|
|
|
|
|||||
|
||||||
Subject: Re: Sextant Accuracy
From: Fred Hebard (Fred@XXX.XXX)
Date: Tue Mar 18 2003 - 14:56:29 EST
George Huxtable wrote:
> Thanks to Fred for providing data of his land-based observations,
> which any
> sextant observer would be proud of. It would be interesting to compare
> those figures with what Fred could get out of shore-based
> observations up
> from a horizon, or those taken in real-life from a small boat.
>
>
> It's notable that much of the scatter in Fred's positions occurs in his
> altitudes of the Moon. Is Fred taking all the minor corrections for
> the
> Moon into account, such as augmentation of semidiameter, reduction of
> parallax, I wonder? Normally, one would only bother with such matters
> when
> seeking the ultimate precision of a lunar, but Fred is achieving such
> remarkably consistent answers that it may be worthwhile taking these
> minor
> matters into account.
First, thank you George for giving me a reason to go sailing! That may
be related to the ultimate purpose of my exercise...
The moon is definitely more "variable" than the other bodies in my
observations. Here is a summary (I know it's incorrect to average the
standard deviations, but it's accurate enough with these data to
demonstrate that the moon is more variable):
~~~~~~~~~~~~~~~~~~~~~Ho-Hc
Object~~~N~~~mean~of~~~~~~mean~of
~~~~~~~~~~~~~block~~~~~~~~block~std
~~~~~~~~~~~~~means~~~~~~~~deviations
Jupiter~~3~~~(-)0.53~~~~~~0.35
Moon~~~~10~~~(-)0.10~~~~~~0.66
Sirius~~~5~~~(-)0.63~~~~~~0.32
Sun~~~~~~6~~~(-)0.25~~~~~~0.30
Regarding moon errors, I am just using the standard Nautical Almanac
corrections, and keying data in from the Almanac, rather than formulas,
such as George outlined for Young's method of clearing lunar distances.
The Almanac's corrections include reduction in HP with altitude; I
can't remember offhand whether it still includes augmentation. I note
that the Almanac's corrections vary from those of the U.S. Naval
Observatory site.
I'm not sure why the moon is so untrustworthy for me, although when
it's good it's very good for use in a liquid artificial horizon. It
doesn't heat the liquid and the wind tends to be calmer at night,
allowing removal of the glass cover. It's also bright enough to see
easily. In the larger data set, when the standard deviation of the
moon shots is small, the position is pretty close to that of the other
objects.
A big part of an exercise as I'm doing is to train oneself to know when
a sight has been muffed, and to get some feel for good and not-so-good
sights, although this consistent deviation to the south is still
aggravating me.
I note in the data summary above that the sun is perhaps a bit better
behaved than the other bodies. With regard to Bruce Stark's mentioning
of irradiation, he says that his measurements of solar diameter tend to
come out on the large size. Mine come out on the small side, often by
about 0.1' of arc, which would move the sun's deviation closer to zero
in this case.
Another component of my consistent deviation may be deflection from the
vertical, as pointed out by Doug Stephen. We live in the middle of the
Appalachian Mountains, with the tallest mountains in Virginia about 10
miles away. Although not as massive as the Rockies, they could be an
important factor.
I welcome Doug's participation. Clearly he has been been very
seriously researching the location of his house! I wonder whether he
could share with us links to the programs for determining deflection
from the vertical. Some of his data comparing the Astra to the Wild
also would be a delight to see. I can well imagine that aacurate
timing would be critical for the Wild.
|