NavList:

Alex, you asked:

"Why did not you use the Sun?"

 

Because I get excellent results with the Moon, and it eliminates any issue of shade error. But I repeat, if you want real verification, use a table-top measurement of index error. If you have a large enough work area (a couple of hundred feet) and a decent laser, it should be infallible.

And you wrote:
"Adding and averaging gives 64.275.
While according to the Almanac, true Moon SD
for October 29 was 15.8, which gives 4DS=63.2.
So you are off by the whole minute.

So much of the Lunar IC correction.
Or do I confuse something?"

 

Actually, the correct horizon SD at the time of the observation was 15.9 (calculated from 27.27% of the HP --you should always use a calculated SD when doing lunars). Then you need to augment it for altitude which gives 16.0 minutes of arc. The result is then quite a bit closer, but yes, a little off --an error of 0.14 minutes of arc in the Moon's diameter.

Based on your similar observations you wrote:

"Now I derive from my data IC=0.7 which also seems absurd."

 

I checked my notes, and that was the IC of that sextant last time I checked. Then again, you've adjusted the sextant a bit so don't count on it.

You concluded:

"So I am very skeptical about using the Moon for IC correction."

 

Keep at it, Alex! In my experience, I know of no better simple observation than this one. Also, ask yourself this question: how could it possibly go wrong?! This is one of the simplest observations you can make with a sextant. If it doesn't work, what on Earth could explain the error??

 

By the way, for those on the list who are not crazy lunarians (not yet crazy lunarians, I should say <g>), there is, of course, nothing wrong with the tried-and-true method for getting index correction by lining up the direct and reflected images of the sea horizon. The reading on the micrometer is the "index error" and the negative of that number is the "index correction". Equivalently there's the navigator's mnemonic: "if it's off, it's on; if it's on, it's off". More explicitly, if the reading on the sextant is above zero when the direct and reflected horizons are lined up, then that value should be subtracted from every observation made with the instrument; if it's below zero, that value should be added to every observation. This horizon test will yield an index correction accurate to perhaps 0.25 to 0.5 minutes of arc --plenty good enough for standard celestial navigation, but not quite up to the task for lunars.

 

For the experts on the list, I'm spelling things out here because there are, in fact, some navigation 'newbies' on the list, and I do hope they'll find something of value in our occasionally obsessive/compulsive discussions of tiny errors.