LSC/VIRGO Stochastic Analysis Review

Navigation

Review Home
StochReview List/Archive
Stochastic ilog
Stochastic list/Archive
LSCsoft CVS
LSC, LIGO
VIRGO
ALLEGRO

Docs

Conferences 2007
Edit these pages

Review

Overview
Members
Minutes
Status of Reviews

Agenda and Minutes, 06 February 2007

Agenda: Tues, 06 February 2007

I'd like to continue reviewing the Free Spectral Range paper. There is a new 
version posted. There is also a new timing technical document, and I've 
linked the calibration and the previous two timing technical documents on our 
review page at:

http://www.lsc-group.phys.uwm.edu/sbwg/review/projects.html

Next week I'd like to get the Rochester group and whoever else from the 
stochastic group is interested to discuss the analysis with us. This week, I 
want to get a list of questions to start addressing this week.
  

Minutes

+---------------------------------------------------------------------------
| 20070206-minutes.txt - minutes from the Feb 06 Stochastic Review telecon.
|
|  Author        : Warren G. Anderson (warren@gravity.phys.uwm.edu)
|
|  Last Modified : Tue Feb 06 03:00 PM 2007 M
+---------------------------------------------------------------------------

Attending: Warren, Nelson, Robert, Harry

Issues from last meeting:
- Calibration is the biggest issue
    - calibration group expects to have statement by March Meeting
    - no hardware injections - how can calibration issues be identified?
    - photon calibrator? probably not that good unless you happen to get on a
      resonance, which will make calibration useless.
    - what are the limits on what can be injected? other optical stimuli? use 
      resonant side bands?
- Timing jitter not measured in S5
    - new timing jitter result
    - Robert suggests using phase os a sharp resonance in band for S4.
- Why is there even a result at 37.5 kHz?
- This is an H1-H2 analysis - can we trust it?
  - What is the resolution. Robert would like to see coherence at an order of
    a mHz.  Broad band S5 coherence analysis?
  - PEM channel injections at those frequencies and look at the
    coupling/downconversion - Robert has a plan where he could inject high
    frequency acoustic, EM, etc noise at level that register in fast channel 
    and then compare to ambient levels.
- are there really no conceivable correlations at these frequencies?
- What are the lines in fig. 3?
- Will executive committee even be receptive?
Last action items:
- Contact calibration team.
- Use frequency modulated light?
- We will look at H1-H2 part.
- Robert will take a look at the fast channels.
This weeks AIs:
- have Rochester group discuss what they have looked at as possibilities for
  direct calibration of/injections in the fast channel.

Related Email

Subject: Two Questions
From: Warren G Anderson 
To: Adrian Melissinos , 
    Stefanos Giampanis , 
    Tobin Fricke 
CC: stochreview@gravity.phys.uwm.edu
Date: 2007-02-06 15:15

Hi Adrian et al,

The review committee continued our early discussions of the FSR paper again 
today, and two questions arose on which we would like to hear your feedback:

1) at what bandwidth do the coherence analysis in Fig. 7?

2) what methods were considered for doing a direct calibration or hardware 
injection at the FSR frequencies, and why were they rejected?

Thanks,
Warren

-- 
+================[ WARREN G. ANDERSON ]====================+
| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
| Calgary AB, T3H-3C4, CANADA                              |
+----------------------------------------------------------+
| P.O. Box 413,                     (414) 559-5366 US CELL |
| Dept. of Physics                  (414) 229-3323 OFFICE  |
| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
+==========================================================+

Subject: Re: Two Questions
From: Adrian Melissinos 
To: Warren G Anderson , 
    Stefanos Giampanis , 
    Tobin Fricke 
CC: stochreview@gravity.phys.uwm.edu, 
    Vuk Mandic 
Date: 2007-02-06 16:00

Hi Warren,

(1) The bandwidth for the coherence plot (as for all of the data analysis) is
at 1/32 Hz.

(2) As to the calibration, as mentioned in our document
http://www.ligo.caltech.edu/docs/T/T050180-01.pdf
it was done at the beginning of the run using the standard calibration
lines. It validates the fast channel.
      No injections were done at the fsr because they are not reliable (one can
not assume at these high frequencies that the mirror moves as a real surface
and and that there is good overlap between the test mass mode and the
laser mode). In addition the drive to the coils is so high that it often 
causes
the interferometer to get out of lock; thus it was out of the question to do
injections at the fsr during the S4 run.
      A surrogate for injections are the acoustic resonance lines which we can
(and do) monitor during the run. We believe that we have assigned correct
errors to our calibration measurements. The calibration committee is now
reviewing our procedures and it will be helpful to have their conclusions as
well.
      It is good to raise these questions, especially since it is important 
to clarify
them with respect to the S5 run before it is completed.

Cheers,
Adrian

At 05:15 PM 2/6/2007, Warren G Anderson wrote:
>Hi Adrian et al,
>
>The review committee continued our early discussions of the FSR paper again
>today, and two questions arose on which we would like to hear your feedback:
>
>1) at what bandwidth do the coherence analysis in Fig. 7?
>
>2) what methods were considered for doing a direct calibration or hardware
>injection at the FSR frequencies, and why were they rejected?
>
>Thanks,
>Warren
>
>--
>+================[ WARREN G. ANDERSON ]====================+
>| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
>| Calgary AB, T3H-3C4, CANADA                              |
>+----------------------------------------------------------+
>| P.O. Box 413,                     (414) 559-5366 US CELL |
>| Dept. of Physics                  (414) 229-3323 OFFICE  |
>| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
>+==========================================================+
>

Adrian 
Melissinos 
meliss@pas.rochester.edu
Department of Physics and Astronomy                        Phone  585-275-2707
University of Rochester, 
Rochester  NY                      FAX    585-276-0018

Subject: Re: Two Questions
From: Adrian Melissinos 
To: Warren G Anderson 
CC: Stefanos Giampanis , 
    Tobin Fricke , 
    Vuk Mandic 
Date: 2007-02-08 07:58

Hi Warren,
        Thanks for the comments. I think the best thing is for me to
write a little note on calibration in general and in particular at the
fsr where the situation is simpler, even though not as readily
intuitive. I will also attach links to direct "mirror shaking" work
that has been done at the fsr by our group as well as by Rick Savage
and collaborators over the past four years.
       I will try to work on it over the weekend so that I can send it to
you next week. Is that OK with the committee's schedule?

Adrian


At 01:50 PM 2/7/2007, you wrote:
>Hi Adrian,
>
>On Tuesday 06 February 2007 16:00, Adrian Melissinos wrote:
> >         (2) As to the calibration, as mentioned in our document
> > http://www.ligo.caltech.edu/docs/T/T050180-01.pdf
> >       it was done at the beginning of the run using the standard
> > calibration lines. It validates the fast channel.
>
>Yes, we've looked briefly at that document. We were more interested in what
>other options you tried.
>
> >       No injections were done at the fsr because they are not reliable (one
> > can not assume at these high frequencies that the mirror moves as a real
> > surface and and that there is good overlap between the test mass mode and
> > the laser mode). In addition the drive to the coils is so high that it
> > often causes
> > the interferometer to get out of lock; thus it was out of the question to
> > do injections at the fsr during the S4 run.
>
>This is more what we are interested in. We, of course, understand your 
>reasons
>for rejecting driving the optics mechanically. Were any other ways of putting
>a signal into the system considered but did not implement for some reason?
>
> >       A surrogate for injections are the acoustic resonance lines which we
> > can (and do) monitor during the run.
>
>Yes, this is also good to know. I don't recall seeing this in the companion
>tech document - if it's not in there, it should be.
>
> > We believe that we have assigned
> > correct errors to our calibration measurements. The calibration committee
> > is now reviewing our procedures and it will be helpful to have their
> > conclusions as well.
>
>Yes, we know. But we believe it would be good to consider having some direct
>measurement, as opposed to a measurement which is extrapolated. And the
>experimentalists in our group think that it is not inconceivable that
>something clever could be done. For other analyses, hardware injections have
>provided a critical confirmation that everything is well understood. We
>understand the difficulty of doing this for the fast channel, nonetheless, we
>would like to know what, if any, other options for driving the system at the
>FAR frequencies have been considered and why they haven't been implemented.
>
>Thanks,
>Warren
>
>--
>+================[ WARREN G. ANDERSON ]====================+
>| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
>| Calgary AB, T3H-3C4, CANADA                              |
>+----------------------------------------------------------+
>| P.O. Box 413,                     (414) 559-5366 US CELL |
>| Dept. of Physics                  (414) 229-3323 OFFICE  |
>| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
>+==========================================================+
>
>

Adrian 
Melissinos 
meliss@pas.rochester.edu
Department of Physics and Astronomy                        Phone  585-275-2707
University of Rochester, 
Rochester  NY                      FAX    585-276-0018

Subject: Re: Two Questions
From: Warren G Anderson 
To: Adrian Melissinos 
CC: Stefanos Giampanis , 
    Tobin Fricke , 
    stochreview@gravity.phys.uwm.edu, 
    Vuk Mandic 
Date: 2007-02-07 11:50

Hi Adrian,

On Tuesday 06 February 2007 16:00, Adrian Melissinos wrote:
>         (2) As to the calibration, as mentioned in our document
> http://www.ligo.caltech.edu/docs/T/T050180-01.pdf
>       it was done at the beginning of the run using the standard
> calibration lines. It validates the fast channel.

Yes, we've looked briefly at that document. We were more interested in what 
other options you tried. 

>       No injections were done at the fsr because they are not reliable (one
> can not assume at these high frequencies that the mirror moves as a real
> surface and and that there is good overlap between the test mass mode and
> the laser mode). In addition the drive to the coils is so high that it
> often causes
> the interferometer to get out of lock; thus it was out of the question to
> do injections at the fsr during the S4 run.

This is more what we are interested in. We, of course, understand your reasons 
for rejecting driving the optics mechanically. Were any other ways of putting 
a signal into the system considered but did not implement for some reason?

>       A surrogate for injections are the acoustic resonance lines which we
> can (and do) monitor during the run. 

Yes, this is also good to know. I don't recall seeing this in the companion 
tech document - if it's not in there, it should be.

> We believe that we have assigned 
> correct errors to our calibration measurements. The calibration committee
> is now reviewing our procedures and it will be helpful to have their
> conclusions as well.

Yes, we know. But we believe it would be good to consider having some direct 
measurement, as opposed to a measurement which is extrapolated. And the 
experimentalists in our group think that it is not inconceivable that 
something clever could be done. For other analyses, hardware injections have 
provided a critical confirmation that everything is well understood. We 
understand the difficulty of doing this for the fast channel, nonetheless, we 
would like to know what, if any, other options for driving the system at the 
FAR frequencies have been considered and why they haven't been implemented. 

Thanks,
Warren

-- 
+================[ WARREN G. ANDERSON ]====================+
| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
| Calgary AB, T3H-3C4, CANADA                              |
+----------------------------------------------------------+
| P.O. Box 413,                     (414) 559-5366 US CELL |
| Dept. of Physics                  (414) 229-3323 OFFICE  |
| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
+==========================================================+

Subject: [warren@gravity.phys.uwm.edu: [Stochreview] Two Questions]
From: Stuart Anderson 
To: Warren Anderson , 
    Harry Ward , 
    Robert Schofield , 
    Nelson Christensen 
Date: 2007-02-06 15:34

My apologies for missing the meeting today.

In the interim I spoke to Steffan who indicated that due to the timing
uncertainties in S4 the analysis assumed that that where arbitrary time
jumps between every 256 frame file, i.e., each 256s segment was coherently
analyzed on its own and the results incoherently added for a significant
reduction in sensitivity.

If this is true, then I believe the answer to question 1) below should be
1/256 Hz or less.

Does this agree with what others have found out for this 1 out-of 3
potential show-stoppers?


Thanks.


----- Forwarded message from Warren G Anderson  -----

Hi Adrian et al,

The review committee continued our early discussions of the FSR paper again 
today, and two questions arose on which we would like to hear your feedback:

1) at what bandwidth do the coherence analysis in Fig. 7?

2) what methods were considered for doing a direct calibration or hardware 
injection at the FSR frequencies, and why were they rejected?

Thanks,
Warren

-- 
+================[ WARREN G. ANDERSON ]====================+
| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
| Calgary AB, T3H-3C4, CANADA                              |
+----------------------------------------------------------+
| P.O. Box 413,                     (414) 559-5366 US CELL |
| Dept. of Physics                  (414) 229-3323 OFFICE  |
| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
+==========================================================+

Subject: Re: [Stochreview] [warren@gravity.phys.uwm.edu:  Two Questions]
From: Warren G Anderson 
To: stochreview@gravity.phys.uwm.edu
Date: 2007-02-07 12:01

Hi Stuart,

On Tuesday 06 February 2007 15:34, Stuart Anderson wrote:
> My apologies for missing the meeting today.
>
> In the interim I spoke to Steffan who indicated that due to the timing
> uncertainties in S4 the analysis assumed that that where arbitrary time
> jumps between every 256 frame file, i.e., each 256s segment was coherently
> analyzed on its own and the results incoherently added for a significant
> reduction in sensitivity.

I believe this is the gist of the new technote they've posted, with a factor 
of 5/6 increase in the upper limit corresponding.

> If this is true, then I believe the answer to question 1) below should be
> 1/256 Hz or less.

They actually divide each 256s segment into eight 32 second subsegments and 
generate Y's and sigmas for each of them, so I think 1/32 Hz is the relevant 
number, although in principle, I guess they could have done the coherence 
study with arbitrarily sized segments.

> Does this agree with what others have found out for this 1 out-of 3
> potential show-stoppers?
Yes, although your version (or Stefan's) seems much clearer to me than what I 
got from attending the stochastic telecon last week.

Cheers,
Warren

> --
> +================[ WARREN G. ANDERSON ]====================+
>
> | 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
> | Calgary AB, T3H-3C4, CANADA                              |
>
> +----------------------------------------------------------+
>
> | P.O. Box 413,                     (414) 559-5366 US CELL |
> | Dept. of Physics                  (414) 229-3323 OFFICE  |
> | Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
>
> +==========================================================+
>
>
>
>
>
> _______________________________________________
> Stochreview mailing list
> Stochreview@gravity.phys.uwm.edu
> http://www.lsc-group.phys.uwm.edu/mailman/listinfo/stochreview
>
>
> ----- End forwarded message -----

-- 
+================[ WARREN G. ANDERSON ]====================+
| 15 Sierra Vista Terr. SW          (403) 212-1426 HOME    |
| Calgary AB, T3H-3C4, CANADA                              |
+----------------------------------------------------------+
| P.O. Box 413,                     (414) 559-5366 US CELL |
| Dept. of Physics                  (414) 229-3323 OFFICE  |
| Milwaukee WI, 53201, USA          (414) 229-5589 FAX     |
+==========================================================+
$Id: 070206.html,v 1.1 2007/02/14 18:27:59 warren Exp $