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

Do Calibration Frames Degrade?

  • Please log in to reply
5 replies to this topic

#1 My 2 Stars

My 2 Stars

    Vendor - "Listen To The Stars" Audio & Sketch Book

  • -----
  • Vendors
  • topic starter
  • Posts: 882
  • Joined: 13 Dec 2015
  • Loc: SE Penna.

Posted 21 April 2017 - 05:56 AM

Many imagers build a calibration library (as I do) and they talk about using the DARKS or BIAS for about six months before reshooting them.

What is the reason for this? Am assuming that it's NOT the library that has changed but the LIGHTS and possibly changes within the camera itself over this time period.


Edited by My 2 Stars, 21 April 2017 - 05:58 AM.


#2 Brett Waller

Brett Waller

    Sputnik

  • *****
  • Posts: 36
  • Joined: 18 Aug 2006
  • Loc: Staunton, VA

Posted 21 April 2017 - 06:08 AM

For most cameras, the dark noise and hot/dark pixels will change slowly over time.  Since the purpose of the calibration frames is to remove both random noise and patterned noise generated by your sensor and the electronics reading it, the calibration frames will only be effective it they represent the noise being produced.  How fast it changes is going to depend on your sensor and, to some degree, how much you use it.  When you start to notice your calibration frames aren't removing all of the noise, then it is time to shoot new frames.

 

In similar fashion, your flat frames remove artifacts generated by your optical system, such as vignetting, or dust on your filters and sensor.  This too will change over time and you will need to reshoot your flats.

 

In neither case do the light frames you take contribute to the need to re-shoot your flats, although certain light frames will more easily highlight defects in your calibration process and make it obvious they need to be redone. 


  • bobzeq25 and My 2 Stars like this

#3 pedxing

pedxing

    Surveyor 1

  • *****
  • Posts: 1914
  • Joined: 03 Nov 2009

Posted 21 April 2017 - 07:08 AM

I will add that because flats are sensitive to orientation and transient things like dust specs, they have to be re-taken much more often than bias or darks.

 

In fact, it is generally accepted as good practice to re-take them every time you modify any part of your optical train, for instance remove/re-orient your camera.



#4 xiando

xiando

    Soyuz

  • -----
  • Posts: 3974
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 21 April 2017 - 07:32 AM

Unless you have a fixed setup in an observatory, the idea of a static library of cal files seems pretty dumb to me. Maybe not so much darks, which tend to be time dependent (given a fixed operational temperature), but definitely with respect to flats, due to natural dust accumulation if nothing else.



#5 james7ca

james7ca

    Skylab

  • *****
  • Posts: 4291
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 21 April 2017 - 08:52 AM

Well, with CMOS cameras the darks and calibration files can change by the day or hour, even with regulated temperature. This happens because the sensor re-calibrates with every change in exposure or gain, or with any reset/repower of the camera. So, using a dark library with a CMOS camera is pretty challenging unless you combine your darks with dithering to help to control the changes that happen over time.

 

IMO, the only way you can get perfectly matched darks with a CMOS camera is to keep the camera running right through the change from lights to darks, but that's often not very practical since you may need to alter exposures for focus or when changing filters or targets. However, I have done sequences where I keep the camera running and just have a "dark" filter so that I can interspace the lights and darks (five lights followed by two darks, then repeat the cycle without stopping the camera or making any changes to its settings). When I do that the darks match exceedingly well and the calibration is very clean.

 

In any case, note that the OP lists only CMOS cameras, which in my opinion require different calibration practices than CCDs.


  • pedxing and bobzeq25 like this

#6 My 2 Stars

My 2 Stars

    Vendor - "Listen To The Stars" Audio & Sketch Book

  • -----
  • Vendors
  • topic starter
  • Posts: 882
  • Joined: 13 Dec 2015
  • Loc: SE Penna.

Posted 21 April 2017 - 03:33 PM

I guess when using SGP Pro, you could run sequence events (for DARKS), in between your lights.

Example,...run ten lights, three darks, ten lights, three darks and so on. Little more work this way but would match better as you suggested.




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.







Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics