Need help dechipering "telescope design" performance

[vlaiv]

This post is about "new" telescope design rather than one of exiting ones (so mods, please move topic if needed).

Actually, I'm not sure if it is indeed new design - I've not encountered it before and just came up with idea, but it might been long known - and known not to work at all

It is sort of catadioptric design, consisting out of both mirrors and lens elements.

There are two parabolic mirrors - primary and secondary, arranged either in Cassegrain or Gregorian configuration - of the same curvature, spaced exactly at sum of focal lengths.

This arrangement, I believe, will act to produce "denser" collimated beam (at least it will on-axis, not sure for off-axis). One just needs to add small doublet/triplet lens to convert it into telescope.

Here is diagram (showing Gregorian configuration):

It looks like mirror system is acting as a sort of light amplifying device. Lens element can be positioned on either side of primary (subject to field width, vignetting, baffling and other parameters).

Size of secondary can be adjusted, so we can have rather small central obstruction.

I'm trying to figure out answers to following questions:

- What will off axis performance of this system be (if we for example take lens element to be perfect lens element) - or rather what sort of aberrations can one expect off axis from amplifying part alone - I did not do any calculations, but it looks like beam will be collimated both on and off axis.

- What will be resolving power of such system provided that it works. Reasoning alone suggest that it will be order of magnitude provided by secondary / lens element rather than primary, but I can't be 100% sure on that one.

- Provided that system works - what would be field of application of such system. I'm inclined to say that it would be really good low resolution astrograph / low power telescope (due to possible issues with resolving power).

I also "envision" following features for such system:

- the worst design out there - it will feature all the elements people usually object with current designs - diffraction spikes, difficult collimation, chromatic aberration (this largely depends on lens element used, but I believe it can be well controlled because lens can be made out of ED glass within reasonable budget due to small size, it can also be rather slow - this would facilitate use of regular glass as well), low resolving power, it will simply be ugly to look at , (depending on lens element position and focal length)

- On the other hand there are bunch of benefits - "modular" design - depending on lens element used (and this can be as easy as unscrewing one lens and screwing in another) - change in focal length of system. Slow beam is easy on the eyepieces. Really short focal length with "massive" aperture - If we assume 8" primary and 2" secondary and add 50mm F/10 doublet, we will effectively have F/2.5 instrument - excellent nebulae / comet hunter type scope, as well as very fast astrograph. Due to lens element being F/10 I think that field curvature will be no issue at all - very well suited for latest generation CMOS sensors - small pixels.

I was thinking about modelling this in some software like Modas NG to get spot diagram / vignetting / psf, but being complete Noob with such software, I don't even know how to start ...

What would be your thoughts on this design?

 

[Peter Drew]

Try "Cloudy Nights"(if you dare!), there are some top optical designers on that forum who would be willing to advise. 

[Stub Mandrel]

5 hours ago, vlaiv said:

Size of secondary can be adjusted, so we can have rather small central obstruction.

Just a thought, secondary needs to completely shield the lens system, or you will get a two-scale image.

[vlaiv]

3 minutes ago, Stub Mandrel said:

Just a thought, secondary needs to completely shield the lens system, or you will get a two-scale image.

Yes, probably even a bit larger than lens - otherwise, vignetting would start almost from center (again not 100% on that, this is why I need a ray trace of sorts). It also depends on position of the lens - further away, bigger chance that angled rays will vignette.

I did not provide baffling in above diagram, but it would be similar to RC scope - so secondary would be "sunken" in shield, and lens would also be embedded inside tube to prevent stray light coming directly from sky.

[Astrobits]

I have been contemplating something similar for some years now but other things have just got in the way of tackling it seriously. My thoughts have centered around a Naysmith design where the primary is not perforated and a flat sends the focus out through the alt axis in an alt/az mounting. This is to allow a stable platform for, in particular, spectroscopy.

I have used the program "BEAM 3" which runs in DOS ( yes it is that old ) to get close to a workable design. I have more recently tried "WINLENS" but had a problem with it as it persisted in sending one ray off at 180 degrees to the correct path making nonsense of the output. This is a free/paid for  program. The free one is limited in number of surfaces and the data cannot be saved so you would have to use screeshots or pencil & paper. It seemed to be the Naysmith flat causing the problem for me so perhaps you will have a better experience if you try it.

As Peter says "Cloudy Nights" might be a useful source of info.

Nigel