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

Flats are over-correcting

EAA DSO Imaging
  • Please log in to reply
15 replies to this topic

#1 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 22 March 2025 - 11:03 AM

I need help with my flats. 

 

I have been capturing and using flats in sharpcap for a long time now. It worked well. But recently I have run into this problem. 

 

I captured flats using both sky illumination as well as my LED panel. I confirm that the white curve in the histogram is in the middle, and sharpcap also says "OK" for histogram. However when I use flats, I get this weird reverse-corrected image. See the first attachment. I'm also attaching the camera settings file that sharpcap saved during this capture.

 

Even weirder: when I disabled darks, the image became much better! The image was almost uniformly lit, as if the flats did correctly correct the exposure. Now I'm really lost! My camera is a QHY 533C (cooled camera, so I have been using its darks for several months now). 

 

The only change I can think of is that I'm using a night owl 0.4x reducer with my C8. This does produce more vignetting than my 0.63x reducer, which is precisely why I need my flats! I do not remember for sure if this problem has started since I started using the new reducer, but I do not believe so.

 

Any insights?

Attached Thumbnails

  • Capture_00001_WithDisplayStretch-reduced.png

Attached Files


Edited by amitshesh, 22 March 2025 - 11:05 AM.


#2 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 22 March 2025 - 11:04 AM

Here is the master flat from sharpcap I used, and the camera settings used for capturing it.

Attached Thumbnails

  • MasterFlat_18_29_51_offset=4.979%-reduced.png

Attached Files


Edited by amitshesh, 22 March 2025 - 11:06 AM.


#3 steveincolo

steveincolo

    Skylab

  • *****
  • Posts: 4,078
  • Joined: 26 Sep 2017
  • Loc: Boulder, Colorado, US

Posted 22 March 2025 - 11:57 AM

According to your settings, you are using darks.  Perhaps they’re not matched?   SharpCap manual states:

 

Note: failure to use a dark frame when capturing the light frames will tend to result in the flat frames over correcting (previously dark areas become brighter than average).

Edited by steveincolo, 22 March 2025 - 11:59 AM.

  • Dynan likes this

#4 hyiger

hyiger

    Fly Me to the Moon

  • *****
  • Posts: 6,535
  • Joined: 05 Sep 2021
  • Loc: East Bay, CA & South East, VA

Posted 22 March 2025 - 12:18 PM

I need help with my flats. 

 

Any insights?

Are you calibrating with bias or flat-darks? For this sensor, bias is easier. 



#5 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 22 March 2025 - 12:35 PM

According to your settings, you are using darks. Perhaps they’re not matched? SharpCap manual states:


Wait, what?

I am supposed to use darks when capturing flats? Sharpcap warns me against it when I try to capture flats and have them enabled!

Besides, won't using darks while capturing flats mean I have to use the same gain as my lights while taking flats?

#6 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 22 March 2025 - 12:38 PM

Are you calibrating with bias or flat-darks? For this sensor, bias is easier.


I do use bias.

#7 steveincolo

steveincolo

    Skylab

  • *****
  • Posts: 4,078
  • Joined: 26 Sep 2017
  • Loc: Boulder, Colorado, US

Posted 22 March 2025 - 12:45 PM

Wait, what?

I am supposed to use darks when capturing flats? Sharpcap warns me against it when I try to capture flats and have them enabled!

Besides, won't using darks while capturing flats mean I have to use the same gain as my lights while taking flats?

No, the quote from the manual says you need to use darks when capturing lights.  



#8 hyiger

hyiger

    Fly Me to the Moon

  • *****
  • Posts: 6,535
  • Joined: 05 Sep 2021
  • Loc: East Bay, CA & South East, VA

Posted 22 March 2025 - 01:02 PM

No, the quote from the manual says you need to use darks when capturing lights.  

Is this a feature of Sharpcap? For my IMX571 sensor cameras (ASI2600 etc) which are the same sensor as the IMX533, I never take flats with a dark library. I do on occasion use bias, flats and darks when live stacking but mostly don't bother since I'm just going to throw it away after a few minutes. 


Edited by hyiger, 22 March 2025 - 01:16 PM.


#9 steveincolo

steveincolo

    Skylab

  • *****
  • Posts: 4,078
  • Joined: 26 Sep 2017
  • Loc: Boulder, Colorado, US

Posted 22 March 2025 - 01:13 PM

My recollection is that for live stacking, SharpCap does allow you to use a bias or a flat-dark when creating a flat master.  I also use bias for flats for my 533-based cameras, due to the low thermal noise. 



#10 rgsalinger

rgsalinger

    James Webb Space Telescope

  • *****
  • Posts: 17,285
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 March 2025 - 01:14 PM

I use bias frames and they seems to work just fine. That's because the latest Sony CMOS chips have extremely small amounts of dark noise. There is no difference that you can measure between a 10 second dark exposure and a bias frame. So, why bother creating a set of dark exposure that, for each filter, match the exposure time of the flats?

 

I noticed that my 90 second dark frames had a median of exactly 1ADU over my bias frames using an ASI6200 last night. So, the real issue with these is getting even illumination, IME.

 

On the other hand, some cameras being sold - the 294 is the prime example - have amp glow. For any camera with amp glow you have to match the darks with the flats. 

 

At the same time, since I use 3 minute exposures, I do capture sets of darks to use in calibrating the lights. I've done the experiment and at 3 minutes there is definitely a visible improvement shown in my calibrated images. I'm not even sure that's necessary in all use cases but I do it. 



#11 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 22 March 2025 - 01:27 PM

No, the quote from the manual says you need to use darks when capturing lights.


Right. I do that. Always have. So that doesn't explain why now removing the dark actually causes the picture to be better corrected with flats.

I will retake my darks and try them the next chance I get.

#12 rgsalinger

rgsalinger

    James Webb Space Telescope

  • *****
  • Posts: 17,285
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 March 2025 - 01:42 PM

Right. I do that. Always have. So that doesn't explain why now removing the dark actually causes the picture to be better corrected with flats.

I will retake my darks and try them the next chance I get.

Well, without the data, my guess would be that the gain and offset were different between the flats and the dark flats. Since, dark flats aren't really necessary, when you removed the "bad ones" the images improved. Again, just a guess.  



#13 artik

artik

    Apollo

  • ****-
  • Freeware Developers
  • Posts: 1,160
  • Joined: 14 Mar 2021
  • Loc: Israel

Posted 23 March 2025 - 05:02 AM

I am supposed to use darks when capturing flats? Sharpcap warns me against it when I try to capture flats and have them enabled!
 

Yes with same settings you take flats with. For example if you take flats with 5ms exposure and 200 gain same should be applied for dark flats.

 

Bias is just an assumption that for "short" exposure there is no other input.

 

Anyway another important point - make sure you reach the focus on the "stars" first before you taking flats. Position of focus affects flats



#14 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 26 March 2025 - 10:12 PM

I think some of the terminology I used on this thread may have been confusing:

 

Before imaging:

 

1. Turn "subtracting darks" off.

2. Turn "Flats" off.

3. Take flats at low gain.

4. Take *dark flats* at the same gain and exposure as flats OR take bias (lately I have switched to bias because that saves me another trip outside to put the lid on for dark flats).

 

5. (Optionally) Retake darks at the same gain and exposure as the lights I intend to take. My camera is cooled, so keeping temperature the same (0 C) is easy.

 

Imaging:

 

1. Turn "subtracting darks" on and ensure that the darks I use are taken at the same gain and exposure time as my lights.

2. Turn "Flats" on to use the master flat above.

 

----

 

I will retake the darks and try this again the next chance I get (busy days and cloudy nights for me right now!)



#15 amitshesh

amitshesh

    Apollo

  • *****
  • topic starter
  • Posts: 1,425
  • Joined: 25 Sep 2020

Posted 26 March 2025 - 10:14 PM

According to your settings, you are using darks.  Perhaps they’re not matched?   SharpCap manual states:

I posted my question on the Sharpcap forum and Dr. Glover suspected the same. Although I cannot think of why my darks have suddenly stopped working, I will retake darks and check all my settings in my next session (whenever that will be!)



#16 Clouzot

Clouzot

    Apollo

  • -----
  • Posts: 1,120
  • Joined: 09 Jul 2018
  • Loc: French Riviera

Posted 27 March 2025 - 03:56 PM

I posted my question on the Sharpcap forum and Dr. Glover suspected the same. Although I cannot think of why my darks have suddenly stopped working, I will retake darks and check all my settings in my next session (whenever that will be!)

One common cause of suddenly mismatched dark is Sharpcap's "Brain" (it should rather be called "Pinky" in that case), which has a tendency to change the offset/backlevel/brightness setting from its default value... without telling you that it did, of course.


  • steveincolo 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





Also tagged with one or more of these keywords: EAA, DSO, Imaging



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics