|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee
Agenda for Monday August 21, 2000 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items: LSC meeting reprive, discussions with NSF
No report received. LSC meeting this week.
WBS 1.2 LIGO Operations--Administration
PROPERTY MANAGEMENT (Chargois)
From: Ed Chargois <chargois_e@ligo.caltech.edu>
| Packages | Faxes | |
| In |
|
|
| Out |
|
|
I've been training Jim Covington all week to take over my mail/fed-ex duties while I'm out on vacation (Aug. 24 - Sept. 8).
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>
Reconciled pcard today. Have several more pcard orders to place.
Worked on Triad (an internal mod to release funds on PMA.01500 1.3. Working on Louisiana Tech, Support Services, and several backup documentation on MMR, Richard Price, and Butler.
From: Florence Kaufman <fkaufman@ligo.caltech.edu>
I am gathering the details (open encumbrances and the actuals) for the Advanced R&D accounts, again... (I had them completed and formatted yesterday, but my computer crashed and I lost EVERYTHING). I will email them to the Task managers as soon as I have completed them.
Continue to review the open encumbrances on the Construction and Advanced R&D accounts to see if the requested removals have been made. I will continue to review these accounts weekly until all encumbrances have been removed.
The financial reports on the web provide supporting detail.
http://docuserv.ligo.caltech.edu/~fireport
http://docuserv.ligo.caltech.edu/~finance
From: Ed Jasnow <jasnow@ligo.caltech.edu>
The Town of Livingston has sent an alternative lease to the one we proposed for the storage space. As opposed to our three page lease, theirs is eight pages long. Ed Jasnow and Ruth Brambila are discussing this with representatives of the town, and hope for a quick resolution.
Rita Torres
Worked with a number of outside vendors in Hanford and was able to arrange with them to accept my P-Card for payment of non-card holders expenses. Have started working with vendors in Livingston to also accept my card for payment of non-card holders hotel bill. With these expenses now being pre-paid for the traveler, either by his/her P-Card or mine, I am now issuing Advance Checks (written by me) to the traveler for per diem and miscellaneous. We no more longer have to wait to get you your advances. With Criselda's fantastic help we're working out the kinks as we go.
Started ten (10) new trips this past week. There are approximately an additional six (6) new trips in various stages pending completion of travel arrangements.
The SURF Students are starting to depart for their home universities. All arrangements and last minute changes have been completed. I should start getting receipts for their reimbursement and final close out on their individual trips.
Worked on, and completed twelve (12) Expense Reports. Rita was able to complete seven (7) reports and has an additional six (6) that she is working on. I have two (2) Expenses Reports being held which require a check from the Traveler before processing can be completed. Due to meetings, training and preparations for the new P-Card Program I'm backed up on Expense Reports at the present time. I have twenty-three (23) outstanding at the present time. I should be able to get these going this coming week.
Prepared the Travel/Vacation Itinerary for the Week of August 14, 2000. Performed normal recording and filing associated with Travel and Reimbursement. Worked on several problem issues with Travel Audit. Also performed miscellaneous duties as requested by various members of the LIGO Project here at Caltech as well as from members of the staffs of each of the two sites. I continue to do MIT's travel to the sites for installation activities and also to assist them wherever possible.
Elizabeth K. Wood
No report this week (attending LSC Meeting).
From: Kris Duncan <kris@ligo.caltech.edu>
From: Ed Jasnow <jasnow@ligo.caltech.edu>
The design for the new Livingston Staging Building is nearing completion. The schedule still calls for the issuance of an Invitation for Bid by the end of August.
General Items:
--------------
(F. Raab)
We had company this week. About 120 people attended the LSC meeting for three days. On Sunday, Monday and Tuesday, we had a visit by John A. Wheeler for outreach activities; Wheeler also attended the plenary sessions of the LSC meeting. During his tour of duty with the Manhattan Project, Wheeler was the cognizant scientist for B-Reactor, the world's first production reactor, which is being preserved as a historical site at Hanford. The B-Reactor Museum Association had a dinner for Wheeler, his former coworkers at B-Reactor and some LIGO folk on Sunday. A few lucky LIGOans accompanied Wheeler on a tour of B-Reactor Monday morning and then we reciprocated with a LIGO tour for the B_Ractor folk. The main event on Monday night was the LIGO Public Lecture by Kip Thorne and John A. Wheeler, celebrating Wheeler's contributions in engineering and science, from fission to black holes. About 350 people attended the standing-room-only event.
Despite all the company, work continued on ongoing projects.
Computing:
----------
(C. Patton)
Set up 4 new Sun Ultra 10s. Two for long-term visitors and 2 in
the
common areas. Fixed a path problem when mounting files from the
license
server to each Sun workstation. The gnu compilers now work on
all Suns
except for rainier. Rainier is in the process of becoming a file
server
only. In the not too distant future user login will be turned
off on
rainier.
OPTICS/COC/SEI INSTALLATION: We used the Coarse Actuation System (CAS) on BSC-2 for the first time last Friday in an attempt to correct a position error of the beamsplitter. After making coordinated moves in two axes, we found that the system introduced a yaw of ~750 u-rad to the BSC table!! An attempt to return home in a single axis exaggerated the problem. We've contacted HYTEK (the vendor) and after many exchanges and several days effort have identified several problems with the delivered software. An attempt was made yesterday to move the table using low level software, but after several home moves we're at the extreme limit of our adjustment range in the -X direction. This is an unacceptable position to place ourselves in, in the future. We've decided to return to nominal center of the Actuator System's range and physically move the beamsplitter on the optics table. This is in progress, and we're attempting to orient the table near its original position in order to preserve the COS alignment work performed 2 weeks ago. (The orientation of the table, yawing by hand, was successfully completed so that the COS alignment is preserved. Mark Guenther has come from LLO to assist us making sofware changes to the HYTEK software. While he is here we will also attempt to characterize the performance of the CAS system by exercising it while simultaneously making measurements of the translation stage positions and optical measurements of the payload positions. - Mark Coles)
(Riesen / Kern)
No report this week.
No report this week.
| Installation
& Commissioning:
Livingston |
Other Science/Engineering
Activities:
Issues/Concerns |
See also the Installation web page
Along the way:
At the end of last week, four of our six SURF students gave final oral reports on their summer research projects. Keisha Williams of Southern University reported on her work with Szabi Marka on GDS and on seismometer temperature sensitivity. Josh Smith of Syracuse described his work with Peter Saulson and Mark Barton on OSEM transverse sensitivity and on suspension diagonalization. Richard Yamada of Yale discussed his work with Joe Giaime on tiltmeter temperature sensitivity. Christie Sayes of LSU reported on her work with Joe Kovalik on amplitude noise in the PSL. While SURF doesn't end until this coming Friday, the talks were held early because of advisors' travel to the LSC. The remaining students, Kevin Tubbs of Southern and Dan Fabrycky of Caltech, will give their reports at the end of this week.
Measurements characterizing the modified current shunt on the laser in the PSL Lab are currently underway. This is relevant for designing the intensity stabilization servo.
Rich Abbott
Preparations continue including
orders of parts for the fabrication of three more PSL wiring harnesses.
Began the process of working on the new intensity servo. Preliminary testing of the new current shunt is underway with a focus on understanding the limitations of its gain in the face of periodic increases in diode bus current as the lasers age. The concern is that as the laser bus current is increased, we might move into a more saturated regime and cause the relationship between a change in diode current to a change in optical output to show a reduction in gain. This, of course, will be an issue for any proposed servo.
General Optics has delivered the first of the re-polished SPETMs. We will measure it to confirm the ROC as soon as we are through with 4ITM07.
Janeen has completed the assembly drawings for these parts and will work to complete the assembly specification next week.
Jay Heefner
6 Satellite amps have been
received from LHO and are being modified for use with the new OSEM heads.
They should be tested and ready by 8/25.
Since each PCB takes approximately 24 man-hour to build and CDS is currently shy of available manpower, the rest of the circuit boards will be subcontracted out to AccuAssembly to build. Sander and Paulmyself went to Accu-assembly (a board assembly house in Duarte) to tour the facility and we left them with a PCB and drawings to give us a quote.
We are currently waiting for the arrival of additional two pin Lemo connectors to complete our orders.
No report this week.
Report of the work on the Thermal Noise Interferometer (TNI) and the
Photothermal experiment (FROM SEIJI)
July 13, 2000 - August 14, 2000 by Seiji
Kawamura
1. TNI (with Luca Matone and Eric Black)
The frequency stabilization system for the TNI interferometer has been
successfully developed. The system consists of the NPRO laser and the
0.5m
triangular mode cleaner. The laser frequency is directly stabilized
by the
mode cleaner using the PZT actuator of the laser, while at very low
frequencies (below 300mHz) the mode cleaner follows the laser frequency.
This system is much simpler than the originally-planned optical and
servo
configuration. It uses two-step stabilization: the laser frequency
is first
stabilized by a rigid reference cavity using the PZT and the Pockel
cell and
then further stabilized by the mode cleaner using the AOM placed before
the
reference cavity.
We verified that this simplified system still satisfies the requirements
as
follows:
a) the lock acquisition and holding is very reliable,
b) the required frequency stabilization gain can be obtained,
c) the shot noise and electric noise of the system are small enough
for the
required frequency stability,
d) the required frequency stabilization gain is not limited by the
spurius
path effect.
See appendix for more detail. Also the document about this is in
preparation.
2. Photothermal experiment (with Shanti Rao and Eric Black)
The control system of the laser locking has been improved. The control
system had only the laser PZT path, so the lock was maintained only
for a
few minute. We implemented the thermal actuator path below 50mHz and
the
lock holding became very stable.
We measured the displacement signal of the probe beam when we modulated
the
amplitude of the pump beam. We found that the signal was comparable
with the
spurius signal we obtained when the cavity was not locked. This indicates
that the cross-coupling between the probe beam and the pump beam is
too
large.
[Appendix]
Detailed Description of the Frequency Stabilization System for TNI
1. Required Frequency Stability
The final goal sensitivity for the TNI interferometer is 1e-19m/rHz.
It
means that the frequency stability of the light coming out of the mode
cleaner should be 30mHz/rHz, assuming that the common mode rejection
ratio
is a factor of 10.
1-1. Frequency Stabilization Gain
Since the original frequency noise of the laser is 200Hz/rHz at 100Hz
and
20Hz/rHz at 1kHz, the required loop gain is 6,000 at 100Hz and 600
at 1kHz.
The laser light is currently locked to the mode cleaner with the PZT
actuator with the following condition:
a) the unity-gain frequency is 80kHz,
b) the phase margin at 80kHz is more than 45degree,
c) the loop transfer function has one pole at 100Hz, and
d) the loop gain is 800 at 100Hz and 80 at 1kHz.
It indicates that the required gain can be easily attained by boosting
up
the current gain below 10kHz.
1-2. Shot Noise
The shot-noise limited frequency noise was calculated to be a factor
of 10
lower than the aimed frequency noise with 1mW light power on the photodiode.
1-3. Electronic Noise
The electronic noise of the photodetector and the demodulation system
was
measured to be comparable with the shot noise with 1mW light power
on the
photodiode. Thus it is negligible for the aimed frequency noise.
This noise right after the mixer was measured to be 50nV/rHz. Therefore
the
input equivalent noise of the filter/amplifier after the mixer should
be
below this level. For this purpose the low-pass filter is implemented
at the
high voltage amplifier output instead of before it.
1-4. Feedback to the Mode Cleaner
Since the feedback to the mode cleaner at low frequencies reduces the
effective frequency stabilization gain, the relative gain of the mode
cleaner path to the PZT path should be small enough so that the required
frequency stabilization gain is maintained. The cross-over frequency
of
these two loops is set to be around 300mHz, which ensures no degradation
in
the effective frequency stabilization gain above 100Hz.
1-5. Spurius Servo Path
The spurius servo path from the feedback voltage applied to the PZT
to the
demodulated voltage via amplitude modulation coupled with the 12MHz
amplitude modulation could limit the frequency stabilization gain.
The gain
of this spurius path was measured to be 3e-5. It was analized that
this
spurius path will limit the loop gain below 300,000, which is large
enough
for the required gain.
2. Locking Acquisition and Holding
2-1. Acquisition
For acquisition the frequency of the laser should be adjusted thermally
so
that the TEM00 mode appears within the PZT range. Once the mode cleaner
is
locked with the PZT servo path, the mode cleaner servo path should
be
activated.
Incidentally we found that we can acquire lock with a servo bandwidth
that
is a factor of 10 less than our initial estimates, which gives encouragement
for acquiring lock in our test cavities.
2-2. PZT Range
The PZT feedback voltages fluctuate by 40Vpp typically. It exceeds 60Vpp
every few minutes to tens of minutes, depending on the seismic activities
around the lab. This is caused by the mirror motion of the mode cleaner.
The
motion is dominated by the 1Hz pendulum swing and 2-3 Hz stack resonances.
The contributions for the rms feedback voltages from these two components
were found to comparable, which indicates that the damping of the pendulum
motion is nearly optimal.
At present we have only a 60Vpp amplifier for the PZT driver, so it
can hold
locking only for a few minutes to tensof minutes, but we plan to replace
it
with a 100Vpp amplifier, which is expected to hold locking for much
longer
time.
2-3. Drift Compensator
Since the PZT driver has a limited range, the frequency drift should
be
handled by other actuators. Two possibilities are to change the laser
frequency by the thermal actuator and to change the mode cleaner length.
We
measured the frequency stability of the laser and the length stability
of
the mode cleaner below 100mHz and found that the laser is more stable
than
the mode cleaner by a factor of 10. Therefore we decided to implement
the
mode cleaner path to compensate for the drift.
2-4. Feedback Path to the Mode Cleaner
The mode cleaner path is split from the output of the high voltage
amplifier, since it is the output of the high voltage amplifier whose
drift
should be compensated.
3. Other Specifications
3-1. Finesse of the Mode Cleaner
The cavity pole of the mode cleaner for the TEM00 mode was measured
to be
about 60kHz by the following method. The mode cleaner was locked to
the
laser only with the mode cleaner servo path with a unity-gain frequency
of
about 200Hz. A transfer function from the voltages applied to the the
Pockel
cell to the demodulated voltages was then measured. The transfer function
had additional 45degree phase shift at 60kHz. This corresponds to a
finesse
of 2,500.
The same measurement was done with the TEM01 mode, and the cavity pole
was
measured to be 18kHz, which corresponds to a finesse of 8,000.
>From these measurement we suspect that the mirrors might have been
contaminated around the spot for the TEM00 mode.
3-2. Coil Driver
The DC efficiency of the coil driver from the input voltage to the length
change was measured to be 13um/V by the following method. The mode
cleaner
was locked to the laser only with the mode cleaner servo path with
a
unity-gain frequency of about 2kHz. A voltage is then applied to the
PZT
input of the laser at 100Hz and the corresponding signal in the feedback
voltages to the coil driver input was measured. Knowing that the pendulum
frequency is 1Hz, and the PZT efficiency is 4MHz/V, the DC efficiency
of the
coil driver was obtained.
LASTI (Zucker, Shoemaker, Mason, Smith, Kruzel)
----------------------------------------------
MZ and DHS worked on documentation for review at LSC
meeting, noise & servo calculations.
K. Mason worked on the optomechanical layout and
planning for the upcoming Phase I SEI installation.
M. Smith and E. Kruzel worked on preparing the
lab and obtaining parts and hardware for the
SEI installation.
* Matt is visiting at LHO from Aug. 17 to work on the lock acq.
* Biplab is doing some systematic runs to study lock-acquistion time
of
LIGO in the presence of misalignments.
At LHO
* Biplab, Malik and Hiro are working to make a simple set-up
of 2k IFO in order to examine various effects easily using e2e.
* Biplab worked with Rick and Brad on Reference cavity.
Ran time-simulation of Reference cavity with
cavity motion data supplied by Cella from his mechanical
simulation. A few confusions arose. Had a telephonic
conversation with Cella. Some filtering in "data-reader"
module may be included by Hiro later to avoid reading beyond
Nyquist frequency.
* Rai claimed that it is necessary to have a scientist
at the site
who knows about e2e well in order
to make e2e useful. (GHS - We must achieve this capability at both sites.)
* Hiro talked with Cristine and several things were straightened.
Due to a misunderstanding about the software maintenance
procedures at
LHO, Hiro and Ed installed gnu software in the e2e user
area,
which introduced extra burdons on Cristine.
Cristine offered to maintain necessary softwares updated
timely with the collaboration and necessary input from
us.
Alfi
* A new version 4.0.14 was released and is installed at LHO as well.
* Renaming capability is back to function.
* Ed and Bruce are working to fix bugs which cause segmentation faults.
Software Systems (Blackburn)
------------------------------------------------A long lasting bug with ILWD data socket test code was isolated this week
to be in a TCL test script. New ILWD data socket tests using threaded versions
of the socket communications code identified a race condition in the threaded
version of exposed TCL commands found in the genericAPI. These have been fixed
and are now under testing.A bug in the frameAPI which was causing a newline character translation for
binary data was found and fixed this week.Attempts to run the dataConditionAPI on a quad Sun 450 to test the thread
capabilities of the code failed to thread more than one call chain. This is
even fewer than the two that was occasionally seen on the quad pentium which
was limited to no more than two threads. The race condition that was fixed in
the genericAPI may impact this result and will soon be tested.Data communications performance for the wrapperAPI was yet again improved by
another factor of 10x. This brings the overall performance increase on our
SMP boxes up by a factor of 100 over the first draft of the code. This is now
in the ball park of what is needed for target LIGO beowulf platforms.The new versions of GDB and DDD, used for debugging, fixed the problem we were
seeing with debuggers detaching themselves from our LDAS APIs. This should
greatly improve the debugging cycle in the traditional LDAS enviroment.LDAS installation documentation on the web page was updated this week in
response to problems discovered by LSC collaborators while building versions
of LDAS on home institutions.The versions of LDAS installed at the sites were updated this week to include
the latest set of code fixes and new APIs. This will allow more concurrent
testing of LDAS software across systems while pressing hardware upgrades are
scheduled for upgrading servers and such.The problem with the operator socket used for LDAS commands blocking during a
length process which was discovered and fixed during the MDC has been migrated
from the dataConditionAPI to all LDAS APIs.The managerAPI has been enhanced to more reliable startup and shutdown other
APIs in the system. In addition, it will auto-sense the location of the
anonymous ftp, no longer requiring ftp to be installed in a particular way.A good percentage of the group attended the LSC meeting this week in Hanford
and have held multiple side bars with collaborators on issued ranging from
future engineering runs to adding new modules to the dataConditionAPI.
Hardware Systems (Anderson)
Data Analysis Activities
MIT:
Nothing to report.
Livingston:
We have received two Sun Ultra 10 workstations, which we are using
for
testing in our upgrade to Solaris 8.
Hanford:
Set up 4 new Sun Ultra 10s. Two for long-term visitors and 2
in the
common areas. Fixed a path problem when mounting files from the
license
server to each Sun workstation. The gnu compilers now work on
all Suns
except for rainier. Rainier is in the process of becoming a file
server
only. In the not too distant future user login will be turned
off on
rainier.
CIT:
(Lisa)
- Working on the Cadence installation. Most of this work is being done
in the
evenings because of the need to move the licenses around for the new
s/w install
which do conflict with the present production system.
- Still building the the new Ultra 10's that have arrived.
(Samantha)
- Worked a number of PC issues and getting data for new computer video
equipment.
(Suresh)
- Still on vacation.
(Larry)
- Working a number of procurement issues. Getting quotes for a new
plotter.
The order for the pc's to work with the projectors has finally been
placed.
Now trying to resolve a number of problems with the SUN orders and
billings.
Getting quotes from FORE to get more ports on our existing network.
- Resolved a number of printer issues. Most were just things needed
to be
cleaned.
- Working with Rita on getting a new pager service going for the group.
- Contacted Bellsouth (Ganel Young) to get more information for a OC3
connection
to the Observatory in Livingston.
- Installing equipment into new rack.
- Worked with Ed C. and had a number of old pieces of equipment turned
over to
Caltech.
(Barbara)
- Continued efforts on the Roster database. Added new fields,
forms,
reports, etc. Have been working through new features in Access
2000.
- Updated Publication and About LIGO pages on LIGO site. Made
minor change
to the LDAS site.
From: Mike Smith <smith@ligo.caltech.edu>
LIGO 2
Scattered Light Requirements:
An analysis of the scattered light requirements for LIGO 2 is proceeding.
PO/APS Telescope:
Budgetary RFQ has been placed for PO/APS telescope parabolic mirrors.
PO Mirror:
Budgetary RFQ has been placed for PO mirrors for LIGO2
ETM Telescope:
The current LIGO 1 design for the ETM telescope will be used for LIGO
2.
Mike
From: Peter King <pking@ligo.caltech.edu>
3.1 LIGO II PSL
The second draft of the PSL WBS Dictionary has been completed. The
additions being the high power laser R&D sections.
A first cut at the PSL schedule has been done, without any serious
thinking done about the sequencing of tasks. I will be checking the
schedule to make sure that no obvious tasks have been omitted.
Two more quotations for PSL components trickled in.
Some noise measurements on the Stanford high power laser will be
performed next week. Not sure which ones just yet but the most likely
one(s) will be intensity noise measurements.
From: Garilynn:
As reported last week there are preliminary homogeneity measurements
from CSIRO. These are measurements of the first of two 15 cm diameter
sapphire pieces from Crystal systems. Jordan has presented data from
the best of the two pieces at the LSC meeting.
For any who did not attend:
These are measurements of a piece of m axis sapphire, surfaces have
been subtracted. The piece is probed with a 690 nm polarized laser
in a Fizeau interferometer. The orientation of the glass is the same
in both data sets, in effect only the polarization is rotated.
0
degree measurement, the color scale is in nanometers
Cross
section of 0 degree measurement, vertical units are in micrometers,
horizontal units are millimeters
polarization
rotated to 90 degrees, the color scale is in nanometers
Cross
section of the 90 degree measurement, vertical units are in micrometers,
horizontal units are millimeters
The current best-guess about the fine structure is that the crystal axis is wandering slightly and causing localized changes in the beam polarization. I would like to have CSIRO probe one of these pieces with an elipsometric measurement to verify that our guess is correct. They will be providing a quote for that measurement.
For additional information about this report, contact sanders@ligo.caltech.edu
