NavList:

Marty Lyons, you wrote:
"The entire unit is much lighter thean a sextant which makes holding it steady difficult. This level could never be used on a rocking boat. I have clamped it to a ball head on a tripod for steadiness, just to see if the procedure is even doable for fixes."

I think what you need is "moment of inertia". If you can attach this level to something long and light with a weightt at the far end, it will be much more stable directionally. For example, get a broom handle (or something with similar dimensions and weight. Lash your level to one end. On the other end attach a counter-weight so that the thing balances close to the level. Attach some padding and you can rest it on your shoulder. Because that arrangement has good moment of inertia, it is rotationally stable -- slow to react to torques whether from shaky hands, accelerations of a boat's deck, or a gust of wind. Or, if the tripod attachment that you mentioned worked reasonably well, do that again, but hold the tripod up off the ground (and legs together but extended to full length). Maybe tie a weight to the bottom of the tripod. This is basically the same as the broom handle but vertical instead of horizontal, also probably harder to hold since you can't rest it on your shoulder. But it would provide similar moment of inertia and thus stability. By the way, backyard astronomers have been making rigs for higher-power binoculars along these lines for decades. It's a good trick!

You also wrote:
"Similar to not having to worry about dip with a bubble horizon sextant or aircraft sextant, I would not use dip in any calcualtions. I do think , however, that if using a natural horizon to establish the index error of the instrument, I would have to figure dip in."

Yes. But you don't necessarily need the horizon for this. What I was suggesting previously is that you can just shoot stars as normal. Measure a dozen altitudes. Compare with predicted Hs values (like from my app set for bubble sights). The errors you find should show a consistent pattern. Average those, and that's your index error. You may find it easier to work with stars that have slowly changing altitudes. This week Polaris, Kochab, and Zubenelgenubi are all close to the meridian in early evening twilight. And tomorrow the Moon is close to the meridian in evening twilight for many of us (not everyone -- depends on longitude). But of course you don't have to do this in twilight, so you can wait later into the night to find brighter stars on the meridian.

What's the angular range on this level? Can you measure up to 90° high? You may also want to test to see if the error in the altitudes is itself altitude-dependent. On a marine sextant, this would indicate arc error in addition to index error. But in an instrument where index error is locked or unadjustable, there's really no difference. It's just different errors at different altitudes. How to assess that? Same as above, but limit yourself to sets of stars in certain altitude ranges. For example, shoot a few Venus and Moon altitudes when low near the horizon (Venus and Moon because they're visible right down to the true horizon --90° from the zenith). Those sights would give you the error near 0°. Then shoot a dozen sights close to 15° high (maybe 12°-17° for a range). What common error do they display? Repeat at 30° and 45°. And so on, every 15° up to 90°. Do you find a different correction at higher altitudes? I would be very interested to hear what you find. :) 

Frank Reed