›

Flat Darks or Bias in WBPP - questions to your Calibration Frames videos

Pschemek Pierscinski
in PixInsight
Hi Adam
I reviewed in detail your 2 "must see" videos about the calibration frames and have a few understanding questions for my own process. 
As background: I use a modern OSC astro cam (ZWO 2600 MC Pro). I had set up a library with Dark frames for various combinations of Exposure / Gain at my usual sensor temperature of -10 Celsius and I always make Flats with a dimmable light Panel with the same fixed exposure of 1s. I also had set up a library of Bias Frames of the same exposure as my Flats (1s) so technically they are Flat Darks if I understood your explanation. However they are all named Bias as coming out of my AsiAir. 

Question 1: given my Bias frames are technically like Flat Darks, should I put them into Bias Tab or the Darks Tab in the WBPP? Currently I put them into Bias Tab - is this a problem? In your video you put Flat Darks into Darks Tab, together with your Darks
Question 2: does it matter what name my Dark Flats have? They are named automatically Bias by AsiAir so will it be a problem if I put them into the Darks Tab in WBPP without changing name into Darks?
Question 3: given I am able to always match my Darks with my Lights (same exposure, gain and temperature), does Dark Master Optimisation make any sense? My understanding is that - because of the matching - it would not make any difference.
Question 4: you talk in your videos about exposure and temperature, but I did not hear you talk about Gain in context of Calibration Frames. Does Gain matter, and if so for which Frames? My understanding is that Gain matters for Darks if you want to match them with your Lights.  Maybe also for the Flats but what about Bias or Flat Darks? 

Big thanks for your perspective on these points and kudos for the Fundamentals Course which I am enjoying a lot

Comments

  • BlockHeadBlockHead
    HI Pschemek,

    1. Technically no matter what you call them- they are subtracted from the light frame. So functionally it doesn't matter where you put them. BUT...in terms of what they *mean* - a Bias frame has two purposes. It can be used as a "dark" frame that doesn't have much dark current (which is just the bias level) and you can use it as a tool for removing the bias of darks when you want to scale the dark current. This latter purpose no longer really makes sense for CMOS cameras! (not in a straightforward way) There is one reason to maintain the semantics of biases and darks in PixInsight. The logic of WBPP is that if there exists a matching dark frame it will always use this first. If there is no matching dark frame it will use a bias (a frame you put in the bias section).

    2. Naming can have an effect. However, the keywords for image type found in the header take precedence over file naming conventions. So WBPP will read the headers first..then it will try to figure out image type by looking at the filename. Putting "Master" in a file name (or path) will make any file into a master unless you uncheck this feature in WBPP. So it is the IMGTYPE in the header that matters...not the file name. I suspect the ASIAIR will be setting the IMGTYP as well as the filename (at least it should). I would suggest not trying to do an end-run around the "system" and put things in the proper place.

    3. No, Dark Optimization FOR CMOS does not make sense. However to answer your question- yes, applying optimization to CCD darks of the *same* time was a thing. This is because temperature fluctuations and other electronic stuff meant that it is sometimes possible to find a better estimate of the dark current between your light frames and darks even if they had the same exposure time. Usually this difference was so small it did not matter. However, say you are using an uncooled camera. You might have the same exposure time ...but some minor fluctuations in temperature.

    4. I did not talk about GAIN because I made the videos *before* that was a thing to worry about. CMOS cameras were just starting to come out. It was important to address the idea of matching darks for the stupid cheap CMOS cameras that required it. These were a minority of cameras...but so many entry level users were buying them and not understanding how to calibrate their data. It goes without saying that you need to always match the GAIN setting off all calibration data with light frames.

    -the Blockhead   
  • Pschemek Pierscinski
    Thank you Adam for your very thorough answers and your thoughts on this. Much appreciated!

    Based on yours here's what I will adjust in my process:
    • for my well-behaved CMOS - stop using Bias frames as such and focus only on FlatDarks for my Flats
    • use Asiair to shoot and label those FlatDarks properly both in the Fits header and as file name to avoid any confusion within WBPP
    • put the FlatDarks together with my Darks in the same WBPP tab, counting on the script to match automatically the FlatDarks with my Flats and Darks with Lights respectively. Leave the Bias tab blank
    • ensure Gain to be the same for all my calibration frames
    Hope that this is in line with your recommendation? Thank you!


  • BlockHeadBlockHead
    For that first point...
    If you have a well behaved CMOS... then you CAN use a single BIAS frame to calibrate all of your Flats.
    That is my recommendation. You do not need to acquire darks to match each exposure time. That is just a waste of time and effort (and files).

    The third point is correct if you ignore my point above.

    The fourth point is golden.

    -the Blockhead 
Sign In or Register to comment.