For years I have always taken dark and light images at temperatures at seasonal ranges in 5 degree increments such as -30, -25, -20, and so on. I typically take light frames in 15 minute exposures for LRGB and 30 minute for Ha, O3, and S2 so the darks are timed accordingly. On occasion, depending on the object such as the Orion Nebula (M42), I might take a series of 2, 5, and 10 minute images to assemble a HDR image but I'll also take matching dark frames to use. I create a library of master darks that usually last the season and reuse accordingly. I'll take massive amounts of bias frames at those same temperatures as I was advised years ago that my camera (STL-11002M) does better with matching temperature bias frames and that seemed reasonable to me and easy enough to take in bad weather or even in the day as my observatory is very dark when closed. Flat frames are acquired using a FlatmanXL panel mounted on my southern wall directly in front of my parked scope and its brightness levels are controlled by software by settings I determined from experimenting. I take dark frames with a minimum of 2 seconds exposure to avoid shutter influence although I admit I've never seen this problem, better to be safe and no harm. Due to using a rotator on my system and the camera not being absolutely centered to the optics I take a series of flats at my required PA, say 90 degrees, and then rotated 180 degrees from there (270) to accommodate E/W exposures. I take a series of 16 flats, 8 each for E/W and create the masters. I usually take anywhere from 50-200 bias frames and darks a minimum of 20 but sometimes as many as 100 if in a stretch of bad weather. Hard drive space is very cheap these days so why not. So now we have what and how I take calibration frames, on to how I use them.
I'm not sure if it's still posted on PI's website but there was a writeup about creating master frames that I follow to this day that listed the settings to use to create master bias, dark, and flats. Actually it is found here
http://pixinsight.com/tutorials/master-frames/index.html. The only exception to the settings is I don't have "calibrate" checked under the bias and dark frame for some reason. I'll have to go back and see what difference there was for me. Most likely it's because I'm already using master frames. Maybe I need to change this. So after creating the master frames I save them to a library for future use. I usually find the flats work for a very long time
if I haven't changed anything in the image trane and the optics haven't been removed for say recoating and so on. Collimation on my RC stays good between mirror re-coatings so that isn't an issue and I don't change the optical path with reducers or barlows although if I did I'd take flats to deal with that.
The only time my bias frames are used is for flat frame calibration. Once they are calibrated I make the master flat, save to the library and continue to make master darks. In calibrating my light frames I use the appropriate master dark (temperature and exposure length) and flat frame then combine using the proper method depending on how many aligned subs I'm combining. I reject images that have obvious issues or low S/N usually by visual inspection first (guiding/focus/cloud interference) and then look at the stats to further cull. I usually reject data that is above 2.5-3 arc seconds. I typically try for deep images taking 10-15 hours luminance and 5 hours each of RGB with 15 minute exposures. That usually gives me a good amount of data to work with. On narrowband I go with 30 minute exposures and usually 15-20 hours per filter. I use ACP Expert so this is automated and can be carried over several seasons if necessary. It may seem excessive but again, when it's automated you simply acquire and process. The target database is controlled by you and each target is given a priority level by you. It looks for the best time (closer to the Meridian) and then picks the target, takes the images (dependant on you time grouping, mine is set to 1 hour blocks) and takes the images, looks again for best target, and continues.
So the point of all this is to see if there are any improvements to my calibration procedures that I can make or flaws that need correcting. I have not used the SuperBias feature as of yet and may try that on my next set of bias frames, actually collecting them now.
Thanks for your time.
Steve
