February Commissioning and March Persistence Coax Improved Performance from Hanford Interferometers
Two weeks of Mid-February commissioning on LIGO's H1 and H2 detectors produced sensitivity breakthroughs for each instrument. The S5 science run resumed at the completion of the commissioning interval; LIGO scientists and engineers followed with several weeks of stabilization activities. The results are very encouraging. During the final week of March H2 operated at an 87% duty cycle with a typical binary neutron star inspiral range of 23 million light years (7 Megaparsecs). H1 demonstrated a 90% duty cycle during the same period with an inspiral range of 46 million light years (greater arm length yields greater strain sensitivity). The commissioning team from LIGO Caltech and LIGO MIT will undertake similar activity at the Livingston Observatory in early April to improve the performance of the L1 interferometer.
Tackling the List
A list of targets for February's work arose after the start of S5 in November, 2005. The commissioners joined forces with LHO staff members to improve most of the major subsystems on each machine. On H1, better software filter banks were installed for the large optic suspensions (shown at right for H2) and for the optical levers. Upgrades to LIGO's radio frequency distribution system reduced RF noise on H1's photodiodes. Wavefront sensors (WFS, shown in the top-view photo below), the heart of LIGO's angle sensing and control subsystem, received an increase in bandwidth that significantly improved H1's abililty to maintain angular control of its mirrors. Better angular control led to improvements in length sensing and control (LSC), resulting in a higher signal-to-noise ratio at H1's dark port. On H2, progress centered on the LSC subsystem. Several length control loops were tuned to yield lower noise, lower levels of oscillation and better inter-loop performance, reducing the extent to which the behavior of one loop acts as a noise source in another.
Thermal compensation system (TCS) lasers and associated hardware, necessary for fine control of the shapes of the inner Fabry-Perot mirrors (ITM's), have historically added to the detectors' noise profiles through mechanical vibrations and TSC-induced backscattering of the main laser light. Both detectors benefited from noise mitigation on their TCS units.
H1 and H2 are now sensitive enough that seismic and acoustic effects from the site's quiet HVAC system manifest themselves as instrument noise. The obvious solution of turning off the air conditioning delivers an undesirable trade-off. Stringent environmental temperature control (i.e. leaving LIGO's air conditioning on) prevents thermally induced drifts in the mirror positions that degrade an interferometers' alignment over time. One post-commissioning task has been to further tune the air handling system in the large equipment rooms to provide adequate temperature control while preserving the interferometers' elevated sensitivities.
Prior to the commissioning period H1 operated with 5.4 Watts of input laser power at night and 4.4 Watts during the day. One evening's return to 5.4 W is marked on the right-hand plot below (click to enlarge). The morning power drop was necessary to keep the instrument in lock in the face of elevated daytime seismic activity, but the switch removed about 15% of the detector's sensing range. Tightening H1's angular and length servos not only increased the average inspiral range but also allowed the interferometer to run in a stable condition at 5.4 W throughout the day. Late one evening near the end of the commissioning period, a computer script failed to heed a line of code that set the input power at 5.4 W. The power instead rose to 6.4 W, a level that would have quickly destabilized the machine in January. Instead, H1 yielded solid performance at the elevated power and 6.4 W became the new default, adding several percent to the inspiral range.
For several years Hanford's 2-kilometer detector has displayed remarkable stability but a sensing range that suggested unfulfilled potential. Commissioning activities produced a welcomed makeover, bringing H2's displacement sensitivity to a level that challenges H1 and L1 at most frequencies. H2's binary neutron star range of 7 Megaparsecs now approaches the inspiral sensitivity that H1 displayed during the S4 run of 2004.
The Rest of the Story
The commissioning team departed late in February with excellent sensitivity gains in hand. LHO staff members then faced the challenge of returning the interferometers to smooth 24-hour operations. Initially the machines resisted, displaying a variety of temperamental behaviors in early and mid-March. In response, scientists have sought to characterize and moderate the instabilities. Engineers have constantly monitored and adjusted HVAC air flows on the instrument floors. Control room operators have deduced new techniques for maintaining stable optical alignments and for diagnosing the control systems. The 03/29/06 plot shown above (left) attests to the success of these efforts. As March draws to a close and the interferometers approach 90% duty cycles, LIGO looks forward to similar commissioning progress on L1 and anticipates a highly sensitive and robust search for gravitational waves in the next installment of the S5 science run.
Last modified March 30, 2006
For problems or suggestions about Web material, contact firstname.lastname@example.org
For information about LIGO, contact email@example.com
LIGO is supported by the National Science Foundation
Any opinions, findings, conclusions or recommendations expressed here are those of the
author(s) and do not necessarily reflect the views of the National Science Foundation