| The LIGO
LIGO Livingston Observatory NewsRecycled Recombined Interferometer at LLO
Feed Forward Reduction of the Micro-Seismic Motion at LLO
Construction Continues on Staging Building
The Livingston Observatory has achieved locking of the full interferometer in its final configuration. The interferometer consists of a series of resonant optical cavities whose length must be controlled to a very high degree of precision--a few hundredth's of a wavelength of light (1.06 Ám in our case).
The two main cavities are the four-kilometer arms of the interferometer which both must have light resonant in them. Next, there is a Michelson interferometer formed by taking the light reflected from each arm and recombining it at as a beamsplitter. The difference in length between the near mirror in each arm and the beamsplitter must be kept at a constant half a wavelength of light so that the output of this Michelson interferometer is on a dark fringe. Finally, the recycling mirror which takes the light reflected by the Michelson interferometer and reinjects it into the interferometer must be controlled so that it forms a resonant cavity with the near mirrors of each arm. A general schematic of the configuration looks like this:
Control signals to push the mirrors to their proper position are extracted from various places in the interferometer. There are four degrees of freedom that must be controlled in the end: the common arm length, the differential arm length, the Michelson differential length, and finally the recycling cavity length. The gravity wave signal will be seen in the differential arm length.
We currently can keep the interferometer locked in its full configuration for about forty minutes during the night. We see light build-ups from the resonance of the optical cavities of over one thousand. Since we input about one watt of light into the interferometer, this means that we have about one kilowatt of laser light circulating in the interferometer arms.
Figure 1 below shows a screen shot of parts of the interferometer before it is locked. The upper left quadrant is an image of the end mirror on the X-arm, and the upper right is an image of the end mirror on the Y-arm. The lower left is an image of the light reflected by the interferometer, and the lower right is an image of the light at the dark port of the interferometer.
Figure 2 below is a screen shot of parts of the interferometer while locked. You can now see a bright spot of light on the upper left and right images of the end mirrors of the X- and Y- arms. This is from the kilowatt of laser light circulating in each arm cavity. Since the interferometer is locked, most of the light is circulating in the interferometer and there is little light reflected by the interferometer. This can be seen in the dim image at the reflected port on the lower left. On the lower right is seen an interference fringe on the Michelson dark port.
We are now working on keeping the interferometer locked for longer periods. We also want to understand how to maintain the interferometer lock during the day when ground noise in the area is ten to one hundred times stronger than at night.
The surface of the earth is constantly vibrating with a period of about six to seven seconds. This effect, known as "micro-seismic" motion, is observed worldwide. It is believed that most micro-seismic motion is weather induced, caused by the interaction of the wind with the ocean. Huge storm waves change the loading on the ocean floor. This time varying force produces waves which couple into deep geologic strata and from there they propagate around the world. Here at Livingston, Louisiana, the amplitude of the micro-seismic peak is typically around a few microns, varying with weather conditions at sea. The propagation velocity of the micro-seismic peak in Livingston is around two kilometers per second, so there can be a significant phase difference in the displacement of the input and end test masses of one of the interferometer arms depending on the relative direction of propagation of the micro-seismic wave with respect to the interferometer orientation. The micro-seismic motion can therefore be larger than a wavelength of light and must be compensated for in order to maintain the interferometer in an optically resonant state.
The "Fine Actuator System" (FAS) incorporated into the LIGO design provides a means of compensating for this motion. Piezo transducers with stroke lengths of about 90 microns allow displacement of the input and end test masses along the direction of each arm. Joe Giaime and Ed Daw have developed a scheme by which the earth's micro-seismic motion is sensed with high performance Streckheisen STS-2 three axis seismometers at the location of the input and end test masses. A mathematical model was developed that provides a best estimate of the expected displacement along the interferometer axes of the suspended test masses due to the ground motion.
This model was decomposed into the interferometer degrees of freedom called "CARM" and "DARM." CARM refers to the common mode stretching of the interferometer arms, while DARM is the differential change in arm length. These are the degrees of freedom used by the interferometer's length sensing control system to keep it optically resonant and sensitive to gravity waves. In operation, this scheme, referred to as a "feed forward" scheme, senses the earth motion using the various seismometers, estimates the appropriate compensations to be applied, and then moves the mirrors by actuating the piezo transducers to displace the entire seismic support structure from which the test masses are suspended. This reduces the amount of force that needs to be locally applied by the OSEM's (see our Summer 2001 newsletter for a discussion of OSEM's) in order to maintain the interferometer length.
This system was tested during the E6 Engineering Run held November 16-19, 2001. The figure below shows the results of implementing this scheme.
The figure on the left shows the amount of displacement ("amplitude spectral density" or ASD) as a function of frequency for both the common mode and differential servo systems with and without the feed forward correction as a function of frequency. The suppression of the micro-seismic motion is dramatically evident. The figure on the right shows the same data as a function of time with and without the feed forward correction.
Construction of the Livingston Staging Building is continuing. As shown below, the new front entrance is taking shape. This improves the appearance of the building considerably, as do the added windows on the east side.
The new entrance is being glassed in (left), and the sheet rock work in the auditorium is completed (right). Painting of this room should commence shortly.
On the second floor, the future LIGO Data Analysis System (LDAS) room, seen below, is coming along well. It is collocated with a LDAS office, and the communication room is next to this office.
On the far side of the building, the hi-bay is now closed in (left) and lots of sheet rock has been installed over the insulated wall panels. Also, the last of the air conditioning ducts has been lifted up into the roof truss for installation (right).
Tying the two upper floors together will be the "drawbridge" that allows handicapped access to both upper floors (below). Since this bridge partially obstructs the roll-up door, it has been designed to be lifted out of the way to gain the full height of the door opening when it is needed.
Under our present construction rate of progress, completion of the new staging building can be expected sometime early in the spring next year. Watch this column for full photo coverage of the final product!