I’ve been struggling a bit lately with my QHY9 and getting quality results. Some of these have been driver issues (I accidentally cracked the screen on my netbook which usually runs my system, so switched to my other laptop, and something isn’t installed quite right). I’ve come to two conclusions in the last day or so, and they revolve around the importance of quality flats.
I built a lightbox a while ago for flats, and while its big and bulky, it works well. I didn’t realize just how well until I was lazy and took some sky flats on my setup, and could not figure out why my narrrowband images were “wonky”. I re-ran everything using the lightbox, and well, the results speak for them selves. Here’s the Oiii and Ha subs (onlyu 90 mins and 80mins respectively) before and after flats.
Here’s a great example of the difference in detail level you can achieve with proper flats. The images on the left are with skyflats, and the images on the right are with the flatbox. This is with the auto STF stretch in pixinsight applied. (The images are from the central area of NGC 2264). Ha on top, Oii on bottom. The right side images clearly have way more detail and much less vignetting. In all cases, dark flats have been applied in order to account for the read noise and sensor noise.
You may not be able to see the full difference in the screenshot as the jpg compression plays a bit of havoc, but you can see (moreso in the Oiii data) that vingetting is mostly gone, and there are no harsh gradiants. Once I have more data, this should make for a much better result.
I’ve seen lots of people nix the concept of putting a lot of effort into flats, but I think that they are wrong – a proper flat goes a long, long way to the best possible final image.