NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Bris Sextant
From: George Huxtable
Date: 2005 Nov 8, 09:56 -0000
From: George Huxtable
Date: 2005 Nov 8, 09:56 -0000
Frank Reed sent a posting yesterday, most of which I agree with, but not this bit- > Also, I notice with my mockup that the > reflected images are all doubled (presumably from the non-zero thickness > of the > glass). This effectively "blurs" the reflected image though with careful > observation it should be possible to deal with this. Alex got it right in responding- >Or from prismatic effect. The ray is reflected from BOTH surfaces >of the glass pane outer and inner surface, so if these surfaces are >not perfectly parallel you will see a doubling. ======================= As long as the two surfaces of each glass are parallel, there should be no doubling of the image due to reflections from its fore and aft surfaces. Those reflections should coincide exactly. However, if it's low-quality glass, with some tapering of its thickness, that would mean the two surfaces were non-parallel, and you would see a doubling, exactly as Alex suggests. That would show up if you just took a single piece of that glass, and looked at the Sun reflected in it. It's the taper, not the non-zero thickness, that causes the doubling. But there are other possibilities with a Bris device, depending on the combination of subtended angles that were chosen. It the two gaps were chosen with a nominally-equal angle of parting, then small differences between the actual angles would show up as such a doubling. Similarly, if one angle is nominally (but not precisely) set to be twice the other, then light rays undergoing a double-bounce between one pair would almost coincide with rays making a single-bounce between the other. No doubt there are many such possibilities; which implies that certain angle-combinations (including equal, and also 2:1) are perhaps best avoided, because otherwise confusion will result. That must mean avoiding angle-pairs in which n x A1 equals, or nearly equals, m x A2, where n and m are small integers and A1q and A2 are the subtended angles. In musical terms, that's calling for a discord rather than a chord. What were the nominal subtended angles, when Frank saw that doubling? ================================== On the question of collimation error, and the necessary rocking to minimise it, the three of us (Alex, Frank, George) now seem to be in accord, it's pleasing to note. George. contact George Huxtable at george@huxtable.u-net.com or at +44 1865 820222 (from UK, 01865 820222) or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.