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

Collimating the RC - out of the frying pan and into the fire

reflector imaging
  • Please log in to reply
12 replies to this topic

#1 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 16 July 2019 - 12:59 PM

I've finally managed to collimate my GSO 10" RC for imaging.

 

 

it seems like now that the scope is colimated, the focus is soft. Much softer that it was before

 

anyone experience this or have any input as to what to look at next?

 

 

 

 

The 3D collimation view from about 50 frames shows me collimation is the best i've gotten it so far. 

 

collimated 0719.JPG

 

curvature 0719.JPG

 

BUT - now that i have round stars across the field, the telescope doesn't focus as well as it used to. it's just soft

 

Previously I could get FWHM 3, but now FWHM is more like 7-8. 

 

I use automated focusing routines in SGP and have an FLI atlas focuser. I've been using both for quite a while, so i don't think it's an error in focusing procedure. the scope just seems soft.

 

 

 

 

also FYI i've been talking about collimation for a while but I think i have a handle on it now.

 

Collimation was painful and took many nights and hours of experimenting. After using many (all?) of the various tools and techniques (glatter laser collimator, hotech, takahashi, etc.) what worked for me is the real-time collimation tools in CCD Inspector.


Edited by bvalente, 16 July 2019 - 02:40 PM.


#2 TOMDEY

TOMDEY

    Skylab

  • *****
  • Posts: 4268
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 July 2019 - 03:07 PM

There is indeed one optical parameter that could cause that. If the spacing between the PM and SM is not right... that would manifest as spherical aberration (over the entire field, of course). This assumes that the PM and SM are built to match for the nominal designed conjugates and spacing. Recommend that you run through focus and look, specifically for classic spherical aberration (Bright edge with diffuse core on one side of best focus and bright center with diffuse halo on the other side. You then adjust the spacing twixt PM and SM to drive that out... Depending on your metering structure... that could further require that you also re-tune your original field-symmetric adjustments!

 

[Typical Anecdote: years ago I was called-in on a system that manifested problems. I suspected that a prolate ellipsoidal reflector in the imaging chain was not being used at its stigmatic conjugates. Pulled it entirely out and found the spherical-free conjugates out in the lab, and back-solved for spacings in the system. Turned out that the adjustment range on the metering structure did not reach the needed conjugates?! I fixed that with added spacers... put the thing back together --- and it was perfect! The factory rep (who flew in to set us straight) entirely missed that blunder!]

 

Anyway --- I'm convinced your problem, as described, can be fixed by empirically tweaking the PM-SM separation --- to drive out the spherical.    Tom


Edited by TOMDEY, 16 July 2019 - 03:09 PM.

  • Mike Spooner and Timo I like this

#3 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 16 July 2019 - 04:51 PM

Thanks for the tip Tom,  kind of what I was thinking

 

i just did the focal length calculations via astrometry.net and the actual focal length is 1997.921mm vs reported 2000mm. do you think that's enough of a difference to warrant this?

 

 

i'm partly asking because I'm not sure at all with this GSO RC scope if adjusting the PM-SM distance is even possible

 

 

Brian



#4 TOMDEY

TOMDEY

    Skylab

  • *****
  • Posts: 4268
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 July 2019 - 06:39 PM

Thanks for the tip Tom,  kind of what I was thinking

i just did the focal length calculations via astrometry.net and the actual focal length is 1997.921mm vs reported 2000mm. do you think that's enough of a difference to warrant this?

i'm partly asking because I'm not sure at all with this GSO RC scope if adjusting the PM-SM distance is even possible

Brian

Hi, Brian! That depends on the actual conics (vs nominals) of your mirrors, as-built. You describe the image as ~much softer than before~ which indicates that the spacing somehow changed when you adjusted things; and that there is indeed a spacing that will reduce the spherical aberration (and coma) to zero... which is what the Ritchey–Chrétien is all about. (Classical Cass eliminates only spherical, but not coma.) RC optical alignment includes (should include) tweaking the PM-SM spacing, to drive out the spherical aberration.

 

[One of my assignments, at work, was the optical metrologies on Ritchey PMs, SMs and combined alignments and certifications. The PM-SM alignment is most certainly a ~full~ 5-DOF (Five Degrees Of Freedom) of the SM, relative to the PM. Those five adjustments are decenter-x, decenter-y, tilt-x, tilt-y and despace-z (overt or implicit). One way to finesse those is by adjusting Hexapod Struts that meter those two mirrors, relative to each other.]    Tom



#5 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 16 July 2019 - 06:46 PM

Tom - copy that

 

i can calc the focal length for an earlier and better focused image and see what if any is the difference. that should determine it pretty quickly

 

Brian



#6 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 16 July 2019 - 07:18 PM

UPDATE: I calculated the lower FWHM focal length and it's identical. 

 

i'll look into how to adjust PM-SM spacing for the GSO 10" RC

 

Thanks

 

Brian



#7 John Miele

John Miele

    Gemini

  • *****
  • Posts: 3366
  • Joined: 29 May 2005
  • Loc: North Alabama

Posted 17 July 2019 - 11:29 AM

I believe you can either turn all three screws on the secondary mirror or all three screws on the primary by the exact same amount to move the mirrors closer or farther part. But you need to be super careful not to destroy the collimation you just worked so hard to achieve.

 

Just a word of caution...when I tried to adjust the mirror separation distance on my AT8RC it started a downward spiral of mis-collimation that I never fully recovered from. You may be better at it than I am though!

 

Good luck!

 

cs...John



#8 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 17 July 2019 - 12:31 PM

I believe you can either turn all three screws on the secondary mirror or all three screws on the primary by the exact same amount to move the mirrors closer or farther part. But you need to be super careful not to destroy the collimation you just worked so hard to achieve.

 

Just a word of caution...when I tried to adjust the mirror separation distance on my AT8RC it started a downward spiral of mis-collimation that I never fully recovered from. You may be better at it than I am though!

 

Good luck!

 

cs...John

 

Wow that sounds like an even bigger nightmare. Not sure i want to take that step! lol

 

thanks for the input

 

Brian



#9 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 17 July 2019 - 01:35 PM

 One way to finesse those is by adjusting Hexapod Struts that meter those two mirrors, relative to each other.]    Tom

 

Tom i have the carbon fiber tube, so I don't have the struts to adjust. but now that I know it's possible, maybe the truss version is in the cards for next time

 

 

Brian



#10 Timo I

Timo I

    Sputnik

  • -----
  • Posts: 48
  • Joined: 20 Sep 2010
  • Loc: Finland

Posted 18 July 2019 - 12:15 PM

Thanks for the tip Tom,  kind of what I was thinking

i just did the focal length calculations via astrometry.net and the actual focal length is 1997.921mm vs reported 2000mm. do you think that's enough of a difference to warrant this?

i'm partly asking because I'm not sure at all with this GSO RC scope if adjusting the PM-SM distance is even possible

Brian

Tom knows perfectly what he's saying there for the spacing between those mirrors (PM and SM). You have now wrong spacing there and that change is probably caused by your own collimation process and turning those collimation screws in one direction only. If you have only tightened all three screws with your adjustments and you have done that several times in a row, then you have gradually increased the spacing between PM and SM!

Only way of fixing that is turning those collimation screws into the opposite direction (maybe 90 degrees per /screw for starters, but I think you need something like 360 degrees movement per screw to fix that much "curvature" CCDI shows to you now). And each time you have turned all those screws back equal amount, then you need to fine-tune your scope's collimation back to that "collimation centered" position.

 

That change in mirror spacing also causes the difference you see there for the actual focal length vs. scope default focal length. In addition it changes the scope's backfocus distance too.

 

Here's an image how that secndary mirror holder is shown from the side:

https://astrokuva.ga...eilinpidin2.jpg

It helps you to understand what had happened there, when you have turned collimation screws into one direction only. The correct way of doing adjustments is loosing one screw very slighly and then tightening the other two very slightly. That changes the tilt of SM, but not the spacing PM/SM too much. 

 

Here's two CCDI measurements, where the spacing between primary and secondary mirrors in a RCT scope is about correct.

https://astrokuva.ga...on_21092018.jpg

https://astrokuva.ga...results_15s.jpg

 

You should acquire something like 18%...22% CCDI curvature measurements from your RCT scope when the spacing is correct, but even 24%...26% curvature measurements give very good star fields too, if you have a medium sized CCD sensor (like KAF-8300) in your scope/camera combination.



#11 bvalente

bvalente

    Mariner 2

  • -----
  • topic starter
  • Posts: 217
  • Joined: 04 Jan 2011
  • Loc: Los Angeles, CA

Posted 18 July 2019 - 02:38 PM

Thanks Timo - that's super helpful.

 

wow i see you have tons of experience with exactly the RC that i own = the GSO 10" carbon fiber

 

i was very careful to tighten/loosen in pairs, but I did tighten it all down to begin, so i'll have to take this next step

 

 

just to make sure I understand, What I should now be looking at is the amount of curvature (in addition to proper collimation) as an indicator of correct (or at least better) SM-PM distance, correct?

 

 

 

i suspected there was a spring in there, which makes sense to see the pic

 

thanks a million

 

Brian


Edited by bvalente, 18 July 2019 - 02:53 PM.


#12 Chucke

Chucke

    Viking 1

  • *****
  • Posts: 573
  • Joined: 12 Mar 2010
  • Loc: AZ

Posted 18 July 2019 - 02:41 PM

I think that spacing affects spherical aberration so you could use a ronchi at the focus to test for straight lines.  If I am wrong someone can correct me.  I have heard that Planewave supplies a ronchi with their systems.



#13 Timo I

Timo I

    Sputnik

  • -----
  • Posts: 48
  • Joined: 20 Sep 2010
  • Loc: Finland

Posted 19 July 2019 - 01:55 AM


i was very careful to tighten/loosen in pairs, but I did tighten it all down to begin, so i'll have to take this next step

just to make sure I understand, What I should now be looking at is the amount of curvature (in addition to proper collimation) as an indicator of correct (or at least better) SM-PM distance, correct?

Hi Brian, pleased to hear that I could help you a bit.

To your questioning, yeah... you need to get good results for every aspect of that CCDI measurement to get a decent starfield. CCDI's "curvature" measurement flattens out when you get nearer to the accurate collimation point (without mirror spacing adjustments), but the accurate spacing between RCT mirrors will be the limiting factor there too. You can read more info of that from this blog writing:

http://interferomete...retien-gso.html

Scroll down until you can read this:

"Like all Cassegrain Systems, in the Ritchey-Chrétien ist the spherical aberration is dependent on the distance primary to secondary, there is only one optimal distance. At the same time the backfocus (distance focal plane to tube) also changes with that distance."

 

After that you can find a practical table of mirror spacing:

"...distances given are from secondary holder to spider (means I changed the secondary only, inverse to mirror distances):
6mm - 99%
8mm - 99%
9,5mm - 98%
12mm - 93%
14mm - 79%
With decreasing distance secondary to primary, the undercorrection increases. Interesting, there ist no linear behaviour, first it goes slightly off, then rapid. If you ever need to adjust the distance, I advice to use a Ronchi grating, as zone errors can be deceiptive if you plan to use the star test."

 

As you can see, there's also a reference for the usage of Ronchi grating while collimating a RCT scope, so I can verify that here.

But I personally have used only CCDI for those measurements (with averaged test image measurements) and the excellent DSI RCT collimation procedure described in this document:

http://www.deepskyin...ure_Ver_1.0.pdf

Adjust the mirror spacing according that blog quide and then finish the final RCT collimation without changing the mirror spacing too much (possibly with that DSI guide).

Then you should be just fine with your 10" RCT scope's star field ie. no disturbing optical abarrations can seen in your seeing limited test images. Good luck there as that needs finally only very minimal collimation adjustments with those screws to get significant changes into your star images smile.gif




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





Also tagged with one or more of these keywords: reflector, imaging



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics