Re: [atm_free] Re: Schmidt corrector fabrication
- I would finish the corrector before putting the hole in it. In my experience, the hole doesn't affect the figure at all, nor does the slight bending from the weight of the secondary.The axial corrector position is typically not very critical. You should have +/- 1/2 cm to work with, but check it with a ray trace.DaveIn a message dated 7/3/2013 12:54:10 P.M. Pacific Daylight Time, acklin@... writes:
I ordered a bunch of books today- thanks for the name and lead, Mel.
If I built a spider and then supported the corrector just in front with it's own ring possibly- or padded from the vanes, then I guess I could polish, test, repeat using your autocollimation method or an artificial or real star- and I remember reading somewhere that the corrector position in the system is not super critical.
My mounting method could be then refined to attach the secondary to the corrector by making a perforation in the disk, and getting the spacing from secondary to the primary the same. What I was having trouble with is the concept polishing after that 2" hole is drilled (no way to use the vac pan)
I doubt that an inch or two would affect the corrector to a huge extent (?)
--- In email@example.com, aplanatic@... wrote:
> Hi Larry,
> I've made two Schmidt cameras and two Slevogt (concentric Schmidt
> Cassegrain) cameras ranging from 9.25" to 12" in clear aperture. In each case I
> tested the Schmidt plate in the actual system using a light source at focus,
> and inspecting the image with an auxiliary test telescope. For the Schmidt
> cameras the light source was placed at the focus of the camera and the image
> was inspected using a high-quality 10" f/6 Newtonian reflector. The two
> Slevogt cameras were tested the same way, but with the light source at the
> Cassegrain focus of the camera. The smaller Slevogt (CSC) was tested using the
> aforementioned 10" f/6 Newtonian and the larger CSC was tested using a 16"
> f/5 Newtonian that I borrowed for the purpose.
> The light source for the Schmidt cameras was a difficult-to-align
> arrangement of a laser shining on a ball bearing, all carefully centered in the
> camera housing. For the CSCs, the light source was a spherical wave beam
> expander and a laser, comprised of a microscope objective and a 6 micron
> pinhole. The pinhole was impossible to align until I made a simple fine-adjustment
> stage for it.
> Focus is very important when testing fast optics. To focus the light
> source I first focused the test telescope at infinity under the stars, I then
> brought the test telescope inside, set it up and aligned it, and moved the
> light source axially until focus was achieved at the locked focus position of
> the test telescope. If this sounds a bit complicated, it is, until you've
> done it a couple of times. It becomes easy with a bit of practice.
> At the focus of the test telescope I used a knife edge, various Ronchi
> gratings, and the "star test," i.e., simple inspection of the extra-focal
> images. One could also employ a Hartman mask if desired. In the end, I mostly
> used the Foucault test to determine the amount of spherical correction I
> needed to add or subtract to the corrector plate, and I used the star test
> extensively to estimate astigmatism and other non-symmetrical aberrations.
> (There were lots of problems in the first plate that I made until I understood
> how to use the vacuum pull-down method in detail.) Of course, these tests
> are all null tests, making the analysis and inspection particularly simple.
> Yes, it's a lot of work to test a Schmidt plate by this method, but because
> you're testing it in the finished telescope you can achieve nearly perfect
> correction. Of course, one must have a test telescope of known high
> quality. I choose this method because I could use it during the day, and on
> cloudy nights, of which there are many at my house.
> Another way, at least for a Cassegrain, is to make, borrow, or buy a good
> flat and test the finished telescope in autocollimation.
> Finally, one could just use the real star test. Estimate the amount of
> over- or under-correction, and then estimate and modify the amount of vacuum
> pull-down needed for perfect correction. If you've got a lot of clear
> nights, this surely is the easiest of all. It's interesting to note that the
> Foucault test works just as well using a real star as it does with a pinhole
> light source. To get an analytical result you'll need a dial indicator on
> your focuser. Then one would just use the "fixed light source" option in
> FigureXP (for instance) and you'll get an estimate of the waveform error through
> the fictitious "surface height error" of the fictitious mirror. By knowing
> the overall correction, you can easily estimate the fractional error in the
> corrector plate. This fractional error in spherical correction is all you
> need to adjust the vacuum pull down.
> Dave Rowe
> In a message dated 7/2/2013 8:06:43 P.M. Pacific Daylight Time,
> acklin@... writes:
> Hi all- I have found and used Dave Rowe's vacpan.xls and understand the
> process of pulling a vaccum then "polishing to a sphere". By question is how
> to test? The roc of the tool to grind/polish would be 1200 inches(!) a bit
> long to use a ronchi test. Or do i just polish until i get sick of it and
> figure thats good enough? 1/4 inch optical glass, 9 inch dia. 7.65 inch
> primary f10 scope.