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

Help with Darks & Flats

  • Please log in to reply
39 replies to this topic

#1 Cey42

Cey42

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 17 Aug 2020
  • Loc: Kirkland, WA

Posted 15 October 2020 - 02:31 PM

I am new to EAA and I am trying to figure how to use flats and darks. I have tried several times and read various posts and I still cannot figure out what I am doing wrong.

 

My set up is a Celestron C6-S GT XLT with a 6.3FR using a ZWO 294MC camera.

 

I have made sure that I am applying the flats and dark at the same resolution, exposure and gain.

 

Here is what I am seeing. This is an image with no flat or dark applied. I modified the display histogram to make the vignetting more obvious. This is live view, no stacking.

1-Before.jpg

 

 

Now I apply the flat.

2-FlatOnly.jpg

As expected, the vignetting is greatly improved.

 

 

 

Now I add the dark. This is both the dark and flat on live view.

3-Flat+Dark.jpg

The vignetting looks a little worse.

 

 

 

Now I try live stacking with both the dark and flat applied with 16 seconds subs. This what I get.

4-Stack.jpg

 

Horrible vignetting.

 

 

Any idea on what am I doing wrong?

 

 

Thanks in advance.



#2 Forward Scatter

Forward Scatter

    Viking 1

  • -----
  • Posts: 887
  • Joined: 22 Jul 2018
  • Loc: Wandering the PNW

Posted 15 October 2020 - 04:22 PM

HI Cey

How did you do your flats? You mention "applying the flats and dark at the same resolution, exposure and gain". That goes for darks, but flats are a different story. 

 

Typically flats are done by getting your image train into focus in its final orientation, aiming at a blank piece of sky with a white t-shirt over the front of the scope, or by using a light box. The resolution and binning should be identical to what you would be imaging with. After bringing up the histogram in SC, turn down the gain to 0 and adjust the exposure until the white peak is centered approximately at the 50% point of the histogram. The R, G & B peaks don't need to be adjusted (ignore them) as the flats are done in monochrome. adjusting the RGB setting on the flats before capture may result in all sorts of colorful vignetting. Once the white peak is above the 50% mark, then click the "Start" on the flat capture box. BTW...if you are using a filter for imaging, that filter should be in place for the flats. Not having the filter in place for the flats or the converse will also create weirdness you are seeing.

 

I got a USB-powered LED white light tracing box for my flats. It works pretty well. Mounted it to a free 1 gallon plant pot (6" wide with the end cut off!) that slips over the front of the scope. Works like a charm.

 

Cheers



#3 Cey42

Cey42

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 17 Aug 2020
  • Loc: Kirkland, WA

Posted 16 October 2020 - 03:48 PM

The darks are done at same resolution, exposure and gain as what I did for viewing. Bias is turned off when creating darks.

 

For the flats, I used the white t-shirt method.

 

No filter used in either the flats or darks or imaging.

 

After doing more investigation I discovered two potential issues. According the SharpCap user manual "The FITS file produced can be opened by suitable software." This implies I should have FITS darks but when I looked at darks they were PNG. Not sure if that is the problem but I am recreating my darks as FITS.

 

The second potential issue is my flats. According to this CN post, they recommend using 2 second plus exposures when using the 294. I definitely was not doing them that long. So now I need to redo my flats.

 

 

 

 

 



#4 Forward Scatter

Forward Scatter

    Viking 1

  • -----
  • Posts: 887
  • Joined: 22 Jul 2018
  • Loc: Wandering the PNW

Posted 16 October 2020 - 04:25 PM

Interesting. I wonder if the problems the poster was having was only a funkiness with their camera.  Can you the histo to the 50% point without issues?

Once I did see the extreme vignetting you had when first using my 533; it turned out the Red & Blue sliders were each set to "auto" when I did that flat. Other have had issues when the gain was set to auto when doing flats and darks.



#5 garyhawkins

garyhawkins

    Ranger 4

  • *****
  • Posts: 312
  • Joined: 16 Feb 2019
  • Loc: San Diego

Posted 17 October 2020 - 09:35 AM

Hi Cey,

 

I store my darks and flats as .png files, this should not be a problem.

 

According the SharpCap manual, for flats these should be done at low gain and the exposure set to bring the histogram towards 50%.  The 'white t-shirt' method works fine if you're pointing the scope at constant illumination - sky with no sun.

 

A filter should be in place for darks and flats if you use it during you EAA imaging session.

 

Also, once you've captured darks and flats the imaging hardware cannot be moved/changed (with the exception changing filters with a filter wheel).

 

CS Gary

 

 

The darks are done at same resolution, exposure and gain as what I did for viewing. Bias is turned off when creating darks.

 

For the flats, I used the white t-shirt method.

 

No filter used in either the flats or darks or imaging.

 

After doing more investigation I discovered two potential issues. According the SharpCap user manual "The FITS file produced can be opened by suitable software." This implies I should have FITS darks but when I looked at darks they were PNG. Not sure if that is the problem but I am recreating my darks as FITS.

 

The second potential issue is my flats. According to this CN post, they recommend using 2 second plus exposures when using the 294. I definitely was not doing them that long. So now I need to redo my flats.



#6 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 17 October 2020 - 09:49 AM

On the subject of darks, I would disagree with the suggestion that you need to match the use of a filter with whether one is used for your lights - in my experience it makes no difference. I have taken darks with or without filters with similar results -  the parameters that you need to match for darks are camera gain, temperature, exposure and binning.  Dark characteristics are determined by the sensor's parameters, not the optics.

 

For flats it is true that you can't alter the imaging configuration, e.g. rotate camera or change the scope,  but for darks it makes no difference if you rotate the camera or change the scope. My dark library for each camera is based on just the above 4 parameters (gain, temp, binning and exposure) and I use the same dark library for each scope.


  • Starman27, BinoGuy and garyhawkins like this

#7 garyhawkins

garyhawkins

    Ranger 4

  • *****
  • Posts: 312
  • Joined: 16 Feb 2019
  • Loc: San Diego

Posted 17 October 2020 - 10:26 AM

Hi Errol, Cey

 

Those are good points relating to darks, which, as Errol's say's only characterize the sensor.  Thus for darks don't worry about the filter or changing the optical chain in front of the senor, it does come down to only gain, temp, binning and exposure.

 

CS Gary

 

On the subject of darks, I would disagree with the suggestion that you need to match the use of a filter with whether one is used for your lights - in my experience it makes no difference. I have taken darks with or without filters with similar results -  the parameters that you need to match for darks are camera gain, temperature, exposure and binning.  Dark characteristics are determined by the sensor's parameters, not the optics.

 

For flats it is true that you can't alter the imaging configuration, e.g. rotate camera or change the scope,  but for darks it makes no difference if you rotate the camera or change the scope. My dark library for each camera is based on just the above 4 parameters (gain, temp, binning and exposure) and I use the same dark library for each scope.


  • alphatripleplus likes this

#8 GaryShaw

GaryShaw

    Viking 1

  • *****
  • Posts: 653
  • Joined: 28 Nov 2017
  • Loc: Massachusetts and Wyoming

Posted 17 October 2020 - 11:51 AM

Only issue I’ve noted is the need for more exposure and/or higher gain when using some filters, hence the need, in my case anyway, to do darks based on filters in use at the time. It’s easy anyway since I have a ‘dark’ filter in the wheel.
Gary


Edited by GaryShaw, 17 October 2020 - 11:52 AM.


#9 DSO_Viewer

DSO_Viewer

    Apollo

  • -----
  • Posts: 1,129
  • Joined: 03 Jan 2016

Posted 17 October 2020 - 12:04 PM

Only issue I’ve noted is the need for more exposure and/or higher gain when using some filters, hence the need, in my case anyway, to do darks based on filters in use at the time. It’s easy anyway since I have a ‘dark’ filter in the wheel.
Gary

How could filters make any difference for your darks when taking the darks there should be absolutely no light entering the scope to the sensor??

 

Steve



#10 donstim

donstim

    Mariner 2

  • -----
  • Posts: 289
  • Joined: 13 Aug 2018
  • Loc: Seattle Area, WA USA

Posted 17 October 2020 - 07:02 PM

One other parameter that needs to be held constant for the lights, darks, and flats when using SharpCap with ZWO cameras is the flip setting.


  • roelb likes this

#11 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 17 October 2020 - 08:31 PM

Interesting that SharpCap requires that. In ASILive and Starlight Live the flip settings are not used for darks as far as I can tell using those programs.



#12 donstim

donstim

    Mariner 2

  • -----
  • Posts: 289
  • Joined: 13 Aug 2018
  • Loc: Seattle Area, WA USA

Posted 17 October 2020 - 10:21 PM

They should be. If you flip the image, the hot pixels will be flipped as well.


  • mikenoname likes this

#13 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 17 October 2020 - 10:39 PM

I've checked my recent darks in ASILive and the amp-glow in the saved fits dark files seems to be always in the same location regardless of the flip settings on the ASILive screen. So I'm pretty certain the flip setting is not used in the same way that SharpCap makes use of it. I'm not sure why that dependency exists in SharpCap, as the physical orientation of the sensor is independent of any flip setting.

 

The absolute position of hot pixels and amp-glow on the physical sensor is independent of any flipping that the viewing software on the user's screen interface makes.



#14 donstim

donstim

    Mariner 2

  • -----
  • Posts: 289
  • Joined: 13 Aug 2018
  • Loc: Seattle Area, WA USA

Posted 17 October 2020 - 11:17 PM

Obviously, the flip setting is applied by SharpCap before flat/dark correction. The orientation of vignetting and hot pixels changes, and if not aligned with the flip setting used for the lights, the adjustment will be incorrect. 

 

I assume you haven't used the flip setting in SharpCap with flats/darks, or you don't need much of an adjustment. It is very evident, especially with flats, in my Celestron 8" with 0.63x reducer and ZWO ASI 294MC Pro camera if I forget to use a consistent flip setting. FYI, I use a flip setting of "both" when not using a star diagonal and "horizontal" when using a star diagonal to maintain a north up image. So, I am frequently changing the flip setting.

 

https://forums.sharp...opic.php?t=1470


Edited by donstim, 17 October 2020 - 11:29 PM.

  • roelb likes this

#15 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 18 October 2020 - 07:21 AM

Yes, I don't use the flip setting in SharpCap or I would have been aware of this behaviour. My point is that in other examples of EAA software, such as ASILIve and Starlight Live,  the application of dark subtraction is done before  any orientation change (such as a rotation or flip) that the user chooses on the screen is applied to the frame. As a result, the user does not need to worry about flip orientation when taking darks with ASILive or Starlight Live.

 

I have no idea why SharpCap was implemented to apply dark or flat subtraction to light frames after a flip setting is applied to the frame - as noted the other EAA software that I have used is not implemented in this way.

 

Given this difference between different EAA software, you are correct that it is important to be aware of this issue when using SharpCap and the flip feature. 



#16 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 18 October 2020 - 07:42 AM

So to  get back to the OP's issue with SharpCap, he should probably check the flip orientation on his light frames vs the darks and flats he used. I think this is unlikely to be the issue for him - If the flip orientation was different in the calibation frames, one would expect to see evidence of amp-glow from the darks  in a couple of the corners, and I don't see that. 



#17 GaryShaw

GaryShaw

    Viking 1

  • *****
  • Posts: 653
  • Joined: 28 Nov 2017
  • Loc: Massachusetts and Wyoming

Posted 18 October 2020 - 09:14 AM

How could filters make any difference for your darks when taking the darks there should be absolutely no light entering the scope to the sensor??

 

Steve

Filters impact the exposure time you’ll need for your lights. Darks need to be made using the exposure you’ll use on the lights. Hence the need for using a filter impacts how you need to make your darks.



#18 DSO_Viewer

DSO_Viewer

    Apollo

  • -----
  • Posts: 1,129
  • Joined: 03 Jan 2016

Posted 18 October 2020 - 12:02 PM

Filters impact the exposure time you’ll need for your lights. Darks need to be made using the exposure you’ll use on the lights. Hence the need for using a filter impacts how you need to make your darks.

How in the world does filters determine your exposure time when there is no light entering? The darks are for dark current noise only and you just match the darks exposure time with your lights exposure time along with other parameters. This is why using a filter or not using a filter when taking darks are not an issue.

 

Steve



#19 Cey42

Cey42

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 17 Aug 2020
  • Loc: Kirkland, WA

Posted 18 October 2020 - 02:53 PM

 Can you the histo to the 50% point without issues?

Jay, can you explain more what you mean by this?

 

 

Here are answers and comments to questions asked since my last reply.

 

Red and Blue sliders are not set to Auto. They have a value of 50. Gain not set to auto.

 

Good to know that PNG/FITS is not an issue for the darks.

 

I don't own a filter (it is on the way). Thanks for info about needing to have it in when taking flats and planning to use it during the session. I agree that it should not make a difference on the darks.

 

The imaging train was not changed after capture of darks/flats.

 

I don't use the flip setting so that is not an issue.

 

Given the answers above I don't think I have done anything "wrong." Maybe I did by accident and don't know it. I will try again (making sure I do all the steps properly) the next time I get some clear skies.

 

 

A potential issue is the camera temperature. It was stated above for darks, one needs to match camera gain, temperature, exposure and binning.The 294MC is not a cooled camera. Thinking about when I took the darks/flats and when I was doing the imaging, there probably was a 15F drop in temperature. Could this cause the problem? Does temperature matter for the flats?

 

 

Thanks for all the input



#20 Cey42

Cey42

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 17 Aug 2020
  • Loc: Kirkland, WA

Posted 18 October 2020 - 04:57 PM

I think I have a good idea on what may have happened. I ran the following experiment.

 

Given the perfectly grey and drizzly day here in Seattle, I pointed the scope out the window, added a white t-shirt and captured a flat. Then I added another t-shirt and started live stacking. Here is what I got.

 

First.PNG

 

Only some slight vignetting. Basically as expected.

 

 

 

Then, I twisted the camera about 1/4" to 1/2". I wanted to simulate the image train moving. I wondered how much I would have to move it to notice it in the live stacking. Here is that image.

Twist.jpg

 

This looks a little like my OP image on live stacking. I was surprised that it only took a very small twist.

 

My new theory is I may have bumped the image train or it was not tight. On my next night out, I am going to make sure it is tight and I don't bump it.

 

Thank you all for your help.

 

 

 

 


  • BinoGuy likes this

#21 Forward Scatter

Forward Scatter

    Viking 1

  • -----
  • Posts: 887
  • Joined: 22 Jul 2018
  • Loc: Wandering the PNW

Posted 18 October 2020 - 05:02 PM

Jay, can you explain more what you mean by this?

 

 

Here are answers and comments to questions asked since my last reply.

 

Red and Blue sliders are not set to Auto. They have a value of 50. Gain not set to auto.

 

Good to know that PNG/FITS is not an issue for the darks.

 

I don't own a filter (it is on the way). Thanks for info about needing to have it in when taking flats and planning to use it during the session. I agree that it should not make a difference on the darks.

 

The imaging train was not changed after capture of darks/flats.

 

I don't use the flip setting so that is not an issue.

 

Given the answers above I don't think I have done anything "wrong." Maybe I did by accident and don't know it. I will try again (making sure I do all the steps properly) the next time I get some clear skies.

 

 

A potential issue is the camera temperature. It was stated above for darks, one needs to match camera gain, temperature, exposure and binning.The 294MC is not a cooled camera. Thinking about when I took the darks/flats and when I was doing the imaging, there probably was a 15F drop in temperature. Could this cause the problem? Does temperature matter for the flats?

 

 

Thanks for all the input

Hi Cey

To quote the CNer posting on that you linked to :

 

"- that camera is known to behave badly when using short exposures (the ones we typically use when shooting flats). There are posts in the AP section explaining the nitty-gritty details (all of which I'm not sure I want to understand), but bottom-line is:

     (1) try to target a sub-exposure longer that 2 seconds, minimum. That means using multiple layers of t-shirts, fabric, a cheap solar filter, or one of those car window tinting films for a perfect 80s look..."

 

My question was whether you are seeing the same bad behavior with you 294, specifically when trying to get the histo peak to 50% by adjusting the gain and exposure needed to record flats.



#22 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 118,240
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 18 October 2020 - 06:33 PM

I think I have a good idea on what may have happened. I ran the following experiment.

 

 

 

My new theory is I may have bumped the image train or it was not tight. On my next night out, I am going to make sure it is tight and I don't bump it.

 

Thank you all for your help.

Please let us know if things works out, as I'm sure a few people are curious.



#23 Cey42

Cey42

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 17 Aug 2020
  • Loc: Kirkland, WA

Posted 18 October 2020 - 09:44 PM

I will post an update the next time I get out. Unfortunately the forecast does not have any clear nights until Saturday.

 

My plan is to capture two flats. One with a low gain and short sub as suggestion in the SharpCap manual. Another with a sub over 2 seconds. Then make sure there is no change in the image train.

 

I am curious to find out what the issue is.


  • alphatripleplus likes this

#24 dmanuel

dmanuel

    Sputnik

  • *****
  • Posts: 29
  • Joined: 17 Jul 2020
  • Loc: Knoxville, TN

Posted 19 October 2020 - 01:00 PM

I'm also considering capturing some flats, but am still unclear on a couple of things:

1) Using a cooled camera, do I need to cool to the same temperature for flats as I do for darks?

2) I see that the image train needs to be left unaltered from the time the flats are taken until they are used.  I usually remove my FR/Spacers/Camera and store them separately.  So I'm guessing my only options are to either a) start storing my optical system intact or b) take new flats during the day before an evening that I'll be imaging.

 

Thoughts?

David



#25 BinoGuy

BinoGuy

    Viking 1

  • *****
  • Posts: 705
  • Joined: 11 Jul 2014
  • Loc: Michigan

Posted 21 October 2020 - 06:28 PM

I read that many people grab flats in the AM as they are packing up.  I keep mine train assembled.  Not only does this make my rig more of a G&G but I worry about the impact of repeated disassembly; my extension tubes are metal but the threads are not exactly super duper robust.

 

 

Clear skies, BG   {°¿°}




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