Transcript Baseline Fit: What Went Wrong

Baseline Fit of the
Interferometer: What Is
Happening
By: Michelle Simon
W.M. Keck Observatory
Mentor: Jorg-Uwe Pott, Ben Berkey
Home Institution: Pacific Lutheran University
1
What You Will See
Today
The vocabulary you may
need to know
What was supposed to
happen with the baseline
What actually happened
before and after I arrived
What can be done to
make the program work
as desired
2
Vocabulary
Interferometer
Baseline
bFit
Python Script
3
Why this project is needed
and what an Interferometer
does
 An interferometer provides the ability to
resolve objects that may not be resovable in a
normal viewing with a single telescope.
 A new instrument (ASTRA) is being added to
the interferometer and requires more precise
measurements of the baseline.
4
How the program bFit
Works
 Takes the measured fringes
positions for each star
which are found when the
delay line matches the
geometric delay.
 Then derives the baseline
vector from these positions,
assuming a perfect rotation
of each telescope.
 The precision of the
baseline fit is good enough
to find the fringes with in
50 microns.
5
What is Happening
(Physically)
•
The rotation of each
telescope is imprecise and
has yet to be predicted.
1. The azimuth motion can
be repeated for each
telescope but is not the
same.
2. The rotation axis can
change from night to
night.
3. The up portion of the
baseline vector may
depend on the thickness
of the azimuth oil bearing.
6
What is Happening
(Programming)
1. Incorrect
measurements
that are
extremely off.
2. bFit as is, is not
designed for
detailed
monitoring of the
interferometer
baseline.
7
What I have done
 Created multiple
Python programs to
ease the baseline
analysis
 Each program is
documented to allow
the future user to
explore what has
been done.
 The programs help to
establish the cause of
bad data on the
results of bFit, and
allow to optimize the
data selection.
8
Graphs Showing the Impact of
Different Baseline Fits
9
What can cause bFit to
go bad in a single night
 Only observing a
single star
 Observing any
number of stars
but in small range
of azimuth.
 The variation in
the up and
constant terms of
the baseline
vector.
10
What's next...
The team will
further
investigate what
exactly causes
the baseline to
vary.
See if the
baseline variation
can be predicted.
11
Acknowledgments:
 Jorg – Uwe Pott
 Ben Berkey
 CfAO
 W.M. Keck
 Sarah Anderson
 The Akamai Internship Program is funded by
the Center for Adaptive Optics through its
national Science Foundation Science and
technology Center grant(#AST-987683)
12