|
|
From: Frank Reed (no email)
Date: Mon Nov 08 2004 - 01:57:30 EST
Alex E asked:
"Which one in Perseus? Beta, Epsilon?"
Algol = Beta Persei
And:
"What does it mean "eclipsing binary star"? Sometimes you see one star,
sometimes two?"
Yes, but they're so close together that they cannot be resolved in ordinary
telescopes. All you see is a cyclic drop in brightness. As one star passes in
front of the other, even if it's smaller, the net light from the system sent
our way decreases. The star is like a light house, blinking at us every three
days. Algol is the prototype of the eclipsing binary. Some eclipsing binaries
visible in small telescopes have considerably sharper eclipses that would allow
an observer to get UT (conceivably) within five minutes or better.
And:
"How long are the eclipses, and what is the exact period (you only give to m
inutes, this is not enough to predict for a whole year. One needs to seconds at
least)."
The rules I gave would allow you to predict eclipses to the nearest couple of
minutes for some years into the future (the rule that you subtract another
minute every seven weeks amounts to giving the orbital period accurate to about
one part in 100,000). The eclipses last something like ten hours, if memory
serves. A visual observer can get the time of minimum accurate to about half an
hour without instruments. Not awful for finding longitude, but not much better
than awful. Again, before some tells me this is totally impractical, yes, I
know. This is in the spirit of a "game".
By the way, the seasonal rule, +/- 7 minutes in certain months, arises from
the finite speed of light. The Earth is something like 250 million kilometers
further from Algol around May 20th than it is on November 20th.
And:
"If these eclipses are seen with a naked eye, they should be known since
antiquity."
Perhaps surprisingly, it is unlikely that Algol's highly predictable eclipses
were known before 1782 though its variability was discovered earlier. Here's
an article I found that discusses some of the history and the astronomy, too:
http://www.surveyor.in-berlin.de/himmel/Bios/Goodricke-e.html
And asked:
"Has anyone ever proposed to use it for longitude?"
Probably not in any formal sense! Eclipsing binaries were discovered a few
decades too late for the early historical discussions of longitude (fully 15
years after the publication of the lunar distance tables in the Nautical
Almanac). I would imagine the concept has occurred to many people, but I doubt there's
ever been any reason to put it in print. Incidentally, some years ago, there
was a proposal briefly to re-define time standards in terms of pulsar
observations. This would be roughly equivalent to using variable stars like Algol to
get an absolute time standard, but of an entirely different level of accuracy
and requiring the highest end astronomical equipment for observations.
Frank R
[ ] Mystic, Connecticut
[X] Chicago, Illinois
|