Jump to content


Photo

Refiguring a Cassegrain

  • This topic is locked This topic is locked
43 replies to this topic

#1 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 02:28 AM

I've been given a 16" Cass to evaluate and possibly refigure. I've done a very quick test of the primary and it has only about 30% correction toward a parabola. It's also got a rough surface and a TDE. Pics to follow. The owner believed it to be a Classical Cass. but it is evidently closer to a Dall Kirkham. I assume the secondaries (there are actually two of them, giving different focal ratios) are of similar quality. Startest actually shows the overall correction is reasonable but in focus star images are poor due to the TDE and surface roughness. It also shows a lot of coma.(Dall Kirkham?)
I've very little experience with polishing cass. optics and don't even have a spherometer so I have a question for the experts.
It seems the easiest option is to repolish the primary to fix the TDE and smoothen the surface but keep the overall correction as is. Recoat it and then repolish the secondaries using the startest to test or possibly test through the back. Use a comacorrector.
The other option is to refigure the primary to a parabola and refigure the secondaries to a hyperbola using startest. Is this feasible using the existing radii of curvature?
Here's the foucault.

Dave

Attached Files


 

#2 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 02:34 AM

Ronchi 133 lpi

Attached Files


 

#3 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 02:47 AM

Star at focus

Attached Files


 

#4 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 12 January 2011 - 07:05 AM

What's the focal ratio of the primary? And do you know what surface is on each of the secondary mirrors? I have heard that Dalls can be quite good, though the correction you seem to have doesn't match what I'm accustomed to seeing. Generally the correction of the Dall primary has to be in the area of 60-70% with a spherical secondary mirror.
Mark
 

#5 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 07:23 AM

The primary is 1230mm focal length so just over f/3. The 60-70% correction you mention is what I've been told is usual as well. I've read that the D K needs to be kept with a longer focal ratio to keep coma at bay. It's main advantage is it's easier to make. I have no idea what surface is on the secondaries at this stage. I'm confident I can improve the primary as there's really nothing to beat and I can take it to say 65% then polish the secondaries spherical. The owner wants to use it for imaging though (deep sky) and prefers the secondary giving the smaller f/ratio hence the bad coma.

Dave
 

#6 Mike I. Jones

Mike I. Jones

    Gemini

  • *****
  • Posts: 3239
  • Joined: 02 Jul 2006
  • Loc: Fort Worth TX

Posted 12 January 2011 - 09:16 AM

Nasty, zoney figure on the primary for sure. Put up all the Cass parameters for this system (amplification, radii, separation, back working distance, etc.) so we can determine the right amount of primary correction.

Of the RC, classical and DK mirror shapes, the DK has the most coma, but the flattest field and least sensitivity to misalignment. The RC has zero coma, but has the most curved field and strongest hyperboloidal secondary shape, and thus has very tight alignment tolerances. The paraboloid/hyperboloid classical Cass might be the best of both worlds for you. But let's play with some numbers here and see what's what.
Mike
 

#7 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 12 January 2011 - 04:54 PM

I'm sure Dave will fix that primary!!! Don't think he'd ever let one like that get by......
Mark
 

#8 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 06:35 PM

Ok, things are getting interesting.
Just did some measuring and have :-
primary 400mm dia. r/c 2460mm f/r 3.075
secondary 162mm dia r/c unknown at this moment.
secondary is 490mm inside primary focus
primary-secondary spacing 740mm
secondary to final focus 1035mm

Am I right that this means an amplification factor of 1035 ÷ 490 = 2.112 which makes the final f/r 6.49!!

I measured the primary central baffle and it is 80mm dia and extends 250mm in front of the primary surface. Looks like it might vignette to me.
I haven't swapped the secondaries at this stage but the other one is 140mm dia. and looks to have a spacing of 860 to the primary.
I will have access to a spherometer shortly so can measure the secondary r/c's in the next day or two.
Btw the 3 mirrors are plate glass and are currently silvered.

Dave
 

#9 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 06:46 PM

I'm sure Dave will fix that primary!!! Don't think he'd ever let one like that get by......
Mark


Yes Mark, I'm confident I can improve the primary, although I've never worked a mirror with a central hole before. Nothing to beat there. It's more the secondaries and the system as a whole. I'm sorting through to see if I want to take this on as it's new ground for me. I'll be doing this for basically nothing so don't want to regrind anything or make test plates for the secondaries.

Dave
 

#10 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 12 January 2011 - 06:48 PM

Forgot to mention, the secondaries are very rough/scratched on the back so testing through the back will not be possible.

Dave
 

#11 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 12 January 2011 - 07:05 PM

It's been something of note before; testing thru the back. But I notice with PCX lenses, there is a convergence of a reflection when looking thru the convex side; seeing a larger image of a lightbulb for example. I'm sure Mikey could figure a way to test it from either side.
Mark
 

#12 Dick Parker

Dick Parker

    Viking 1

  • *****
  • Posts: 765
  • Joined: 17 Aug 2007
  • Loc: Tolland, CT and Chiefland, FL

Posted 12 January 2011 - 09:02 PM

Dave -

If you are going to consider a Classical Cassegrain

Convex aspheres have one of their conjugates within the reflection, i.e. virtual. It is difficult to have access to it without auxiliary optics.

Without auxillary optics, your best bet might be a star test, or star test with artificial star at a known distance. Before 1859 (Foucault test) that is the way many, if not all, telescopes were made.

There is an article in Telescope Making , issue 31, pg 24 where the author describes a method where he tested the secondary by adjusting the spacing between the primary and secondary of the telescope and established conjugates in front and behind the telescope where the focuser is. The light source was at the conjugate in front and the knife edge was at the back conjugate. This system resulted in known spherical aberration which could be tested by measuring zones. Some ray trace analysis was required. The knife edge was replaced by a wire where the wire was a human hair.

Testing through the back might require that the back be quite flat, or some other curve (for best results). That seems to me like making an auxiliary optic.

Some other auxiliary optic choices are:

1) Spherical concave test plate, test per Texereau. ROCs have to be an exact match. Be careful of the shape of the intereference fringes.

2) Hyperbloidal concave test plate, i.e. the silvertooth method.

3) A Hindle sphere. You would only need a 12 inch or so Hindle sphere, but the curve is deep.

4) A 16 inch flat. You could figure the primary, then have it aluminized, then figure the secondary by testing the whole telescope as a system using autocollimation.

Don't know if that helps, but a good 16 inch Cass is worth going after.

Dick Parker
 

#13 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1628
  • Joined: 02 Mar 2007

Posted 12 January 2011 - 10:29 PM

primary 400mm dia. r/c 2460mm f/r 3.075
secondary 162mm dia r/c unknown at this moment.
secondary is 490mm inside primary focus
primary-secondary spacing 740mm
secondary to final focus 1035mm

Am I right that this means an amplification factor of 1035 ÷ 490 = 2.112 which makes the final f/r 6.49!!



Yes, this is a very fast system. For the secondary radius, you can use a simple two-mirror system relation: R2=mkR1/(m-1), where "m" is the secondary magnification, "k" is the relative size of the axial cone at the secondary (in units of aperture diameter), and R1 is the primary r.o.c. With
m~2.1 and k~0.4, that gives R2~0.76R1~1880mm.

In the classical Cassegrain, secondary conic would have been close to -8; in the Dall-Kirkham, the primary would have been about -0.54 (with lots of coma: 1.2mm, or little over 1/40 degree diffraction-limited field radius).

Vla
 

#14 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 13 January 2011 - 12:04 AM

Thanks Dick.
As I've mentioned, it looks like any work done on the secondaries is going to be tested on a star. Perhaps initially using a ronchi to get in the ball park then startest to finish. I'm picking that since the easier optic (primary) is so poor, the secondaries are likely worse.

Dave
 

#15 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 13 January 2011 - 12:16 AM

Thanks Vla.
I'm wondering what thought process was in the mind of the original maker. I have met him and I believe he does use the startest when figuring. The obvious thing is to give him a call but I'm hesitant to do that as he's very hard to talk to and would likely take offence to any questions.

Dave
 

#16 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 13 January 2011 - 06:51 AM

An addition to Dick's fine post, do -NOT- test the secondary by looking thru the concave testplate. The curve would be strong enough to mess up the bands to give an unreliable indication of actual fit.
Mark
 

#17 Mike I. Jones

Mike I. Jones

    Gemini

  • *****
  • Posts: 3239
  • Joined: 02 Jul 2006
  • Loc: Fort Worth TX

Posted 13 January 2011 - 07:56 AM

I'm wondering what thought process was in the mind of the original maker. I have met him and I believe he does use the startest when figuring. The obvious thing is to give him a call but I'm hesitant to do that as he's very hard to talk to and would likely take offence to any questions.



If he's so uppity and hard to talk to, show him the focogram you made of his "wonderfully" made, way undercorrected, ultra not-smooth mirror ;)
Mike
 

#18 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1628
  • Joined: 02 Mar 2007

Posted 13 January 2011 - 09:51 AM

If the primary is about 30% corrected, it takes a strong oblatre ellipsoid to compensate for the undercorrection (in this case, secondary conic should be about 4, optimally around 4.4). No wonder the optics is rough. Seems as if the maker exclusively relied on the star test while figuring - that can take you anywhere. Accidentally or not, this arrangement has zero astigmatism, but the coma is even a bit stronger than in the DK.

Vla
 

#19 Mike Lockwood

Mike Lockwood

    Vendor, Lockwood Custom Optics

  • *****
  • Vendors
  • Posts: 478
  • Joined: 01 Oct 2007
  • Loc: Usually in my optical shop

Posted 13 January 2011 - 12:28 PM

I've made several classical Cassegrains.

I'd recommend refiguring the primary to a parabola with your standard techniques.

Testing the primary and secondary in combination will be very difficult unless you have a good flat. Star testing will take a long time and will probably be frustrating.

Don't feel constrained by the secondary that came with it - if a different system is desired/required, then you may need to use a different diameter or regrind to a different radius.

I'd recommend a simple test for the secondary. A concave test plate can be figured to the proper hyperbolic shape using Foucault testing, and then you can figure the secondary to match by straightening the interference fringes between them. This method had a learning curve, but it works. Alternatively you can make a Hindle sphere and create a null test for the secondary.
 

#20 KerryR

KerryR

    Gemini

  • *****
  • Posts: 3068
  • Joined: 05 Dec 2007
  • Loc: SW Michigan

Posted 13 January 2011 - 03:20 PM

How do you grind and figure a small surface, such as the secondary of a (smaller) DK? Is it necessary to use a spindle polisher, or can such small(ish) surfaces be done by hand using 'normal' atm methods?
 

#21 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 13 January 2011 - 03:44 PM

I've polished secondary mirrors for DKs down to about 40mm diameter, with a little forethought and care. Fairly similar to big stuff. Biggest mindset to alter when polishing- The item is far smaller in area, so lighten up on the pressure!!!!! Grinding goes fast, and can be done with 220 or finer grit on long radii. Common sense combined with a couple regular mirrors work well here.
FWIW, M.
 

#22 gregj888

gregj888

    Viking 1

  • -----
  • Posts: 594
  • Joined: 26 Mar 2006
  • Loc: Oregon

Posted 13 January 2011 - 11:25 PM

Just some ramblings…

DKs are generally f/14 or longer to get reasonable performance.

You can add 2 lenses near the focal plane and make a corrected DK. I know you can get good performance to f/8. I don’t know if a 3-4 lens corrector would get to f/6 or not. Anyway, the CDK might be worth a look. Mark may want to comment.

If you can find a PCX lens with the right ROC it can be used as a secondary.

The primary of the DK is difficult to correct for prime focus work if that’s an issue.

A Classical Cassegrain is probably the most flexible. Different secondarys can give you different final focal lengths, back foci etc. If you add a 2 element corrector it’s better than a CDK. The primary is a parabola so it’s good to go.

If the CC’s secondary is made of optical glass, it can be figured from the back to a null. See the “Advanced Telescope Making Techniques” page 57 (Leonard). The secondary can also be figured kind of directly with the Gaviola test: ATMT, page 59. Both of these should also work for an RC.

Have fun,

Greg
 

#23 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 16 February 2011 - 02:09 AM

Time for an update with this project.
After a lot of discussion with the owner it was decided to refigure the optics as a Dall-Kirkham using the larger f/6.5 secondary. Here's a pic. taken with the existing optics. The main faults look to be focus and tracking. No doubt coma will be more noticable when those are improved but still not too bad. The spikes on one side of the stars are from the TDE but the primary baffle's not aligned 100% and is vignetting the edge on one side. When I did some star testing I removed it and the spikes were visible all around. We will look at coma correction later.

Dave

Attached Files


 

#24 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 16 February 2011 - 02:15 AM

I managed to take a ronchi (85 lpi) through the back of the smaller (f/ 9.5) secondary as it isn't as scratched as the bigger one. Another rough figure and turned edge and I assume the larger one is similar.

Attached Files


 

#25 dave brock

dave brock

    Viking 1

  • -----
  • Posts: 615
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 16 February 2011 - 02:21 AM

I've started polishing the primary, concentrating mainly on the TDE. Still a little edge left but it's looking a lot smoother. Diffraction ring is finally starting to appear. Not showing so well in the pic. though. I'm now close enough that I'll make a couder mask and start taking measurements. The conic works out to be -.555 btw.
Foucault:-

Attached Files


 






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics