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

Reflection Artifacts in the ASI6200

  • Please log in to reply
457 replies to this topic

#51 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 12:18 AM

The halo is still definitely there, but its the type of thing that if you hadn't mentioned it would never had noticed. Those donuts pre-processing on the other hand.......... Its too bad about the ZWO 6200MM, hopefully they will correct this - you cannot be the only one with this issue. Great FOV!!

That is fair. I am sure most would not even pick up on it. I am the star Nazi as Tolga likes to call me, so I see these things. lol.gif

 

The FOV is perfect for a full frame chip on a lot of targets. That is why I bought this scope. 1720mm FL with plate solving. Great optics, really good! The thermal system is something that takes some time to get dialed in, its not set and forget like the RCOS or Planewave smart systems are, but its a pretty good scope.

 

Better be, it was not cheap. ;) 


  • GeneralT001 likes this

#52 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,730
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 13 September 2020 - 01:51 AM

I still see square halos on every star... obviously the ones with the bad window i see a circle in it... i wonder if he's running square filters? if you look at every image around stars with spikes, the star glow is square regardless of the other artifacts. I can't unsee it now.

The "square halos" in Richard Crisp's ML8300 report are probably caused by the usual microlens diffraction reflecting off the sensor coverslip.  The Kodak KAF-8300 sensor is well known for microlens diffraction although at the time of writing that report it wouldn't have been recognised as such.

 

Mark


Edited by sharkmelley, 13 September 2020 - 02:09 AM.

  • rockstarbill likes this

#53 xthestreams

xthestreams

    Explorer 1

  • -----
  • Posts: 66
  • Joined: 18 Feb 2020
  • Loc: Melbourne, Australia

Posted 13 September 2020 - 03:06 AM

Am I to conclude that I should be staying away from the 6200 fir the time being? (12” RC)


  • rockstarbill likes this

#54 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 03:09 AM

ZWO made an incorrect claim that the problem is with the optics. So I guess we'll need to source more repros of this problem.

Edited by rockstarbill, 13 September 2020 - 03:09 AM.


#55 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 03:09 AM

Am I to conclude that I should be staying away from the 6200 fir the time being? (12” RC)


I would, yes.

#56 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,730
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 13 September 2020 - 03:22 AM

ZWO made an incorrect claim that the problem is with the optics. So I guess we'll need to source more repros of this problem.

ZWO must be aware of microlens diffractions - these donut reflections arranged in a grid are quite well known.  Here's just a couple of illustrative examples:

https://www.cloudyni...-new-edge-hd-8/
https://www.cloudyni...nd-bright-star/

 

It just requires some simple arithmetic (like I did in post #2) to deduce which reflective surface is involved - cover slip, camera window or filter. 

 

Mark


Edited by sharkmelley, 13 September 2020 - 03:24 AM.

  • rockstarbill and RossW like this

#57 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 06:15 AM

ZWO must be aware of microlens diffractions - these donut reflections arranged in a grid are quite well known.  Here's just a couple of illustrative examples:

https://www.cloudyni...-new-edge-hd-8/
https://www.cloudyni...nd-bright-star/

 

It just requires some simple arithmetic (like I did in post #2) to deduce which reflective surface is involved - cover slip, camera window or filter. 

 

Mark

I relayed that information to them. Appreciate your time looking at this and the analysis. 



#58 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 13 September 2020 - 07:20 AM

Here are two stacks of the OIII data, one was done with Drizzle and one without.

 

https://1drv.ms/u/s!...2tBTCQ?e=u9P4Cb

 

The offset halo effect I can still see, and see much better than before. The donut grid is not present. 

Bill,

 

Looking at stars from corner to corner there is something not right (either tilt or collimation) and that is likely a major contributor to the offset halo you are seeing.  It looks like some softness on the edge of stars that causes the light to scatter in that direction.  It's easier to see on smaller stars than the bigger brighter offenders.    Have you 1) put another of your large sensor cameras on this scope to verify the CURRENT state of collimation to rule out even a tiny bit of miscollimation?  (Using a sensor that you know to be tilt-free).  2) Verified that the CURRENT tilt adjustment you have made with the CTU is valid on another scope such as the GTX to ensure there is not something compounding the issue you are having specific to the AGO? 

 

It is good to see that the donut issue didnt reveal in the integration, although this is not definitive considering there are not really any very bright stars in the field.   I couldnt see the big donuts in your single sub for this target, not like gamma cyg. 

 

The airy pattern doesnt seem too bad in the integration either, which is a good sign. 



#59 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 07:41 AM

I am well aware of the field aberration and I can assure you it is tilt. I cannot move the Gerd any further to fix the portions of the frame that are still not squared. I plan to refactor my spacing logic to move the Gerd closer to the camera, which will allow me to fully correct the field. The collimation of the scope is excellent. Both myself and Tolga reviewed it together (thanks Tolga! :) ), and we made a few adjustments prior to any of this data here for PacMan being taken. The Wizard data was before the collimation changes. Those changes were rather minor anyhow, and the more important one was a fix to the spider to get the spikes to be more clean. Dall Kirkham telescopes are not difficult at all to collimate. Especially since their secondary is spherical. 

 

I would not have this field aberration problem if the camera was made with a sensor not skewed so terribly. It sucks, but for this project I had to make a tradeoff on wasting more nights on turning screws on the Gerd or getting the field close enough and getting some data. After all, we don't buy telescopes for endless nights of troubleshooting. :) 

 

I have compiled another set of masters, one Drizzled and one not Drizzled with the entire OIII data collection. The previous masters were 33 frames, and this is 153 frames. I do want to tell folks that the newest version of PixInsight is a much better performer than the previous, even on Windows. To have 153 frames in Drizzle finish in 12 minutes is pretty solid performance over what I have seen in the past on the same system, which was about 20 minutes.

 

The new masters are in the same share as before. 

 

https://1drv.ms/u/s!...2tBTCQ?e=u9P4Cb



#60 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 13 September 2020 - 07:44 AM

Well, until the tilt can be addressed, either by moving the CTU closer to the sensor or some other method, the jury will still be out on whether the offset halo issue is a noticeable issue with well corrected imaging train. 



#61 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 07:46 AM

Well, until the tilt can be addressed, either by moving the CTU closer to the sensor or some other method, the jury will still be out on whether the offset halo issue is a noticeable issue with well corrected imaging train. 

It will not be relevant, in my view. The on-axis portion of the frame confirms that. 

 

The bigger problem, is going to be the effect that occurred on Gamma Cyg, which this thread should probably shift its focus toward. 


Edited by rockstarbill, 13 September 2020 - 07:47 AM.


#62 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,730
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 13 September 2020 - 07:55 AM

The airy pattern doesnt seem too bad in the integration either, which is a good sign. 

Can you clarify what you mean by the Airy pattern?  The rings around Gamma Cyg are not the Airy pattern because the radius of each successive zero in the Airy function increases by less than 1.22*wavelength*f_ratio/2 

 

For H-alpha(656nm) at f/5.6 this means the spacing of successive rings in the Airy pattern is less than 2.2 microns which is less than the pixel pitch i.e. they would not be resolved in an image.

 

Mark


  • rockstarbill likes this

#63 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 08:07 AM

Have you 1) put another of your large sensor cameras on this scope to verify the CURRENT state of collimation to rule out even a tiny bit of miscollimation?  (Using a sensor that you know to be tilt-free).  2) Verified that the CURRENT tilt adjustment you have made with the CTU is valid on another scope such as the GTX to ensure there is not something compounding the issue you are having specific to the AGO? 

I should probably respond directly to these two questions:

 

1. The size of the sensor for the sake of collimating a DK is irrelevant. Its secondary is spherical, so its not like an RC where you have to play the secondary tilt and primary tilt dance to correct the far reaches of the image. Directly centered on-axis performance of even a tilted sensor should be immune to tilt entirely. 

2. The CTU adjustments on the GTX would NOT be a good test because tilting the camera will adjust the overall spacing. The DK scope here is like an FSQ insofar as spacing is concerned. Once the camera is at the focal plane and well-focused, you are spaced correctly. The corrector does not move. Its in a fixed position in the donut hole of the primary mirror. 


Edited by rockstarbill, 13 September 2020 - 08:08 AM.


#64 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 13 September 2020 - 08:12 AM

Can you clarify what you mean by the Airy pattern?  The rings around Gamma Cyg are not the Airy pattern because the radius of each successive zero in the Airy function increases by less than 1.22*wavelength*f_ratio/2 

 

For H-alpha(656nm) at f/5.6 this means the spacing of successive rings in the Airy pattern is less than 2.2 microns which is less than the pixel pitch i.e. they would not be resolved in an image.

 

Mark

Sorry mark.  The "airy disk like" pattern.  I was just using the descriptor to specify which issue I was making that statement about as there are several being discussed in this thread.


  • sharkmelley likes this

#65 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 08:14 AM

Sorry mark.  The "airy disk like" pattern.  I was just using the descriptor to specify which issue I was making that statement about as there are several being discussed in this thread.

I think you mean the ((o)) pattern? The concentric rings appearing on bright stars? 



#66 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 13 September 2020 - 08:15 AM

I think you mean the ((o)) pattern? The concentric rings appearing on bright stars?


That is correct.

#67 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 13 September 2020 - 08:18 AM

That is correct.

That seems to be the cost of doing business as it appears across a lot of different cameras and chips and filters. I do wonder though if it would be minimized by a larger NB width, like 5nm or 8nm. The same effect, including the matrix of donuts seen in the Gamma Cyg frame do not occur even on a L frame at the same exposure time. 


Edited by rockstarbill, 13 September 2020 - 08:18 AM.


#68 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 13 September 2020 - 08:21 AM

That seems to be the cost of doing business as it appears across a lot of different cameras and chips and filters. I do wonder though if it would be minimized by a larger NB width, like 5nm or 8nm. The same effect, including the matrix of donuts seen in the Gamma Cyg frame do not occur even on a L frame at the same exposure time. 

Thats a good point about wider band-pass. 

 

I recently ordered a set of 8nm Chroma filters which I will be using with the 6200.  As I also image with a large aperture long focal length scope (Native EDGE 9.25) I'll do some testing when everything arrives to see if I can reproduce this/these issues.  Will be a few weeks though I am sure before I have everything.



#69 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,730
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 13 September 2020 - 02:21 PM

The bigger problem, is going to be the effect that occurred on Gamma Cyg, which this thread should probably shift its focus toward. 

There have been threads on this subject before, regarding the ASI1600:

 

It looks like I contributed to the second thread with the interim conclusion that it's some kind of diffraction effect because the ring spacing varies with wavelength.

 

It gets us no closer to the exact cause though.

 

Mark



#70 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 15 September 2020 - 07:05 PM

I've seen the effect on the 1600 before but not the huge matrix that appeared on the HA test image I shared.

#71 RossW

RossW

    Mariner 2

  • -----
  • Posts: 265
  • Joined: 15 Jun 2018
  • Loc: Lake Biwa, Japan

Posted 15 September 2020 - 09:26 PM

I've seen the effect on the 1600 before but not the huge matrix that appeared on the HA test image I shared.

It's quite common when imaging a bright source at long focal lengths (e.g. ASI1600+EdgeHD1100@f/10).

Attached Thumbnails

  • H_master_50_subs-Edge1100-stretch-1000.jpg

  • sharkmelley likes this

#72 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 15 September 2020 - 09:54 PM

The 1600 has a known slip cover coating deficiency. So I'd expect to see some strange behavior on that. The Sony chip slip cover is coated so I wouldn't expect to see that. I've done similar shots with a different camera and that pattern was not present.

#73 bugbit

bugbit

    Vendor - Octopi-Astro

  • *****
  • Vendors
  • Posts: 203
  • Joined: 28 Apr 2015
  • Loc: west of seattle

Posted 15 September 2020 - 10:06 PM

This reminded me of a pic I had put away awhile back.

 

01_Dot_Array.jpg

 

 

Made from reflection from the sensor.

 

tilter.png


  • sharkmelley, ezwheels and RossW like this

#74 rockstarbill

rockstarbill

    Cosmos

  • *****
  • topic starter
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 15 September 2020 - 10:45 PM

Yeah, the Starlight Xpress rig for tilt correction. ;) 



#75 bugbit

bugbit

    Vendor - Octopi-Astro

  • *****
  • Vendors
  • Posts: 203
  • Joined: 28 Apr 2015
  • Loc: west of seattle

Posted 15 September 2020 - 11:12 PM

Exactly. I found it interesting that a bright concentrated light hitting a cmos at a angle would react like that so it gives me cause to think that too high of intensity and/or a tilt in the incoming light is the prime suspect for those out of focus ghosts to appear.Maybe the AR coating needs to be improved or the exposure duration lessened. BTW, that item we discussed is almost done.




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