You said ....
"Traditionally, most navigators have considered full dark-adaptation a
pre-requisite for star sights in order to see the horizon clearly, but
that may work against star-star sights."
Sorry, I must disagree. Traditionally, by which I mean from at least the 1930s onward, navigators sought to take star sights for position determination purposes as early into evening twilight and as late into morning twilight as possible while there was a distinct horizon visible. This practice was facilitated by the precomputation of altitude + azimuth,
which, in conjunction with a suitable telescope, permitted observation of 1st and 2nd magnitude stars, as well as the planets, before they became visible to the naked eye.
As I have posted before, this method produces the most accurate results as respects position based on the sea horizon. Assuming no cloud interference, it was and is quite possible to shoot a full round of from four to six stars before loosing a clear horizon. As strange as it sounds, I have yet to find this practice clearly spelled out in most standard texts - it's
something that is learned at sea.
--- On Fri, 7/3/09, frankreed@HistoricalAtlas.com <frankreed@HistoricalAtlas.com> wrote:
From: frankreed@HistoricalAtlas.com <frankreed@HistoricalAtlas.com>
Subject: [NavList 8933] Re: Star-star distances for arc error
Date: Friday, July 3, 2009, 8:22 PM
Bill Morris, you wrote:
"What you say about his lack of data is fair enough, but isn't estimating index error by using a star the simplest case of a star-star sight? If one accepts this, may we then accept that sights using two stars are likely to have similar errors?"
For short distances, less than five or ten degrees, sure, of course, they're similar, and identical at zero degrees, but there's a substantial difference
with larger angles. You 'swing the arc' (as George also has noted in a couple of recent messages) when you bring the two star images together, and it's much easier to see the alignment in those cases.
"In my case, using a SNO-T sextant with x6 telescope clamped
atop a theodolite tripod the standard deviation of index error observations
using a star was 0.17 minutes (n = 30). "
Sounds quite reasonable! Just about what I would expect with a SNO-T. Have you tried this unclamped? Also, what do you find when you average sets of four? Does your s.d. decrease by a factor of two? And may I ask, how are you "clearing" these observations (just out of curiosity)?
"I am not a fan of Galilean 'scopes and Gordon seems to have been particularly
unlucky with his if he found "...a star image will superimpose over a range
of 5'." "
Yeah, that's an amazingly bad result (Gordon's 5' range, that is). It
could either be an unusually bad telescope or possibly that issue of dark adaptation that Bill B. and I have experimented with. It turns out that many people see bright stars as distorted, flared "blobs" when their eyes have dark-adapted but the images are much sharper when we first step into the dark. Traditionally, most navigators have considered full dark-adaptation a pre-requisite for star sights in order to see the horizon clearly, but that may work against star-star sights. Also, Gordon's daylight observations of altitudes are much better than this 5' error range he finds with stars. Something is clearly amiss.
I don't think we should worry too much about Gordon's results way back in the early 1960s. He was at that time a young, newly-graduated engineering student with a passion for boating and an enthusiasm for celestial navigation. His results are what they are: the observations and ruminations of one individual using (mostly) one
sextant. There are some nice points in his paper, but also some very weak points.
Navigation List archive: www.fer3.com/arc
To post, email NavList@fer3.com
To unsubscribe, email NavListemail@example.com
What is NavList?
You can also join by posting. Your first on-topic post automatically makes you a member.
Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.