AAS 203rd Meeting, January 2004
Session 24 Observatories and Observing Conditions: Optical to X-rays
Oral, Monday, January 5, 2004, 10:00-11:30am, Centennial III

[Previous] | [Session 24] | [Next]

[24.02] Longitudinal Dispersion Compensation for a Long Baseline Optical Interferometer

D.H. Berger, W.G. Bagnuolo, T.A. ten Brummelaar, H.A. McAlister (CHARA / Georgia State University)

Long baseline stellar interferometers with air-filled light paths are limited at visual wavelengths (\lambda=0.4 to 0.8~\mum) by longitudinal dispersion --- a chromatic path length mismatch caused by light travel through unequal lengths of media with wavelength-dependent refractive index. Consequently, there is a degradation in the degree of coherence as measured by the fringe visibility amplitude. A solution for the CHARA Array, a six-element optical interferometer, will be to use a pair of glass prisms to match the dispersion in each of the six input beams. Without longitudinal dispersion compensators (LDCs), long baseline optical interferometers resort to a choice amongst limiting bandwidths, using shorter air-filled delay lines, and/or observing with shorter baselines resulting in decreased sensitivity, confined sky coverage, and/or lower resolution, respectively.

Dispersion models that included the refractive indices of air in the path length compensation facility and a choice of ~200 glass types was developed to significantly narrow the search for a suitable LDC prism pair, which was optimized for high-dispersion R band observations (\lambda=0.6 to 0.8~\mum), but was required to not be detrimental to the near-IR wavelengths. The components of each LDC unit are: optomechical stages and mounts used for positioning and alignment, a high precision motion control translation stage for thickness adjustment, and prisms made of SF 10 glass --- a commonly used dense flint.

The throughput of the LDCs is 98% in R, 95% in I (\lambda=0.9~\mum), and 75% in K' (\lambda=2.2~\mum). The ratio of the fringe position shift to the change in glass thickness was 0.77 and confirmed the translation stage resolution exceeds the predicted 1.5 \mum. Dispersion control was demonstrated and verified the accuracy of the model.

Improvement to sensitivity by the LDCs allowing wider bandpasses will be further extended by a method called wavelength bootstrapping, a technique of simultaneously tracking in wavelengths where the fringe contrast is high and observing were the fringe contrast is low allowing for longer integration times. This is conducive to measuring the angular diameters and limb darkening of stars with high precision. Of particular interest are the cool, low-mass main sequences stars --- a severely under-sampled, yet the most numerous, population of stars. The CHARA Array, with the aid of LDCs to improve the visible wavelength capabilities, will measure the angular diameters of approximately a dozen stars cooler than our Sun within the solar neighborhood (d<10~ pc).

The author(s) of this abstract have provided an email address for comments about the abstract: berger@chara-array.org

[Previous] | [Session 24] | [Next]

Bulletin of the American Astronomical Society, 35#5
© 2003. The American Astronomical Soceity.