Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Oil Flat Autocllimation Test

  • Please log in to reply
186 replies to this topic

#1 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 10:27 AM

Introduction

 

An oil flat can be used to test the quality of optics in a finished telescope. It can be used for Newtonians, Cassegrains, refractors, photographic lenses, etc.

 

Liquid flats

 

From a practical point of view, liquid flats are "perfect" reference flats since they assume the curvature equal to the averega earth's radius, which is about 6,400 km or roughly 4,000 miles (distance from New York City to Berlin, Germany). 

 

Materials needed

 

To make and use an oil flat is simple and cheap.

 

You will need:  

 

(1) plywood

 

(2) a means to cut a hole in it, and attach three leveling screws

 

(3) a pizza pan and

 

(4)  motor oil. If your place is subject to vibrations, pick heavy oil such as SAE 85W 140,  or chain saw oil.

 

Setting up the test

 

1. Clean the pizza pan and make sure it's dust and lint free. Pour the oil slowly into the pan and stir it gently with a plastic stick to remove as many air bubbles as possible. Scoop up any remaining air bubbles form the surface with a plastic spoon.

 

2. Place the plywood stand over the oil pan and place your telescope on it, the sky end facing the oil flat.

 

3. In the eyepiece holder inset a DIY eyepiece that has an LED and a Ronchi grating. Adjust your focuser so  so that the Ronchi grating is close to the telescope's infinity focal plane.

 

4. Turn on the LED and look down the eyepiece.  You will most likely initially see some strange pattern but don't panic!

 

Testing

 

1. Using leveling screws (the finer the better, but 1/4-28 or even 1/4-20, with large knobs made of toy wheels, is more than sufficient), tilt and tip the platform until you see the reflection of the aperture move into the view. You should see numerous Ronchi bands fill the POV at this point.

 

2. Using your focuser knob, bring the number of viisble bands to 3-4 per field and observe the pattern.

 

Assessment

 

This oil autocllimation test is a double-pass test, which is to say it's twice as sensitive as a single-pass collimation tests. 

 

Closing comments and reminders

 

1. The biggest problem you'll encounter will be the air bubbles in your oil flat.

 

2. The biggest advantage of of the oil flat test is that the flat size is limited only by the size of your oil pan, and the cost is pretty much negligible.

 

 

Mladen


Edited by MKV, 06 March 2015 - 10:40 AM.

  • GJJim, Live_Steam_Mad, LarsMalmgren and 3 others like this

#2 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 10:29 AM

Pictures

Picture 1: Oil poured into the pan. There are some water bubbles (9 o'clock position). I didn't remove the water bubbles, but simply pushed them aside, as that part of the flat was not used.

 

Picture 2:  Plywood platform over the flat with the overhead light on.

 

Picture 3: A C8 placed over the flat

 

Picture 4: The eyepiece with LED and 133 LPI Ronchi screen. The eyepiece housing  is a 1.25" PVC coupling and the LED/Ronchi housing is a section of a 1.25" wooden dowel.

 

Mladen

 

PS Change made in LPI rating of the screen used, as determined with a microscope, and thanks to Danny's comment.

Attached Thumbnails

  • 20150306_065011.jpg
  • 20150306_065155.jpg
  • 20150306_065507.jpg
  • 20150216_104538.jpg

Edited by MKV, 06 March 2015 - 05:04 PM.

  • Pierre Lemay, LarsMalmgren and allardster like this

#3 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 10:31 AM

Pictures (cont'd)

 

Picture 5: Ronchi bands, outside focus, showing a number of issues, such as undercorrection, turned up edge, and central area distortion.

 

Picture 6: This is what happens when you have air bubbles.

 

Mladen

Attached Thumbnails

  • 20150306_080228.jpg
  • 20150306_070318.jpg


#4 Pinbout

Pinbout

    Hubble

  • *****
  • Posts: 17413
  • Joined: 22 Feb 2010
  • Loc: Montclair, NJ

Posted 06 March 2015 - 10:46 AM

 

Picture 5: Ronchi bands, outside focus, showing a number of issues, such as undercorrection, turned up edge, and central area distortion.

 

 

looks typical for an sct. even the CO distortion. tsk tsk.

 

good idea with the 300 lpi. to show the zones. they probably wouldn't show on 133.

 

I know ronchiscreens.com sells 250 but where did you get 300?

 

nice platform.


Edited by Pinbout, 06 March 2015 - 10:48 AM.


#5 DAVIDG

DAVIDG

    Fly Me to the Moon

  • *****
  • Posts: 7332
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 06 March 2015 - 10:47 AM

Mladen,
  Great  work and you beat me to it ! It was on my list to do a thread almost exactly like this. 

     A minor word of caution to anyone trying the test when your testing a Cassegrain like  C-8 were the focusing is done via moving the primary mirror. There is only one location of the primary that results in spherical aberration being fully corrected so you want  to have the primary at or near that location which should be when the telescope is focused at infinity using the factory eyepiece adapter and star diagonal. So if the primary was positioned fully forward or fully back,  spherical aberration will be added to the test and since this is a double pass test it will be doubled as well.  So for critical testing  you want to focus the Ronchi screen by pulling  it, in and out of the focuser and not by using the focuser knob on the scope which will move the mirror.
 
                       - Dave


Edited by DAVIDG, 06 March 2015 - 11:06 AM.


#6 Ed Jones

Ed Jones

    Gemini

  • *****
  • Posts: 3494
  • Joined: 06 Apr 2004
  • Loc: Sin-sin-atti

Posted 06 March 2015 - 11:40 AM

Good work Mladen.  I'm guessing the fuzzy looking image is from your Ronchi screen?

 

Ed



#7 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 05:20 PM

 

 

Picture 5: Ronchi bands, outside focus, showing a number of issues, such as undercorrection, turned up edge, and central area distortion.

 

 

looks typical for an sct. even the CO distortion. tsk tsk.

 

good idea with the 300 lpi. to show the zones. they probably wouldn't show on 133.

 

I know ronchiscreens.com sells 250 but where did you get 300?

 

nice platform.

 

 

Danny, thank you. I checked the screen with my microscope and it is 133 LPI, not 300. The screen is an old sample I got back in the '80s at some Edmund sellout fare, and never used it until now. It came in a small wax paper bag with "300" written on it. I presumed the number represented the screen's frequency, and never bothered to actually check it. However, the fact that it is 133 LPi (or 5 lpmm) only shows that some SCTs are far from what their advertised quality -- which only makes the oil autocollimation flat test that much more useful for an amateur.

 

However, let's not forget that the error seen in an autocollimaiton test is twice exaggerated and that the actual wavefront is considerably smoother.

 

Mladen



#8 Pinbout

Pinbout

    Hubble

  • *****
  • Posts: 17413
  • Joined: 22 Feb 2010
  • Loc: Montclair, NJ

Posted 06 March 2015 - 05:53 PM

but what's up with the CO distortion, I have several I tested and they have the same, some more or less.

 

here's a mak:

 

mak pinched.jpg

 

sct

 

6insct.jpg


Edited by Pinbout, 06 March 2015 - 05:55 PM.


#9 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 07:53 PM

Mladen,
  Great  work and you beat me to it ! It was on my list to do a thread almost exactly like this. 

     A minor word of caution to anyone trying the test when your testing a Cassegrain like  C-8 were the focusing is done via moving the primary mirror. There is only one location of the primary that results in spherical aberration being fully corrected so you want  to have the primary at or near that location which should be when the telescope is focused at infinity using the factory eyepiece adapter and star diagonal. So if the primary was positioned fully forward or fully back,  spherical aberration will be added to the test and since this is a double pass test it will be doubled as well.  So for critical testing  you want to focus the Ronchi screen by pulling  it, in and out of the focuser and not by using the focuser knob on the scope which will move the mirror.
 
                       - Dave

Hi Dave, I hope you will still post your work on the subject, and add to this preliminary introduction. Your caution is spot on. I did not have an EP holder for my T-thread extension tube, in order to move the Ronchi screen, but the procedure, as you describe it, is the correct one -- one moves the Ronchi screen and not the mirror.

 

The only problem is, how do you know where the actual infinity focus is for a SMC such as C8? There is no marker for the infinity focus, which will depend on the design, and can be anywhere. I presume the infinity focus is approximately where the default Celestron 1.25" star diagonal ends, but even that leaves too much uncertainty, since even the lightest turn of the focusing knob changes the configuration.

 

To test this conjecture I machined a 1.25 inch EP holder with T-thread to accept a standard camera adapter. Even at 6 inches from the previous  position the Ronchi pattern doesn't look much different! From several unofficial C8 configurations, the bfl is about 200 mm or 8 inches behind the primary, and I don't have a longer draw tube. I will repeat the test on Sunday, as per your description.

 

Maden

Attached Thumbnails

  • 20150306_181859.jpg
  • 20150306_182451.jpg

Edited by MKV, 07 March 2015 - 06:10 AM.


#10 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 07:58 PM

but what's up with the CO distortion, I have several I tested and they have the same, some more or less.

 

here's a mak:

For SMC such as Celestron with its thin corrector, I'd say it's probably the weight of the secondary, but in a Mak with a thick meniscus, I doubt it. Perhaps it's the way the central hole was cored (release of trension, etc).

 

Mladen 



#11 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 08:08 PM

Good work Mladen.  I'm guessing the fuzzy looking image is from your Ronchi screen?

 

Ed

I'd say yes. The screen is really old (at least 30 years) and it was on a surplus sale at Edmund in the 1980's where evreything was priced garage-sale style. :) I have newer mounted screens from ronchiscreens.com but I don't feel like cutting them. I also used my phone camera, handheld, and the picture is obviously overexposed.

 

 

Forgot to mention, the bottoms of the leveling screws have acorn nuts. Normally I machine a rounded end with a radius tool on my lathe, but acorn nuts were a quick fix. 

 

Mladen

Attached Thumbnails

  • 20150306_185727.jpg

Edited by MKV, 06 March 2015 - 08:11 PM.


#12 Pinbout

Pinbout

    Hubble

  • *****
  • Posts: 17413
  • Joined: 22 Feb 2010
  • Loc: Montclair, NJ

Posted 06 March 2015 - 08:16 PM

here's my ronchiscreen.com's 133 screen under 100x microscope. not bad for ~$7

 

20140728_103828.jpg


Edited by Pinbout, 06 March 2015 - 08:17 PM.

  • sopticals likes this

#13 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 06 March 2015 - 08:24 PM

here's my ronchiscreen.com's 133 screen under 100x microscope. not bad for ~$7

 

No, not bad at all. I have two from ronchiscreens.com (a 2" by 2" inch and a 4" by 4" inch 133 LPI, mounted). I didn't use them for this test because I didn't feel like cutting them, since they're utilized elsewhere.

 

Mladen


Edited by MKV, 06 March 2015 - 11:28 PM.


#14 sopticals

sopticals

    Viking 1

  • *****
  • Posts: 931
  • Joined: 28 Mar 2010
  • Loc: New Zealand

Posted 06 March 2015 - 09:12 PM

Introduction

 

An oil flat can be used to test the quality of optics in a finished telescope. It can be used for Newtonians, Cassegrains, refractors, photographic lenses, etc.

 

Liquid flats

 

From a practical point of view, liquid flats are "perfect" reference flats since they assume the curvature equal to the average earth's radius, which is about 6,400 km or roughly 4,000 miles (distance from New York City to Berlin, Germany). 

 

Materials needed

 

To make and use an oil flat is simple and cheap.

 

 

Assessment

 

This oil autocllimation test is a double-pass test, which is to say it's twice as sensitive as a single-pass collimation tests. 

 

Closing comments and reminders

 

1. The biggest problem you'll encounter will be the air bubbles in your oil flat.

 

2. The biggest advantage of of the oil flat test is that the flat size is limited only by the size of your oil pan, and the cost is pretty much negligible.

 

 

Mladen

 

Very informative post. Thanks Mladen. :waytogo:

 

Stephen.(45deg.S.)



#15 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 08 March 2015 - 05:06 PM

Thanks for all your comments. This is a follow-up based on Dave G's recommendation that in the case of Celestron and configurations using similar focusing  mechanisms (the mirror moves rather than the focusing tube) one should set the system to the infinity focus, and move only the Ronchi test eyepiece. The test was repeated this morning:

 

1. The C8 was set to infinity focus using the diagonal and a 25 mm Plossl that came with the scope.

 

2. Only the illuminated Ronchi eyepiece moved along the optical axis to increase or decrease the number of visible bands.

 

3. A rheostat was used to dim the light source to avoid overexposure.

 

4. The Ronchi gram was made using a hand-held phone camera, since this is only for illustration purposes. In a test intended to for optical assessment of the

scope, a fixed camera mount is required.

 

5. For assessment purposes, use a new, clean, Ronchi screen for the test. (I used a very older one because it was handy).

 

6. Remember that wavefront errors are doubled by autocollimation, so that 1/10 wave error will look like 1/5 wave.

 

7. When pouring oil into the test pan, do so as close to the pan surface as possible. Pour it slowly and in a steady thin stream to avoid creating air bubbles. If done correctly, the oil will spread in a circular pattern (like a pancake) until it reaches the edge of the pan, and will have no air bubbles. needless to say the pan should be clean and dust and lint-free before any oil is poured.

 

8. The oil level does not have to be very deep. In my test it was roughly 3/16" or 5 mm deep. 

 

Mladen

Attached Thumbnails

  • 20150308_084742_LR.jpg
  • 20150308_085040_LR.jpg


#16 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 08 March 2015 - 05:37 PM

And here is the schematic of the test, for illustration purposes.

 

Mladen

 

 

Attached Thumbnails

  • OIL_FLAT.jpg


#17 BigC

BigC

    Fly Me to the Moon

  • *****
  • Posts: 5646
  • Joined: 29 Sep 2010
  • Loc: SE Indiana

Posted 08 March 2015 - 08:11 PM

might be a bit of a chore with large dobs and newts-check those mirror clips!



#18 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 08 March 2015 - 10:02 PM

might be a bit of a chore with large dobs and newts-check those mirror clips!

Yes, of course, but if you can mount them on a rocker, you can mount them on a test stand. :)

 

Mladen



#19 allardster

allardster

    Messenger

  • -----
  • Posts: 409
  • Joined: 01 Jul 2011
  • Loc: Toronto, Ontario, Canada

Posted 08 March 2015 - 10:09 PM

I am wondering that how would that work with a standalone mirror though? I imagine that the clips will deform the mirror too much. Any ideas?



#20 Pinbout

Pinbout

    Hubble

  • *****
  • Posts: 17413
  • Joined: 22 Feb 2010
  • Loc: Montclair, NJ

Posted 08 March 2015 - 10:31 PM

its great for testing refractors...



#21 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 08 March 2015 - 11:39 PM

its great for testing refractors...

Danny, it works just as well with Newtonias. 

 

Mladen

Attached Thumbnails

  • OIL_FLAT_2.jpg


#22 Mark Harry

Mark Harry

    Fly Me to the Moon

  • *****
  • Posts: 6935
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 09 March 2015 - 06:53 AM

"6. Remember that wavefront errors are doubled by autocollimation, so that 1/10 wave error will look like 1/5 wave."

A ronchigram cannot be used to assess error in waves DIRECTLY. Using a grating is self defeating, concerning any advantage gained by double-pass. Loss of sensitivity is actually decreased by a huge amount; dependent on how many bands are used in the Ronchi image, and its frequency.


 This is a good test, up until the point a grating is recommended to be used with it. 



#23 MKV

MKV

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5980
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 09 March 2015 - 09:07 AM

"6. Remember that wavefront errors are doubled by autocollimation, so that 1/10 wave error will look like 1/5 wave."

A ronchigram cannot be used to assess error in waves DIRECTLY. Using a grating is self defeating, concerning any advantage gained by double-pass. Loss of sensitivity is actually decreased by a huge amount; dependent on how many bands are used in the Ronchi image, and its frequency.


 This is a good test, up until the point a grating is recommended to be used with it. 

In principle, the AC test doubles the wavefront error and makes it more discernible, whether it's qualitative or quantitative. Each method benefits from the doubling of the waferont error.

 

Mladen



#24 kingjamez

kingjamez

    Apollo

  • *****
  • Posts: 1285
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 09 March 2015 - 09:20 AM

 

its great for testing refractors...

Danny, it works just as well with Newtonias. 

 

Mladen

 

Good luck supporting a thin mirror in that arrangement.

 

-Jim



#25 DAVIDG

DAVIDG

    Fly Me to the Moon

  • *****
  • Posts: 7332
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 09 March 2015 - 09:30 AM

 As stated, the wavefront is doubled so the result is that  it is easier to see the defects. Also there are  no measurements errors from reading zones,  errors in the mechanics of the  tester, the procedure in making the  measurments or in getting any critical spacings correct. The result is a test method that has very little inherent error. When the Ronchi line are straight then one has a very well corrected system and one can always switch to a knife edge and also  use other test methods to cross check the results.

   I have never seen a mirror or telescope that tested well using double pass autocollimate, perform poorly when used on the sky, but I have seen many telescopes and mirrors that were stated to be  "1/8" to "1/20" show gross errors when tested via double pass and in reality are only good to 1/2 wave or worse.

 

                 - Dave




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.







Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics