Return to the Cloudy Nights Telescope Reviews home pageAstronomics discounts for Cloudy Nights members
· Get a Cloudy Nights T-Shirt · Submit a Review / Article

Click here if you are having trouble logging into the forums

Privacy Policy | Please read our Terms of Service | Signup and Troubleshooting FAQ | Problems? PM a Red or a Green Gu… uh, User

Equipment Discussions >> ATM, Optics and DIY Forum

Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | (show all)
kfrederick
Carpal Tunnel
*****

Reged: 02/01/08

Re: Herrig *DELETED* new [Re: Mike I. Jones]
      #5564898 - 12/10/12 09:02 PM

Post deleted by kfrederick

Post Extras: Print Post   Remind Me!   Notify Moderator  
Mike I. Jones
Post Laureate
*****

Reged: 07/02/06

Loc: Fort Worth TX
Re: Herrig new [Re: kfrederick]
      #5564899 - 12/10/12 09:03 PM

OHHH - "Never mind"!

Post Extras: Print Post   Remind Me!   Notify Moderator  
ccaissie
professor emeritus


Reged: 09/13/10

Loc: Whitefield, Maine
Re: Herrig new [Re: MKV]
      #5565032 - 12/10/12 10:32 PM

I ran several scenarios of the file in OSLO. Relatively INsensitive to spacing, radii, and angles.

C


Post Extras: Print Post   Remind Me!   Notify Moderator  
ccaissie
professor emeritus


Reged: 09/13/10

Loc: Whitefield, Maine
Re: Herrig new [Re: Mike I. Jones]
      #5565060 - 12/10/12 10:55 PM

Quote:

I'm not following the issue of transmission. The Herrig is all-reflective. It is true that bubbles that grind through will scatter some light. But the internal transmission of the mirror substrates is immaterial.
Mike




Testing the convex surface through the back as a concave?

A.Leonard wrote in ATMT 2 about this in making convex hyperboloids. pp57,58, and in making mak correctors.

Right, needs to be homogeneous to a high degree for testing through the back.


Post Extras: Print Post   Remind Me!   Notify Moderator  
MKV
Carpal Tunnel
*****

Reged: 01/20/11

Re: Herrig new [Re: ccaissie]
      #5565066 - 12/10/12 11:03 PM

Quote:

I ran several scenarios of the file in OSLO. Relatively INsensitive to spacing, radii, and angles.

C



The spacing, angle and radii were never even mentioned. If you read Mike Jones' archived thread from 2006, he and Ed Jones mention the "cons", one being that you need a 12.6" and an 11.2" mirror for an 8-inch clear aperture and a substantial loss of transmission on each reflection x 4 Ed put is succinctly: "What do you like about it? Four reflections, 2 oversize optics, doesn't have much going for it."

I think four reflections will mulitply even the tiniest errors on the mirrors wavefront and make them stand out. It's a well know fact that in autocollimation testing you have doubling of the errors even with fewer reflections.

Mladen


Post Extras: Print Post   Remind Me!   Notify Moderator  
Dave O
sage
*****

Reged: 12/21/11

Loc: Sri Lanka
Re: Herrig new [Re: MKV]
      #5565177 - 12/11/12 12:04 AM Attachment (29 downloads)

Well, let's be fair here ...

1. Yes, there are four reflections; but only two from each of two mirrors. (And they are offset from each other ...)

2. Surfaces are spherical, so easier to get smooth by most people's experience.

3. Many SCT's use 3-mirrors (including the diagonal); so really only ONE extra reflection ... and an even number of reflections will orientate the image the same as what you'd see in a Newtonian.

4. As there are only two mirrors, one could go with the enhanced coatings and improve the image brightness.

5. Yes, some glass is worked that is not used ... however, the quality of the image speaks for itself -- I think it is a good trade. (And one gets the 'option' of orientating the mirror to present its best face to the incoming light, so to speak.)

6. Four 'optical surfaces' -- but only two elements to align with each other (+ the focusser). As they are both spherical, there is some leeway on 'centering' -- just need to get the separations and tilts right.

7. In my mind, the real difficulty in the design is supporting that upper mirror ... but, other designs have gotten around this (Schupmanns, Yolos, etc.).

8. It does have some field tilt (generally <5 deg. from what I have seen) -- so, this will have to be accounted for as well.

Attached is the OSLO file I had sent to Kevin. It is larger than I'd build one, but as most of us know, Kevin likes big scopes.

The Herrig can be as fast as f/9; but aberrations do increase with faster focal ratios.

I'd also like to extend a big thanks to Dave G for getting those Basic routines up and running. The Herrig is a very easy scope to design using those routines -- enter your desired aperture and focal ratio, and it spits out a well-corrected design, which can be ported to Winspot for analysis. From there, it is an easy thing to enter into OSLO.

I had mentioned to Kevin that Dave G had some ideas for testing that long CX surface; was happy to see he mentioned it here ... Dave G is a pretty clever guy who knows his stuff.


Post Extras: Print Post   Remind Me!   Notify Moderator  
MKV
Carpal Tunnel
*****

Reged: 01/20/11

Re: Herrig new [Re: Dave O]
      #5565270 - 12/11/12 12:57 AM

You make good points, Dave O, but we all know that each reflection doubles, or nearly doubles, the wavefront error.

The mirrors are spherical, but OSLO shows that conic constants of both mirrors, to a reasonable degree, do not really affect the image that much. One has to really go off the scale to begin to see deterioration that cannot be refocused.

I think the best and easiest way to test a convex surface is using a test plate. One easy way to do that is to use a full size glass tool as your test plate. Nulling a front surface through the back requires raytracing to determine the exact position of the light source. But it also requires polishing and figuring the back of the mirror flat (assuming ou have a quartz mirror). You might as well make a tests plate from cheaper glass instead. Testing through the mirror doesn't really save on effort, imo.

Mladen


Post Extras: Print Post   Remind Me!   Notify Moderator  
Mark Harry
Vendor
*****

Reged: 09/05/05

Loc: Northeast USA
Re: Herrig new [Re: MKV]
      #5565434 - 12/11/12 06:11 AM

Making a convex spherical surface is not difficult. If certain accepted methods/parameters are used, very little change in ROC ocurrs, and the glass will readily adopt a smooth zone-free surface. Very little testing will be required; likely not until near the very end of the job. Preliminary curve could be done with a good spherometer. A 1/2 diameter testplate could be used if preliminary setup is done correctly as outlined above.
******
What I'd like to know, is what the variance of the ROC of the long convex surface could withstand- could it vary by 5% (+ -) without a bad effect?
******
This design has a particularly fascinating quality about it.
M.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Mark Harry
Vendor
*****

Reged: 09/05/05

Loc: Northeast USA
Re: Herrig new [Re: Mark Harry]
      #5565440 - 12/11/12 06:21 AM

I see that DMG's version is actual 6.5" diameter. I sized it for 6" and a 1 degree field- still very high quality spots, and smallish errors!

Post Extras: Print Post   Remind Me!   Notify Moderator  
Dave O
sage
*****

Reged: 12/21/11

Loc: Sri Lanka
Re: Herrig new [Re: Mark Harry]
      #5565470 - 12/11/12 07:22 AM

Quote:

What I'd like to know, is what the variance of the ROC of the long convex surface could withstand- could it vary by 5% (+ -) without a bad effect?



The ROC of the first mirror largely determines the system's final focal length, so if that 'spec' is loose, then one could fab the first (CX) mirror to get as smooth a sphere as possible, somewhat close to the desired ROC. With that mirror done, the system can be re-optimized to the measured ROC (with the resulting change in focal length) and the second (CV) mirror could be figured to match without any appreciable change to the system performance (other than focal length/ratio).


Post Extras: Print Post   Remind Me!   Notify Moderator  
DAVIDG
Post Laureate
*****

Reged: 12/02/04

Loc: Hockessin, De
Re: Herrig new [Re: Dave O]
      #5565504 - 12/11/12 08:09 AM

The issue with testing the convex surface with a test plate is the convex surface has a very very long radius so the concave test plate needs to match this. So you need to make that long radius concave test plate. So your back to the same problem of testing the tertary mirror in Tri Schief ie having the mirror many yards away and testing out doors. The accuracy of the concave test plate determines the accurary of the convex piece your testing with it. If you test the convex surface by testing thru the back and using a much short radius sphere to setup a null condition then the accuracy improves. To me there is nothing better then actually see the surface under a knife edge/ ronchi type null test.

- Dave


Post Extras: Print Post   Remind Me!   Notify Moderator  
kfrederick
Carpal Tunnel
*****

Reged: 02/01/08

Re: Herrig new [Re: DAVIDG]
      #5565570 - 12/11/12 09:14 AM

The 6 inch f 12 looks like a easy risk .The concave is a 8 inch f 10 spherical [160 inches] .And the convex is 503 inches RC spherical . I think some times it will need a 90deg diagonal like a cass. does .A sloted interlocking CNC water jeted aluminum box will cost around 100 dollars. Big question the there are 4 tilts and two mirrors I need to know the tilt of each optic .Just talking. Told the wife the house will be finished before any more telescopes .

Post Extras: Print Post   Remind Me!   Notify Moderator  
MKV
Carpal Tunnel
*****

Reged: 01/20/11

Re: Herrig new [Re: DAVIDG]
      #5565743 - 12/11/12 11:15 AM Attachment (16 downloads)

DAVIDG is right. For relatively long radius surfaces, testing through the back surface may be the only practical approach, which is by no means as simple as it seems. It involves grinding and finishing both surfaces to optical standards, and calculating exact position of the setup. It also requires a nulling lens and a diagonal mirror, etc.

The Herrig seems rather insensitive when it comes to the sphericity or asphericity of the mirrors, so a smooth surface rather than an exact figure appears to be a priority given that there are four relfections, imo.

Below is an example of the Herrig configuratuion provided by Mike Jones. The upper set of spots is as provided by Mike. Both mirrors are spherical (cc = 0). The lower set of spots represent an altered setup, where the convex primary is a rather strong prolate ellipse (cc = -0.5), and the secondary is a full parabola (cc = -1).

Mladen


Post Extras: Print Post   Remind Me!   Notify Moderator  
kfrederick
Carpal Tunnel
*****

Reged: 02/01/08

Re: Herrig new [Re: MKV]
      #5565777 - 12/11/12 11:37 AM

Good to know that a long focus newt mirror can be used for the secondary .But keeping the surfaces spherical makes things easy.I saw one that had two mirrors instead of the one secondary .Guess that free it up for a different RC on the last surface..

Post Extras: Print Post   Remind Me!   Notify Moderator  
Mark Harry
Vendor
*****

Reged: 09/05/05

Loc: Northeast USA
Re: Herrig new [Re: MKV]
      #5565851 - 12/11/12 12:26 PM

You guys are making this more complicated than it actually is to do. CXs are relatively easy; I've made thousands of them. They're easier than flats, or long CVs. (-IF- you have initial parameters correct when you start off, and prepare the polisher correctly.) If you spend too much time worrying whether you 'theorized' enough, it can literally stop any meaningful progress in making a (good) CX surface, period. A 1/2 diameter TP can be made, polished/dummy-shined in 15-20 minutes, and with reasonable common sense can ascertain whether the CX surface is smooth or not, has no edge issue,zones, etc. By the time you figure out a null test thru the back side to take a look, a lot of progress could have been made in the meantime- up to you. Dave is somewhat correct about TP determining the accuracy of the surface; but it doesn't necessarily have to be as precise as the final tolerance desired. Ideally it shouldn't have a convex fit to the optic, and be just reasonably smooth. If not, you can use the optic to 'test' the TP, but this might be a bit frightening for the inexperienced to contemplate and get a handle on.
BTW, CXs normally don't go prolate during normal polishing. -THAT- is rather difficult to accomplish even deliberately. This can be a big help trying to get a sphere.
*******
How sensitive is that CX first radius ??????????????????????????????????????????????????????????????? Can you get to within 5%???

MVK, it almost looks like if you deformed one or both of the optics somewhere between (less) than what you did above, it could make it better than an all-spherical scope....?
M.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Dave O
sage
*****

Reged: 12/21/11

Loc: Sri Lanka
Re: Herrig new [Re: Mark Harry]
      #5565928 - 12/11/12 01:22 PM

Quote:

How sensitive is that CX first radius ??????????????????????????????????????????????????????????????? Can you get to within 5%???




Perhaps I am not understanding the question -- the answer depends on how much variance you can tolerate in the final focal length -- similar to a Newt. For any primary (CX) ROC, there is a matching secondary (CV) that will produce similar images -- but, the system focal length will change.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Dave O
sage
*****

Reged: 12/21/11

Loc: Sri Lanka
Re: Herring new [Re: Mike I. Jones]
      #5565971 - 12/11/12 01:41 PM

Quote:

Although, I want to improve on Herrig's mailbox-like enclosure design. Seems very leaky from a stray light standpoint, and the aperture stop is steeply slanted rather than perpendicular to the optical axis, which is better.




When I saw that picture, I was thinking that what it really needs is a 'tilt up' top that when 'closed' covers the optics, but can be opened for use. It would be hinged back by the eyepiece end and could include baffles as well. I think it would be very neat -- folded up into a nice box for transport.


Post Extras: Print Post   Remind Me!   Notify Moderator  
DAVIDG
Post Laureate
*****

Reged: 12/02/04

Loc: Hockessin, De
Re: Herrig new [Re: kfrederick]
      #5565992 - 12/11/12 01:51 PM

Mladen,
When you have long radius optics, if you see any signs of an aspheric surface at all on the test stand, the conic constant will be very large, like on the order of 100 to 1000 or more. The conic on the back convex surface my Schupmann objectives is -1128 and the radius is 302" convex. I didn't have to chisel the glass away to achieve that conic. It only took some mild polishing and on the order of what is done to parabolize a much shorter radius surface. So your analysis is somewhat misleading to the insensitivity of the Herrig design to figuring errors. If you don't get to smooth and spherical surfaces you could have problems.

- Dave


Post Extras: Print Post   Remind Me!   Notify Moderator  
kfrederick
Carpal Tunnel
*****

Reged: 02/01/08

Re: Herrig new [Re: DAVIDG]
      #5566080 - 12/11/12 02:42 PM

At 800 inch radius cx could a flat be used? Not much difference.

Post Extras: Print Post   Remind Me!   Notify Moderator  
DAVIDG
Post Laureate
*****

Reged: 12/02/04

Loc: Hockessin, De
Re: Herrig new [Re: kfrederick]
      #5566096 - 12/11/12 02:50 PM

Quote:

At 800 inch radius cx could a flat be used? Not much difference.



Kevin,
No you need that convex surface, how ever long it is.

- Dave


Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | (show all)


Extra information
12 registered and 19 anonymous users are browsing this forum.

Moderator:  ausastronomer, richard7, Starman81 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 4963

Jump to

CN Forums Home


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics