A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Brad Morris
Date: 2016 Jul 19, 22:20 -0400
The real trick would be mounting the scale concentric to the index arm axis of rotation and providing an appropriate mounting surface for the readerhead such that it was within tolerance for mounting. That would require a competent machine shop. The cost of that would exceed the encoder price. Not to mention the mechanical engineering required to come up with the set of prints.
Francis asked how to make it better, so here is the next idea. Put the sight reduction right in the electronics. That is, why not come up with the numerical lat long given the observations? Take a few observations, click, click, click and out pops your lat long. Why bother with the fussy sight reduction tables, extensive forms, error prone arithmetic and the interpretation of the cocked hat.
Don SeltzerI toyed around with this idea a few years ago. Not for making a high resolution sextant, but to come up with something very easy to read (big digital display) with time-coded recordings of the readings.I believe that a 15 bit encoder was necessary to achieve 1 arc minute resolution, and the cost of such encoders that I looked at was a bit too high, even using cheaper incremental encoders.On Tue, Jul 19, 2016 at 3:23 PM, Brad Morris <NoReply_Morris@fer3.com> wrote:
The most obvious improvement would be to replace the sextant arc with a rotary encoder.
An encoder consists of two parts. A readerhead and a scale. As one rotates past the other, a measurement of angle is produced. There are incremental (not directly suitable) and absolute encoders. The absolute encoder "knows" where it it at the moment it is powered on with out the need to reference the device.
The index arm would rotate the scale, while the reader head would remain fixed to the body of the sextant.
On a recent device, I installed a rotary encoder from Heidenhain which has ~134,000,000 counts per revolution. A count is the smallest integral unit of measure. Therefore the resolution of that high quality encoder is ~0.00967 arc SECONDS per count, or very roughly one one hundredth of an arc second.
While this encoder may be a bit pricey for some, lower resolution encoders are available.
Press a button and get your reading. No more gear lash or micrometer nonsense. The reading is directly off of the index arm, similar to vernier arc sextants.
Now add a little bit of memory and a stop watch feature to the electronics, and you can get time and altitude at the touch of a button, recorded into memory. Viola, you are ready for the next observation