From the Palomar Testbed Interferometer

July 27, 1995

Michael Shao, leader of the Palomar Testbed Interferometer Project, conveyed this message to his colleagues at 0h 19m this morning:

"The guys got fringes on alp cyg about 15 min ago. We're going for a 2nd star (and maybe 3rd, 4th) so we can solve for the baseline vector."

No other details have been released, but more news will follow this afternoon, when he and his team wake up (I'm assuming that they went to sleep when the sun came up, but they may still be celebrating!) Congratulations are due to Mark Colavita, Brad Hines, Jeff Yu, Kent Wallace, Dean Palmer, Xiaopei Pan, Fabien Malbet, Jean-Francois Leger, and Yekta Gursel. We're on our way to detecting extra-solar planets!

Last Updated: Sept. 12, 1995

First fringes were detected by the Palomar Testbed Interferometer on Alpha Cygni on July 27, 1995. Fringes were tracked on the star for several minutes, and four other stars were observed that night in order to solve for the instrument baseline.

The baseline of the instrument currently in use is a 110-meter baseline that is close to North-South (103 m N-S, 37 m E-W). The 7 weeks since then have been used to increase the level of automation of the instrument, tune system gains, coordinate transformations, and other parameters, and perform some early science measurments. At this point, V^2 values of 0.5 or better have been observed for small sources, and fringes have been tracked on stars down to about 4.5 V-magnitude. The magnitude limit at present is due to the fringe detector; many more photons should be available, but no work has been done to date to try to track this down.

Operation of the instrument in single-star mode is now fairly routine. On Sept. 8th, fringes were tracked on 13 different stars repeatedly for a total of about 35 observations over the course of the night. Data recording and analysis software is all in place, with about 500 MB of fringe and delay data being produced on a highly productive night. Baseline solutions using the data are typically ready for the next night's observing.

At the instrument's current level of efficiency, it is possible to switch to a new star and acquire fringes in approximately 5-10 minutes. This time is dominated at present by the fringe search time; fringes are typically found within +/- 10mm of the expected position. We should be able to improve this to just a few seconds once the baseline is known precisely - we believe that there may be some problems in the way atmospheric refraction is currently being accounted for.

Once fringe search time has been reduced to a few seconds, the limiting factor in observing speed will be the wide-angle acquisition camera system, which is used to intially position the star image directly on the quad-cell APD detector of the fast tip-tilt sensor. The camera currently being used will probably be limited to about one minute per acquisition. So the final observing efficiency (in single-star mode) is expected to be about 1-3 minutes per star required to acquire and track the fringe, down from 5-10 minutes now.

The instrument is designed to work in a dual-star mode, observing both a bright primary star and a nearby faint secondary star simultaneously, taking advantage of isoplanatism to allow measurements made on the bright star to be used to phase the instrument for the secondary star. Optics such as siderostats and delay lines are largely shared between the two interferometers in order to avoid systematic errors.

As of this writing, installation and commissioning of one of the dual-star feed systems has commenced, and light from both stellar beams is now feeding into the main optics lab. Initial testing of the secondary star selector mirror (which performs the offset pointing from the primary to the secondary star) and the feedforward from the primary tip-tilt control system to the secondary will be performed over the next two weeks. It is hoped that we will be able to acquire secondary star fringes by the end of October.