As the Oliver Observing Station was taking form on Chews Ridge at the beginning of the 1980s, a broad revolution was taking place in astronomy.
For about 100 years, astronomers had depended on photography for both images of the sky and for recording spectra. Photography had been a revolution of its own. Replacing the human eye and drawings, photography permitted astronomers to collect light for hours, recording objects too faint to be seen with the eye. The photographs, called 'plates' by astronomers because most emulsions were deposited on glass for stability, also provided a permanent record of the observation.
Spectra of two stars recorded with a photographic plate. The vertical dark lines in the top star are due to hydrogen atoms.
In the 1980s, however, astronomers around the world were shifting from the traditional photographic plate to electronic detectors. These new detectors, based on the same materials that are used for the transistors in modern electronics, were a hundred to a thousand times more sensitive than photographic emulsions and produced results that were much more accurate than could be obtained with photographs.
Unfortunately, this technology was so new that, in order to reduce the electronic "static" or noise in the measurements, astronomers and observatory engineers had to develop new electronic circuits to take full advantage of the wonderful potential of these detectors. For MIRA, with no staff of electronics engineers, this meant enlisting the help of Friend Gary Love, who spent hundreds of hours working with us as we tried to develop a tiny array of 512 electronic sensors, each 1/1000 of an inch wide, into a usable astronomical sensor. The Reticon Corporation had donated the sensor and the Hoover and Hewlett Foundations had funded the development.
Even as the observatory was still being completed, I was using software Bob Bogardus had quickly written to record the data on cassette tapes to make the first measurements of stellar spectra on Chews Ridge.
The Reticon chip sits atop the thin-wall truss designed by Dr. Frank Melsheimer to keep the sensor at liquid nitrogen temperatures inside its vacuum fixture. The Hewlett-Packard calculator provides scale and acknowledges the support of H-P's founders in developing the detector system.
Friend of MIRA Gary Love adjusts electronic test equipment before starting a night of testing the Reticon.
Supporting equipment changed over the years (e.g., an early PC was pressed into service to provide easier control, floppy disk storage, and immediate spectral display) but the Reticon, safely ensconced in its vacuum dewar, remained firmly attached to the telescope until this spring.
One of the first calibration spectra: a 15-second exposure of Vega in the near infrared.
During the more than two decades it was on the telescope, the Reticon was used by a score of astronomers to take thousands of spectra of comets, planets, stars, nebulae, and galaxies. Research projects spanned young stars and old, hot stars and cool, and unusual objects never measured before. Research done with this detector has been published in every major astronomical journal in the world.
One of over a thousand fully-calibrated spectra observed with the Reticon for the Torres-Dodgen & Weaver Near Infrared Stellar Classification System. The large dips are due to the many molecular absorption bands in this cool supergiant star.
Although the Reticon sensor is as sensitive as any CCD sensor of today, the design is intrinsically electronically noisy. An uncertainly of over 300 electrons is included in each measurement compared to 5 to 10 with a modern CCD. Also, modern CCDs have become so ubiquitous that excellent products can be purchased for about $15,000, saving years of electronics work in the MIRA shops. Reluctantly, I took the Reticon system off the telescope for the last time this spring. Its place on the spectrograph was taken by a modern CCD system that will let us observe stars nearly ten times farther away than was possible with Reticon. The new CCD also has twice as many sensors-over a thousand-so that twice as much of the spectrum can be observed at one time.
Although it will take many months to fully understand how to properly calibrate the new sensors, initial measurements have been very exciting. But I can't help but feel a little sad about leaving an old friend, albeit a bit cantankerous lately, with whom I shared hundreds of nights under the spectacular skies of Chews Ridge.
| Fall 2002 Contents | Newsletter Index | Mira Home Page |
Last updated 12/15/02 DMC