|
 |
 |
By Sean Adkins
|
 |
| Photo: Keck engineer Tom Nordin works on
MAGIQ camera testing and calibration inside the clean
room. A computer display of camera imaging and control
software is shown in the foreground. Photo by Sarah Anderson. |
For small telescope users, direct visual observation and a
star finding chart are usually sufficient to find an object
in the sky. For astronomers who use large telescopes like those
at Keck Observatory, many of the objects they wish to observe
are extremely faint. Even instruments equipped with the most
modern and sensitive electronic detectors require long exposures
to obtain useful images.
To obtain the long exposures astronomers need to observe faint
objects, they must constantly move or “guide” the telescope
to keep the object in alignment with the telescope, as the
telescope systems compensate for the rotation of the Earth.
On a large telescope this guiding motion must be very precise,
so as to avoid blurring the high-resolution image. Because
the science target is so faint, other brighter objects near
the science target, referred to as guide stars, are used as
the basis for making measurements of where the telescope is
pointed. The guide star is imaged using a dedicated electronic
imager commonly called a guide camera. Measurements of the
position of the guide star are used to control telescope pointing
and tracking, in order to keep the fainter science target properly
positioned in the field of view during the observation. Large
telescopes like those at Keck use a closed loop system for
guiding, where the error signal for the closed loop is based
on the position error derived from measurements of the guide
star.
Keck Observatory’s two 10-meter telescopes are currently the
world’s largest optical and infrared telescopes. The development
of the Keck telescopes presented many challenges, due to the
sheer size and mass of the telescopes and their mirror systems.
The Keck design team needed to build highly accurate motion
control systems for pointing the telescope, as well as systems
to adjust the telescope optics and compensate for the effects
of gravity as the telescope is moved during an observation.
While the original guiding system provided excellent performance
at the time it was developed, this system can now be improved
to offer even better performance and reliability, by taking
advantage of new and improved technologies.
The Keck I telescope began science operations in 1993, and
the guide cameras on Keck I use small format (288 x 384 pixels)
detectors that have poor sensitivity when compared to currently
available detectors. These cameras are obsolete and, in many
cases, spare parts are no longer available. The guide cameras
on Keck II are also no longer manufactured, but they are equipped
with larger detectors (1024 x 1024 pixels) with better sensitivity.
The age and obsolescence of the current guide cameras result
in relatively frequent “faults,” or errors. Over a one-year
period (July 2004 to July 2005), faults due to the acquisition
and guiding systems of the Keck II telescope resulted in the
loss of a total of 46 hours of observing time - a significant
amount over the course of a year. While the Keck I acquisition
and guiding system fared better, faults resulted in the loss
of nearly 8 hours of observing time in that same year. In general,
acquisition and guiding system faults are responsible for the
largest amount of lost time at the observatory, excluding weather.
 |
| Photo: View of the MAGIQ prototype camera
showing the new CCD detector. Photo by Sarah Anderson. |
Keck Observatory has recently completed the preliminary design
phase of a project to implement a major upgrade to our guiding
system. This new system is called MAGIQ, the Multi-function
Acquisition, Guiding and Image Quality monitoring system. The
MAGIQ project will replace the guide cameras and software for
existing instruments, in order to improve the process of finding
objects (acquisition) and guiding performance. The MAGIQ system
will become the observatory standard for new instruments.
The new MAGIQ guiding system is also expected to contribute
to improvements in telescope image quality. Current practice
is that every night before observing starts, observing assistants
run a special software program which allows the telescope optics
to be precisely adjusted to deliver the best image quality.
The figure below illustrates the improvement in image quality
after one such adjustment of the telescope optics.
A major factor in image quality is telescope focus. Changes
in focus occur throughout the night because of changes in temperature
and because of the effects of gravity as the telescope is moved.
While the telescope control system provides some compensation
for these effects, image quality usually deteriorates during
the course of the night, so that we are forced to interrupt
observing to readjust the focus. Each of these adjustments
can take up to 15 minutes, and they are often performed two
or three times each night. To eliminate these interruptions,
we are designing MAGIQ to not only guide the telescopes, but
also to monitor image quality and make real-time adjustments
of telescope focus, a capability that was not incorporated
into the system when the telescopes were built.
MAGIQ will provide a range of selections for focus control.
These include “on demand” image quality verification and telescope
focusing. This will take less than five minutes and will allow
guiding and observing to continue while the focus is measured
and adjusted, if necessary. MAGIQ will also provide a continuous,
automatic mode that will operate in conjunction with guiding
and that will not interfere with the observation or affect
the resulting science data. During the course of the observation,
MAGIQ will provide information about image quality that can
be displayed on a computer screen, logged into a data file,
and included in the header of the observation data file.
 |
|
 |
| Photo: Before adjustment, 0.8 arc second
image size. |
|
Photo: After adjustment, 0.5 arc second
image size |
Upgrading the acquisition and guiding system
at Keck is well aligned with the observatory’s strategic
plan, which has as a major goal the achievement of highly
efficient operations. The expected benefits to science
include increased efficiency for observations, improved
quality for imaging observations, and valuable supplementary
data on delivered image quality during all observations.
The MAGIQ project is funded by a grant from NASA, a partner
in the operation of the observatory. NASA has supported a number
of instrumentation developments at Keck Observatory. In 2004
we submitted a proposal to NASA for a new three-year grant
to develop a replacement for our guide camera system. When
our proposal was funded and we were ready to start work, Rich
Matsuda, the observatory's electronics engineering manager
and other members of the electronics group came up with a new
title for the project that resulted in the acronym "MAGIQ."
Like almost every technology development in astronomy, MAGIQ
is a multidisciplinary project. The project team is made up
of observatory staff from several groups: instrumentation,
observing support, optics, mechanics, electronics, and software.
We also have collaborators at our partner institutions on the
Mainland. Professor Judy Cohen of Caltech serves as our project
scientist. As an experienced observer, Judy provides guidance
on scientific issues and ensures that the user’s perspective
is adequately addressed in the design and implementation of
the project.
To establish the requirements for MAGIQ, we followed an approach
that considers science, user and functional requirements, compatibility
with existing instruments, and integration with the telescope
control system. At Keck Observatory each of our instruments
has its own acquisition and guide camera, with a unique optical
and mechanical configuration. Installing new guide cameras
and providing for image quality monitoring within the constraints
of the limited space available in the instruments and for the
variety of optical designs is a significant challenge. We also
have to interface with existing software systems that control
the telescope and collect the science data.
Working with the observatory’s support astronomers and observing
assistants (telescope operators) we initiated a “brainstorming” process
to identify user and functional requirements. The support astronomers
and observing assistants have nightly experience with the performance
of the existing guider system. Their experience was essential
to frame the improvements and new functionality desired for
the new guider system. Through the brainstorming process, some
400 detailed requirements were collected. This list included
various aspects of the problems with the existing system and
the improvements and added capabilities that a new acquisition
and guiding system might provide.
 |
| Photo: MAGIQ camera electronics connected
to the prototype camera and equipped with a lens for
laboratory testing. Photo by Sarah Anderson. |
Using this bottom-up approach helped us to gain insight into
the scope of the problems that we needed to solve with MAGIQ.
The next step was to examine the system from a high-level perspective
in order to identify the science and user-driven requirements
that would provide the framework for designing the MAGIQ hardware
and software. We developed the high-level user requirements
by considering the perspectives of three distinct user domains:
the astronomers, the observing assistants, and the engineering/maintenance
staff.
The astronomers are primarily interested in efficiently and
accurately acquiring the science object and then monitoring
their observation until the exposure is complete. The quality
of the data is a primary concern, so the astronomer needs to
be assured that the image quality is maximized given the observing
conditions. Also, when post-processing the science data, the
astronomer may wish to know the measured image quality during
the observation and analyze its effect on the science data
recorded.
The observing assistants operating MAGIQ and the telescope
are interested in being able to easily access the capabilities
of the system so that the observation desired by the astronomer
is carried out as efficiently as possible, with the highest
possible quality. This requires MAGIQ to offer system features
that are flexible and straightforward to employ and monitor.
The engineering and maintenance staff is responsible for assuring
that the telescope and instruments perform reliably and at
peak performance. Peak performance includes ensuring that the
closed-loop control systems, for example guiding and focusing,
operate properly. The condition of the equipment must be monitored
to maintain consistent and reliable performance. Repairs and
maintenance must be done safely and efficiently to make the
best use of limited observatory resources.
The chance to improve the performance of the telescopes and
the instrumentation system for all three of these user
groups is what I like most about working on MAGIQ.
My entire career has involved interdisciplinary projects, and
I enjoy working on the development of real systems that combine
optics, mechanics, electronics, and software.
 |
| Photo: Sean Adkins, instrument program
manager at Keck Observatory. Photo by Sarah Anderson. |
MAGIQ offers something for everyone to get
excited about. Astronomers, observing assistants, and the
engineering/maintenance staff will appreciate the improvements
in observing efficiency with MAGIQ, which will result from
the increased sensitivity of the acquisition cameras, the
improved performance of guiding and focusing, and the more
efficient acquisition and setup of observations.
After the first three-year phase of the MAGIQ project is completed,
there are five more instruments and several more cameras that
still need to be upgraded. To make a charitable investment
in Keck Observatory’s MAGIQ instrumentation project, contact
Debbie Goodwin, Director of Advancement, by calling her at
808.881.3814.
This material is based in part upon work supported by the National
Aeronautics and Space Administration under Grant No. NNG05GC95G
issued through the NASA Astronomy and Astrophysics Research
and Analysis program. 
Click here to return to the main page. |
|
|
 |
|
