This past year has seen great progress in two distinct detector activities:
installation-defined as getting working subsystems into position-and
commissioning-making the subsystems work together to operate as a
complete detector with full sensitivity.
Our installation and commissioning plan has evolved into one where
each of the three interferometers has a well-defined role, and the
scheduling of the work on each one has been tailored to its role.
The Hanford 2 km interferometer is the first in line and serves as
a "pathfinder" to identify problems early. The Livingston 4 km
interferometer follows about 6 months behind, and is used for
problem resolution and detailed characterization. We will initiate
coincidence testing as soon as the first two interferometers are
operational, but we will deliberately delay installation of some
elements of the Hanford 4km interferometer (primarily control
electronics) to enable lessons learned from the first two interferometers
to be realized in redesign before installation. The LIGO I science
run will begin when reliable and calibrated coincidence data on
three interferometers can be taken while keeping the configuration
stable for substantial periods of time. The improvements to reach
final design goals in sensitivity and reliability will be alternated with
data running in a way that optimizes both the early running and
obtains integrated high sensitivity data taking before the completion
of the initial LIGO science run.
On the installation front, the fabrication of the detectors was completed
(with the exception of some electronics components), and most detector
components have been delivered to the Observatories for installation.
Installation of the Hanford 2 km interferometer was completed in May 2000.
Installation of the Livingston 4 km interferometer is being completed as this
Newsletter goes to press (September 2000). As mentioned above the
Hanford 4 km interferometer installation has been intentionally delayed,
but substantial progress has been made: the seismic isolation has been
installed and the infrastructure (networking, data acquisition, monitoring
equipment) has been installed and tested.
Commissioning the LIGO detectors began even before the installation was
complete. On both the Hanford 2 km interferometer and the Livingston 4 km
interferometer, the pre-stabilized laser has been integrated with the
mode cleaner (a suspended cavity to stabilize the laser beam before it
enters the interferometer). Initial characterization of the laser/mode cleaner
system has been completed and show that the combination is already
very close to meeting their performance requirements.
In December 1999, we began a four month test of the 2 km interferometer
in which each arm of the interferometer was separately locked to the laser.
This test was performed to measure optical properties of the arms, to test
the interferometer sensing and control electronics, to gain information about
the environmental noise sources and to exercise the data acquisition and
control networks. Lock sections up to 10 hours were obtained and all planned
investigations were successfully concluded. At the end of the testing, a
24-hour
stretch of data was taken and archived for use by groups developing
software and techniques for data analysis and detector characterization.
During the past summer, we have gradually begun to bring the entire 2 km
interferometer on-line. We began by operating it in the recycled Michelson
configuration, (without the long arm cavities). This simple configuration
allowed us
to test the control systems: verifying the myriad connections, measuring
transfer functions, setting modulation/demodulation phases, all of the nuts
and bolts
of precision interferometry that must be in place before everything will work.
Most recently, we have locked the power-recycled Michelson with one
Fabry-Perot arm also locked on resonance. This initiated another set of
control system measurements which should lead to locking the full
interferometer
early this fall.
Of course, a number of problems have been encountered along the way, but many
have been solved, solutions to others are in the works, and none will
jeopardize
the performance or schedule significantly. We are on track to initiate the
first triple coincidence science runs by early 2002.