NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Index mirror adjustment with offset mirror
From: Ken Gebhart
Date: 2014 Jan 15, 16:44 -0600
From: Ken Gebhart
Date: 2014 Jan 15, 16:44 -0600
The way I explain it thus: with the silvering of the mirror offset from the axis of rotation (usually a front surface mirror) the image of the arc and reflected image of the arc changes with the eye position. The only correct view is when the eye is level with the plane of the arc. But then you can see neither the arc nor its reflected image. Using objects with a flat top simply raises the plane so that you can see both objects, and their flat tops cue you to the correct eye position. The devices supplied with the SNO-T do not have flat tops and are inferior because the viewer must calculate the correct eye height on his own. With mirrors not offset (usually a rear surface mirror), the arc and reflected image do not change with eye variation, and no devices are needed. So, the first thing one should do is check for changes with eye variation. If so, then go get some flat top devices.
Ken
On Jan 15, 2014, at 2:48 PM, Luc Van den Borre wrote:
On 15/01/2014 20:48, Frank Reed wrote: > Yes, and as I mentioned in a post last night, a really useful trick is > to use objects (could still be dice) that are different colors. Marker > caps, legos, and other small toys work great for this. LEGO bricks, that's a great idea! They have a nice sharp edge (so the top face is clearly defined) and I hear they have very precise dimensions. > You also wrote: > "* It's not correct to use the top _edge_ of the dice. One must be > looking along the top plane." > > I'm not sure I understand the distinction you're drawing here. You sight > in line with the top edge of the two objects. Do you mean that one > should not be sighting from "above" trying to get the top faces of the > two dice to appear aligned? If that's what you meant, then yes, that's > an important distinction. The goal is to get the two top edges to align > in a straight line when viewed from exactly the height of the dice above > the arc. Yes, I meant 'edge' as the line segment for the intersection between 2 faces of the dice. Rather than 'top plane' I should have said 'top face'. This is one of the things I did wrong - I was matching up the edges without checking the height from which I was looking. > Yeah, I don't understand those peculiar shapes. They may have been > modeled on a bit of testing gear created for the optics lab. Clearly > simpler objects work just fine. I found several posts in the archive, made by Alexandre Eremenko in 2004 about this same situation, e.g. http://fer3.com/arc/m2.aspx/Adjusting-perpedicularity-Index-Mirror-Eremenko-oct-2004-w18186 http://fer3.com/arc/m2.aspx/Perpendicularity-SNOT-Eremenko-oct-2004-w18298 Since the SNO-T has a very offset index mirror, it is great that they put the correct procedure in the manual, but sad that they then shipped with these 'diopters' that don't have a top surface. Strange. Today I have mocked up a sextant arc and an index mirror in POV-Ray, a ray tracer, and it has confirmed what I described in my earlier post. It's a really fun tool to play with and surprisingly easy to use (the objects are defined in a text file with a simple syntax). I can introduce errors in alignment and increase the mirror offset and see the result on the reflection. > By the way, it's far better to call that Russian sextant a "SNO-T". Will do. I copied it from the PDF posted earlier today. I guess the importer preferred a less snotty sounding acronym for this noble instrument. Thank you very much for your reply, Luc
