From desai@gravity.psu.edu Fri Jan  4 22:15:23 2008
Date: Fri, 4 Jan 2008 22:15:20 -0500 (EST)
From: Shantanu Desai 
To: Keith Riles 
Subject: Proposed H2 DQ flag intervals for ISCT10 problem in Sept. 2007

Hi
Keith,
As you know there were a few H2 glicthes in Sept. 2007 because of
ISCT10_ACCX. (See my email below to the glitch group in Sept. 07 which I
have appended below, as a reminder of that.) Anyhow I would like to flag the
time intervals in the attached text file (maybe +/-1 sec can be padded
around these intervals) as an H2 DQ flag.
It could be called ISCT10_glitchiness or something similar.

One caveat obviously is that this is somewhat adhoc and not quantitative
and it is based on what is seen in the event display in the ISCT10 channel
around a given time, rather than any quantitative criterion (based  on the
trends in ISCT10 or something similar). But since this is a known problem
and documented in the ilog, I think it is ok to flag these intervals
somehwat crudely. Moreover, since  this is based on scanning of B-N outliers
, I may have missed some other times with this problem (which may have
caused low or moderate value transients)

Of course once someone does a more thorough study of this channel, then this
flag can be superseded by that.
But until then could the above intervals be flagged? Also on the DQ flag
page , maybe the following 2 weblinks can be added (as a reference)

http://tinyurl.com/2m8wgu (Link to the ilog which explains the problem)
https://gravity.psu.edu/~psurg/events1/evtdisplaylink-H2-2007-09-04.html
(Links to Event displays of these events)

Thanks
Shantanu


---------- Forwarded message ----------
Date: Thu, 6 Sep 2007 12:21:33 -0400 (EDT)
From: Shantanu Desai 
To: ligo-glitch@mit.edu
Subject: Lots of "band-limited" glitches seen in B-N in H2 due to
ISCT10_ACCX

This has alredy been noticed and diagnosed by Cheryl,
Gerardo,Corey  Robert,Keita-san et al.
http://tinyurl.com/2m8wgu
But today these ISCT10 events were seen in Block-Normal  as outliers with
power threshold > 1000.
https://gravity.psu.edu/~psurg/events1/evtdisplaylink-H2-2007-09-04.html

if you click on to the E-D of any one of these events, you will see a
glitch every 10 secs or so in ISCT10_ACCX/ISCT10_MIC

one characteristic feature of these glitches is that they appear in a
narrow frequency band (between 200-500Hz) unlike most glitches which
usually extend over the whole spectrum.

I presume  this is the first such occurrence of glitches and given that
right now these events have not been caught by any other online DQ flag,
this may need to  be DQ flagged.

Shantanu

    [ Part 2, ""  Text/PLAIN (Name: "isct10glitches")  5 lines. ]
    [ Unable to print this part. ]