I will
be
moving some of the user accounts to the new disks and cleaning up some
of the old application software that is no longer needed. Apex
is
currently a 300 MHz Ultra10 with 64 MB of memory, it will be replaced
with a 440 MHz Ultra10 with 256 MB of memory and will have an external
36 GB disk drive. This is the first step to turning apex into
a better
web and anonymous FTP server as well as email. Both machines
will be
upgraded to the latest software and patches.
OPTICS/COC/SEI INSTALLATION: We've inventoried and shipped the COS installation equipment, as well as our own ISC instrumentation to LHO for support of the LHO 4k installation. This week has been spent completing the installation of cameras at the end stations, and vertex. With David Shoemaker's assistance we located ICST-4 at the darkport, and temporarily installed a camera and photo detector to use while aligning the Michelson. Joe Hanson is cleaning and baking the last 2 BSC bellows for MIT, and we're moving the machine shop as opportunities present themselves. (Jonathan Kern)
We are also preparing for the imminent opening of the gate valves along the arms so that we can look at the alignment over 4K.
The bulk of this week's activities have been consumed with commissioning
and preparation for the PAC meeting. We also had a successful bid
opening for the addition to the staging building.
No report.
The Caltech group had its Xmas party today. It was a great success and there was quite a crowd. Kudos to Linda for a successful party!
| Installation
& Commissioning:
Livingston |
Other Science/Engineering
Activities:
Issues/Concerns |
See also the Installation web page
The effort has been on aligning the 2 km interferometer, making the calibrations needed to feed the lock acquisition code, and to get back to making the full recycled, recombined interferometer work.
The major piece of news was the finding that the digital phase shifting part of the software had a small bug, but one that affects the signal separation on the two phases of the reflected light signal. This was easily fixed, once it was diagnosed, and the corrected code loaded. Unfortunately it meant that all the previously determined lock acquisition parameters had to be remeasured.
By the end of the week, everything seems to be in place and we are again seeing the servos try to lock the full interferometer. Durations are about what we had observed before without using the gate valves to spoil the arm cavity Q's, typically 0.3 - 1 second. We are continuing to work on improving this.
Ben Abbott
Started working with Rick
Karwoski regarding the PSL. RF photodiode board is tested, and I
have designed the frequency sensitive portions which should enjoy final
installation/testing today.
Then we did a more careful alignment. We adjusted the angle of the Recycling Mirror to be parallel to ITMX, as judged by seeing bright flashes in the poor Fabry-Perot formed by those two mirrors. This is also a test of the alignment of the input light to the Michelson. When this alignment is done, the FP cavity formed by RM and ITMY also flashes as expected. Next, the BS was adjusted so that the spots from the two arms were superposed at the anti-symmetric port, and fringes were observed. This could be accomplished while leaving the alignments of the two ITMs undisturbed, as required to leave them aligned with the long arms.
A number of issues remain. We needed to make substantial changes to the orientations of the RM and of the BS to accomplish this alignment, and this is something of a mystery. There is also a problem with the BS controller that needs to be fixed.
We are working to be prepared to take a look down the X arm (at least), perhaps as soon as next week.
A new, lower bandwidth, PMC has been fabricated and located on the PSL table. The PMC was roughly aligned to the laser - no attempt was made to move the mode-matching lenses despite a large number of changes to the laser - and locked to the laser. Optimization of the mode-matching will take place when I get back from LLO and when Lee returns from LHO.
Lot's of action in the lab this week, as we are trying to put all the
pieces together and lock our cavities in the TNI, with assistance from
Seiji Kawamura. We've run into several difficulties, however,
so progress
has been slow. Some of our problems seem to have their origin
in our
laser, but this is not yet certain.
No report this week.
Studied lock-acquisition in presence of misalignments
using new version (full001107) of Han2k model.
The upper limit on the rms value of misalignment noise
in ALL MIRRORS, for which stable lock could be achieved
without ASC, still stands at about 4e-08 rad.
The average lock-acquisition time is found to be about
3.3 sec in scalar case and about 11.6 sec in multi-mode
case with misalignment noise of rms 3.4e-8 rad in all mirrors.
Modifying document no. LIGO-T000106 written on this topic.
As a test of both primitive modules and recycling summation cavity,
ran the Han2k model replacing its recycling summation cavity
by the corresponding arrangement of primitive modules like
mirrors and propagators without changing any parameter in
servo, suspension or anything else. After 7 days of continuous
run
with a time-step of only 1e-8 (as compared to a time-step of
6.7e-6
in standard Han2k which uses the recycling summation cavity),
it could be confirmed that these two versions produce the same
result and acquire lock at same point of time.
Introducing violin modes for generating noise curves from Han2k.
Facing some problems in generating such sharp lines from time-series.
Various testings are being pursued.
* Lock acuqisition:
Matt has improved the lock acuisition code to be implemented in his
visit
to LHO this and next week. This makes the calibration easier.
* simulation engine code:
Tavio and Matt worked on the global thread support in adlib. Tavio has
developed a testbed which can be used to test the thread safe queue
and
develop the algorithm for optimizing the usage of thread. The overhead
is
the first thing to be tested to have a good view of the thread-based
simulation.
Hiro worked on to implement thread in the summation cavity which spends
most of the time. Multi threads can be activated for each sideband.
But,
somehow, the overhead is too large, and now is trying to solve the
problem.
* Alfi :
Most of the reported bugs are fixed. Code change are going on for further
stablization and for easier development/maintenance. The new features
planned next are (1) copy & paste and (2) support of connection
node.
Software Systems (Blackburn)
The communication problem between the mpiAPI and the wrapperAPI being used
to report messages such as status, progress and perform load balancing has now
been fixed. It did take the better part of the week to resolve. The reversal of
the client and server side of the communications however broke the problem lose
and allowed the fixes to be implemented. The next step is to integrate a search
engine (dynamically loaded shared object).The problem with namespaces in the new perceps used to document the LDAS C++ code has now been fixed. This brings LDAS one significant step closer to
being in a position to have a new release.The change over to the new Unified Data Type continues to make progress. A
version of LDAS in which the new UDT is implemented in the dataConditionAPI is now on the development system. Only limited algorithm functionality has been
demonstrated to function, with various minor bugs and issues being identified by testing.The testing verification of the dataConditionAPI mock-data-challenge scripts
has continued this week. A few of the tests are not providing the expected
accuracy in the results as expected. These are being look into more closely. Also, a
few of the performance tests were not able to be streamed through the system due
to memory resources issues these are being documented with the expectation that
the MDC closure will be soon.The frameCPP and frameAPI were found to lack interfaces to the full detector
and static data structures. This functionality has now been implemented and
exposed to the TCL layer of the frameAPI. It has yet to be tested though.New webpage interfaces into the LDAS database statistics with indicate the
total number of rows, insertion rates, and errors have been added to LDAS using
the db2utils toolkit.The controlMonitorAPI's log page views have been cleaned up and the filters
used to select information from the log files cleaned up and unified.Peter Shawhan drafted a detailed schedule for the Database Mock Data
Challenge. The schedule was discussed during the SoftwareCommittee teleconference, where some suggestions were made about involving additional people and further defining what LDAS computing resources will be needed when. The general plan has activities continuing from now through late January.
Hardware Systems (Anderson)
LDAS news from CIT
------------------1) The new LDAS 16 node Beowulf and server have passed their acceptance tests
and are now ready for LDAS software development and upcoming Mock Data
Challenges.2) Anderson and Lazzarini took a trip to BNL in order to make an evaluation of the HPSS system being used at BNL (RHIC Computing Facility, or RCF). Our findings have positive implications for LIGO. Briefly:
- Their HPSS experience started out bad and has become rather good; they
grew their own experts from zero to FOUR in 3 years; NOTE THAT they have 4
FTEs supporting the HPSS installation (which supports all 4 RHIC experiments).- RCF is different from us in one key area: they are relying on HPSS for
REAL-TIME data ingestion. Even IBM/HPSS suggested against this when RHIC
first started out.- The prices they are getting from the same hardware mfrs. seems to be in
line with what we are receiving as quotes.- We learned about how/what types of tape drives they use and their
maintenance problems. It probably changes how we will want to do things a
little.- An earlier report that RHIC/RCF (RHIC Computing Facility) was
having difficulties with HPSS seems to have been skewed by bad experiences from the one group only. And there has been some finger-pointing over this. Apparently some of the experimental code brings down the unix filesystems (WHICH NO PROGRAM SHOULD BE ABLE TO EVER DO). So one side is saying: bad programming; the other says: bad systems administration.... The fact is
that they have isolated the issue to one specific set of code running.- One thing we found surprising was that the RCF provides COMPLETELY REDUNDANT AND NONOVERLAPPING resource support to each of the 4 experiments: some times precious resources go unused by one group while another group may be struggling .... this is NOT the model LIGO wants to pursue.
LDAS news from LHO
------------------1) Since the Nov E2 run, a new configuration for the SCSI SONY AIT 2
tape drivers have
been implemented and tested. The new configuration appears to greatly
reduce
problems with SCSI resets. Remaining resets can be overcome by
software. More
can be learned about the configuration by reading the man pages for st.
The configuration added to /kernel/drv/st.conf is:tape-config-list=
"SONY SDX-300C", "SONY 8mm AIT",
"SONY_AIT",
"SONY SDX-500C", "SONY 8mm AIT",
"SONY_AIT";SONY_AIT = 1,0x34,0,0xd679,4,0x13,0x8c,0x8c,0x8c,3;
2) The following enhancements were made to the tapecontrol software
script:
(i) The script will backup from disk onto tape arbitrary amounts of data
from files or directories matching a pattern. The script emails the
operator when more tapes need to be added to the tape library.
(ii) The script will perform a tar diff and report difference between
data on tape and disk.
(iii) The script will restore from tape onto disk arbitrary amounts of
data. The script emails the operator
when more tapes need to be added to the tape library.These enhancements have been used to backup reduced data sets from the
Nov E2 run on fortress.3) An HP 9340i CD-Writer has been installed on linuxbox1
Other
[Bruce Sears]
*Ilog:
- Continuing work associated with MIT installation, including
needed upgrades at Hanford and Livingston.
MIT:
Working on getting a fix for the new disk drive.
Livingston:
(Nothing to report at this time)
Hanford:
(Christine)
Reminder - Starting 12/27/00 until 12/29/00, the LHO file and NIS+
server, rainier, and the LHO email/web server, apex, will be undergoing
hardware upgrades. These machines will be shutdown for several
hours on
the 27th and may require periodic reboots over the 28th and 29th.
-I am adding more disk space and more memory to rainier. I will
be
moving some of the user accounts to the new disks and cleaning up some
of the old application software that is no longer needed. Apex
is
currently a 300 MHz Ultra10 with 64 MB of memory, it will be replaced
with a 440 MHz Ultra10 with 256 MB of memory and will have an external
36 GB disk drive. This is the first step to turning apex into
a better
web and anonymous FTP server as well as email. Both machines
will be
upgraded to the latest software and patches.
CIT:
(Sam)
Nothing to report, just busy with school.
(Lisa)
- I spent a lot of time working on Rolf's development computers over
at Wilson
House. Cdssol10 is having problems. It got moved yesterday
and now its
storedge disks won't come back on line. Larry is going to help
me with that
problem.
- Upgraded a couple of sun boxes to Solaris 8.
- Tried to swap out Hongyu's workstation but the replacement had a
bad power
supply. I'm waiting on a replacement from sun.
- Working on resolving some file sharing list issues for Rick Karwoski.
- Worked with Larry on modifying a rack for equipment installation.
- Worked with Larry on setting up new print drivers for the the new
printer
being used by the DCC.
(Suresh)
-Fixed the desktop application menu problem with Mark Barton's Sun
workstation.
-Helped setting up a PC for a new grad student in third floor bridge.
-Prepared documentation for setting up printers/plotters to be included
in
LIGO internal Web pages.
-Resolved Rich Savage's dial-up PPP account problem.
-Fixed Irena's web browser accessing problem.
(Barbara)
-Working on DCC conversion to Office 2000.
-Setup Timetarget on the new DCC server.
-Installed the DCC software including changes for multiple electronic
files
per document.
(Larry)
-Spent a great deal of time this past week helping Liz and others on
getting out
the LIGO proposal document. Mostly, working through the logistics and
getting
the equipment to perform the tasks needed.
This little exercise showed a number of weak points that we are now
going to
resolve, such as; more memory for the printers, larger spool areas
and a sorter
for the DCC printer.
-Continued work on purchasing of new equipment and software.
-Checking out some of the new virus attacks. It looks like the new
ones are
getting to be more malicious.
-Working on setting up some new equipment in the server room.
-Working with the group to get more documentation on the internal web.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Silicate Bonding
Phil and I visited Sheila Rowan at Stanford.
We learned and practiced the technique for bonding fused silica to
sapphire.
Also, met with Ken Bower, who with J. Gwo, researched and made the
silicate bonds used in the Gravity Probe B experiment.
We learned about UV cleaning approaches and discussed mechanical strengths.
Mr. Bower was very approachable and open, sharing with us some of his
practical knowledge.
Advanced LIGO PSL
Peter King
I finally received a quotation
for the pump laser diodes, the
delivery fibers and associated power supplies from Coherent
Semiconductor.
Surprisingly the price of laser diodes has increased from 12 months
ago
by
approximately 16% to ~$7500 ea. Certainly the lion's share of
the cost
of
the laser is in the pump diodes. Each laser is expected to require
around
60 laser diodes, not including spare parts.
Advanced LIGO SUS
Janeen Romie
Advanced LIGO - Planning a meeting Tuesday morning, December 12 at
9am
pacific to talk about the Suspensions Design Requirements Document.
For additional information about this report, contact sanders@ligo.caltech.edu