The LIGO Executive Committee Agenda for Monday July 5, 2004 will be:
No Meeting due to holiday
Special Items:
no report
Status of LSC/MOU Research
Updates and Program Reports (Petrac)
(LSC Research Updates through August 2004, and Progress Reports
through February 2004)
no report.
LIGO Weekly Site Telecon (Lindquist)
There was no site teleconference scheduled Thursday, July 01,
2004.
PROPERTY MANAGEMENT (Chargois)
>From: Ed Chargois <chargois_e@ligo.caltech.edu>
Nothing significant to report.
DOCUMENT CONTROL CENTER (Turner, Mak)
>From: Linda Turner - turner@ligo.caltech.edu>
No report.
> From: Cleveland Mak <mak_c@ligo.caltech.edu>
packages:
in - 12
out - 9
faxes:
in - 24
out - 25
FINANCIAL (Cunningham, Brambila, Kaufman,
Salone)
>From: Esther Cunningham <esther@ligo.caltech.edu>
For ACCOUNTS PAYABLE HISTORY DATA: http://www.ligo.caltech.edu/%7Ephil/Weekly/AccountsPay.htm.
>From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
>From: Gina Salone <gsalone@ligo.caltech.edu>
No report.
>From: Florence Kaufman <fkaufman@ligo.caltech.edu>
Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport.
(For passwords contact Florence)
SUBCONTRACTS MANAGEMENT (Petrac, Jasnow)
>From: irena@ligo.caltech.edu (Irena Petrac)
No report.
>From: Ed Jasnow <jasnow@ligo.caltech.edu>
SUPPORT (Baldon, Lloyd)
>Irene Baldon
>Dorothy Lloyd
ADVANCED LIGO (Cost
Schedule Control Systems) T. Frey
>From: Thomas Frey <tfrey@ligo.caltech.edu>
See Advanced R&D Section.
Reports (Lindquist)
Operations Annual Report--the Annual Report for Operations is
due at the end of June. I have prepared and distributed an outline and
assignments, and requested contributions by Friday, June 18, 2004. A
number have not yet been received.
We will be scheduling the preparation of a proposal for FY 2007-2011 Operations
later this year.
Change Control/Contingency (Lindquist)
I have saved a copy of a change request submitted by Albert
Lazzarini at:
http://www.ligo.caltech.edu/~phil/ChangeBoard/CR040013.pdf
The request is for budget needed to cover LDAS costs at Hanford. Budget
was initially retained in planning packages for the various LDAS maintenance
accounts to provide aggressive targets and fiscal flexibility. I propose
to discuss this change request during the next normally scheduled meeting of
the LIGO executive committee, Monday, July 12, 2004.
Human Resources (Akutagawa)
>From: Cindy Akutagawa <cindy@ligo.caltech.edu>
Various personnel/payroll /HR related work.
Quality/Safety (Tyler)
>From: Bill Tyler <tyler@ligo.caltech.edu>
Scheduled and sent out notices for the LIGO SURF student laser safety
class. Also scheduled baseline eye exams for the students that will be
working with lasers.
.
Summary of Commissioning Activities at LIGO Hanford Observatory (Landry)
Activities this week centered around a vent of the 2K corner, to install a new
Faraday isolator. Vent preparation
started Tuesday, followed by the main Wednesday activity,
including documentation by a film crew with subsequent fly-bys.
Vent followup
showed that the vaccum was being restored as expected, the 2K has had its modecleaner
locked, albeit at low power, and the beam tubes have been exposed to the 4K
(only), so that it is back
in business.
Photon Calibrators (Evan Goetz, Corey Gray)
All four Photon Calibrators have been installed. Calibration
measurements have been run (posted in e-log and will post on Photon Calibrator
webpage:
http://www.ligo-wa.caltech.edu/~cgray/photon_calibrators.html). Evan will
continue w/ commissioning of the Photon Calibrators as his SURF project this
summer.
The cosmic ray detector comprises two paddle scintillators viewed by four photomultiplier
tubes, resident under BSC1. The detector received its own IOC,
and a dedicated DMT monitor was being commissioned this week.
4K IFO
The ITMs received revised
anti-imaging boards (rev. C instead of rev. B), to accomodate DAC differential
inputs. This means the ITM calibrations will need to be revised,
important when propogating calibration to the ETMs. The current ETM cals are fine however.
A measurement
of the mode cleaner length found it to be 12.24599541 m.
Super-polished beam splitters partition
the light equally between the four ASPDs.
Last week it was noted that the installation of the new crystal oscillator only
reduced our high frequency bump in AS_Q, and did not eliminate it. More work
has been performed: the 25MHz distribution system's phase
noise was analyzed. This, and careful perusing of the RF distribution
schematics led to a new RF distribution
scheme. Better
high frequency noise performance was the result. More oscillator phase
noise measurements are described here.
A bold model
for mis-alignment-to-AS_Q coupling in the output mode cleaner (OMC) was
formulated. The observed coupling was larger than anticipated, and it is
suggested here that an order of magnitude improvement could be made by modifying
the OMC configuration from a triangular cavity to one composed of four mirrors.
2K IFO
Like the 4k, the 2K ITMs also received
new rev C boards, to accompany new DACs. The associated software upgrade
had some startup
trouble, however.
Prior to the vent, the PSL Faraday isolator was removed.
DAQ
A fast ADCU is being commissioned.
Safety (Rich Riesen)
LASER SAFETY
1). Met with LLOs laser eye exam ophthalmologist concerning the new exam
requirements referenced from ANSI Z136.5-2000, Section 6, Appendix G. The
changes have been made and are now in effect
2). Editing the LLO laser safety sop, correcting typos, incorrect
information, etc. that were found during our annual safety audit.
3). Responded to a concern requarding high noise levels produced by the
new CDS rack fans. Referencing OSHA regulations (Standards-29 CFR),
Occupational noise exposure. 1910.95. OSHAs permissible noise exposure is
90dBA slow response for an 8-hour day. The levels measured in the corner
station electronics rack area (racks were empty and sides off), was 84dBA max.
The reading at the Y-end station with the racks stuffed and closed was 74dBA.
This is below OSHAs permissible levels. Note: Earplugs are readily available
for anyone who wishes to use them.
SITE SAFETY
1). I found no safety concerns during the weekly site tour.
2). Installed proper warning signs on the flammable storage area doors.
3). Installed safety glasses dispensers with clear safety glasses next to
the 3 entrance doors to the machine shop and 1 next to the first aid kit in the
staging building HEPI buildup area.
Other:
Working with Ed Chargois and Tom Evans in the disposal process of computers and
monitors that are no longer in service. Shipped a crate of alignment fixtures
to Myron MacInnis at MIT.
LDAS (Igor Yakushin)
1) Reinstalled DB2 and ldas-dbcentral to be able to use administration server.
2) Replicated LLO_1 from LLO's metaserver to ldas-dbcentral through VPN to test
that it works.
3) Running replication test between ldas-pcdev1 and ldas-pcdev2 for 4 days
successfully. 3.8 million rows were inserted into a table at ldas-pcdev1 and
appeared in a replicated table in ldas-pcdev2 a few seconds later. Preparing
similar test with the real production database between LLO and ldas-dbcentral.
Networking & GC (Shannon Roddy)
* Current bandwidth usage can be seen at http://teche.ligo-la.caltech.edu/mrtg/LLO-Router/130.39.245.1_1.html
* Archived bandwidth usage can be seen at http://teche.ligo-la.caltech.edu/mrtg/archive/
under the dated folder for the week of interest.
* Spent a lot of time trying to get the new backup server running this
week. Thursday I tried to install an external SCSI disk on the machine
and it would not install. Solaris did not see it at all. It spewed
errors when I tried to do a probe-scsi-all from the OK prompt. We have
had SCSI problems in the past with this machine and always suspected the
disks. However, since I had errors with the new disk, I then assumed that
it was a problem with the machine. I put a call into Sun and they had me
run several diagnostic routines, etc. They also had me update the boot
PROM. None of their suggestions fixed the problem, so they sent out the
local Sun guy to replace the mother board. This didn't fix the problem
either. After trying many things over the period of a few days, I finally
got the machine running with a different external disk. Turns out that it
was probably a combination of errors. Unfortunately, the amount of time
spent trying to get this machine running put me behind on some other items, but
I needed to get the backup server running.
* Installed the new licensed copy of Arkeia on the new backup server. Installed
the tape library and labeled 30 tapes so that I can start bcking up with it
tonight. I will start with the home directory/mail server and the web
server (general web & ilog) since those are the most critical. I will
continue to add machines next week.
* Had several problems while trying to upgrade ghost script & ghost view.
Not sure if it is the compiler or something else that it does not like, but I
could not build the shared object version of ghost script.
* Installed a central syslog server and started several machines logging to
it. I plan on installing some log analysis software on this machine once
I find a suitable package. This will help me keep tabs on things around
here.
* Was informed by the local Sun guy that any machine that was bought >6
months ago and is running Solaris 9 should have the boot PROM updated. He was
rather surprised that we were successfully running Solaris 9 on with the older
revs of the OpenBoot PROM. He said there are numerous bugs in the old
versions. This means that the web server and some other servers will have
to be rebooted in the next few weeks.
* Have not been able to replace the failed switch in the new building yet due
to the difficulties with the backup server. This will have to be done
soon.
* Had the usual round of support issues and student account requests.
* upgraded Mathematica on the Suns to version 5.
Ken Franzen:
1) Still writing on my draft documenting the recent WaveMon S2 veto work.
2) Converting the on-line produced S3 WaveMon triggers from ROOT format to text
format. This was requested by the glitch group during last week's telecon in
order to make the triggers more accessible for all members. Due to the large
data volume I have had some problems, but the triggers should be posted within
in a few days.
Rupal Amin:
HPLF:
Late last week, our laser experienced an explosive malfunction with its fiber
optics/electronics. This laser had less than 48 hours of total usage and
failed while its beam was being dumped on an HPLF style beam block. There is no
diagnosis from IPG Photonics as to why the device failed. Ken Franzen and I
could not determine where the damage was located due to the sealed nature of
the package.
WFS:
I am helping Ken Watts continue repairing and modifying the WFS heads as per
request of Rai Weiss. Last week we found several blown capacitors on WFS
3 and 1. All WFS heads (except for the spare) were found to have
under-rated or incorrect capacitors on board. We will check the all caps
with respect to the blueprints. Perhaps they were altered for some
reason. After this repair, G. Gonzalez is free to test the optical
response of the heads.
Phase Noise:
Since the laser failure, I have been focusing on improving the phase noise
measurements from the new Wenzel crystal oscillators. Thus far, I have
re-engaged in my battle against RF and 60 Hz noise. I hope I can obtain decent
numbers today or early tomorrow.
PSL: No report. (Chiller filled = laser happy).
Modeling (Dodda, Jamal, Rogillio and Yoshida)
We continued characterization of HAM1 table motion using e2e. We took DAQ
signal of HAM1 accelerometer, integrated it twice in time using Matlab (after
removing the DC offset), and gave the resultant displacement time series to HAM
stack box. The HAM1 table-top motion computed by HAM box was then given to MC1
box to compute the position and yaw osem signals. The power spectrum of the
position osem signal shows good agreement with the DAQ signal of MC1 position
recorded simultaneously with the above-mentioned accelerometer DAQ signal. (We
shifted the U-U and V-V transfer stack resonance peaks of the HAM box from 1.6
Hz and 1.8 Hz to 1.45 Hz and 1.6 Hz, respectively, so that they may be
consistent with measured stack transfer function) The yaw osem power
spectrum appears to be higher than the corresponding DAQ signal by a factor of
8 near the 1.6 Hz resonance.
The construction of Mode Cleaner box was also continued. To test the MC box
under a realistic condition, we gave HAM table-top motion that we previously
estimated to all the three MC mirrors. The MC length sensing mechanism seems to
work but the MC cavity has not been locked stable yet. It seems that the filter
used in the length sensing control needs to be modified for steady lock. We
look into the problem. The IOO box except for the MC box has been combined with
Core Optic Box of SimLIGO.
HEPI:
All HEPI pier-top assemblies are now in place. The two end stations are
also fitted with electronics, and wiring of the corner station is under
way. Gary, Harry and Danny were able to reposition the ITMY table using
spring tensioning, and have continued installing actuators. Joe Hanson
has been wiring in the LVEA, and has began bleeding the actuators in HAMs 1, 2
and 4.
Chethan and Rolf wrote and installed an improved watchdog system, allowing the
user to specify pre-filters to be applied before an absolute value and
thresholding.
At the X end station, the the system has been operated in its 'crudest'
mode. We servo the payload to the eight displacement sensors, using
feedback directed along x, y, z, the three angles and the two over-constrained
modes. The displacement error signal in x, y and z is 'corrected' for ground
displacement using filtered signals from a floor-mounted STS-2 seismometer.
We are seeing about a factor of ten reduction in noise in the 0.1 - 3 Hz band
using this setup. We are adding some noise below about 30 mHz and at
about 10 Hz.
The overall commissioning scheme requires two more techniques to be applied: 1)
we need to make use of the small L-4C geophones that are mounted near each
actuator, in order to get a bit of inertial-feedback vibration reduction at a
few hertz, decoupling the feedback loop from the displacement sensor noise, and
2) we need to improve the low-frequency high-pass filter in the sensor
correction, probably using an FIR filter designed using a method that Wensheng
Hua studied and tested.
For example, a set of graphs from the July 1 LLO elog shows the y-direction
transfer function from the floor to the crossbeams (measured using a witness
GS-13 up there), the coherence of that coupling, and the displacement noise on
the crossbeams. Each is shown with and without the HEPI system servos
closed. The transfer function is lowered by more than an order of
magnitude between about 0.1 and 2 Hz, while the measured noise decreases by
approximately a factor of 10. (Performance in X was similarly measured using
the GS-13s. Z performance has not yet been measured with the the GS-13s,
but looks to be good based on L-4C measurements.)
(installation: Abbott, Spjeld, Traylor, Overmier, Hanson, Hayden, Carter,
Sellers. commissioning: Mittleman, Coyne, O'Reilly, Wen, Busby, Lantz, Giaime.
and more.)
---
A team working on the test mass thermal compensator was at LLO this week, and I
imagine that their report will appear elsewhere.
see also the CDS weekly meeting minutes in the commissioning archives:
John Zweizig
The week before last, I installed a new DMT version (rev 2.9) at LHO and set up the sixth Blade-1000 as a DMT machine for online use. The new DMT version includes support for the new gds trigger table format, a more efficient data vector template and other improvements. It is compiled with root version 4.00-06 (which has a Solaris bug in the io streams fixed) and the lastest release of frameCPP
I also reconfigured the data quality monitor that produces the Figure of Merit plots at LHO to exclude the line features (power lines, violin modes, calibration lines, etc). Since returning to Caltech I have continued to verify the new version, specifically the new trigger table format.
no report
Jay Heefner reporting
Fiber Optic Timing Link for HEPI EMI (Sander)
==========================
- Bench testing is complete and has been summarized in a report. The report will be submitted to the dcc.
- We will now test the link on the 40 meter end crates. Sander will order the necessary fibers.
Timing System Redesign (Flavio)
============================
- On hold for ISS
Anti-Image Filter for Low Noise DACs (Jay, Mohana)
======================================
- Boards are being fab'd. They are due back 6/28.
40 Meter (Ben, Jay)
====================================
- Installation of MZ servo and controls is complete.
- Database and screen testing in progress.
HEPI (Ben)
===================================
- The remaining Pier Interface Modules have been completed
and tested, and will be shipped off to Livingston soon.
LSC PD Redesign (Ben)
===================================
- Putting the final touches on the schematic for the RFPD. The layout should begin shortly.
ISS (Flavio)
====================================
- Review today at CDS Meeting
LLO EMI Upgrade (Mohana, Jay)
=====================================
- System drawings and cross connects lists are complete and have been sent to LLO.
- Cables have been ordered. Most are 3-4 week lead, but the 64 conductor cables are currently being quoted at 13 weeks by the vendor (Wilco). The spectra-strip cable is not available for 6-8 weeks. Wilco will try to expedite.
- EMI feedthroughs are on order.
AdL SEI (Jay)
====================================
- Capacitive position sensors have been ordered. We have recently discoverred a problem with the vacuum compatibility of the epoxies in the probes. Jay and Larry Jones are working with ADE.
- The low noise coil driver sent to Stanford has been tested and meets the requirements.
AdL SUS (Jay)
===================================
- MC triple electronics and controls are installed at LASTI and are working.
TNI (Jay)
====================================
- Working with Eric Black on an automated arm cavity servo and locking system.
Other
====================================
- Ben has issued a document with guidelines for RFI connectors, feedthroughs and cables. The document can be found at http://www.ligo.caltech.edu/~babbott/RFI_Considerations.pdf
PeterKing
I have run into some problems with the event handling in the power amplifier diode current state notation controller code. Timing in the second state set does not get triggered with the event flag that I declared. I am still trying to sort this out. In testing of the code so far the IOC seems to hang (or at fails to respond to keyboard input) but does not crash and still updates MEDM screens.
Mike Smith
Optical Bench: The fixed polarizer arrived from II-VI and
the mounting hardware was received from CIT and Southern Enterprises.
One CO2 laser optical bench was aligned in the optics lab. The infrared pattern
was bounced off of 2 folding mirrors to achieve a 20 m projection path length
in the optics lab and projected on a screen on the wall. The aiming beam cross
pattern was focused and roughly aligned with the infrared annular pattern. A
simplified alignment procedure was developed to align the raw beam, the central
heating spot, and the annular pattern.
LVEA installation: The TCS enclosures were received and the support base was notched to clear the dog clamps on the optical lever pier. Leveling feet were added to the support base, because the LVEA floor was not level at the intended location of the TCS enclosure near.
EPICS wiring: Dave O., Stefan, et al at LHO are developing requirements and schematics for the interface electronics and wiring between the control room and the TCS table. Stefan has volunteered to design, fabricate and test the circuit boards for LLO and LHO.
Cooling for CO2 laser: A proposal was made to use commercial chillers and to place them on the x and y arm exit slabs near the optical lever piers. Dave O. is in the process of getting price and delivery quotes for a water-cooled version of a commercial chiller that would use the site cooling water to remove the heat instead of a cooling fan.
Note by D. Coyne: Fred Raab and I decided that LHO personnel will test the effect of the 2k TCS chiller on/off on IFO performance before a Technical Review Board discusses the above proposal.
CO2 laser power monitor: A decision was made to use the same type of PD that is used to monitor the power in the projected heating pattern to monitor the power sampled directly at the output of the externally polarized laser. A cost growth summary is being prepared to reconcile the system costs based on the recent changes: purchase of new chillers; use of rotating polarizers instead of AOM to control the heating power; cost of EPICS interface board; addition of long-term DC power monitor for the CO2 laser output.
Commissioning schedule delay: Commissioning of the TCS will be delayed until August due to the following undetermined issues: fabrication of the TCS interface circuit board and completion of the EPICS wiring; installation of a permanent water chiller for the CO2 laser; completion of the HEPI installation and the availability of a fully-locked interferometer.
A wire mesh attenuator was designed and parts are being
built; they will be shipped to LLO on 6-28. Two versions are being made: 36%
transmissivity, and 49 %. These attenuators are intended to be placed in the
beam leading to the PD to reduce the power density to a safe level. Mounting
posts for the attenuators were ordered and are being shipped to LHO and LLO by
6-28.
2 fixed polarizers were ordered from II-VI and shipped to LLO. Only one rotatable mount was in stock, so I designed and am having made, a rotatable mount for the 2nd polarizer.
An alignment mirror assembly for use at LLO in the optics lab was designed and will be shipped to LLO on 6-28. The plate-glass 8 in Edmund mirrors are at LLO.
2 crossline aiming lasers were ordered from World Star for use at LLO. The 2 crossline aiming lasers from Edmund at LLO were shipped to LHO.
Extra beam dumps were ordered to catch the pick-off beam that will be used to monitor the CO2 laser power, until a decision is made re what power meter will be ordered.
Ge mirrors were ordered for LLO and LHO for the CO2 laser pick-off.
I am in the process of designing a finned beam dump for the rotating polarizer assembly.
Lee Cardenas, Liyuan Zhang
OTF Lab. (W. Bridge)
We are having hard time to get the right set of clean mirrors in order to get a previous transmission and ring down as well as beat frequency. So far we have tried two sets and now we are in our third set. Hopefully this set will do the job as right now cavity chamber is pumping down. we have a new sample to test which is a white ceramic. I suggest to put this sample in the other chamber where the blue epoxy is at. We can do this tomorrow and by Friday, we can tell if this sample is dirty. So far we are working with the chamber in OTF lab.
Absorption Test Measurement prototype in standby
The 30 watt laser enclosure.
Enclosure completed and parts are going to be anodized.
Scatterometer system in standby
OTF Lab at Lauritsen ROOM 38
Cavity #3
The cavity is locked and we have optimized and improved its output power and transmission. We have introduced (6) disks of TRA-BOND #2151 color blue epoxy.
We are taking ring down and beat frequency measurements every day. So far we can tell the parts are clean! Because our ring down (loss) did not change and the beat frequency(absorption) have not either. BUT in order to really have very real measurement, the cavity needs to be running al least 1000 hours.
Cavity #2 Test cavity optical set up is still in standby
PSL.
In STANDBY for now!
no report
no report
Simulation and Modeling
(Hiro)
**** Commissioning support (Biplab)
Biplab has been working at LHO and LLO sites in the past two weeks.
- At LHO, helped Joe to set up phase-camera at the asymmetric port. Until now
the camera set-up for demodulation with (fibre-carried) 65MHz carrier was
available only at POY port. This is a better way for getting proper information
about individual sidebands and can be directly compared with FFT simulation.
With antisymmetric port set-up we will also get better information about the
field that enters the Output ModeCleaner and could be causing some of the
problems.
- At LLO, with Andri Gretarsson discussed ways to extract field information out
of phase camera with 25 MHz and 50 MHz demodulation but there are too many
ambiguities to get proper information about individual sidebands and concluded
that a set-up like the one at LHO is the thing to do, especially when thermal
compensation will start working.
- The Wave Front Sensors (WFS) at LLO had a number of problems in operation.
With help from Andri got all measurements for the lens-distance set-up on
tables for WFS. Earlier when WFS was made to work, these set-ups were adjusted
to get maximum signals for target degrees of freedom but the actual values of
their gouy phases were unknown. The Gouy phases calculated from my Matlab codes
seem to be different from theoretical maxima for the target degrees of freedom.
Finished calculation of some of the WFS signals for LLO from FFT runs and
compared with those in my note on "The effect of Thermal lensing on
wave-front sensor signals"(T040066) which is based mainly on H1
interferometer. No significant difference with H1 WFS signals observed.
- Interacted with Sany Yoshida and his students from Southern Louisianna
University on use of e2e and SimLIGO.
*** Improvements of the Mechanics modeling (Virginio)
This is a preparation to study the lock loss of IFO at LLO.
Found that the implemented BSC TF in SimLIGO do not match the measured TF by
Hytec S (HYTEC-TN-LIGO-36) 12/29/98. Tried to contact Hytec to get
the t.f. measurements they made in 98. Those ones seem to match quite well
experimental observations.
Using PEPI and seismometer signals to inject noise into SimLIGO. Results under
investigation
Abandoned the use accelerometers signal acquired during S1 to inject seismic
noise into SimLIGO.The signal is completely dominated by some excess noise at
low frequency. All the LVEA acc. signals also have a very high coherence up to
5Hz which is quite improbable. 721-207 Wilcoxon acc. Datasheets suggested
that they were good down to .5 Hz and even some sensitivity down to .2Hz.
These suggest that the common multichannel acc.s driver is probably responsible
for such a noise.
Working to Introduce the right topology on the BSC stack to account for the
suspension position respect to the stacks optical tables
Writing few matlab scripts to produce standard output to monitor simulated
subsystem to speed up a more systematic debugging process
*** Seismic noise (Sany and his students)
Sany kept working to measure the pitch and yaw motions of mirrors to compare
with the simulation result. (This is an important part to use e2e for the lock
loss study using actual seismometer and accelerometer signal).
Pitch signal has been reproduced reasonably well, but the yaw motion revealed
some problems. Some possible issues to take into account to do the actual data
vs simulation have been pointed out, and Sany is revisiting.
Virginio pointed out the accelarometer signal is not reliable because of the
electronic noise. When Virginio studied coherence among seismic signals, strong
coherence was observed among those which should be completely independent. This
is the reason why Vitginio is suspicious about the electronic noise. But Sany
said his measurement of coherence does not show those spurious coherence.
Virginio commented that the electronics could have been revised since S1, whose
data were used by Virginio.
*** LIGO I Input Mode Cleaner modeling (Sany Yoshida and his students)
Sany and his students have been working to build an e2e setup to simulate the
input mode cleaner. At this week's e2e meeting, the current status has been
presented and the problem of locking was discussed. Suggestions were given to
find if the locking filter used at the site is to be used in the simulation or
if a different filter needs to be written.
*** FFT related work (Hiro, Biplab, Xiao)
All HR side reflection phasemaps have been processed and are available from
www.ligo.caltech.edu/~e2e/FFT. Beam splitters have been analyzed, and a plot of
cross sections and curvatures is given in http://www.ligo.caltech.edu/~e2e/FFT/LIGO_I_data/BeamSplitter.pdf
Using the modified FFT program which handles curvatures of the BS in the same
way as other mirrors, the effect of the curvature has been studied for the
Hanford 4k IFO. The asymmetry of upper and lower sidebands is affected by the
curvature. The distributions are shown in a plot,
http://www.ligo.caltech.edu/~e2e/FFT/pdfs/darkport.pdf. The ratio of side band
gains in a hot state (mode matched by thermal lensing) changes (Gain(lower side
band) / Gain(upper side band) ) from 1 to 1.3 when the beam splitter curvature
changes from flat to 200km, while the ratio of dark port output powers (power
of lower sideband / power of upper side band) changes from 1 to 1.14. These are
the results using the locking algorithm of FFT. Plans to understand this result
is to use e2e and to apply length locking to FFT in the same way as the real
IFO.
*** Modal Model (Hiro)
In order to study the noise of the output mode cleaner, the modal model in e2e
has been improved. To simulate a beam whose waist in x and that in y are off by
several cm, it is not clear if modal model works. It is now being studied
analytically how the approximation works to represent such an astigmatic beam
using limited number of modes. If a modal model does not work, FFT-based mode
cleaner will be built.
*** Modeler code (Hiro, Biplab, Melody)
Biplab and Hiro worked to revise the code to handle the beam splitter curvature
issue in modeler.
Melody resumed work to process a .sim file, the modeler file which contains
consolidated box files information.
*** Alfi (Bruce, Melody)
Bruce is working to implement a new functionality called BUNDLE. A version
tried before revealed some problems, and Bruce has revised the architecture to
make it more robust.
Melody Implemented Hiro's suggestions for the edit menus. The changes
have been committed to CVS.
Melody finished regression testing using JGo 5.0, Java 1.4 and Java 1.4.2. The
new version of JGO has been committed to CVS and is a part of the newest ALFI
release (5.9.0).
LIGO Data Analysis System
Software Systems (Blackburn)
Coded up bug fixes for errors with respect to cacheCheck failures
(PR2603). Tested on tandem and dev yesterday, and committed changes to
CVS yesterday. Will test from nightly build today, then close this PR.
Finalized C++ code for speeding up frame queries at the C++ layer (PR2390). The
new API has been tested on tandem-5 this week and on dev last night. We will do
one more test as requested by Stuart before this code gets committed to CVS later
today, in preparation for weekend tests.
The bug associated with mail delivery as seen at LHO and MIT has been fixed by
reqorking the TCL code to use sendmail instead of direct socket communications
with the mail port.
The version of LAM has been upgraded from 7.0.4. to 7.0.6 on the LDAS-DEV
system and all tandem systems.
Modified the cacheDump utility to properly report malformed diskcache API
resource files and to accept a cache filename as an argument.
Added code to display frame by site and type in the diskCache viewer graph of
the controlMonitorAPI's client. Also added a button to list details of gaps in
the frame cache.
Added new webpage to list tests to run to validate LDAS database functionality
in case the database or related software is changed. This was to support recent
changes being made by LLO staff.
Updated the system/integration test website with the LDAS development build
version 1.1.49 and committed the new webpage to CVS.
Discussed making changes to the createRDS command to "break" the
currently enforced one-to-one symmetry between raw frames and RDS frames. This
will impact a significant amount of software both within LDAS and outside. But
the goal of longer RDS frames should facilitate data movement.
Hardware Systems (Anderson)
Caltech
-------
(Dan Kozak)
* Continued the transfer of files from HPSS to SAM-QFS. Also continued
deleting files in HPSS after confirmation that they were successfully
transferred.
* Harped on ITS router guy to get samtest access to ACSLS, configured SAM
on samtest to use tape robot.
* Moved tapes from ldas-archive catalog to samtest catalog and labeled
them for testing purposes.
* Met with IPAC re: their SAM-QFS configuration.
* Did some detective work to discover the whereabouts of some missing S3 L0
data (there was a tape in the LLO robot that was no longer in any catalog). Got the tape back
in the catalog, verified the data at LLO and had Igor ship the tape to CIT
(should be here today). Need to update procedures to prevent this from
happening again.
* Tried to get Igor to release some pinned to disk data so that LLO's
/samrds won't stay at 98% full for the next two months. So far
without success.
* Dealt with stuck staging (particularly from saiph) on the ldas-archive
system several times.
* Got a stuck tape out of one of the 9940A drives that samtest is using. This
drive appears to be bad and is off support as far as I know. This is OK,
we're phasing these out anyway.
(Hari Pulapaka)
* Working on getting grid cvs server up and running.
* Installed Condor and Ganglia on the test cluster.
* Modified the publishing script to make metadata additions one atomic
operation.
* Developed some temporary scripts for deleting and adding data to the
LDR.
* Helped Keith debug the GSI-SSH server at MIT.
* Updated the LDR installation on ldas-gridmon to enable Scott to use his new
tool GatherLFN designed for Griphyn Pulsar search.
(Al Wilson)
* Fine tuning tripwire for solaris.
* Working with Mary on data directories for testing.
* Replaced disk drive in datacache4.
(Stuart Anderson)
* Testing the next next version of SAM-QFS (4.1) which fixes a number of open
issues with the current version we are running.
* Upgraded several computers to the latest Solaris 9 patch set.
* Testing IDE-RAID NFS server performance.
* Setup ldas-dbcentral to connect to the LDAS VPN.
Livingston
----------
(Igor Yakushin)
* Reinstalled DB2 and ldas-dbcentral to be able to use administration
server.
* Replicated LLO_1 from LLO's metaserver to ldas-dbcentral through VPN to
test that it works.
* Running replication test between ldas-pcdev1 and ldas-pcdev2 for 4 days
successfully. 3.8 million rows were inserted into a table at ldas-pcdev1
and appeared in a replicated table in ldas-pcdev2 a few seconds later.
Preparing similar test with the real production database between LLO and
ldas-dbcentral.
Hanford
-------
(Greg Mendell)
* I worked with Lazzarini and Worden to purchase an additional 5 tons of AC for
the LDAS room. This should allow us to use
the entire cluster using condor. I also worked on future floor
plans for data archiving and cluster hardware at LHO.
* Currently working update the createrds scripts for the next release of LDAS.
(Ben Johnson)
* Reconfigured firewall on ldas-grid to allow all traffic on ports 1024 and
above. This was necessary to get the gridftp server to serve requests. All
lower ports except 22 (ssh) are still blocked.
* I've helped some users, SURF students and their sponsors, on the site access
trend and RDS data. The samfs filesystems have been exported to computers in
the computer users' room next to the control room.
* Removed the "forcedirectio" mount option on /frames for fb1. This
has apparently cured fb1's several times a day reboots. As fb0 has
rebooted with the same problems, our guess is that it's a loading issue on the
framebuilder, not particularly a QFS issue.
* Adapted UWM's smartmontools config for the LHO cluster. The smartd daemon and
weekly extended self-tests are now running. They have revealed 8 nodes that
have HDDs with bad sectors. All nodes are still running presently.
* Got LDRdataFindServer running properly on ldas-gridmon again. Looks like some
kind of "environment variable not being set" issue.
* Started work on Gamma Ray Burst archiving scripts. The basic function stubs
have been created. I only need to fill in the details now...
Data Analysis Activities (Lazzarini)
Creighton:
Continuted working on the pulsar Condor script, and modifying ComputeFStatistic
to read frame-formatted SFTs.
Shawhan:
* Compiled "Astrophysics and Data Analysis" section of LIGO Lab
Annual Report / Work Plan, with contributions from Peter Fritschel,
Michael Landry, Albert Lazzarini, and Igor Yakushin.
* Helped move along the S2 untriggered burst search with bookkeeping and
algorithm considerations.
* Helped SURF student Matthew Wroten start to learn how the LALApps inspiral
search code works, and helped him with Matlab and LaTeX for his first progress
report.
* Reprints of my article in American Scientist ("Gravitational Waves and
the Effort to Detect Them") have arrived. Contact me if you would
like a copy. (Soon, we should get a pdf file in addition to the paper
reprints.)
Mendell:
As discussed at the CW group F2F meeting at Tufts in June 2004, I have put
together with Yousuke Itoh and Xavier Siemens a document that discusses
parameter estimation in the frequency domain, including how to handle
nonstationary noise. I have committed it to the lscdocs pulgroup cvs docs
directory as fdParameterEstimation.tex. I plan to return to StackSlide work
next week.
Lazzarini:
Continued to work and read up on deconvolution techniques in preparation for
developing the framework for building a stochastic GW background map of the
sky. All references lead me to VLBI techniques (...not surprisingly). These use
maximum entropy or iterative deconvolution rather than direct SVD matrix
inversion, which, for noisy data tends to be fragile.
General Computing (Wallace)
Livingston:
(Shannon)
-Current bandwidth usage can be seen at
http://teche.ligo-la.caltech.edu/mrtg/LLO-Router/130.39.245.1_1.html
-Archived bandwidth usage can be seen at
http://teche.ligo-la.caltech.edu/mrtg/archive/
under the dated folder
for the
week of
interest.
-Spent a lot of time trying to get the new backup server running this week.
Thursday I tried to install an external SCSI disk on the machine and it would
not install. Solaris did not see it at all. It spewed errors when I
tried to do a probe-scsi-all from the OK prompt. We have had SCSI
problems in the past with this machine and always suspected the disks.
However, since I had errors with the new disk, I then assumed that it was a
problem with the machine. I put a call into Sun and they had me run
several diagnostic routines, etc. They also had me update the boot
PROM. None of their suggestions fixed the problem, so they sent out the
local Sun guy to replace the mother board. This didn't fix the problem
either. After trying many things over the period of a few days, I finally
got the machine running with a different external disk. Turns out that it
was probably a combination of errors. Unfortunately, the amount of time
spent trying to get this machine running put me behind on some other items, but
I needed to get the backup server running.
-Installed the new licensed copy of Arkeia on the new backup server. Installed
the tape library and labeled 30 tapes so that I can start backing up with it
tonight. I will start with the home directory/mail server and the web
server (general web & ilog) since those are the most critical. I will
continue to add machines next week.
-Had several problems while trying to upgrade ghost script & ghost view.
Not sure if it is the compiler or something else that it does not like, but I
could not build the shared object version of ghost script.
-Installed a central syslog server and started several machines logging to
it. I plan on installing some log analysis software on this machine once
I find a suitable package. This will help me keep tabs on things around
here.
-Was informed by the local Sun guy that any machine that was bought >6
months ago and is running Solaris 9 should have the boot PROM updated. He was
rather surprised that we were successfully running Solaris 9 on with the older
revs of the OpenBoot PROM. He said there are numerous bugs in the old
versions. This means that the web server and some other servers will have
to be rebooted in the next few weeks.
-Have not been able to replace the failed switch in the new building yet due to
the difficulties with the backup server. This will have to be done soon.
-Had the usual round of support issues and student account requests.
-Upgraded Mathematica on the Suns to version 5.
Hanford:
(Christine)
Network usage can be seen at
http://www.ligo-wa.caltech.edu/~christin/mrtg/
198.129.208.1_198.129.78.122.html
- Lots of user support for the SURFs and their mentors.
- Purchased two more Matlab licenses. Installed Matlab release 14
on the license server and started installing the Windows version on several
PCs.
- Mounted the LDASdata directories on 5 more GC computers for SURF students to
do data analysis on S3 data.
- Thanks to Larry, finally have completed the renewal of my Cisco maintenance
contract.
- Built a new laptop for a new staff member. Created user and e-mail
accounts for the new person.
- There have been some printing problems when trying to print from a Sun
computer using Acrobat Reader 5.0. Don't know yet what is causing the
problems, it seems to be something in the user's preferences file.
CIT:
(Lisa)
- Looked into a networkproblem on mintaka. I'm still monitoring it to see
if there is a hardware problem.
- Did monthly backups.
- Worked on the application server. I've been having problems getting all
of the X11 dependencies/applications compiled.
- Attended a sendmail webinar on e-mail sender authentication. It looks
like all of the major ISP's will have this implemented by the end of 2004 with
large scale implementation being mandated by the end of 2005.
(Mike)
-Mounted the PDMWorks server in our pc server rack, located in the GC server
room.
-Larry and I moved around office equipment from Florence Kaufman's office, to
make room for locker storage of GC equipment.
-Larry and I swapped out Ed Jasnow & Ed Chargois's Monitor's. These
monitors were replaced with LCD flat screens.
-llpdmworks server: I had a lot of work on this server transferring data over
to the new server. I ended up having to copy over one directory at a time in
order to get the proper directory hierarchy. I also am looking into some backup
software to get the internal tape drive working.
-I updated four loaner laptops with OS/office security patches. I also updated
the General Computing Software.
-Setup the ITRP meeting over in East Bridge conference room on 6-28 to 6-30;
this consisted of running network cables, setting up network switches, wireless
access point and a network printer. Larry and I also supported these users with
connectivity, and other pc support issues to get these users up and running on
the LIGO network.
-Troubleshot Oracle login issues for Gina Salone & Dorothy Lloyd, this
turned out to be java applet problem. The J-Initiator ended up getting
corrupted after Caltech upgrade their Oracle server. After I reinstalled the
J-Initiator software, they were able to login to do their work.
-Worked up on the second floor troubleshooting a network problem. This turned
out to be a bad UPS that was connected to the switches on this floor. I swapped
this out with a surge protector. Larry is ordering a replacement UPS for these
switches.
-This week I installed additional software for the following users: Juri
Agresti, Michael Perreur and Yani Chen.
-I finish loading a visitor's workstation with OS and all General Computing
software. This workstation will be setup on the 2nd floor of W/B.
-Other Misc. onsite/phone user support.
(Veronica)
- CaJAGWR: Installed the slides of the last talk after it received JPL
clearance. Burned a CD of Michele's talk per his request.
- LSC website: Working on the website for the upcoming August meeting. Posting
information as it arrives. Updates to other parts of the LSC (observational
results).
- LIGO website: Installed the finalized versions of the LIGO GC Information
webpages. Installed the May newsletter. Updated various pages. Added a query to
the roster database for generating the author list. Troubleshooting the html
scripts of the roster. Videotaped and compressed the 2nd part of LIGO SURF
talk. LIGO seminars webpage now has pointers to video streams of both lectures.
(Larry)
-Worked with Mike and Donna on setting up things for the ITRP meeting. There
was a lot of user support the first day but after that things were pretty
normal.
-Went through a number of contract items. CISCO is still getting things worked
out on their end.
-Started evaluation of another spyware detection program. It looks to be better
than the one we are using. We will decide which one to purchase sometime next
week.
-Rebuilt one of the E2E servers. Just a few more items to configure on the unit
and then Virginio will be able to start loading applications.
-Worked a couple of printer issues. Worked with Mike on setting up the new
conference printer.
-Assisted the DCC and Ed J. on a couple of document issues.
-Working with the local shop on the computer room expansion. Yes, they have
started and we are down to getting the electrical installed, the wall has been
removed.
-Assisted a number of surf students with different issues.
-Addressed a number of mail alias issues. Most were additions needed in the
main file.
Progress Period from 06.25 to 07.01
·
·
See http://www.ligo.caltech.edu/~tfrey/index.html
for a complete listing of all project related cost and schedule data.
Accomplishments:
·
·
Sub-system
PLANNING activities
§
§
ISC - Continued data
input / changes. Will send questions to Dennis and Peter next week.
§
§
Continue working on
AOS and IO concurrent with ISC regarding scope movement to ISC.
§
§
Executed a telecom
with Albert and Carol regarding LDAS. Provided materials to Albert and
Carol.
§
§
Sent out progress
data requests for progress through June 30. 2004. Data requested by the
8th.
§
§
Assisted Janeen with
compiling cost data requested by Caroline Cantley as needed for a meeting with
Pparc next week.
·
·
ROSTER DATABASE:
§
§
Assisted Irena as
needed with email address related tasks.
§
§
Work with Irena
regarding the author ship list changes provided by Peter Saulson.
§
§
Posted updated
source files for Albert's use in generating the authorship list.
·
·
COST BOOK DATABASE:
§
§
Modified the labor
summary calculation on the detailed cost sheets, so that FTEs are calculated
relative to the task duration.
From: Larry Jones <ljones@ligo.caltech.edu>
SEI Structure:
Work continues on making detailed drawings and addressing the action items from
the Design Review.
The HAM structure Design Review is now scheduled for
December, 2004
Actuators:
The winding of the prototype actuators has been completed. They are expected to
be delivered at Caltech on July 6. Mechanical inspection and electrical tests
are planned, as well as a thermal vacuum test. Difficulties are delaying the
start of the optical cavity contamination testing of Ceramabond 835M, which was
used as a potting compound for the prototype winding terminations.
Displacement sensors:
ADE has told us that we have 3-4 weeks to determine whether either or both of
the Tra-Bond epoxies they use need to be replaced, and to identify an
acceptable replacement. The Tra-Bond 2151 (blue) epoxy has been in test for a
week, and (so far) appears good. We may decide to shorten or interrupt this
test to get a check on the Tra-Bond 2902 (silver) epoxy.
Seismometers:
Nothing new.
Galling/Dusting Test:
Nothing new.
Other:
Planning is proceeding on the design and sourcing of a larger vacuum bake oven
for the quad structure.
From: Janeen Romie <romie_j@ligo.caltech.edu>
AdLIGO
Suspensions
Working with Ian, Calum and Mike P-L on the ETM blade drawings. Gina
has extended the quote due date to July 16th. RFQs have gone out to 4
companies. We worked with Gina yesterday to collect any more questions the
companies may have had and answered all questions with a fax and an extension.
Coordinating with suspension team on MC installation and testing at LASTI.
Working on the ETM Upper Structure Product Design Specification.
Working with Caroline and Justin on costs and quantities of penultimate and
reaction mass blanks for the final design and for fabrications.
From: ctorrie <ctorrie@ligo.caltech.edu>
Advanced LIGO MC
I have been preparing all of the mode cleaner files for release. All of the
files will be moved to the PDMWorks Vault and to the DCC later this week. The
DCN E040303-00 will cover the initial release of the mode cleaner in its as
built configuration for LASTI, June 2004.
It is our plan to send each drawing to the DCC as a . PDF with an embedded
.STEP file. The native SolidWorks files, drawings, parts and assemblies,
will be stored on the PDMWorks vault along with the .STEP's and .PDF's.
An earlier example of this can be found on the DCC at: -
http://www.ligo.caltech.edu/docs/D/D020205-02.pdf
Quad ETM
Mike, Janeen and I have created a set of requirements for the upper half of the
structure. We have also had several discussion related to the various suspended
masses. Several bench experiments are planned over the next couple of weeks.
Blades
Janeen and I have been discussing the blade quote process with the folks in
procurement. Several questions were posed to us by one of the vendors and the
quote process has been extended until the 15th of July.
PDMWorks Vault
We now have a data management vault set up at Caltech. The name of the vault is
llpdmworks and its IP address is 131.215.155.22. User names and passwords can
be obtained from me or from Mike Pedraza. More to follow.
MC at LASTI
The MC has now been installed and aligned at LASTI. Ken Mason has taken a new
set of photographs which can be found at: -
http://emvogil-3.mit.edu/~kmason/triple_installation/
Pre-stabilized Laser
From: Peter King <pking@ligo.caltech.edu>
The GAP976 photodiodes have arrived. A pair of circuit
boards for them have been stuffed in preparation. I hope that this
particular photodetector circuit will run closer to the design limit than the
previous implementation.
From: Riccardo DeSalvo <desalvo@ligo.caltech.edu>
Weekly-1-July-04
Barbara
In this week we built the wood support for the vacuum pipe, positioned the pipe
on the supports, assembled the invar cavity structure inside the pipe and
suspended it from its suspesion system. In the process we damaged two of the 4
MGAS springs (We were too compacent when installing and bent the nose of two of
the blades. The two damaged blades still work properly but sit crooked and will
not accept the actuators for Maddalenas electromagnetic corrective springs
without replacement. We will redesign them slightly to avoid this installation
problem when replacing them.). After the assembling we found that all the
blades were bistable, either because of the much colder temperature in OTF, or
because while we visibly damaged two, we may have modified the working point of
the other two as well. For this we had to retune the blades frequency of all
four by increasing the distance between the clamps of the blades by a
millimeter. We have fixed the frequency around 0.6 Hz and the cavity is
pleasantly suspended.
Some images in
www,ligo.caltech.edu/~desalvo/assembly-sequence.ppt
-Nicky
I am still waiting for the rest of the supplies needed to build the oven for my
project.
In the meantime, I have been assisting in the assembling of Barbara's
interferometer while she is in
AnaMaria
This week I continued Maddalena's low frequency measurements; and re-did my
setup for the experiement to prevent enegry transfer between the spring and the
payload pendular oscillations. Now my blades are smaller, they have a smaller
center support piece and the wire that's holding the payload is shorter and thinner.
Also I did a presentation of what I have done so far to the people in my group
and submitted it as a SURF progress report.
Juri
During this week I used my simulation program to study the misalignment of the
Mexican-Hat cavity with different tilting angles, going from 10^-7 rad to
5*10^-9 rad. The agreement with the perturbation theory for small angles is
worse than for large perturbations. This is probably due to a dynamical effect:
in order to reach the steady-state profile, we have to integrate over more
light round-trips in the cavity for small angles. This point is far from
trivial and needs more study.
I am preparing the presentations for the GR17 conference.
Yanyi
Things done:
Transferring the original circuit diagram of the Amplifier into DXP Schematics.
Transferring most parts of the original LVDT circuit into DXP Schematics and
adding two potentiometers to the receiver coil of the LVDT driver to balance
the two sensing coils
Adding a switch in the 5th part, the Offset Potentiometer, of the Amplifier
circuit so that we can choose the sign of the feedback.
Adding a voltage saturation detector at the outputs of the Offset Potentiometer
and the Mixer. When an internal feedback voltage goes close to the supply
voltage, the op-amp is failing. If this happens, the voltage saturation
detector gives off alarm signals.
Things to be done:
Adding a switch between the Driver and the Receiver in LVDT to change the sign
of the feedback
Adding connectors to the Amplifier in DXP Schematics Combining the LVDT, the
Amplifier, and the Voltage Saturation Detector into one circuit Create a
Netlist in DXP
Create a PCB (Printed Circuit Board) design in DXP
Sent the PCB file to Board House and get the circuit board built
Get the circuit board back and assemble each component of the circuit
on the board
For additional information about this report, contact whitcomb_s@ligo.caltech.edu