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

Spike coming off bright star... Normal?

This topic has been archived. This means that you cannot reply to this topic.
22 replies to this topic

#1 Footbag

Footbag

    Cosmos

  • *****
  • topic starter
  • Posts: 9,115
  • Joined: 13 Apr 2009

Posted 28 January 2014 - 12:19 AM

I'm just getting the first images from my STF-8300m, and I noticed a spike coming off the right side of the brighter stars. It happens 2x2 binning mode, it doesn't seem to appear in 1x1. I think it's blooming, but I'm not sure. Is it normal and to be dealt with in post processing?

Or do I have to do something to mitigate it now?

Posted Image

#2 PGW Steve

PGW Steve

    Apollo

  • -----
  • Posts: 1,359
  • Joined: 03 Oct 2006

Posted 28 January 2014 - 01:45 AM

Yes, that is typical 8300 binning bloom.

#3 Hilmi

Hilmi

    Vendor-Astronomical Solutions Co.

  • *****
  • Vendors
  • Posts: 5,638
  • Joined: 07 Mar 2010

Posted 28 January 2014 - 01:50 AM

Yep, its very normal. My STT-8300 gives me 4 diffraction spikes on my refractor images. Only thing that gives it away is that the spikes are always perfectly aligned with the pixel grid.

#4 Rick J

Rick J

    Cosmos

  • *****
  • In Memoriam
  • Posts: 8,376
  • Joined: 01 Mar 2008

Posted 28 January 2014 - 01:52 AM

That happens with that chip when binned with the gain setting SBIG uses. The read out register isn't bloom protected. Since it is designed to read out horizontally rather than vertically like imaging cells the bloom goes sideways. I know of no cure other than either shorter exposures or don't bin in camera. Binning later is essentially the same if the exposure times are sky limited to keep read noise a non issue.

I have the same issue with my STL-11K but they designed it with an adjustable pot. that can adjust the gain such that the register doesn't overflow when binned. I don't think your camera has this option that I'm aware of. Contact SBIG to see if they have a cure I've not seen.

Rick

#5 Footbag

Footbag

    Cosmos

  • *****
  • topic starter
  • Posts: 9,115
  • Joined: 13 Apr 2009

Posted 28 January 2014 - 02:01 AM

OK. Thanks for the quick info. I'm clouded out after only 5 frames, but at least I got to use it.

#6 Footbag

Footbag

    Cosmos

  • *****
  • topic starter
  • Posts: 9,115
  • Joined: 13 Apr 2009

Posted 28 January 2014 - 10:06 AM

Ok. So I've been pondering this as I slept.

It seems like the best solution would be not to bin. Is that what most 8300 users do?

#7 Ron (Lubbock)

Ron (Lubbock)

    Apollo

  • -----
  • Posts: 1,019
  • Joined: 17 Aug 2012

Posted 28 January 2014 - 12:43 PM

I don't bin any more with my Atik 383 L+ mono. Even when the stars aren't obviously bloomed, 2x2 binning produces a very slight asymmetry in the stars. I got tired of people saying "your scope is out of collimation." The answer is to bin 1x1 and take long enough subs so read noise is less of a factor. How long those subs should be will depend on sensor temperature, the nature of the target, light pollution, and your optics.

One exception is when I am doing quick exposures to focus, or to center a faint target. I use 2x2, 3x3 or 4x4 binning to get bright images quickly. So the binning feature is not useless.




#8 elbee

elbee

    Viking 1

  • *****
  • Posts: 780
  • Joined: 02 May 2009

Posted 28 January 2014 - 01:27 PM

i don't get any spikes like that on my ST-8300 and i always bin 2x2 for RGB data. i suggest you contact SBIG to discuss any adjustment that may be possible. the STF model is newer than my older ST but i think in most respects the design is very similar.

#9 Hilmi

Hilmi

    Vendor-Astronomical Solutions Co.

  • *****
  • Vendors
  • Posts: 5,638
  • Joined: 07 Mar 2010

Posted 28 January 2014 - 01:45 PM

It could be that your telescope is not saturating the stars which would explain why you don't see the spikes

#10 elbee

elbee

    Viking 1

  • *****
  • Posts: 780
  • Joined: 02 May 2009

Posted 28 January 2014 - 02:56 PM

could be, but i don't believe so...i have saturated values in all images. i simply have never seen any blooming issue in my ST-8300. i get very tiny blooms in my STL-11000M all the time with LUM bin 1x1. here is an "analysis" from PI of one of my sets of RGB data with the ST-8300. there are no little spikes on any stars.

Attached Thumbnails

  • 6341224-blink analysis.jpg


#11 elbee

elbee

    Viking 1

  • *****
  • Posts: 780
  • Joined: 02 May 2009

Posted 28 January 2014 - 03:23 PM

if you want to be bored, here is an animation of the unprocessed FITs files. (i particularly like the alien creature in the first frame)

Posted Image

#12 Hilmi

Hilmi

    Vendor-Astronomical Solutions Co.

  • *****
  • Vendors
  • Posts: 5,638
  • Joined: 07 Mar 2010

Posted 29 January 2014 - 12:30 AM

Hmm.. my binned images show some very minor blooming in the form that looks a bit like diffraction spikes, but it only shows on my F6 refractor. on my f8 RC it is difficult to tell if the spikes are there because they are hidden by real diffraction spikes.

#13 Dan Crowson

Dan Crowson

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2,932
  • Joined: 08 Oct 2010

Posted 30 January 2014 - 08:33 PM

I'd add that I've seen these kind of blooms on my ST-8300M if I bin luminance. I shoot a lot of a red lp zone so it isn't unusual for me to see backgrounds in the 15,000's or more.

Dan

#14 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6,925
  • Joined: 20 Jul 2007

Posted 30 January 2014 - 09:21 PM

could be, but i don't believe so...i have saturated values in all images. i simply have never seen any blooming issue in my ST-8300. i get very tiny blooms in my STL-11000M all the time with LUM bin 1x1. here is an "analysis" from PI of one of my sets of RGB data with the ST-8300. there are no little spikes on any stars.


I don't see a max of 65,535 with any sub, so they are not saturated.
BLueman

#15 elbee

elbee

    Viking 1

  • *****
  • Posts: 780
  • Joined: 02 May 2009

Posted 31 January 2014 - 12:31 PM

I don't see a max of 65,535 with any sub, so they are not saturated.


the numbers from the PI analysis are misleading. ccdstack and maximDL display values = 65535 "everywhere". the PI analysis numbers were easy to gather and display so i used them but i don't know what they are doing that never results in values of 65535. below is another example of RGB bin2x2 images. 15min subs on M94 using my TEC160FL @ f/7. believe me, there are saturated stars and galaxy core on all of these images (i combined these with 3min subs to produce a final image). again, PI shows no result = 65535 max but ccdstack and maxim do.

my only reason for getting involved in this thread was the numerous comments of "oh yeah, that's normal for 2x2 binning on a KAF-8300 chip". i wanted to add my data point that i don't see it as normal at all and if someone's KAF-8300 chip is "bleeding out" on 2x2 binning they should have a conversation with their camera manufacturer. i certainly would.

Posted Image

#16 Footbag

Footbag

    Cosmos

  • *****
  • topic starter
  • Posts: 9,115
  • Joined: 13 Apr 2009

Posted 31 January 2014 - 12:49 PM

Well, I did E-mail the manufacturer. Their response was that it was typical of KAF-8300 sensored cameras.

In most CCDs, the Horizontal Shift Register (HSR) and Output Register (OR) have a higher capacity than an individual pixel, so when binning larger than 1×1 you can maximize the dynamic range. Unfortunately this is not true for the KAF-8300 sensor. Kodak does not specify the capacity of the horizontal readout registers but we know from testing they will overflow, bloom, when collecting the charge from a binned super pixel.

Your sample image showing blooming when binned is typical for the 8300 sensor.
To eliminate blooming capture all images full resolution 1X1.



#17 elbee

elbee

    Viking 1

  • *****
  • Posts: 780
  • Joined: 02 May 2009

Posted 31 January 2014 - 12:51 PM

guess i got lucky and i should keep my 8300 camera. it seems to be special. :-)

#18 SL63 AMG

SL63 AMG

    Viking 1

  • *****
  • Posts: 922
  • Joined: 21 Dec 2009

Posted 02 February 2014 - 05:27 PM

Interesting topic. Perhaps Rick is correct, it's an SBIG issue.

I just took this image last night using an FLI ML8300 & 12.5" RCOS, binned 2x2 900s Luminance.

I don't see any blooming in any of the subs.

Posted Image

#19 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6,925
  • Joined: 20 Jul 2007

Posted 02 February 2014 - 06:37 PM

I have seen side blooming with the ML8300, but it was with saturated stars, rather large ones too. But I am not a real fan of binning anyway so I never worried about it.
BLueman

#20 jerryyyyy

jerryyyyy

    Surveyor 1

  • -----
  • Posts: 1,590
  • Joined: 06 Oct 2011

Posted 02 February 2014 - 11:10 PM

I do not bin on my STT-8300M. No downside I see and can always clean up later... better to start with more data than less?

#21 Eddie R

Eddie R

    Vostok 1

  • *****
  • Posts: 104
  • Joined: 02 Jul 2011

Posted 03 February 2014 - 11:35 AM

I have the same problem with bright stars and the Sbig 8300m, it's blooming. Blamed it on a bump in the tracking at 1st then kept an eye on my tracking graph to elimitate that as a possibility. Always happens to the left or to the right and tracking is spot on below seeing levels. See the attached uncalibrated image of a cropped Ha frame of IC1396. As you see, this even happens with narrowband imaging, color filtered frames are much worse. I've not found a workaround for this yet. When it shows up really well I'll just clonestamp them out in PI or Photoshop.

Attached Thumbnails

  • 6353061-Capture.JPG


#22 Eddie R

Eddie R

    Vostok 1

  • *****
  • Posts: 104
  • Joined: 02 Jul 2011

Posted 03 February 2014 - 11:40 AM

Really bad on bright stars like in M45. Here is a stack of blue frames with only 5 minute exposures. Any longer and the spikes are out of control.

Attached Thumbnails

  • 6353069-Capture.JPG


#23 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6,925
  • Joined: 20 Jul 2007

Posted 03 February 2014 - 01:29 PM

I have the same problem with bright stars and the Sbig 8300m, it's blooming. Blamed it on a bump in the tracking at 1st then kept an eye on my tracking graph to elimitate that as a possibility. Always happens to the left or to the right and tracking is spot on below seeing levels. See the attached uncalibrated image of a cropped Ha frame of IC1396. As you see, this even happens with narrowband imaging, color filtered frames are much worse. I've not found a workaround for this yet. When it shows up really well I'll just clonestamp them out in PI or Photoshop.


This is the same as the blooming I saw using 2x2 binning with the ML8300. Richard Crisp explained that the 1000x anti blooming was vertical only and the horizontal registers would bloom.
Blueman


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