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

Star test stacking-- bad stack?

  • Please log in to reply
19 replies to this topic

#1 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 19 March 2023 - 06:07 AM

I am not sure where to place this question, but since the technique is one that is used for planetary, maybe this is okay here. Feel free to move it as needed.

 

So I have a star test video that has frames with steady seeing interspersed with mushy frames, and no matter how I try, the resulting image is much worse than the better frames. (See attached) Is there a way to make sure the stacked image reflects only the better frames? I tried quality > 90% or best 95% frames, and still trash :(

 

Or is the video useless? LOL

Attached Thumbnails

  • Screenshot 2023-03-19 190152.jpg


#2 dcaponeii

dcaponeii

    Fly Me to the Moon

  • *****
  • Posts: 6,763
  • Joined: 01 Sep 2019
  • Loc: Waxahachie, TX

Posted 19 March 2023 - 12:05 PM

Use just a single AP surrounding the image.  Stack the percentage that matches the inflection point of the quality graph (this sometimes is well below 50% of the quality.  Don't drizzle.  You are too far from focus to tell anything from your pattern in any case.  You image planets at focus and you need to collimate at focus.  Try using Metaguide as well.


  • Kim2010 likes this

#3 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 19 March 2023 - 12:14 PM

Thank you! This is not a collimation test though. It is a star test, so outside and inside of focus images.



#4 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 21,172
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 19 March 2023 - 12:16 PM

Using Christmas lights? :lol:

I have no idea what the solution is, I defer to others.
  • Kim2010 likes this

#5 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 19 March 2023 - 12:25 PM

Using Christmas lights? lol.gif

I have no idea what the solution is, I defer to others.

LOL! You know what is frustrating?! There long intervals of PERFECT seeing that night where I can see a perfect Airy and diffraction rings in perfect collimation at 667x LOL (3mm). But suddenly the image is disturbed by firework like turbulence! frown.gif


Edited by Kim2010, 19 March 2023 - 12:26 PM.


#6 RedLionNJ

RedLionNJ

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7,154
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 19 March 2023 - 12:28 PM

Thank you! This is not a collimation test though. It is a star test, so outside and inside of focus images.

I'm with Don, sorry. The ultimate test of optics is an in-focus star. If that's good, there's no need to worry about anything else.


  • Kim2010 likes this

#7 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 19 March 2023 - 12:44 PM

In focus is very fine :) Makes me happy LOL :)

 

But I just want to check the patterns with Suiter's book.



#8 dcaponeii

dcaponeii

    Fly Me to the Moon

  • *****
  • Posts: 6,763
  • Joined: 01 Sep 2019
  • Loc: Waxahachie, TX

Posted 19 March 2023 - 05:29 PM

For planetary you’ve got everything backwards. You get close with the patterns in the book but to achieve the resolution for planetary the ONLY thing that matters is collimation IN FOCUS. Even if those screw positions do not yield an ideal out of focus pattern inside or outside focus. This is why Metaguide is such a valuable resource. Unless and until we see an in focus star image you’ll not get agreement from me that your collimation is adequate.

#9 Kokatha man

Kokatha man

    James Webb Space Telescope

  • *****
  • Posts: 17,328
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 19 March 2023 - 06:35 PM

A couple of points: I presume Kim wants to examine for optical aberrations with the intra & extra-focal patterns but as far as collimation is concerned, if you have an "ideal pattern" 4 or 5 rings out from focus, then for all intents and purposes your scope will be collimated.....focusing at this stage should reveal an Airy disk, but this is something that can be very difficult to detect, hence my point here.

 

In our C14 we rarely "see" the Airy disk and almost solely rely upon the defocused star at 4-5 rings - I cover this aspect briefly in the website notes of our website. (see link in signature below)

 

It does take a bit of experience when examining an onscreen star to appraise it correctly and that is why Frank's Metaguide can be a useful tool for many, but we have no problem with the approach of using 3-5 rings outside focus for a high degree of collimation and subsequent optical resolution...such that imaging Iapetus & it shadow transits of Saturn last year...or planetary outcomes in general for a very long time now! wink.gif

 

I did wonder like Don whether you might have used multi alignment points for stacking a short star video and agree that one should only use a single alignment point if using the video stacking approach. (something we rarely do, relying on the onscreen appraisal a few rings out from extra-focus, accepting that our scope's optics are generally a "warts & all" given...except for the other day which I'll chronicle the next time we collect some images! lol.gif )

 

But looking at your image and leaving aside those "divisions" (for want of a better description that had me thinking you might've used multi-points) the Poisson Point does look slightly skewed, with the inner rings around it more illuminated on the left-hand side than the right...

 

To achieve  the collimation I've referred to the inner rings should all be evenly-illuminated on both sides of the P.P. which I don't "think" I can see in your image. (this isn't absolute, hence the italics, but it does appear that way in the image to me!)

 

The slightly bluish central region is meant to be a feature of white light images due to the differing diffraction of red & blue, but I have never witnessed it as such when collimating in white light personally...

 

KimsStar.jpg

 

 


  • Asbytec likes this

#10 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 21,172
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 19 March 2023 - 07:18 PM

 

The slightly bluish central region is meant to be a feature of white light images due to the differing diffraction of red & blue, but I have never witnessed it as such when collimating in white light personally...

 

attachicon.gifKimsStar.jpg

Interesting. I *may have* noticed it defocused on Arcturus in my 6" MCT. I thought it was quite beautiful and sketched it. 

 

color.jpg

 

Kim, what scope were you using. 


Edited by Asbytec, 19 March 2023 - 07:19 PM.

  • Kim2010 likes this

#11 Kokatha man

Kokatha man

    James Webb Space Telescope

  • *****
  • Posts: 17,328
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 19 March 2023 - 11:01 PM

ps: You asked <"Is there a way to make sure the stacked image reflects only the better frames?">

 

I have never found AS!3 terribly good at stacking defocused stars tbh...the old Registax5 was best imo for stacking these but when seeing creates a lot of mis-shapen images, what you are trying to discern if you are collimating in poorer conditions is to "see the tree in the forest" so to speak to determine what adjustments you make.

 

Many find this too difficult and a good reason to use Metaguide, but it can be learnt and this is the way we do it always.

 

What you could do is hand-pick the best-looking frames if you take a video to stack because it is these that will be what you are looking for and should reflect the true state of your collimation: going through a few hundred frames might sound daunting but if you really want to see the resultant stack it will often be the best approach..! lol.gif

 

As to your: <"You know what is frustrating?!There long intervals of PERFECT seeing that night where I can see a perfect Airy and diffraction rings in perfect collimation at 667x LOL (3mm). But suddenly the image is disturbed by firework like turbulence!"> 

 

The view through an ep always looks much better imo btw - I mentioned in my earlier post here about collimating problems where - after changing the secondary screws - the secondary mirror came off its central pivot point...which I was unaware of until I went to collimate with the camera - it was producing abhorrently skewed star patterns and meant I had to dis-assemble the secondary to fix the problem...

 

Not wanting to go to the trouble of setting up the camera, laptop etc even though I was certain I had fixed the problem, I used ep's to check and make initial collimation adjustments...as mentioned I will expand upon this the next time we collect images (in the possibility that someone might like to know a bit more grin.gif ) and of course imagers should always collimate "in camera" - but getting everything back into the ballpark so to speak was easier with an ep - but they are much more forgiving at the 100x and then 350x I employed than "in camera" appearances onscreen. wink.gif

 

Btw we've been at this game for a while now wink.gif and I have never seen "PERFECT" seeing at anytime! lol.gif


  • Kim2010 likes this

#12 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 20 March 2023 - 04:06 AM

Folks,

 

I think this post has been misread. This is NOT about collimation. This is simply posted here because I thought the techniques used in stacking for planetary is useful for stacking stellar images in and out of focus. And I was asking if there was a way to stack a video of star images that are interspersed with mushy images and excellent moments in one video.

 

That is all. Anyway, I will need to ignore all posts that are veering off topic. 

 

Thank you!


Edited by Kim2010, 20 March 2023 - 04:06 AM.

  • RedLionNJ likes this

#13 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 20 March 2023 - 04:09 AM

ps: You asked <"Is there a way to make sure the stacked image reflects only the better frames?">

 

I have never found AS!3 terribly good at stacking defocused stars tbh...the old Registax5 was best imo for stacking these but when seeing creates a lot of mis-shapen images, what you are trying to discern if you are collimating in poorer conditions is to "see the tree in the forest" so to speak to determine what adjustments you make.

 

Many find this too difficult and a good reason to use Metaguide, but it can be learnt and this is the way we do it always.

 

What you could do is hand-pick the best-looking frames if you take a video to stack because it is these that will be what you are looking for and should reflect the true state of your collimation: going through a few hundred frames might sound daunting but if you really want to see the resultant stack it will often be the best approach..! lol.gif

 

As to your: <"You know what is frustrating?!There long intervals of PERFECT seeing that night where I can see a perfect Airy and diffraction rings in perfect collimation at 667x LOL (3mm). But suddenly the image is disturbed by firework like turbulence!"> 

 

The view through an ep always looks much better imo btw - I mentioned in my earlier post here about collimating problems where - after changing the secondary screws - the secondary mirror came off its central pivot point...which I was unaware of until I went to collimate with the camera - it was producing abhorrently skewed star patterns and meant I had to dis-assemble the secondary to fix the problem...

 

Not wanting to go to the trouble of setting up the camera, laptop etc even though I was certain I had fixed the problem, I used ep's to check and make initial collimation adjustments...as mentioned I will expand upon this the next time we collect images (in the possibility that someone might like to know a bit more grin.gif ) and of course imagers should always collimate "in camera" - but getting everything back into the ballpark so to speak was easier with an ep - but they are much more forgiving at the 100x and then 350x I employed than "in camera" appearances onscreen. wink.gif

 

Btw we've been at this game for a while now wink.gif and I have never seen "PERFECT" seeing at anytime! lol.gif

 

Just ask Norme. We live in the tropics. Perfect seeing can be common here in some places :)



#14 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 20 March 2023 - 04:16 AM

Interesting. I *may have* noticed it defocused on Arcturus in my 6" MCT. I thought it was quite beautiful and sketched it. 

 

attachicon.gifcolor.jpg

 

Kim, what scope were you using. 

 

This is with the 8SE, Norme. I think the blue is just an artifact of the moments when some sort of turbulence hits the video. The in-focus image was beautiful though with an Airy disc and prominent first ring and faint outer rings. I was floored seeing an in-focus star images look like a textbook Fresnel pattern on the 8SE! I am used to seeing it on the 4SE though :) But for the 8? I finally did the last few weeks. I can't wait to bring the 8SE in La Union where maybe every night would be like that perfect night in NCR! :)


  • Asbytec likes this

#15 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 21,172
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 20 March 2023 - 07:52 AM

Just ask Norme. We live in the tropics. Perfect seeing can be common here in some places smile.gif

Like Chaz...in Florida. :)



#16 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 28,757
  • Joined: 01 Feb 2006

Posted 20 March 2023 - 04:07 PM

I'm with Don, sorry. The ultimate test of optics is an in-focus star. If that's good, there's no need to worry about anything else.

Actually, this is not a very meaningful test unless you have a known perfect sample to compare too and even then, it fails to tell you very much about a variety of errors. It is useful for astigmatism and comatic curve in a mirror but little else.  A comatic hill in the objective might make a small bulge in the ring, and Astigmatism could make several with the cross being the most severe kind, but this is not an error that we see much of. 

 

The eye is not very good at judging small differences in brightness. Suppose there is some spherical (SA) aberration in the system (and this is the most common "severe" error in most reflectors and compound scopes.).  SA will make the first ring brighter, but in reflectors and compound scopes, that ring is already pretty bright because of the diffraction cased by the secondary mirror.  If there there is 1/6th wave of SA, the ring, already being quite bright from the obstruction, will not appear so much brighter that it is easy to see unless you have another perfect instrument to compare it to and even then, it may be difficult to see. The eye is just not a very sensitive radiometer when it comes to brightness and a star is a very high contrast object, so the subtle glow from SA or other errors is not easy to perceive because even if the background brightness is raised by 5%, this is not really enough to jump out at you.     

Next, zonal errors, rough optics, and a turned edge are all very difficult to see on an in focus image.  These errors will draw light from the spurious disk and spread it out around the star, but unless the errors are quite severe, it can be difficult to see this because even the normal sky glow can hide it unless you are under extremely dark skies. That damage will be there though, and on a planet, every point on the planet will scatter that light around that point lowering contrast.

 

Seeing an in focus star with a round first diffraction ring does tell you that there is no coma or major astigmitism, but it leaves you clueless about these other errors, all which are easy to see and estimate severity using an out of focus star.

 

(People used to day double stars were a good test, but this is even more useless.  In the presence of SA, the spurious disk actually gets smaller and this can make a scope with meaningful SA split closer doubles than a scope with no SA, thought the chances of finding a double that would be "just right" to show this are rather small, but the point is that even with meaningful SA, a scope can still split most doubles that the perfect scope could split.  The exception would be if you compared two otherwise identical scopes but one with SA on a "Just right" double, the brighter first diffraction of the scope with severe SA could make a star that sits just under the maxima of the first ring harder to see in the SA scope than in the perfect scope, but once again, this is a Goldie Locks kind of chance to find just the right star in either of these cases.

 

I don't want to discourage anyone from doing whatever they are most comfortable with, but an in focus star is not really telling you much about the overall precision of the optics. Even a so-so telescope can produce nice diffraction pattern.  If you are happy with your scope and use if mostly for just looking at diffraction rings, then that is all that matters I suppose. If though, you want to really know about the quality of the optics, more advanced tests are going to tell you a lot of things that the in-focus star test won't tell you.     


  • Asbytec, KpS and Kim2010 like this

#17 Kokatha man

Kokatha man

    James Webb Space Telescope

  • *****
  • Posts: 17,328
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 20 March 2023 - 05:40 PM

Just ask Norme. We live in the tropics. Perfect seeing can be common here in some places smile.gif

 

Like Chaz...in Florida. smile.gif

Is this a bit like "everything is bigger in Texas"...sorry, but I don't buy either..! :rofl:

 

But if it really is true, why don't you take up planetary imaging like your neighbor Joe Ferret - at times he does very well, but with your confidence you should really excel! grin.gif

 

Meanwhile, back at the ranch.....



#18 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 21,172
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 20 March 2023 - 07:49 PM

I understand Kim means "perfect" seeing is actually very good seeing being regularly Pickering 8/10 or better, and "near perfect" like Pickering 9 or even 10/10 seeing at times. It is not a once in a lifetime experience. Even in the best seeing he and I describe, there are lesser and better moments. That is particularly evident as very fine planetary detail softly rolls in and out of view. But it's not perfect in the pure sense of the word, it just looks perfect compared to other areas where seeing is about average most times and better other times. Especially in the aperture ranges from 4 to 8" we are both using. The seeing is often good enough to very good so we consider it "perfect". He's right, there is an area up north where seeing is amazing on a regular basis. The image above, I suspect, was taken in Manila where seeing is more variable. 


Edited by Asbytec, 20 March 2023 - 07:52 PM.

  • Kim2010 likes this

#19 Kim2010

Kim2010

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,806
  • Joined: 21 Sep 2010

Posted 21 March 2023 - 07:38 AM

"That is particularly evident as very fine planetary detail softly rolls in and out of view" -- sums it up nicely!

 

Yup, that video is taken in Manila-- but what was frustrating was that there were those moments of excellent seeing, but seems like some turbulence hits it at times. I am not sure if it was because of I am at the rooftop and an AC is behind me LOL. But yup, even in this near-Manila place, I saw moments of almost perfect seeing. 667x (3mm) on Arcturus with a steady Airy disc and bright first ring and outer rings in focus! Wow. And Antares looked liked a mini solar system with the double like a small planet near its "sun" and the diffraction rings like orbits. (The crude sketch I sent you, Norme and also Eddgie :)).



#20 KpS

KpS

    Ranger 4

  • -----
  • Posts: 385
  • Joined: 07 Nov 2009
  • Loc: Prague, Czechia

Posted 21 March 2023 - 10:25 AM

I solved a similar problem when trying to evaluate the wavefront quality of my optics using Roddier's method. The key is to properly evaluate the quality of each frame of video. The Gradient or Laplace method used by Autostakkert will certainly pick out as the best just those frames damaged by firework like turbulence. It worked well for me to sort the images by the value of the mean/sigma fraction.


Edited by KpS, 21 March 2023 - 05:01 PM.

  • Kim2010 likes this


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