The LIGO Executive Committee Agenda for Monday, October 2, 2006 will be:
(Meeting time: 10:30 am Pacific Time)
Announcements
Programmatic (Lazzarini)
Comments on weekly report
LSC Issures (Saulson)
LIGO Lab Operations
Enhancements (Adhakari)
Advanced LIGO (Shoemaker)
Change Control Board/Technical Review Board Session as needed
Site and other business issues:
Special Items:
We will review plans for the upcoming NSF Review 23-25 October at LHO
Discussions continue of a variety of issues related to the MOU under negotiation with Virgo. A new proposal has been made for an annual schedule of 5 joint collaborations per year. We are also discussing the question of Virgo participation in the November LSC meeting.
The LSC Election and Membership Committee (consisting of Norna Robertson (chair), Sam Finn, Gabriela Gonzalez, Szabi Marka, and Fred Raab) is off to a good start after its organizational meeting this week.
>From: Rod Luna <rluna@ligo.caltech.edu>
>From: Cleveland Mak <mak_c@ligo.caltech.edu>
>From: "Cronin, Holly" <Holly.Cronin@caltech.edu>
>From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
>From: Florence Kaufman <fkaufman>
>From: Gina Salone <gsalone@ligo.caltech.edu>
>From: Ed Jasnow <jasnow@ligo.caltech.edu>
Nothing to report. Currently working on FY 2007 budgets.
>From: Bill Tyler tyler@ligo.caltech.edu No report.
Summary of S5 Activities at LIGO Hanford Observatory (compiled by M. Landry)
This was a quiet week at Hanford, without major problems. Duty cycles for the week were: H1 - 88% at 14-15Mpc, H2 - 86% at 7-7.5Mpc.
Some S5 highlights from the LHO elog are bulleted below:
4K IFO
2K IFO
DAQ/CDS
A summary of Tuesday maintenance is here
L1 in S5 (MZ for O'Reilly)
---------------------------------------------------------------------
No report
LLO Outreach (Thacker)
---------------------------------------------------------------------
Conducted Photo session on 9/23 - children/parents interacting with exhibitsMoved existing exhibits into SEC; started cleaning and tuning processContacted Louisiana Public Broadcasting's Donna Lafleur about story on a Day at LIGO tentative date Nov. 17thPrepared a sample program to Zachary Community Schools in preparation for school visitSent photo release form and Program Agenda to Westdale Heights Magnet SchoolProvided outreach to 9th - 11th graders at Doyle High School (6 classes), 9/28Posted education material to the SEC webpageWorked on Nov 13th Opening Event issues including speakers' notes, invitations, media objectives.LLO Mech Engineering (Romie)
---------------------------------------------------------------------
*LIGO SEC Kinetic Art Project*
Continuing to status delivery on contracts and POs. Superior Steel contractors are here and have welded the stringers. They should have the bumper assemblies up and installed by the middle of next week. We'll hold a meeting with the Superior Steel superintendent, foreman, and our assembly folks this morning about the assembly and installation procedure. I'll status HPD after the meeting.
Jonas from HPD confirms that Sapa will ship the pendulums today. We should have them here Teusday. The truss weldments should be here tomorrow. The locking arm weldments should be hear the middle of next week. The bolt rings for the bearings should be here tomorrow. The remaining machined parts should be here middle of next week. Winches and magnet housings were delivered yesterday, bearings delivered the day before. All parts and hardware are in the Outreach room, ready for assembly work.
*Advanced LIGO Suspensions
*Working on LASTI schedule for noise prototype work. *
LIGO computing and network security (Roddy)
---------------------------------------------------------------------
Shannon is on travel at CIT
General computing and LDAS admin (Giardina)
---------------------------------------------------------------------
Last two weeks:
Reported under General Computing, see below.
Reported under LDAS System Administration, see below.
Data analysis (Yakushin)
---------------------------------------------------------------------
Storage/Condor/LDAS admin:
Reported under LDAS System Administration, see below.
======================================
Data analysis:
1) Using coherent waveburst to process LIGO-GEO S5 data from June 1 toSeptember 13 to answer the question asked by the GEO team: would it bemore beneficial to continue running GEO as is (if it makes usefulcontribution to S5 search) or should GEO team spend several monthsimproving the detector.http://ldas-jobs.ligo-la.caltech.edu/~igor/LIGO-GEO-S5 2) Made a presentation about LIGO data analysis and computing for LSUstudents.3) Generating final set of coherent waveburst triggers for:
a) LIGO-GEO S4;b) LIGO S4;
c) LIGO S5 (from the beginning to April 3rd);
a) and b) will be used in the method paper comparing coherent andincoherent waveburst searches on S4.c) might be used in the results paper on the first few months of S5(although the current agreement is to use incoherent waveburst+CorrPowerpipeline).
See Advanced LIGO
See Advanced LIGO
The new bridge seems to be working well
_______________
Patrick Sutton:
Late last week I helped Stephen Poprocki with his final SURF report.Since then I've been studying some theoretical aspects of coherentburst search methods. In particular, I've derived the probabilitydistribution of elliptic likelihoods in the noise-only case, allowingme to compare the statistical significance of candidate signals fordifferent sky positions and signal parameters.I've also discussed with Siong Heng some ideas for an undergraduateproject on estimating source locations in coherent analyses.Finally, I'm working with Pia Astone to combine our reviewer reportson the LIGO-Virgo joint burst search methods paper._______________
John Zweizig:
This week I have been assembling Data Quality segment lists correspondingto a couple problems discovered by the glitch group oin the past fewweeks. These include:a) Science mode running in AS_Triggered state
b) DataValid errors, especially in DARM_ERROR.
The LLO flags will be ready pretty soon, although the LHO flags take awhile longer (I'm trying to find a way to work around the slowsecond-trend frame I/O)._______________
Steve Fairhurst:
Continued work on obtaining upper limits for S3/S4 inspiral searches.
Worked with Drew on S5 BNS search.
Refereed a paper.
_______________
Duncan Brown:
- Worked with Diego on implementation of physical template family.
- Working on understanding Chao's generalization of Ryan's 1995
multipole moment paper.
- Changed the domain decomposition of several NR runs that hadcrashed and restarted them.- Cleaned up a bunch of old NR checkpoint files as requested by Stuart.
- Meet with Stuart and Todd from the Condor team (Madison) to discussimprovement to scheduling on our cluster.- Released a new version of glue.
- Fixed a couple of minor segment database issues.
_______________
Greg Mendell:
The past week I have been working with Rejean Dupuis on code to automatesearches of the S5 h(t) data, and auxiliary channels, for thin linesthat can interfere with the pulsar group's search. Our method looks atSFTs over long periods of time with high frequency resolution, tocomplement other line detection methods currently running. The code hasbeen checked into CVS in lalapps/src/pulsar/fscan, and is underconstruction. I am also continuing work on the S4 PowerFlux, StackSlide,and Hough paper._____________________________
Eirini Messaritaki:
I have been working on the S5 BBH inspiral search. Specific issuesthat I have been working on:* Investigations on software injections missed at small distances.
* Statistic equivalent to the effective snr used in the BNS searches.
_____________
Kent Blackburn:
--------------------- Grid Computing Activities -----------------------
TclGlobus:
Adding GPL V2 license into TclGlobus source code and web site. Completed matrix for validating TclGlobus unit test in the following platforms:a. GT-3.2.1, GT-4.0.1 and GT-4.0.2
b. 32-bit Linux and Solaris
c. 64-bit Linux and Solaris
Attended the weekly Virtual Data Systems, Virtual Data Toolkit teleconferences.OSG Workflow Development
Re-running hipe workflow from the first partition with chaining mode at PSU. Last time, hipe workflow failed the third partition and the cause of the problem is unknown but under investigationCompleted the functionality of N-2 data cleanup.
Working on fixing the default VDS behavior for restarting the workflow. Currently, VDS only throws rescue dag at the outer level if there is a failure in particular partition. LIGO workflow planner will overide VDS behavior by keeping the state of each running partition.Using information provided by Michael Samidi to configure a submit host for the LIGO workflow planner, determined that a gsiftp server not on Michael's list of required components is also needed. Determining the best practice to install gsiftp from the VDT.
Received a recommendation at the VDS telecom from Ewa Deelman (ISI) to use the label partitioning method for creating a 24 partition version of the HIPE analysis of S5 data. Requested engineering notes from the vds team for designing a label partitioning post-processor.Running "small" hipe workflow at several OSG production clusters for site validation.
OSG Integration Testbed:
Tested the binary inspiral 470 DAX, (940 DAGnode) LIGO VDS workflow on a CE at UCSD, with a wrapper script employed by the NFS-Lite developer to support NFS-Lite and the use of RemoteInitialDir parameter in Condor-G. Test successfully demonstrated inter-operability of Condor-G's RemoteInitialDir functionality with NFS-Lite as employed by LIGO in the binary inspiral condor submit scripts. All DAG nodes executed successfully with the exception of the chmod DAG nodes which failed due to local site management issues related to making binaries executable on UCSD worker nodes.OSG Executive Activities:
Press release of funding for the OSG took place on Monday of this week.
Worked with OSG Executive Director and contracts assistant to draft the Caltech/OSG Statement of Work.
Visited Fermilab to work with the OSG Executive Directory (Ruth Pordes) and OSG Resources Managers (Paul Avery & Kent Blackburn) on the formulation of the OSG Finance Board, updates to the OSG Management Plan and identification of the OSG projects. Also met with GUMS, SRM and Gracia developers while at Fermilab._______________
Xavi Siemens:
Since this is my first report, here's some of what I've been doing sinceI got here:
- With Dan Kozak and Stuart Anderson, looked at potential mess createdby re-naming of H1 h(t) files from v1 to v2 (which was not supposed tohappen). It was a false alarm; there are no known problems with v2 h(t)
- Worked with Eiichi Hirose of Syracuse (presently in Hanford) on TDcalibration validation
- Worked with Kipp Cannon at UWM on cosmic string search post-processingto produce background and foreground plots as well as efficiency curves
- Wrote and debugged lalapps codes for small and large loops models that take efficiency curves and compute sections of allowed and excluded areas of cosmic string parameter space. This work was presented at a burst telecon (9/19) and a Caltech data analysis meeting (9/25)
- Time domain calibration review preparation. First review was 9/15,second review scheduled for tomorrow 9/29 - Prepared two presentations on cosmic strings for Kip Thorne's weeklyTAPIR meeting. Currently working on a third one. - Re-ran S4 cosmic string search with latest version of code v1.98 oflalapps/src/string/StringSearch.c on nemo at UWM. There was a bug(whose effects we did not spot) but I re-ran the search anyway to makesure results are the same. Kipp is re-running his post-processing pythonscripts on this new set of triggers and injections. We should have theresults very soon. - Worked on stochastic background from cosmic string cusps paper withVuk Mandic. The plan is to finish a short paper with our main results(which we'll submit for LSC review hopefully tomorrow Friday 9/29), andwrite a long one with the gory details to be completed at a later date(we've done some work on this longer one as well)._______________
Rejean J Dupuis:
- Doing scimon shifts at LHO
______________
Shourov Chatterji:
* Continued work on early S5 low threshold Q Pipeline analysis
* Reviewd Rubab Khan's final SURF report on clustering extensions
to the Q Pipeline
* Compiled weekly glitch group report for the S5 run callhttp://ldas-jobs.ligo.caltech.edu/~shourov/reports/glitch/WeeklyDC_20060917-20060924/* Installed QScan on the new GC Linux machine in the LHO control room
* Updated scimon instructions for running QScan to include the new LHO machine as well as information on how to run QPipeline to identify the time of glitches. http://www.ligo.caltech.edu/~shourov/q/qscan/scimon.html-----------------------------
LDAS Software Systems (Maros)
-----------------------------
Efforts to support 64bit builds of LDAS have shown issues with the useof non-reentrant parses used in several places in the code. Changes arebeing made to use the latest version of flex and bison to generate C++class based parsers and lexical analyzers which are reentrant as allvariables are localized to the class. The cntmonAPI has been modified not to reuse variables whencommunicating with the cgi user account script. It has been observedthat with TCL 8.4.13 that TCL does not properly reset the memory spaceof some variables. This results in the newly assigned variable havingtrailing data from the previous value assigned to the memory space. System testing of LDAS was done using version 1.8.312 of the software.All of the tests passed except for the lsync test and two of thedatasocket tests that used the loopback interface on dataserver-dev.After the tcp-fusion option was reenabled and the gateway box wasrebooted, both tests passed. Investigation of the lsync issue currentlyhints to a delay issue as a version of the lsync code with severaladditional print statements did pass.-------------------------------------
-------------------------------------
Caltech
-------
(Dan Kozak)
* Fixed damaged file at LLO for Igor.
* Sent (now) boilerplate "Use LSCdataFind to find data on the cluster disks" mail to Anand and (again) Cokelaer.* Trying to find cause/workaround to files in CIT's /home not getting archived.
* Isolated problem on samtest wherein syslog writes QFS 'unmount' messages over and over again. Solution: kill syslogd.* Determined that the Solaris 10 Update 2 sendmail conf is seriously broken(open relay).
* Got GEO RDS data moving to CIT again (involved explaining problem to those at GEO in a position to republish data).(Phil Ehrens)
* Developed CGI Travel Application for Albert Lazzarini. This task included an html form frontend, a Tcl cgi backend, and installation and debugging documentation. Task is complete.* Unusually large amount of cert related work this week. Several users who had applied for access to only one or another site discovered that they had no access to CIT, others had requested new certs (with new serial numbers in the CN field) had similar problems.* Wrote shell script for John Zweizig to run as a daily cron job on his dmt servers to report imminent expiration of the dmt certs on these machines outside of LDAS administrative control.
* Worked with Fred Donovan at MIT to get him up-to-speed on the care and feeding of LDAS. LDAS and the web server are now running on ldas.mit.edu.* Began work on a cgi based Visitor Appointment application with an html form frontend and a Tcl/pdflatex backend for Alan Weinstein.(Erik Espinoza)
* Rebooted Nodes:
- 9/24: node5, node136, node150, node246
- 9/26: node264
- 9/27: node3 (multiple times)
* Testing sendmail config for Sol10u2, should finalize on Sat
- Determined that default Sol10u2 config is open relay
* Upgraded to Condor 6.8.1.
* Installed libpng-devel on gateways.
MIT
---
(Fred Donovan)
* Inventoried cluster for spare parts.
* Tried to get pcraid 5 started - will not power up.
* Restarted pcraid4, which was off the net.
* LDAS - datacon off net, ldas password need to be reset.
* LDAS nodes 32, 40 down: 32 some disk errors hdb; 40 fsck /dev/md2 issue.
* node61 bad ram, will not boot.
Livingston
----------
(Igor Yakushin)
* Bought 6 more power supplies for the nodes since we just used our last spare couple days ago.* Updated segment database web page. More updates are coming.
* Another discovery regarding the firewall on ldas: when ldas is rebooted or NFS daemon is restarted, NFS might come up at a different port and require firewall reconfiguration.* Installed ipython on ldas-grid and ldas-pcdev1.
(Dwayne Giardina)
* Prepared tapes for shipment to CIT, imported and labeled replacements (both weeks).* Increased NSFD_SERVERS on gateway from 16 to 32 after noticing several scripts hanging.* Reduced the frequency of my nodecheck script from every five minutes to every ten minutes.* Replaced disks in nodes 11, 31, 183, and 196.
* Replaced power supply in node 146.
* Turned sendmail off on the nodes while Igor killed his spray scripts, left it off over the weekend. changed crontab to contain MAILTO="" so that script termination on the nodes won't flood mail servers. cleared mail queues on the nodes after all scripts that were started by cron with the old MAILTO value were stopped. restarted sendmail on the nodes.* Replaced failed disk in T3-25 with spare disk.
Hanford
-------
(Greg Mendell)
* We are now running createrds.tcl version 1.21 at the sites (from the lscsoft CVS repository under dsorun/contrib/createrds/scripts). This version handles "error reading sock... : broken pipe" pipe errors that have started occuring a few times per week on the LDAS gateway box at LLO, and in general has more robust error handling. All S5 RDS generation is up-to-date, and overall RDS generation is running smoothly.(Ben Johnson)
* Discovered "balance" load balancing utility. At the moment it is being used to distribute LSCdataFind requests over multiple processes.* Discovered that LDRdataFindServer can be run on a node, with gateway's host certificate, if gateway uses "balance" to redirect traffic to the node, similar to a proxy server.
* De-bugged the thread locking I introduced into the LDRdataFindServer (for controlling access to the parsed frame cache in memory). A little more testing and it should be ready for prime-time.
* Came up with basic SMF schema for the services on gateway and dataserver. I currently have the web page monitor scripts and FrameCacheDump running under SMF.
* Installed libpng-devel on the cluster for Nick Fotopoulos.
---------------------------
---------------------------
MIT:
(Fred)
-Installed xp for new hire
-Automated the tubera - gall rsync backups
-Automated df /export/home check (email warning) for tubera
-Some more linux desktop work for pf
-Copied 25 cd's for Ken Mason
-Ordered 2 750GB disks for nova (new compute node for Erik)
-Worked on sendmail authentication setup (auth, TLS) - building sasl,sendmail, etc.
-Other misc. stuff ( new accounts, email list)
Livingston:
(Dwayne)
- Packaged a CPU for return (Shannon RMA'd with vendor)
- Auditorium computer would not work through projector. the control panel on the desk needed to be power-cycled.
- Replaced a CPU fan in a server
- Installed MS Office for a user
- Configured our new DNS servers: one master, one slave
- Assisted Shannon with moving some equipment in the CUR racks
- Assisted in moving some of the outreach exhibits to the new area
- Mail server /var disk problems, replaced disk and was able to copy data from the old disk.
- Automatic software upgrade on our spam filter broke the system. after speaking with their tech support, ran some migration scripts and restarted services.
- Continued training our spam filter
- Other usual user requests and support
(Shannon)
- Working with Kent at CIT this week on security issues. Policy and general discussion.
- Installed the OS on an IDS box for CIT this morning. It is live onthe network and will be able to complete the configuration next week.This also means that I need to get the new MySQL box at LLO installed inthe rack and on the network.- Working with LSU to get the new subnet allocation advertised via the
LSU network.
- Received start of authority from Caltech, and transitioned authority.
- Worked with Charter to fix some BGP problems on their end. They were not advertising a default route to me which was going to break multi-homing.
- registered two nameserver addresses with ARIN. Helped Dwayne set up the zones for the nameservers.
Hanford:
(Christine)
- Setup a new PC to replace an old one used as a "community" computer in
the OSB common area.
- Continued setting up a Linux web server for the Einsteins' Messengers web site.
- Renewed the maintenance for four software packages.
- Setup a local account on the Control Room NIS+ Linux for qscan. Shourov is using this account for software install and maintenance. Installed some additional software needed for qscan.
- Helped new staff with email setup and Matlab install.
- Received two of the three quotes requested for the new Cisco router. Need to check with my contacts at PNNL to make sure the router will be compatible with their equipment and future upgrade plans.
CIT:
(Christian)
-Created a backup of Florence Kaufman laptop and Irene Baldon workstation.
-Millikan- Replaced toner cartridges on HP 5500 printer.
-Cindy Akutagawa- Eudora keeps crashing. Cindy created an out of office filter in Eudora before leaving on vacation for 3 weeks. The filter created a loop in the out box that caused Eudora email program to crash. I spent most of the day troubleshooting this problem.
-Kaiko Kokeyama replaced old CRT monitor with new LCD monitor.
-Minor tweaks on the unit in the display case
-Worked on the Spam Filters with Mike and Larry.
-Created new LIGO standard image for parallels.
-Other misc.: Continued onsite software/phone support
(Mike)
-Working on Primavera upgrade. We put together a higher end server torun the current version of the Primavera software plus the upgradedversion of this software. We will also be running the Prism software onthis server. This is an on-going project.-Looked into many issues on the old DCC servers. I have come up with atemporary solution to get Cleveland working.-Had many problems with the mail servers this week. They seem to berunning ok for now.-Ordered some additional memory using the TechMart system, plus a few other items.
-Continued work on the Spam Filters.
-Misc. user/phone support.
-Other sysadmin daily tasks.
-Tested Microsoft patch that came out this week on all versions of windows plus with GC and engineering software.
-Freeing up disk space to be ready for End of Month NTSRV backups.
(Bruce Sears)
-iLog Maintenance:(0.5 days)
General iLog maintenance (user adds, keyword adds, systems work, etc.)
(Veronica)
- LIGO: Updates of the NSF reviews website. An issue has come up toimplement a way to require all visitors to sign a acknowledgement of LIGOComputer Use Policy at the time of registration. It is probably a goodidea to have a list of required items to be included into every meetingwebsite; Kent, Larry and I are working on it.Updates of the PAC meeting website.- LSC: Ongoing work on the LSC November meeting website. Finalized theonline registration form and the integration with the externale-commerce application. The entire website is now up and running.Updates of the database of talks under review.
- CaJAGWR: Website updates/ user support.
- Project Science: Website updates/ user support.
(Larry)
-Continue working procurement issues. Mostly, misc. purchases other thana new workstation which has just arrived.Purchased another access point to check out. This one will handle moreMAC addresses than the others that we have tested. It should be morein-line with what we need for the conferences.-Spent time working on the DCC both the old and new system. The testingof the new system is now on hold until there are some updates made onthe test system.-Worked on a couple of location moves that are going to take place in the next week or so.
-Setup a couple more accounts.
-Working on the tape robot unit. So far it is now working correctly but still having problems with it communicating to the computer.
-Tested out a couple of new h/w items.
-Made some minor changes in the computer room. This is in preparation in getting the racks reorganized.
-Working on the logistics for the up-coming meetings and conferences.We've equipment that will be shipped to each LIGO site this next month.Started testing a system that may be used as a central box for givingpresentations. Still a few more things to test on it but so far it islooking pretty good.-Going over different web issues with Veronica and others. Priorities have been changing on a daily basis.
-Replaced some equipment for a few people and had to troubleshoot one of the E2E units.
-Worked the spam filters and presently working a e-mail stability problem.
From: Dennis Coyne coyne@ligo.caltech.edu
-----------------------------------------------------------
e2e weekly meeting
----------------------------
Osamu gave a lecture how to design control systems. Very nice. The material is available from www.ligo.caltech.edu/~e2e/ME2ET/060927/
AdvLIGO LSC/ASC design using FP arm model with quad suspension (Osamu, Hiro)
------------------------------------------------------------------------------------------------------------------
no progress to report.
Static IFO simulation (Hiro)
-------------------------------------
no progress to report
Fast simulation of Dual Recycled Michelson Cavity (Hiro, Keiko)
-----------------------------------------------------------------------------------------
R(n,t) has been described without any summations orapproximations. With this R(n,t), a part of the FP cavity field wasexpressed by using Taylor expansion and validated numerically. As aresult, for example, when the mirror tilt angle theta=0.1 [rad] and theiteration number N=200, the approximate field values agreed almostcompletely with the field of no approximations. When theta=0.5 and N=50,the field values agreed about 98%. The calculation time was actuallyshortened. The next step is to investigate how the tilt angles appear inthe field equation analytically.Modeler (Hiro, Bruce)
---------------------------------------
Bruce is working to modify modeler to read User Defined Primitive definition to actually create a module on the fly when reading the box file.Melody built several revisions of glibc so that new CPU units (rigel1, nunki) together with saiph can be used to run e2e fast.Unfortunately, no new glibc generate fast executable, and there is some difference between slow version and fast version using old libm.so.6.Left to be investigated.
Mechanical Simulation for advanced LIGO (Kinchen, Quave and Yoshida)
----------------------------------------------------------------------------------------------------
Continued e2e modeling of the effect of reaction from triple suspension to the HAM-SAS table and spring box (part of the horizontal seismic isolation mechanism). We combined our triple suspension e2e model (based on Mark Barton?s mathematica model) and HAM-SAS horizontal model (based on Valerio Boschi?s matlab model), injecting white noise into the suspension point in the X (normal to the optic?s surface) and yaw degrees of freedom, and feeding back the force exerted by the pendulum on the table via the suspension point. With this reaction force, as compared with the other computation we made earlier without feeding the reaction back to the table and keeping the other conditions the same, the optical table?s displacement in the X direction shows about 20 dB, 100 dB and 30 dB higher peaks at the triple pendulum?s resonance at 0.7 Hz, 1.02 Hz, and 1.5 Hz, respectively (in magnitude of the asd). This result is currently examined by the HAM-SAS group.
We also started to work on the vertical model of the HAM-SAS seismic isolation, in the same way. We computed the transfer function from the ground to table top with our e2e model based on the Valerio Boschi?s matlab model. The result is being examined by the HAM-SAS group.
ALFI : e2e front end (Bruce, Melody)
-------------------------------------------------
Melody is continuing in the dialog implemention for the new User Defined Primitive.
Bruce finished to modify the alfi code to save this information in the box.
From: "Joseph A. Giaime" jgiaime@ligo.phys.lsu.edu
Agenda for the weekly SEI telecon
Friday, Sept 22, 2:00 pm Eastern, 1:00 pm Central, 11:00 am Pacific time
Announcements
* As part of his thesis work, Shyang is compiling ground noise statistics from seismometer trends, in order to assess whether there are any trends in ground noise since Ed Daw's paper of several years ago. This study is relevant to Advanced LIGO planning.
HAM prototype progress
* Corwin has delivered a draft of his conceptual design study. He describes three designs, considering the pros and cons of each. The main difference is the position and orientation of the large blade springs. He also discusses a concept for a pod that combines the a vertical and horizontal GS-13 and two actuators. It can be lowered in through a hatch in the optical table. Remaining to study are a better approximation of the bolts and an optimization of the optics table plate.
* The requirements and specimen contract under preparation.
BSC prototype progress
* Damaged actuators have been repaired and are ready to install.
* Position sensor electronics expected early next week.
* Detailed mass budget study indicates that the springs are about 7% softer than expected. After this round of testing we will re-hang with full load (using different spring angle). This may be just possible by mid November.
* The plan is that the vender will assemble the device, then we test it, then it gets shipped to cleaning company. We sample the surfaces there for FTIR analysis. If all is good, the parts get wrapped up for shipment.
* 37 and 46 Hz resonances appear in the H1 collocated actuator-sensor transfer function of stage 1. Laurent is on the hunt.
* The plan is to measure the compliance matrix and then design the new shims.
* Ken will follow up on Dennis's link to the slippery coating product that Dennis listed, plus any other processes or products that may work. Dan suggests ultra-high molecular weight polyethylene.
Feedforward at Stanford
* Matt DeGree has tested rx and ry feedforward. The filter design still needs some optimization in its phase response. The good news is that the rx and ry feedforward doesn't make x, y and z worse, but improvement in rx and ry has not yet been demonstrated. (last week's z feedforward improvement is still present.)
Clean room is clean and ready. A coat of antistatic paint was givenand after that is dry it will be accessible for switching on fans,and furnishing (cabinet, tools, etc.).A vendor for clen room dresses (next day delivery) was identified (inFlorence, they serve Virgo with full satisfaction) Clean room garbsneed to be bought shortly to access.
The 3 meter oven structure is finished and is going to be cleaned bySoldi, probably in situ, without bringing the oven to Soldi's place. Then the thermal blankets and heaters, and the outer structure willbe applied and it will be ready for the first bake.This oven risks to be the critical path, but it is in decentconditions (it is expected to be ready by next week), probably willnot hurt us. A small kitchen electric oven was bought for baking small parts. Asmall Argon pot is being welded to fit inside it.Parts as large as the blades will be processed in this rapid cyclingoven.The argon dewar can be delivered with 48 hours notice, it will bebrought in house when needed.
The four GAS filters have been disassembled from the spring box andare being completely undone for cleaning. Some parts are already incleaning, only lab components for now because the cooking oven isnot ready yet. These parts include the second cart and thestructures and masses to re-assemble and tune the GAS filters.The in vacuum parts should get washed just in time for the ovens tobake them.The full cleaning process is estimated to take at least three weeks.
The Chemical glassware for the FTIR tests has arrived. Need tomachine a couple of corks and a little swagelock piping to get started.Riccardo will be in Lucca in 10 days to help in the startup.Transport carts:
The first one has been cleaned and boxed, together with the opticalbench drilling jigs, and left. It is scheduled to fly on October 3rd on British Airways, it will bedelivered to LASTI to Ken Mason. Please give to G&M the name of acustom broker to indicate to GAVA (the carrier) otherwise they willuse any custom broker of their choice).The second one is being cleaned and will go directly in the cleanroom for local use. It will leave for LASTI together with the rest.New LVDT part fabrication:
Some interference problems found with horizontal LVDTs. Gianni isback from vacation and working on the design of the rollstabilization system inside the central LVDT group as well as solvingthe interference problems. The design is mostly done and will beready by friday morning. Chiara and Gianni will meet friday toinsure that all parts are being modified and/or produced in a timelymanner. We (may be optimistically) expect to have all partsfabricated for test fitting under the IP table by end of next week. Test fitting would be done the next monday when Riccardo is on site.Elevator cart.
The redesign has been delayed to after the LVDT and anti-rollproblems are completely solved.In any case it does not poses big challenges. An irreversiblereduction gear and suitable belt tensioning and anti-slippage partsare foreseen.Electronics -- Ben
Rearranged the cabling to streamline the in-vacuum routing of cabling.The in vacuum breakout box will be built with a modular kaptoncircuit and avoid rat nest cablings.Virginio/Valerio
The LVDT card now shows good performance (~nm resolution) in internal clock mode.
Surprisingly, even with the much better external clock quality, inexternal clock mode the LVDT are almost twice as noisy. This is arelevant issue because, to avoid interferences between channels, weplan to use them with one acting as master oscillator and the othersas slaves. Alternately we have to tune each individual LVDT driverinternal clock to sufficiently different frequency to avoid crosstalk in the cabling. Not a big problem but a nuisance.Proposed a target of two weeks from now to freeze the circuit designand produce it. To guarantee best advance of the work Ben, Virginioand Valerio will meet at least 2 times a week.Simulations
Valerio working with the E2E people
Alberto/Riccardo
Estimated the strength required by the anti-roll device springs.10kN/m for Riccardo, &kN/m for Alberto with all the payload half a
meter above the support point plane.
Since the optical bench and all mass placed on its surface are below that height this seems a reasonable assumption.
In any case these are easily exchangeable springs, we can implement the ones we need during assembly.
After disseminating the latest schematics for the HAM-SAS system which accommodated the new test electronics, Riccardo and I met, at which point he requested some alterations. I have finished the new drawing incorporating these alterations, and am in the middle of changing the module schematics and PCBs to reflect the wiring changes, and the addition of a module. They should all be done soon.
I've received a detailed packet of information from Joe Hanson regarding the new GS-13 board and cable for ISI. From the pictures, and the old cable that he sent, I was able to make a new board and I'll send it out for fab soon.
Virginio has asked me to begin thinking about re-laying out the LVDT Driver board. If all goes well with the testing, we will mostly keep the present design, and re-lay out the board. This is a new task, but one that is not a complete surprise. With luck, I can interleave it with other tasks, and it shouldn't delay other things too badly.
Suspension
From: Janeen Romie romie_j@ligo.caltech.edu
*Advanced LIGO Suspensions
*Working on LASTI schedule for noise prototype work. *
From: Ken Mailand
I'm using lower suspension and support frame dimensions from Ian 9-26, to do the lower quad installation arm shop drawings. The CES shop has completed more than half the parts for the basic arm.
The fixture lift table for LASTI has returned from chrome plate, and is at CES for re-assembly. I'm working on the lift assembly frame for the suspension. The 5 axis fixture is in the clean area
in the synchrotron room, and is being re-assembled with stainless hardware.
We made good progress on determining the allowable low frequency rms roughness for the Advanced LIGO test masses. This is based on using a scaled version of one of our worst actual surfaces from initial LIGO.
Recent information in the form of responses to RFI has assured us of multiple capable vendors for advanced LIGO coating. This comes with very little up front investment on our part for the extension of current technology to larger optic apertures.
PO TEL/TRAMSMON detection bench
Chris developed a SW conceptual model for the ETM TEL suspensions with the TRANSMON in-vac detection bench. The ETM output beam passes through the ETM Telescope, reflects from a dichroic beam splitter, passes through an attenuating beam splitter (BS) on a filter wheel for selecting various attenuation amounts, and is focused by a lens onto an in-vac QPD. The beam reflected from the attenuating BS is dumped into a small, cavity beam dump. The dichroic beam splitter enables the ETM Hartmann beam to be injected through the back of the ETM telescope.
LAYOUT
Mike revised the Zemax Layout of the IO control beams, ISC control beams, PO beams, REFL, and APS beams. They are routed through Brewster-angled viewports placed in nozzles mounted in flanges that seal the vacuum between HAM1-HAM2 and HAM5-HAM6.
Chris completed a SW model of the Brewster-window flanges.
OPTLEV/VIEWPORTS CDR
I am writing the conceptual design requirements document for the AOS OPTLEV & VIEWPORTS subsystems.
Controls, Data systems (R. Bork)
AdvLigo*:
- Lasti Hepi: (Rolf)
- At Dave Ottoway's requests, changed out the Hepi software to allow the HAM X monitors to read in geophones from another chamber.
- _Lasti ISI_:
- (Jay) Fabricated, tested and shipped 2 more capacitive position sensor chassis. LASTI now has enough electronics to run one chamber's worth of sensors.
- Mohana and Rich are converging on a coil driver for the ISIsubsystem. The current drive and noise performance are compatible witheither the coarse or fine actuator requirements, so only one design isneeded to cover both. The last remaining hurdle is to packageamplifiers and achieve good thermal management.- (Ben) I've received a detailed packet of information from JoeHanson regarding the new GS-13 board and cable for ISI. From thepictures, and the old cable that he sent, I was able to make a new boardand I'll send it out for fab soon.
- _Lasti HAM SAS:_ (Ben)
- After disseminating the latest schematics for the HAM-SASsystem which accommodated the new test electronics, Riccardo and I met,at which point he requested some alterations. I have finished the newdrawing incorporating these alterations, and am in the middle ofchanging the module schematics and PCBs to reflect the wiring changes,and the addition of a module. They should all be done soon.- Virginio has asked me to begin thinking about re-laying out theLVDT Driver board. If all goes well with the testing, we will mostlykeep the present design, and re-lay out the board. This is a new task,but one that is not a complete surprise. With luck, I can interleave itwith other tasks, and it shouldn't delay other things too badly.
- _Lasti Squeezing Exp_:
- Alex and Rolf at MIT to begin suspension control installation.All of the rack mount components, DC power and computers and softwareare up and running. Jay and Mohana will follow up next week to finish upthe field cabling.- (Jay) Completed the testing and shipment of 5 new satelliteamps. These satellite amps are replacements for the original LIGOsatellite amps. New features include use of BUF634s instead of LM6321s(obsolete), LED current monitors, use of 50pin Dsub connectors,indicator lights, and BNC output monitors. Adapter boards and boxes havealso been designed and fabricated to make the modules backwardcompatible with the LIGO satellite amps and forward compatible with theAdL SUS vacuum wiring.
- (Jay) Competed testing and shipment of 3 SUS receiver chassis.
*40m Lab*:
- Rich finished the layout of the Tip/Tilt Piezo Driver system andOMC length control board for the DC readout. The boards are at theboard house being manufactured. All needed parts are on order and willbe on hand next week. Some time next week, all the components go to thestuffing house for rapid turnaround, and the following week Rich shouldbe able to test everything. All other items for the DC readout arecomplete with the exception of some minor work Ben is doing on the QPDwhitening chassis. -(Ben) The in-vac wiring for the DC PD was showing signs ofproblems. It turned out that pin 13 of the in-vac cabling is tied tothe shield of all of the in-vac cables. This was only discoveredbecause several cables were touching the vacuum envelope, and wereshorting this pin together, and to ground. I modified the drawings toreflect this, and showed Bob what changes needed doing to a few of thesmall in-vac cables. I then talked to my brother, who changed thepinout of his PZT drivers to match the new wiring diagram. After allthis, the problem is resolved, and should pose us no further hazard.-(Ben) I helped Riccardo Colfi by band-sawing apart his PCBs. They are now ready for populating.
For additional information about this report, contact Albert Lazzarini or Phil Lindquist