Minutes of Detector Characterization Teleconference
(July 12, 2002)
Present:
Caltech: Drever, Lazzarini,
Marka, Shawhan, Zweizig
Florida: Klimenko
LHO: Chin, Ito, Penn, Raab,
Rahkola
LLO: Berger, Coles, Zotov
Michigan: Riles
Penn. State: Schlaufman, Sutton
Syracuse: Saulson
UWM: Brown
S1 Preparation:
- Schedule - Gary Sanders has announced that S1 redux will begin
8:00 a.m. PDT on Friday August 23 (day after LSC meeting at LHO ends) and
run until 8:00 a.m. PDT Monday September 16.
- DMT code deadline - John Zweizig wants all new DMT monitor
code to be run in S1 provided to him no later than Friday August 9 (two
weeks before run begins)..
- Scimon shifts - KR has sent out a solicitation for scimon shift
volunteers. As usual, the goal is to staff both sites with an expert and trainee
24/7. In the event of manpower shortage, priority will be given to owl and
evening shifts, when the interferometers are likely to run best.
- Scientific investigations - A list of
S1 investigations
and team members has been posted. More volunteers are welcome. Persons manning
scimon shifts are encouraged to sign up for an investigation.
- Interferometer status - Fred reported that a spare mirror was
installed in the Hanford 2K to replace the one dropped by a broken wire on
June 28. Pumpdown is underway. The broken wire is attributed to melting by
an input beam tilted by another mirror whose alignment servo went haywire
after jostling from a large earthquake in China. The new common mode servo
feedback configuration has been installed for the Hanford 4K. Conditions
are not yet steady. There is a lingering problem with one end mirror where
unknown saturation leading to misalignment appears to interfere with robust
locking. Mark reported that the Livingston 2K has a new common mode feedback
configuration that appears to give some improvement in noise at low frequencies.
The new PEPI (Piezoelectric External Pre-Isolator) system will begin
installation on Monday for ITMX and ITMY. These are already installed for
ETMX and ETMY.
Mini Data Runs:
- There was some discussion of a proposal now floating around to
hold mini data runs on a regular or semi-regular basis between official
engineering and science runs. One such run was held the night before at Livingston.
All agreed that such runs were potentially valuable for understanding the
progression of commissioning, but there was some disagreement on whether the
added value was worth the potential costs. There was near consensus
that such a set of runs should have flexible scheduling, based on opportunistic
data taking, perhaps tied to achieving certain commissioning milestones.
There was reluctance to require coincidence running between the sites. There
was worry that a too-flexible schedule would mean the runs would never happen.
There was worry that putting an interferometer into a well defined, "neat"
state on a regular basis would be a burden on commissioners. There was worry
that upper limits analysis groups would not have the manpower to analyze
the incoming data and give feedback. Peter Saulson will be bringing this
idea up for discussion at the next upper limits chairs teleconference. All
agreed that the topic deserves more discussion.
DMT Software Status Review:
- DMT infrastructure - J. Zweizig
- John sent around to DMT authors a revised proposal for the
alarm class and received no objections.
- Version 2.1.0 is now frozen. A new version 2.2.0 will be frozen
on August 9 for use in S1.
- DMT Trend viewing with the data viewer is now working correctly
at Livingston.
- After some discussion, Fred & Mark agreed to start the operators
on looking at the preliminary scimon shift procedures to get more familiar
with the DMT tools. It may be appropriate to transfer some of the scimon duties
to the operators. It would be generally desirable for operators and scimons
to have a reasonable understanding of what the "other guys" do.
- DMT summary page - S. Marka
- Szabi has some but not all of the pieces needed from DMT authors
for the global summary page. He agreed to provide a demo version showing
some of the pieces now available. KR offered to help nag authors to provide
what's still missing.
- Monitoring of astrophysical search engine function and trigger rate
- D. Brown
- Duncan has nearly completed the infrastructure for web monitoriing
of DSO's looking for astrophysical signals. He expects to release the
first version with monitoring of the inspiral DSO next week. A set of generic
routines will be made available so that other DSO authors can exploit the
same infrastructure. Alan Weinstein has expressed interest in applying it
to burst DSO's.
Updates from DMT monitor authors (progress since Monday meeting and
prospects for S1):
- Dave Chin: LockLoss and ServoMon
An earlier problem with trending has been fixed.
.
- Ed Daw: blrms and PeakMon (no update)
- Masahiro Ito: glitchMon
All known bugs have been fixed, including earlier trouble with trend outputs.
The monitor is up and running at both sites.
- Sergey Klimenko: LineMonitor and WaveMon
The nine (6 at LHO, 3 at LLO) standard manifestations of LineMonitor are
now running with updated config files. Alarms will be enabled for S1. Three
WaveMon programs are now running, one for each interferometer.
- Szabi Marka: IRIG-B and TimeMon
Trends are now visible with the data viewer at LLO. New config files have
been created.
- Adrian Ottewill: MTLineMon and CorrMon (update by e-mail)
A revised version of CorrMon has been given to Nelson Christensen for testing
before forwarding to John. Adrian is now focussing on MTLineMon.
- Steve Penn: BicoMon
BicoMon has a new, more user-friendly GUI for setting time/frequency parameters,
and a web page introduction has been written that includes examples of various
non-linear artifacts to watch for. Steve will look at some data Robert Schofiled
took at the end of E8 in which upconversion was deliberately induced.
- Daniel Sigg: MultiVolt (no update)
- Patrick Sutton: SenseMonitor and RayleighMonitor
SenseMonitor's previous algorithm for tracking drifts in calibration via
the injected calibration lines proved not to be sophisticated enough because
of short-term fluctuations. Looking at longer trends seems to work better.
Trends are now visible, and calibrated power spectra are dumped periodically
to disk in ascii format as an archive. The present ad hoc method for retrieving
the most recent transfer function measurement will be automated in collaboration
with Mike Landry.
A fast version of RayleighMonitor is now working which uses Daniel Sigg's
"in-place' histogram updating to do fast scrolling in real time. The disadvantage,
however, is that one cannot update anything but the histogram contents once
the program starts, and one cannot interrupt it with a mouse click.
- Natalia Zotov: PTMon
PTMon has been run on some LLO laser channels and does pick out true glitches.
Natalia hopes a version will be ready in time for S1.
- John Zweizig: BitTest, HistCompr, PSLMon, and
SegGener
At Rana Adhikari's request, John resurrected an old program (ListenMon) he
wrote that allows audio listening to any channel -- if one has a DMT workstation
with a sound card. At present the program can be run on the control6 workstation
at LLO. Control6 is a CDS machine, but that runs DMT code and has a Gbit
ethernet connection to the frame broadcaster. It should be suitable for running
interactive programs like BicoMon and RayleighMonitor. ZGlitch (part
of PSLMon) now provides trends of glitch rates (trend files and DMT viewer
plots). The HistCompr config files have been tuned, and the shorthand channel
names made less cryptic. Alarms will be placed on high glitch rates and on
band-limited powers. 0n The documentation for all four monitors has
been updated. SegGener used only physical locks in E8, but will use the new
"run mode" OSC condition for S1 which uses the operator/conlog bits in the
state vector. PSLMon now supports windowing of band limited RMS. ZGlitch
(part of PSLMon) now writes trends of trigger rates.
A.O.B
- Fred requested that when persons update the scimon / operator shift
worksheets, they should remember to include the date of last revision
- Next teleconference: Friday August 9 at 11:00 a.m. EDT (8:00
PDT). This telecon will be devoted to S1 preparations.