NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Meridian-anchored Noon Sun
From: Peter Monta
Date: 2018 Jul 9, 10:38 -0700
From: Peter Monta
Date: 2018 Jul 9, 10:38 -0700
Hi David,
I meant "stable" in the sense that the time sights would have low dependence on the choice of horizon point to the east or west. Unfortunately I think it really doesn't work. I should have thought it through.
The problem is that the plane normal to the observer-sun vector is always inclined to the horizon when looking east or west (unless the sun is on the prime vertical). So there is no tangency or desirable quadratic behavior of the error (as there is when looking north or south). Instead, the error in the time sight is linear in the error of the horizon-point azimuth. With the sun a few degrees from the zenith and with an azimuth error of a few degrees, this results in a sight error of a few arcminutes. Not good enough.
Cheers,
I meant "stable" in the sense that the time sights would have low dependence on the choice of horizon point to the east or west. Unfortunately I think it really doesn't work. I should have thought it through.
The problem is that the plane normal to the observer-sun vector is always inclined to the horizon when looking east or west (unless the sun is on the prime vertical). So there is no tangency or desirable quadratic behavior of the error (as there is when looking north or south). Instead, the error in the time sight is linear in the error of the horizon-point azimuth. With the sun a few degrees from the zenith and with an azimuth error of a few degrees, this results in a sight error of a few arcminutes. Not good enough.
Cheers,
Peter
