mconnelley
sage
Reged: 03/14/06
Loc: Hilo, HI
|
I suppose you'd call it a Schmidt-Gregorian
#2916708 - 02/09/09 12:48 AM Attachment (278 downloads)
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Hello:
I was playing around with some optical ray tracing software (sort of hard to find for us mac folks), trying out an idea I had. Suppose you want a really big scope, but don't want to aspherize that really big mirror. The idea is to make a Gregorian design, and add a Schmidt corrector plate to fix the spherical aberration. The important detail is to put the corrector plate after the secondary mirror, at the image of the primary mirror made by the concave secondary. This optically conjugates to corrector onto the primary mirror, so you can put any figure you want "onto" the primary mirror my putting it on the corrector plate, which is a bit smaller than the secondary mirror. It also so happens that the corrector is very close to the primary mirror's focal plane, so you need to put a small (1" or so) hole in the corrector to let that light through. I tried this out with an aplanatic Gregorian design (where the conical constants are chosen to eliminate coma), and it looks like it worked. The test case was a 24" f/11.6 (sort of chosen at random), and the PSF has a diameter of 2" about 1/4 degree from the center of the field. I don't think you can do this trick with a regular Cassegrain since there is no real image of the pupil.
Cheers
Mike Connelley
|
Mike I. Jones
Post Laureate
Reged: 07/02/06
Loc: Fort Worth TX
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: mconnelley]
#2916752 - 02/09/09 01:35 AM Attachment (233 downloads)
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
I'm getting a lot of spherochromatism that's keeping the blur circles above DL. This is a 24" with 100" ROC primary and f/10.5 overall.
Maybe a slower primary or an achromatic Schmidt plate pair?
Mike
|
Mike I. Jones
Post Laureate
Reged: 07/02/06
Loc: Fort Worth TX
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Mike I. Jones]
#2916768 - 02/09/09 01:59 AM Attachment (230 downloads)
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Yep - the achromatic plate pair worked. N-BK7/N-LAK14 glass, primary ROC=100", f/10.5 overall, 1.5" field, 0.4-0.85um spectrum. Putting a corrector, whether static or dynamic, at an intermediate pupil image works well - it's the basic principle of adaptive optics. But taming the color over a wide spectral band and useful FOV can sometimes be challenging.
Mike
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: mconnelley]
#2918700 - 02/09/09 10:48 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
That is actually viable alternative to parabolizing fast primary. For more common amateur aperture size - 400mm diameter - and f/3/12.9 system, spherochromatism at 400nm (h-line) is 0.12 wave RMS (0.707 neutral zone; with 0.866 neutral zone the blur is twice smaller, but the RMS wavefront error is more than twice greater, and that's what counts). That is almost three times smaller error than in a commercial aplanatic SCT of identical aperture. At 0.25-degree off-axis the error is only ~15% greater than in a comparable aplanatic Gregorian (astigmatism in both).
I don't know how involved is fabricating 100mm Schmidt corrector, but should be easier than parabolizing 400mm f/3 mirror.
Vlad
|
Owen
professor emeritus
Reged: 06/21/07
Loc: New Zealand
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: wh48gs]
#2919073 - 02/10/09 07:32 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
My initial thoughts and questions:-
1 The secondary must be an elipsoid figure to in effect have two focal points?
The use of the ellipsoid will enable a corrector to be used
2 Does this need to be the case?
3/ if a spherical secondary was used - the 'correctors' could be a simpl(ish) APO design?
4/ this strikes me as ideal for use with a Coude focus.
The number of mirrors in the chain will push precision requirements up, but if we can keep them simple, the higher the possibility that will be easier to achieve...
Thoughts..
Owen
|
Mike I. Jones
Post Laureate
Reged: 07/02/06
Loc: Fort Worth TX
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Owen]
#2919248 - 02/10/09 09:56 AM Attachment (244 downloads)
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
This is a prescription for a 24" f/11 Schmidt Gregorian I came up with following Mike's initial configuration. The primary radius is 100", and both the primary and secondary are spherical. The plates are 6.5" in diameter with a 1.5" hole through the middle. Spectral band is 0.4-0.85um and the corrected field is 1.4" in diameter, well suited for both visual use and CCD work up to 1"x1" format.
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Mike I. Jones]
#2919848 - 02/10/09 03:30 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Mike,
Great correction of spherochromatism. But making those aspheres would be a nightmare.
Vlad
|
Mike I. Jones
Post Laureate
Reged: 07/02/06
Loc: Fort Worth TX
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: wh48gs]
#2919919 - 02/10/09 04:01 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
You got that right. Those aspheres would have to be ground in with a flexible tile lap. Lengthening the primary ROC and going to a higher index glass like SF11 for the plates would help soften down the figuring strength of the plates. But the amazing thing is that both the primary and secondary are spheres. You don't see that very often.
I let the conic constant for the secondary float, and it only changed to about -0.03, so it definitely wants to stay spherical.
Put it up on your most-cool telescope optics site?
Mike
|
Gary Fuchs
Pooh-Bah
Reged: 05/22/06
Loc: Easton, PA, USA
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Mike I. Jones]
#2919947 - 02/10/09 04:15 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
flexible tile lap
Mike (or others) - That's not the only part I didn't understand...could you explain them?
Thanks,
Gary
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Mike I. Jones]
#2920380 - 02/10/09 08:22 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Mike,
Thanks, but I doubt my site is cool enough for this system. Anyone that goes under "amateur telescope maker" wouldn't have much of a chance to figure those aspherics properly. But it is definitely exemplary in illustrating unusual and effective way of combining mirrors and lenses.
I was curious what happens in a regular Schmidt-Gregorian arrangement -this same configuration - with full aperture corrector. With both mirrors spherical, there is some residual coma, but significantly less than in a comparable SCT. Better yet, it can be further reduced by moving corrector closer to the primary (not much past primary focus it goes to zero). But spherochromatism is, as you noticed, significant. Not that it makes a system useless - it is at the level of 4" f/20+ achromat - but it could be better. The possibility is to place both, full aperture Schmidt and sub-aperture Schmidt at about the primary focus, and combining the two significantly reduce spherochromatism. Not that 24" Schmidt corrector is easy, but if such arrangement would work, it could be within the ATM reach.
Vlad
|
Owen
professor emeritus
Reged: 06/21/07
Loc: New Zealand
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: wh48gs]
#2920873 - 02/11/09 04:41 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
A final question - as primary and secondary are spheres - surely this could be a symetrical 'tilted' design too (back to thinking around the corrector plate holes...)..?
Owen
|
George Kiger
member
Reged: 05/19/06
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Owen]
#2921293 - 02/11/09 11:18 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Two questions:
1. Is spherochromatism in this design worse than in a Schmidt-Cass? If so, why?
2. Can the spherochromatism be controlled by putting the corrector closer to the final focus?
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: wh48gs]
#2921697 - 02/11/09 03:14 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
The possibility is to place both, full aperture Schmidt and sub-aperture Schmidt at about the primary focus, and combining the two significantly reduce spherochromatism. Not that 24" Schmidt corrector is easy, but if such arrangement would work, it could be within the ATM reach.
Wouldn't make much difference fabrication-wise. The correctors would be weaker than if they are both sub-aperture, but they would be both significantly stronger than non-achromatized corrector, and the full-aperture one is much larger too.
Most people seem to be reluctant to try to make even a single regular Schmidt corrector. The main reason is probably the lack of the how-to information, but the plates are objectivelly more difficult to make than to parabolize same size mirror. If we compare the easiest to make plate - with 0.707 neutral zone - with the easiest parabolizing technic (deepening the middle, while flattening the outer zones), maximum depth of the Schmidt corrector (at the neutral zone) is about four times greater than needed deepening of the center with parabolizing. Making Schmidt corrector is harder than parabolizing; the only reason it's done is to get rid of coma.
Sub-aperture Schmidt in a single-plate arrangement has similar depth as full-aperture corrector, and should be easier to make. The amount of spherochromatism is similar for both.
When achromatising Schmidt corrector by adding an opposite in power element, both correctors need to be much stronger than a single regular type. The stronger plate in Mike's system is about 0.8mm deep, and the weaker one about 0.5mm (that compares to ~0.08mm for a single plate, either full- or sub-aperture). The depth itself is not a problem; but required surface accuracy is also exponentially higher.
As usual, no free lunch. There is a price to pay for leaving mirrors spherical. Since spherochromatism changes inversely to the third power of mirror radius, going with f/2.5 primary would cut it in half. That is nearly 20% less than in an f/2/10 SCT w/spherical mirrors. Most people can live with that much chromatism. For those who can't - and won't be stopped by anything - Mike has found near perfect solution.
Vlad
Edited by wh48gs (02/11/09 03:31 PM)
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: Owen]
#2921707 - 02/11/09 03:23 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
A final question - as primary and secondary are spheres - surely this could be a symetrical 'tilted' design too (back to thinking around the corrector plate holes...)..?
Yes, with off-axis corrector segment. Now, that would be an interesting piece of work, to say the least.
Vlad
|
wh48gs
Pooh-Bah
Reged: 03/02/07
|
Re: I suppose you'd call it a Schmidt-Gregorian
[Re: George Kiger]
#2921718 - 02/11/09 03:28 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
1. Is spherochromatism in this design worse than in a Schmidt-Cass? If so, why?
2. Can the spherochromatism be controlled by putting the corrector closer to the final focus?
It is nearly twice as bad, because the sign of spherical aberration is identical for both mirrors in the Gregorian, when left spherical. So they add up, instead of partly offseting.
Schmidt corrector location has no significant effect on the corrector's surface profile, thus neither on its spherochromatism.
Vlad
|
Previous
Index
Next
Threaded
