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

Using another telescope as collimator

  • Please log in to reply
13 replies to this topic

#1 akulapanam

akulapanam

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,870
  • Joined: 26 Aug 2012

Posted 25 September 2021 - 04:45 PM

I’m interested in doing something like this

https://www.starry-n...een-collimator/

Seems like it would make indoor star testing much easier. So in principle I could go acquire a used SCT, put the 9mm artificial star at the rear cell, and then put the scope to be tested directly in front? I assume the size of the secondary would restrict its use for refractors.

#2 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,527
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 25 September 2021 - 05:03 PM

If the entrance pupil of the refractor is smaller than half the difference in diameters between the SCT entrance pupil and the central obstruction the refractor could be tested without any central obstruction. Otherwise some stitching together of results obtained by measuring overlapping regions of the refractor entrance pupil could be done. 



#3 akulapanam

akulapanam

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,870
  • Joined: 26 Aug 2012

Posted 25 September 2021 - 05:28 PM

How important is the collimation of the SCT used as a collimator

#4 ngc7319_20

ngc7319_20

    Vanguard

  • *****
  • Posts: 2,407
  • Joined: 25 Oct 2015
  • Loc: MD

Posted 25 September 2021 - 06:20 PM

How important is the collimation of the SCT used as a collimator

Yes collimation is important.  You can rotate the two scopes relative to each other by 180 degrees and re-test, and get some idea of the collimation errors.  If you use a Newtonian for the collimator, you can set the collimation fairly accurately with a laser collimator.  (Maybe also for an SCT, though I have not tried.)

 

You can test refractors off the axis of the collimator, and avoid the obstruction.   I sometimes use a 16" Newtonian as the collimator, and can test up to 6" refractors with no obstruction.

 

It is very powerful way to test, since you can avoid nearly all seeing effects.

 

The (minor) disadvantage is that you are (probably) testing horizontally whereas scopes are normally used vertically.  So if there is anything loose or dependent on orientation, you might not see it.  You might also miss any thermal issues that happen during normal observations -- optics that pinch in cold weather, etc.


Edited by ngc7319_20, 25 September 2021 - 07:04 PM.


#5 davidc135

davidc135

    Surveyor 1

  • -----
  • Posts: 1,774
  • Joined: 28 May 2014
  • Loc: Wales, UK

Posted 25 September 2021 - 06:22 PM

Coma due to a mis-collimated sct would affect the Ronchi lines and the ke test as well although you'd still have a good idea of SA and zones in the scope to be tested (using Foucault at any rate.)

 

You'd need to be confident of the sct's optical quality if it's the standard by which other scopes are rated. I've used my smooth, better than 1/4 wave 8'' Meade as a collimator for a number of optics I've worked on but It's not good enough for an 8'' planetary Newt that I'd like to be of 1/10 wave standard.

 

If a fast telescope is being tested that can alternatively be used as the collimator and the wave-front viewed at the sct's focus with the advantage of the f/10 light cone.

 

David


Edited by davidc135, 25 September 2021 - 06:25 PM.


#6 TOMDEY

TOMDEY

    Voyager 1

  • *****
  • Posts: 10,323
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 25 September 2021 - 07:26 PM

Yes indeed! A laboratory collimator is really nothing more nor less than a superb telescope with one or more reference sources / patterns / etc. that can be accurately placed at its focal plane. Note that the collimator itself needs not support a wide field. Rather, the system under test is tip-tilted to access its fields to be characterized/certified.

 

There's a chicken/egg issue here: The fundamental principal of any metrology is that the tool shall be far better than the test articles' requirement. If you are contracted to deliver tenth-wave telescopes to the ultimate customers... then you most certainly would want your collimator the be certified and stable to  --- two, five, ten (?!) times better than that. A fallback position is to at least know the collimator's wavefront errors and then (carefully-meticulously) back them out from the collected wavefronts. This is standard practice for ultra-high performance systems... but gets painfully laborious. Virtually all commercial production lenses and telescopes are just quickly tested vs good collimators or interferometers, no backouts applied. (e.g. Zygo Reference flats and spheres are twentieth-wave PV wavefront. If you need better --- you are on the hook to perform absolute cal on them and apply backouts.)

 

PS: Same comments apply to autocollimation flat mirrors ... where the telescope under test sorta acts as its own collimator.    Tom


  • SandyHouTex, Oregon-raybender, MKV and 1 other like this

#7 akulapanam

akulapanam

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,870
  • Joined: 26 Aug 2012

Posted 26 September 2021 - 02:06 AM

That’s helpful. What I really want to be able to do is adjust a setup durning the day without needing extensive space for an artificial star and without spending $100k+ on a Zygo or Phasecam. Definitely willing to get a optical flat or a Shack Hartmann. What seemed appealing about this approach was the ability to use the final imaging train, no need for a beam splitter, and the ability to address tilt.
  • MKV likes this

#8 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,527
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 26 September 2021 - 03:06 AM

For a near null wavefront a PDI like the SPPDI:

https://kerryos.com/

has the advantage that its a common path interferometer and it can also use an external point source laser (laser module with collimator lens removed will suffice for F/8 or slower beams) at the focus of an external collimator. The dominant residual error of such a laser diode is astigmatism which is easily corrected by taking several interferograms as the source is clocked. This astigmatism is usually negligible for F/8 or slower beams. Typically the axial difference between the 2 astigmatic foci for a red laser diode is about 5 microns or thereabouts.

It also has higher spatial resolution than a Shack-Hartmann wavefront sensor.

A direct green laser diode source also works with the SPPDI.

The laser can be either operate in single longitudinal mode or multimode.

An OTT source would use a laser diode coupled to a single mode (for the laser wavelength) fiber.



#9 lylver

lylver

    Viking 1

  • *****
  • Posts: 961
  • Joined: 15 Feb 2017
  • Loc: France

Posted 26 September 2021 - 08:33 AM

Takahashi artificial star test,

alignement2.jpg.4c1da30e1357437ad1b373b5

alignement1.jpg.24aa106cbad7a9253ed2b810

They collimate their triplets

alignement3.jpg.b512dca0f18c106f3c36c6ee

Yes, a mirror may be more convenient.in a dark room/chamber.



#10 SandyHouTex

SandyHouTex

    Fly Me to the Moon

  • *****
  • Posts: 6,173
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 26 September 2021 - 11:20 AM

Takahashi artificial star test,

alignement2.jpg.4c1da30e1357437ad1b373b5

alignement1.jpg.24aa106cbad7a9253ed2b810

They collimate their triplets

alignement3.jpg.b512dca0f18c106f3c36c6ee

Yes, a mirror may be more convenient.in a dark room/chamber.

That’s how Tak America does it.  They used to use a FS-152, which I witnessed, but probably now use a TOA-150.


Edited by SandyHouTex, 26 September 2021 - 11:21 AM.


#11 ngc7319_20

ngc7319_20

    Vanguard

  • *****
  • Posts: 2,407
  • Joined: 25 Oct 2015
  • Loc: MD

Posted 26 September 2021 - 12:23 PM

What seemed appealing about this approach was the ability to use the final imaging train, no need for a beam splitter, and the ability to address tilt.

Yes I agree.  One very nice feature of collimating with a separate scope, is that you get to see the aberrations as they will appear under the stars during normal observation.  To me (as mostly a visual observer) this is sometimes more useful than having the aberrations doubled in a DPAC scenario.  I have also tested cameras in this mode -- you can move the artificial star around the frame and check for collimation errors, camera tilts, spacing issues, etc., in the lab without expending scarce clear nights.

 

Of course, I can see the DPAC scenario doubles the errors, and is better for measuring small errors.  Also you need excellent optics in the collimator scope -- I use Zambuto and Lockwood mirrors for testing.  And even then, you want to be aware of whatever aberrations, zones, etc., are in the collimator scope.



#12 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 2,207
  • Joined: 09 Aug 2016
  • Loc: Ellensburg, WA, USA

Posted 26 September 2021 - 07:11 PM

We used a 10” Meade SCT with a 20 micron pinhole to collimate the RASAs although a 10 micron pinhole would be better. I thought it would be too dim but it was plenty bright. The pinhole is re imaged at the ratio of focal lengths which on this case is 600/2500. The pinhole goes at the focal plane of course. I’ve also used a large finder scope to test guiding on a mount by illuminating the finder cross hairs and point the main finder at the telescope and mount under test. You’ll see the image of the crosshairs greatly enlarged and then you can test that the mount moves via guide pulses which are teeny tiny movements of an arc second or so, but you will be able to see this since the re imaged crosshairs move.

Also Celestron collimates using a pinpoint source and a reference parabola which I think is better but a bit less convenient.

Edited by 555aaa, 26 September 2021 - 07:13 PM.

  • SandyHouTex likes this

#13 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,756
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 26 September 2021 - 08:03 PM

We used a 10” Meade SCT with a 20 micron pinhole to collimate the RASAs although a 10 micron pinhole would be better

Use lensless 1 to 3 mW 3 V DC lensless laser diode 650 nm with a variable potentiometer (rheostat). Gives gives you perfect brightness control and of course it's monochromatic.



#14 akulapanam

akulapanam

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,870
  • Joined: 26 Aug 2012

Posted 26 September 2021 - 09:48 PM

We used a 10” Meade SCT with a 20 micron pinhole to collimate the RASAs although a 10 micron pinhole would be better. I thought it would be too dim but it was plenty bright. The pinhole is re imaged at the ratio of focal lengths which on this case is 600/2500. The pinhole goes at the focal plane of course. I’ve also used a large finder scope to test guiding on a mount by illuminating the finder cross hairs and point the main finder at the telescope and mount under test. You’ll see the image of the crosshairs greatly enlarged and then you can test that the mount moves via guide pulses which are teeny tiny movements of an arc second or so, but you will be able to see this since the re imaged crosshairs move.

Also Celestron collimates using a pinpoint source and a reference parabola which I think is better but a bit less convenient.


That’s awesome!

Out of curiosity how did you make sure the Meade 10” was collimated first? Would be concerned that the RASA would be in better collimation than the SCT and more sensitive do to f ratio. That seems like it would be one advantage of using a newt because you could get spot on with the scope in place. OTOH a SCT is just so much more convenient. Easy to just line up the 9 micron artificial star.


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.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics