Minutes of Detector Characterization Teleconference
(June 11, 2004)
Present:
Caltech: Ron Drever, Peter Shawhan, Patrick
Sutton, John Zweizig
CSUDH: Ken Ganezer
LHO: Mike Landry, Fred
Raab, Vern Sandberg
LLO: Mark Cenac, Brian O'Reilly, Giovanni
Santostasi
Loyola: John Whelan
Michigan: Keith Riles
Oregon: Ray Frey, Brian
Stubbs
Penn State: Mike Ashley,
Sam Finn
S2 / S3 Investigation Reports
-
Data quality segment repository -- S2 update) (John
Zweizig,Keith Riles)
KR reported that version 5 DQ segments for S2 have been released, with
flags for residual periods of the run when injected calibration lines were
absent. Many thanks to JohnZ and Xavi for tracking these down.
-
High time-resolution calibration information
(Sam Finn, Albert Lazzarini)
Sam presented results from H1 of tracking calibration line amplitude
& phase with a Kalman filter approach. In general, it would be desirable
to track calibration changes faster than 1 minute (see recent detchar presentations
by Gaby Gonzalez and Xavi Siemens), and a Kalman filter offers the potential
of better robustness against noise than a straight heterodyne. Sam uses
a 10-Hz band around the calibration line and reports back the amplitude
and phase of AS_Q w.r.t. the excitation channel every 1/10 second. Sample
time series of the amplitude / phase confirm non-Gaussian fluctuations,
and a power spectrum of the amplitude shows a 1/f behavior up to the Nyquist
frequency of 5 Hz.
KR suggested that Sam talk with Xavi to see if this method seems suitable
for use in generating h(t) and talk with Sergey Klimenko who studied different
line tracking algorithms during S3, including straight heterodyning.
DMT Software Status & Plans:
-
Planning for "Detector Investigation
Camp" following August LSC meeting (Keith Riles)
KR outlined tentative plans for a 3-day detector investigation camp
immediately following the August LSC meeting, where experimenters and analysts
would be invited to learn about a variety of diagnostic software tools,
including how to write DMT monitors. John Whelan wondered in what environment
attendees would be DMT programming. John Zweizig thought that one could
arrange for linux-compatible binaries to be available for laptops, in case
full DMT installation proved difficult for participants.
KR will be contacting prospective instructors soon and alerting the
collaboration to the planned camp.
-
Status of DMT infrastructure:
-
John Zweizig: Version updates, etc
John will be at Hanford next week where he plans to upgrade the online
machines with a new tagged DMT version 2.9 and with upgraded OS and compilers.
The changes will then propagate to Livingston online machines. Most of
the recent infrastructure changes are in low-level routines and should
be transparent to users. The only new monitor he plans to install is Natalia's
latest PTMON. Any other new/upgraded monitors for online installation should
be provided to John no later than Monday.
-
Patrick Sutton: Calibration infrastructure
The primary remaining task is setting up the definition and reading of
reference calibration info (point functions and miscellaneous constants
from EPICS channels at the time of the original calibration measurement).
Patrick is working with John to make this work. John is changing the DMT data
accessor to support the reading of a frame file with the frequency series
and constant info. Patrick will also put in an option to read back alpha
& beta values from the same file, when available (from a previous run).
-
Status of DMT monitors under substantial development / revamping / debugging:
-
Tim Bodiya / Keith Riles: SpecMon
Keith reported that Tim will be arriving soon at LHO for the summer
to finish shaking down SpecMon.
-
Ken Ganezer: seis_blrms
Ken has inserted some code into seis_blrms for checking inconsistent
time stamps, both as a potential fix, and to allow simulation of data dropouts.
The code changes have not yet been tested, however.
Fred Raab mentioned that Szabi's ShapeMon monitor has been alarming continually
for quite a while and requested that it be fixed. KR will e-mail Szabi.
-
Sergey Klimenko: BurstMon
Sergey was unable to make the call, but has found in his BurstMon surprisingly
large non-stationarity in the S2 data, even for H1. He will report at the
next detchar telecon (and the next Burst group telecon).
-
Tim Olson: SpectrumArchiver
No report.
-
Steve Penn: BicoMon and BicoViewer
No report, bur Fred mentioned that a Geneva high school teacher will
be spending part of the summer at LHO and working with Steve to look
at output from BicoMon.
-
Giovanni Santostasi: PulsarMon
Giovanni is coming up the DMT learning curve. He had hoped to work
at Natalia at LLO this month, but she happens to be out of the country
at the moment. Nonetheless, he has written some skeleton code and will
now try to compile and link it.
-
Brian Stubbs / Rauha Rahkola: SuspensionMon
Brian is about to graduate and leave Oregon, but Rauha will take over
maintenance of the SuspensionMon code until a new student becomes available.
There is one remaining problem not yet understood, in which appending to
trend files doesn't work.
-
John Whelan / Mark Cenac: StochMon
Mark has just started as a SURF student at LLO and is getting set up
with computer accounts. As beginners, he and Giovanni may find it useful
to collaborate on their respective monitors. For the time being, the plan
for implementing StochMon is to compute expected sensitivity for a pair
of interferometers, where the actual "local" IFO noise is used
together with a representative noise spectrum from the "other" IFO.
-
Natalia Zotov: PTMon
On vacation.
-
John Zweizig: LSCMon et al
Working on some infrastructure issues in PSLMon. The urgency to get
LSCMon working is less urgent, now that the effect of the low-byte data
corruption in AS_Q is believed to negligible.
A.O.B.
-
Ken Ganezer mentioned that he has been looking with FFT simulations at
the effect of thermal lensing on sideband power in the recycling cavity
and sees promising results. KR suggested he contact Erika D'Ambrosio who
showed some similar simulation results at the March LSC meeting.
-
Next detchar telecon: Friday June 25 at 1:30 p.m. EDT (S2/S3
investigations)