Jump to content


Photo

9.25 inch refractor project

  • Please log in to reply
407 replies to this topic

#151 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 March 2013 - 07:24 PM

Jim,
Here are couple of images of achromats being tested via double pass autocollimation. It takes me about 5 minutes to set them and I can see exactly what is wrong with them.
Here is a picture of 4" f/15 Jaegers lens. Note the Ronchi bands are bowing and also not the evenly spaced top to bottom. The lens is undercorrected and has some astigmatism.

- Dave

Attached Files



#152 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 March 2013 - 07:25 PM

Here is another Jaegers lens, this one is 5" f/5. Not great as well.

- Dave

Attached Files



#153 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 March 2013 - 07:28 PM

Here is a "Zeiss" Telementor lens. We're pretty sure someone swapped out the original lens and put this one in it's place. It well undercorrected.

- Dave

Attached Files



#154 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 March 2013 - 07:31 PM

Here is a real Zeiss lens. It is of excellent quality, notice how straight the bands are.

Attached Files



#155 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 March 2013 - 07:34 PM

Finally here is Selsi 80mm f/11 objective, also of excellent quality from the nice straight Ronchi bands.

- Dave

Attached Files



#156 Dick Parker

Dick Parker

    Viking 1

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

Posted 26 March 2013 - 09:47 PM

Jim -

I have been following this project because I, too, have built refractors. Testing refractors is different from testing mirrors because the design of a refractor objective corrects chromatic aberration (as well as an achromat can), spherical aberration, and coma. This relies on the interplay between both the elements, the surfaces, the thicknesses, refractive indices, and spacing. For best results, it is not a willy nilly affair fixing a hole in one surface with a hill in another.

For best results, the radii have to be close to the design intent as well as the surface figures spherical. The recommended tolerance for R2 and R3 should be within +/- 1/20th % of their design value. R1 about 1/10th % and R4 fairly loose. Keep in mind there are three aberrations being corrected.

Autocollimation testing ONLY verifies that the spherical aberration is corrected, but the color and coma is in the radii.

for best results, a spherometer is not adequate to validate that the radius is correct. There should be some more precise method. Often used is measuring R3 with a knife edge and carefully measured rod. The results from this method should agree with calculation based on a spherometer by only a few thousandths.

Attached is a picture showing measuring R3 of one of my flint elements with rods.

Dick Parker

Attached Files



#157 Dick Parker

Dick Parker

    Viking 1

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

Posted 26 March 2013 - 09:54 PM

Then if your design has R2 = R3 you can, and should test the two together to establish that the radii match. R3 can be validated for exact radius and figure by the knife edge test that we all know. Interpreting interference patterns can be a bit tricky, but is part of the process of making a refractor. If you are lucky, when R3 and R2 match exactly, and you test them againt each other by interference the interference fringes should be straight and parallel as show in the picture of one of my objectives. If they don't match exactly, you will end up with a bull's eye and rings. As you press the side of the top element in the interference test the bull's eye will move from the center to the side. If R2 is exactly spherical, the bull's eye will stay perfectly round. If it turns into a "D" shape or pear shape or anything but round, then you have zones in the surface.

With this test you can validate that R2 is the correct radius and spherical figure.

Dick Parker

Attached Files



#158 Dick Parker

Dick Parker

    Viking 1

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

Posted 26 March 2013 - 10:05 PM

You can get by making a refractor objective with only one other test plate, that being for R1. R1 should test against its test plate with the same result as discussed before. If I could pass on any advice it would be to go make a test plate for R1 being careful to validate that the test plate is the exact radius and a perfect sphere.

With three surfaces out of four validated, you can deduce that any further errors of the lens tested assembled will be in R4 and you can figure that surface.

Then you can assemble the lens and test with autocollimation. Shown is the autocollimation test of my 6 inch f/15.

Again I say, the autocollimation test only validates spherical aberration, not color correction or coma.

One thing I don't see mentioned is that optical glass has horrible thermal properties and you need to wait a long time after you work on a surface before you can test it. A way to minimize the wait is to have a bucket of room temperature water handy and immerse the element in it for about 1/2 hour after working on it before testing. This is big time. Don't overlook it. Ask me how I know.

Dick Parker

Attached Files



#159 Dick Parker

Dick Parker

    Viking 1

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

Posted 26 March 2013 - 10:19 PM

Back to R2 against R3 by interference. Shown is the bull's eye and rings I described when R2 and R3 are not a perfect match.

Dick Parker

Attached Files



#160 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 27 March 2013 - 04:53 AM

Thanks DAVIDG and Dick Parker for your input. I believe this will at least give Jim some idea of what to do if his current attempts to correct the lens by trial and error don't work to his satisfaction.

On the measurement of radii, radius bars are the method of choice to for many optical applications. The picture below shows a setup where precise light source was to be positioned relative to a concave surface of a lens for a knife-edge test. The radius bar used consists of sections that screw into each other to form a rod. the ends have panhaead nylon screws that contact the glass surface and the light source at one point. The length of the rod is then measured either with calipers (I have 24 inch calipers) or on a lathe bed. Even a tape measure can determine the radius to within a mm. The nylon screws serve as fine adjustments for "tweaking" the radius bar length precisely to the desired distance.

Attached Files



#161 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6166
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 27 March 2013 - 05:42 AM

Short answer, no; not if you don't want to saddle yourself with a lot of work.
M.

#162 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6166
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 27 March 2013 - 05:58 AM

I can add one more thing to Dicks excellent posts. If the radii in the IF are not equal, and show a bullseye pattern,

DO NOT VIEW SUCH A PATTERN THRU THE BACK SIDE OF THE FLINT!!!
A round pattern will most likely show up as being flattened somewhat on the side approaching the bevel, IIRC; giving an indication of a ripped edge. Whenever making an IF test, an effort should be made to look thru a positive power; ideally at the focus of the piece.
********
A lot of the angst can be alleviated, if you make a reasonable testplate that's around 1/2D or a bit more. Such a plate, if ground along with the full size surfaces only takes a few minutes of additional effort. And it doesn't have to be completely polished; only spherical and precisely measured.
My 2 millicents worth,
M.

#163 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 27 March 2013 - 07:53 AM

Finally here is Selsi 80mm f/11 objective, also of excellent quality from the nice straight Ronchi bands.

Very telling and indicatuive of the value of autocllimation. Even though these are Ronchi an not IF fringes, the test is a double precision test at the focus, so perfect null even with a 133 lpi Rinchi is smomwhere around 1/8 to 1/10 wave PV if not better.

I would like to add something DAVIDG alerted me to previously, namely that it is very important the lens is properly oriented front to back when installed in a cell. It can inadvertently happen.

This image illustrates the effect of reversing the front to back as opposed to the correct orientation. The reversed orientation showed the lens to be severely undercorrected, but when it was flipped around the lens showed clean straight bands indicating a reasonably well corrected optic (this was a $30 Surplus Shed 80 mm, f/5 air spaced lens)

Mladen

NB The light source was not fully diffused

Attached Files



#164 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 27 March 2013 - 10:50 AM

Autocollimation can be used to check the color correction of an objective as well. As Dick pointed out the color correction is a function of the radii and thickness of the elements along with the refractive index. In the design for most classic achromats the red and blue wavelengths come to focus at the same or nearly the same point. So to check the color correction you set up the lens in double pass mode and use a narrow band light source to produce either red or blue light and find the exact focal position. Then without moving the knife edge or ronchi screen switch to other wavelength. If the color correction is right, there should be no are very little difference in were the focal plane is located at the other wavelength. I use narrow band interference filters and look thru them to selection the wavelength I want to check the color correction with a white light source. I also have a selection of fairly narrow wavelength LEDs that I can quickly change on my tester.
Here is a graph of the longitudinal spherical aberration of a typical achromat. It's hard to see but the green image is perfectly straight and on the y-axis which is the focal plane. The red and blue images focus behind the green but nearly on top of each other. So again one can take advantage of this and use double pass autocollimation to check this. If a radius is off, you still can get the green image to show a null but the red and blue images will no longer focus at the same position, they will move apart from each other. So as Dick pointed out it is critical to get the radii and thickness as close as possible to the design parameters.

- Dave

Attached Files



#165 jimegger

jimegger

    Apollo

  • *****
  • Posts: 1178
  • Joined: 22 Dec 2005
  • Loc: Palmer,Alaska

Posted 27 March 2013 - 11:56 AM

All of these discussions are very informative so thanks to all of you. I am certainly no professional optician and my knowledge base is from what I have read and experienced over the years. All of these discussions have pointed out some gaps in that for me to which I am most appreciative.

That being said, My crown has one heck of a polish ! Also after star testing it last night I have found that it has now gone ever so slightly under corrected again. I had polished on R1 for 45 minutes with the regular lap and it did the job of removing glass from the center but just a hair too much - and I mean a hair ! I was using Sirius as the test star and I was impressed by the color correction as well as the closeness of the optics to putting out an excellent star test. The images are round as well so good news on the astigmatism front as well. Today I am going to put 10 - 15 minutes on the convexing lap again to try and push it back towards a shorter center focus. I believe this may just do it.

Apart from the star test I used the scope to look at detail on the mountain as before. The sharpness of the image it provided was much better than previously noticed. Even with a burbling air mass I could clearly see details on the small pointed rock I have been using for comparison purposes all along. Furthermore there was a diminution of color noticeable along the edges of objects to where it seemed there wasn't any against a bright blue sky.

One other thing I noticed was the lens cell had shrunk up in diameter to where it was starting to squeeze the crown a bit too tight due to pretty cold temperature at around 10 above. So the calculations I did for differential thermal properties was off by a hair as well and need to be remedied by machining the cell diameter out a bit more.

#166 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 27 March 2013 - 12:31 PM

Jim,
Can you discribe what your looking for when you do a star test and how you do it ?

- Dave

#167 jimegger

jimegger

    Apollo

  • *****
  • Posts: 1178
  • Joined: 22 Dec 2005
  • Loc: Palmer,Alaska

Posted 27 March 2013 - 05:56 PM

I find a fairly bright star and look at the image inside focus and outside focus by racking in and out. The images should show identical diffraction patterns on both side of focus with even spaced and same width diffraction rings.

If the diffraction pattern on the inside of focus shows a hole in the center but the outside focus image is full of light with a bright spot in the center then the objective is under corrected / the center has a longer focus than the edge.

If the diffraction pattern shows inside focus, a bright center and flooded with light in the diffraction rings but the outside focus image shows a diffraction pattern of rings surrounding a hole in the center then the objective is over corrected. The over corrected objective has a shorter center focus than edge focus.

My objective now shows a very small hole in the middle of the inside focus but the center with a bright spot outside of focus. It is now ever so slightly under corrected. The rings are all round and symmetrical for good signs of little if any astigmatism.

The diffraction rings should be round and symmetrical to show no astigmatism. They should be of even brightness and width to show no zonal errors.

The in focus star image should be a small bright dot with a couple of diffraction rings the first one brightest and the second quite a bit dimmer on a steady seeing night. My six inch shows perfect diffraction patterns on both sides of focus to a textbook "T". This 9.25 inch is very close to that now.

#168 DAVIDG

DAVIDG

    Aurora

  • *****
  • Posts: 4816
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 28 March 2013 - 10:37 AM

Jim,
Sound good. If your not already doing so I would sugguest that you use a green filter and also an eyepiece whose focal length is same or close to the f-ratio of your scope.
Refractor can be difficult to star test in white light because of the chromatic aberration. A great book on the subject is "Testing and Adjusting Telescopic Objectives " by Cooke. It is an old book but the theory is the same today. You can read an on-line copy from here. http://archive.org/d...ntand00taylgoog

I also highly recommend that you test your fully assembled lens by double pass autocollimation with the objective facing the flat just like it would in the telescope. It will be a cross check of your star test results. If your lens is as close as you think to the proper figure then it should show a smooth figure using double pass and very close to a null. If it doesn't I would recommend investaging both the star test result and the double pass result to find out why they differ. As I have said, the flat would have to be many waves off before it adds any real amount of error to the test results and it is double pass test so it is very sensitive to errors in the objective. I have seen many amateurs believe they are doing the star test correctly but in reality they were not and getting a false impression that their optics were much better then they really were. I personnally know of a number of refractors were the image was bragged about as being excellent but when I tested them I found the objective was backwards in the cell!


Both Dick and I teach optical fabrication and we have won multiple awards for our optics. I'm not bragging, just saying that their is independent data from others to back our test procedures and methods. We were just at the 13th annual Delmarva mirror making class were we helped null figure around 20 mirrors, one as large as 20 f/6 in a few days. Over the years we both have seen optics that the owners stated were "1/10 wave" or better turn out to be no better then 1/4 wave or worse when tested correctly. The more data you have the better the finally results will be.

- Dave

#169 saemark30

saemark30

    Apollo

  • -----
  • Posts: 1145
  • Joined: 21 Feb 2012

Posted 28 March 2013 - 11:30 AM

Seems like you are grinding and polishing by hand?!

May I ask where you got the glass blanks? I would consider such a project in the future.

#170 jimegger

jimegger

    Apollo

  • *****
  • Posts: 1178
  • Joined: 22 Dec 2005
  • Loc: Palmer,Alaska

Posted 28 March 2013 - 12:04 PM

David, I do have that book "Testing and Adjusting Telescope Objectives" by Cooke. It is excellent !
Right now I do not trust my flat so I am going to have it tested that is why the star tests are being done. When I think I have it in the bag the 12.5 inch reflector will be used for a single pass auto collimation test for a null in the Foucault test and a look with the Ronchi grating as well.

When I did the six inch lens some years back I used the 12.5 inch reflector as a single pass collimator and got excellent results to the point of having a nice null in the Foucault test and the star test confirmed the objective was good. Star images are text book and Mars at 500X shows exquisite detail. That being said , until I know how good the flat is I must rely on what I know works. The images now are really sharp. It clouded over last night so the objective was not able to be star tested after the last polishing run. Tonight may clear up and if so I'll test it again.

I read a neat story about a fellow in "Telescope Making Magazine" some years back when it was still being published about a fellow I believe was in Colorado who was making an 11 inch refractor. He had no flat and would literally test by driving to the side of a mountain where he could look at a light some 3 miles distant. He would do a star / Foucault test after polishing using that light. It took a bit but in the end he got a really good objective. There are many ways to skin a cat !

These blanks were obtained from 2 sources, Newport Glass works in California and Schott Glass works back east. They are F2 flint and BK7 crown. The color correction is really good from what I am experiencing.

#171 Mr. Bill

Mr. Bill

    Fly Me to the Moon

  • *****
  • Posts: 6178
  • Joined: 09 Feb 2005
  • Loc: Northeastern Cal

Posted 28 March 2013 - 01:17 PM

WOW....great read. Don't understand everything but feel like I'm taking a crash course using the Socratic teaching method.

Thanks... :bow:

#172 Gord

Gord

    Surveyor 1

  • *****
  • Posts: 1686
  • Joined: 06 Jan 2004
  • Loc: Toronto, ON, Canada

Posted 28 March 2013 - 01:45 PM

Agreed, this is an awesome thread, truly showcasing CloudyNights at its best! Learning so much and really appreciate the detailed explanations, images, and commentary. :waytogo:

It's making me want to go a try it!

Clear skies!

#173 Ed Jones

Ed Jones

    Mercury-Atlas

  • *****
  • Posts: 2695
  • Joined: 06 Apr 2004
  • Loc: Sin-sin-atti

Posted 28 March 2013 - 01:51 PM

Since both internal radii are the same you could oil and tape them together and autocollimate them. This would largely desensitize these internal surfaces. This will introduce overcorrection, however, but you can use a Ross lens to correct this SA.

#174 Mr. Bill

Mr. Bill

    Fly Me to the Moon

  • *****
  • Posts: 6178
  • Joined: 09 Feb 2005
  • Loc: Northeastern Cal

Posted 28 March 2013 - 01:51 PM

It's making me want to go a try it!

Clear skies!


Uhhh...I wouldn't go that far. I'm happy to build my stuff with finished optics but it's nice to have some understanding of what's going on.

:cool:

#175 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 28 March 2013 - 02:12 PM

Jim, you can test your flat by autocllimation using the 12 inch mirror, which you say is a good parabola. If you are absolutely sure the 12 inch is a good parabola then you can test your flat with it, as shown below.

You should be able to observe straight Ronchi bands or a clean Foucualt null with the knife edge. Anything else (i.e. bowed Ronchi bands) will confirm that the flat is unacceptbale.

Double pass autcollimation is a superior test to asingle-pass method you're proposing.

Attached Files








Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics