The LIGO
Web Newsletter
Front Page Hanford Caltech

LIGO Livingston Observatory News

Preparing for the First Science Run

Preparing for the First Science Run

- Contributed by Mark Coles

During the past month, many upgrades to interferometer hardware and software have been undertaken, and these are now close-to or fully completed. This work is in anticipation of the upcoming first Science Run, "S1," scheduled to begin this summer. These upgrades are expected to result in improved interferometer sensitivity and enhanced operational robustness.

A new version of the length sensing and control software was installed by Rolf Bork, which aids the utilization of digital filtering of the interferometer's various servo control signals. This software was field tested on the two-kilometer interferometer at the LIGO Hanford Observatory (LHO) before being installed here at Livingston, so bringing the software into full operation was relatively painless. At the same time, a more powerful Pentium CPU was installed to support the digital filtering capabilities that the software allows. Rus Wooley, Ken Watts, and Mike Fyffe re-routed much of the ASC (Alignment Sensing and Control) and LSC (Length Sensing and Control) wiring to minimize electronic cross talk between various parts of the data acquisition system. This has also reduced power line coupling. New Global Positioning System clock driver modules and timing modules were also installed which further reduced the noise from sampling pulses. This has resulted in significant improvements in the signal/noise ratio in the gravity wave band.

New rack wiring for the LSC and ASC.

The common mode filtering was significantly improved by Daniel Sigg, Nergis Malvalvala, Rana Adhikari, and Rai Weiss. Changes in analog filtering in the input path to the analog-to-digital converter, as well as the addition of a number of digital filters, were the primary means of accomplishing these enhancements. The common mode loop topology was also changed by Rana Adhikari, following a suggestion and first test of this approach by Daniel Sigg at LHO. The new topology makes it possible for the laser frequency to match the average arm length of the two four-kilometer-long Fabry-Perot arm cavities, even for very low frequencies. As a consequence, the low frequency noise spectrum of the interferometer is expected to improve even further.

Late night commissioning activities at LLO.

An intensity stabilization servo was also added to the pre-stabilized laser control system by Flavio Nocera, Peter King and Rich Abbott. This system will result in a more stable laser power input to the interferometer. This consistency of the laser power means that the circulating power in the mode cleaner is also more constant, thus reducing the radiation pressure noise on the apparatus. Otherwise, variations in the radiation pressure causes the length of the mode cleaner to change. This produces frequency noise as the laser attempts to vary the wavelength of the laser light to keep the mode cleaner in resonance. (This is the inner loop servo which stabilizes the intensity delivered to the mode cleaner. A second outer loop servo which closes the loop around the output intensity of the mode cleaner will be implemented after S1 and this should result in yet further improvements.)

The wave front sensor and its associated servo-control loops on the anti-symmetric port were brought into full operation by Gabriela Gonzalez of Louisiana State University (LSU). This sensor tracks the relative angular alignment of the interferometer with respect to its Fabry-Perot arms. It produces a servo control signal which is used to maintain the angular alignment of the anti-symmetric port output (the gravity-wave signal) at frequencies up to around 0.5 Hz. An important finding has been that the alignment servo using this wave front sensor keeps the AS_I signal (which is not controlled directly by any servo) bounded so that the saturation does not occur in the RF amplifiers as the light power is increased. By precisely aligning the output beams of the interferometer at the output port, the carrier contrast defect is reduced. In simpler terms, it stays dark because the output beams from each arm are held in a fixed relative alignment at the maximum destructive interference point. This allows the laser power to be substantially increased, resulting in a reduction in the interferometer noise at high frequency.

Composite photo of the layout of ISCT4.

Another key improvement underway is the implementation of a piezo-electric actuator system (PEPI) to minimize the disturbance on the interferometer due to ground motion exciting resonant modes in our vibration isolation stacks. The new system, being implemented by Joe Giaime (of LSU), Rana Adhikari, and LSU graduate (and soon-to-be Caltech graduate student) Dan Busby, is an enhancement to an early system Giaime and Adhikari implemented prior to the seventh Engineering Run (E7). The new system uses a sophisticated digital control system to sense the motion of the external seismic isolation frame over the frequency range 0.5-5 Hz and apply appropriate feedback via LIGO's peizoelectric fine actuators to reduce it. This is expected to decrease the effects of ground motion on a suspended test mass at the stack resonances by about a factor of ten. Implementation of PEPI should also allow improvements to the already-implemented microseismic feed-forward system, which uses the same fine actuators to correct for the approximately 0.15 Hz microseism disturbance (largely due to deep-ocean storms) as measured by an array of broadband seismometers at the site.

We have also significantly added to our computational capabilities here at LLO. Chethan Parameswariah has installed new Sunblade 1000 computers in the control room. Their greater speed makes it possible to monitor and control a greater number of data points. He has also added software and hardware which increases the overall security of the system and makes it possible to establish remote control rooms at Caltech and MIT so that expert personnel at those locations can assist with the operation or troubleshooting of the interferometer. A new set of hardware is replacing the existing 16 node Beowulf cluster which supports the LIGO Data Analysis system. The new system comprises 48 nodes of rack mounted PC's in a compact array of relay racks. Igor Yakushin is currently testing the new hardware and plans to have it fully operational in time to support S1.

Finally, Gerry Stapfer and Joe Hanson replaced the infamous cattle guard at the entrance to the LLO property! The guard had been a source of seismic noise, rattling the axles of every vehicle entering or leaving the site and generating ground motion that could be seen by both our seismometers and the interferometer itself. The hole where the cattle guard was has now been replaced with a smooth concrete slab.

The 'smooth' new entrance to LLO.

Taken together, these improvements have resulted in the interferometer operating at a level of about a factor of ten more sensitive than it did just a short time ago during E7. Our task during the next month will be to make sure these upgrades are robust and reliable, and that the operating staff are well-trained to use them knowledgeably.


    Onwards! to the LIGO Livingston Observatory     Or  Return! to the Newsletter Front Page