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

Simple flat testing- is this a valid test?

  • Please log in to reply
4 replies to this topic

#1 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1530
  • Joined: 08 Jan 2016

Posted 17 February 2019 - 04:54 PM

I can't use my telescope at the moment as I'm waiting for a replacement part.

 

I decided to use the time to create a master set of bias, darks & flats for the new camera.

 

I can't take an actual star sub at present, however, I needed a way to test if my bias, darks and flats are up to the future calibration task. 

 

Correct me If I'm wrong or just being stupid but if I calibrate the original flat subs against the masters & bias and dark that I created and the output comes out white as in the screen shot does this donate that your calibration integrations have a good chance of working?

 

The screen shot on the right is the field as captured with the camera raw and the picture on the left is the calibration result when a stack of 13 of the original subs are calibrated then integrated. There is a bit of dark spotting in the calibrated frame as I used a 5 minute dark against a 12 second sub but it might illustrate the point.

 

This is a full calibration using bias, dark & dark flat corrections.

 

Let me know your observations.

 

FLAT_TEST.JPG


Edited by pyrasanth, 17 February 2019 - 05:01 PM.


#2 BenKolt

BenKolt

    Surveyor 1

  • -----
  • Posts: 1740
  • Joined: 13 Mar 2013

Posted 17 February 2019 - 10:33 PM

If I followed your description properly, then I think, yes, this is a modest test of the "goodness" of your master flat.  It sounds like you are treating your individual subs as "light" frames and calibrating them as you would normally do.  But of course, in the absence of modulated light frames, your individual flats ought to calibrate out as completely flat.

 

One thing you can do beyond this, just out of curiosity, is to take a frame difference of two uncalibrated flats vs. two calibrated flats to see if there are any residual artifacts present as well as to see how much noise you may be introducing through the calibration process.  Take two images, add a pedestal to one of them, then subtract the other.  The pedestal is there to keep the difference from clipping to zero.  Then, look at the difference image and also measure the standard deviation of the difference image.  Compare this before and after the calibration process and see how they differ.

 

A more thorough check would be to make a Flat Field Photon Transfer Curve (FFPTC), which is admittedly more involved, but it would be a nice exercise to determine the quality of your flat and help determine how many subs to include in your flats.

 

Question: Are you incorporating the bias frame as part of your calibration process?

 

Ben



#3 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1530
  • Joined: 08 Jan 2016

Posted 18 February 2019 - 05:33 AM

Thanks Ben,

 

It is good to know I'm on the right track.

 

I got into the habit of a full calibration workout as this came from the necessity to do this when I had the Atik 11000 camera and the old habit has never left me.

 

This is my process. My camera is set to over-scan so I calibrate all the subs for bias & dark frames with nothing selected other than the over-scan to remove the over-scan area. note- temperature is -20 C for the whole process.

 

1. Take 300 bias subs and integrate into a master bias- I do this for 1x1 and 2x2 bin

2. Take a full set of dark frames 15 of each at all the exposure, temperature & binning I use. I have 5,10,20,30 & 60 minutes 1x1 and 2x2 bin.

3. Calibrate the dark frames with the bias masters and then integrate into dark masters

4. Take a set of flats for all the filters I use 50 subs each and then take a set of dark flats (30 subs) which match the exposure of the flats

5. Integrate the dark flats into dark flat masters per filter

6. Calibrate all the filter flat subs with the dark flat master per filter then integrate the calibrated flats into a flat master file per filter.

7. Calibrate your captured image frames using your bias, dark master (must match exposure of image) and master flat for that filter.

 

This seems to work okay and is a full calibration workout. I don't really do anything else or feel the need to take shortcuts.


Edited by pyrasanth, 18 February 2019 - 06:03 AM.

  • AIP likes this

#4 spokeshave

spokeshave

    Surveyor 1

  • *****
  • Posts: 1862
  • Joined: 08 Apr 2015

Posted 18 February 2019 - 07:58 AM

I think this is only an indication that your individual flats are consistent. It just shows that the individual flat frames are close enough to the same such that when corrected by the master, the vignetting is nulled. It does not necessarily mean that the master flat will properly correct lights. That would be like saying dividing a number by itself and getting 1 proves that the number is 4. 

 

tim



#5 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1530
  • Joined: 08 Jan 2016

Posted 18 February 2019 - 08:04 AM

I think this is only an indication that your individual flats are consistent. It just shows that the individual flat frames are close enough to the same such that when corrected by the master, the vignetting is nulled. It does not necessarily mean that the master flat will properly correct lights. That would be like saying dividing a number by itself and getting 1 proves that the number is 4. 

 

tim

This is understood. I have a few lights which were captured before the mount broke. These lights were captured when the preflash was set, which I'm no longer using, so I can't properly tell how effective the calibration masters will be but they seem to correct most of the vignetting which is what I'm working to solve. When I get the mount fixed I will be able to test properly.




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