Taking Flats with Hyperstar

What is your gain and offset?

Thanks Guys!
 
Yes, I'll share my master flat shortly. (I'm a newbie on this.  Can i share an xisf file via the forum?)
Gain and offset are at the recommended levels for a ZWO294MC.
I'll try 24,000 as well.

After I discovered Russell Croman’s Gradient XTerminator I stopped using Flats!
Stargazer66207

I believe you must use 2" filters because of the steepness of the light cone at f/2.  Also, as @Elmiko mentioned, the 24,000 adu is about the right target for flats.

The dark-flats need to be done as well at the same exposure time as the flats.  The gain/offset for all four elements, light, dark, flats and dark flats all must match.

Flats are indeed really important if you want a nice even background to your shots. Have you ever tried sky flats, from parked position (away from sun), ideally in the blue and just after sunset, or just before sunrise (there's a few 10's minutes ideal where stars won't show) ?

To me the histogram position has proven to be irrelevant, as long as you don't clip anything at either end.

Much easier to do, no gear needed and much less  prone to internal reflection or panel defects. I do 100% of my flats this way, usually automated in the morning while I sleep.

Thanks, all your comments are super helpful and I will try each of them.  I don't know how yet to respond to each comment separately.  (Newbie!) 

Here's more information:  

Attached are the xisf (or jpeg if xisf doesn't load) photos for my MASTER 3 sec flat darks and my MASTER 3 sec flat.

Also, attached is a typical screen shot of the histogram that I see when I shoot my flats.  

The gain I use with my  camera is 121.  The offset is the default value for my 294 which is 30.  

Most of the time I'm shooting in the Bortle 7 light polluted skies of suburban Detroit.

(I'm not sure how to do sky flats with the Hyperstar attached to front of scope? Please advise.  Thx)

Hope this  helps diagnose what I see going on. 

Of course, now that I've written this I see how to respond separately.  

masterDark_BIN-1_4144x2822_EXPOSURE-3.00s.jpg
masterFlat_BIN-1_4144x2822_FILTER-NoFilter_CFA.jpg


​​​​​​​



​​​​​​​

@Michael Ring yeah, I've been using a dark flat master for a while.  I'll try making those for each session for a while.

@Rick Veregin  I'm not dithering (at least not on purpose) but when I look at my subframes via Blink in PI the stars do move around very slightly.  I don't guide and I don't set anything up to dither.  But maybe ASIAIR is dithering for me automatically.

Richard Milford:
@Rick Veregin  I'm not dithering (at least not on purpose) but when I look at my subframes via Blink in PI the stars do move around very slightly.  I don't guide and I don't set anything up to dither.  But maybe ASIAIR is dithering for me automatically.

The point is just that your stars are moving to different pixels from frame to frame. And it doesn't need to be in every frame.  I also do not guide and take relatively short exposures (1 minute), which means I do sample many pixels with my stars over a session, effectively dithering my image. The only things to be wary of is if the shift is consistently in one direction, this can lead to walking noise. Also, dither must sample new pixels, if it is just moving back and forth between a couple of pixels, that is not so effective. But I generally find with polar alignment effects, wind, tracking, that non guided images rarely show walking noise, especially if images are taken over more than one night, and pre/post meridian, or just you have lots of subs.

By the way, DeepSkyStacker is free software for stacking, works really well. Even if you don't want to use it for your final stack, if you stack in DSS it gives you the x,y shift of each frame relative to a reference frame. You can see exactly how much you are shifting each frame, so you can track that you are indeed sampling new pixels. I also use it to make sure I am not drifting too much between subs, which would degrade the quality of my subs.

The reason this  dithering is important is that if your stars are on exactly the same pixel in every frame, then your master dark and master flat have all the pixels matched too with your image. This means that noise in the pixels of darks and flats add linearly in your image, they are not randomly averaged out, as they are always the same for any particular pixel in your light. When the stars shift, they match to different pixels in your masters, and thus, they are uncorrelated, and will randomly average out with enough images. So lets say you have 20 darks and 20 flats And you create a final image say from 100 frames, and every light frame is shifted to a new set of (x,y) pixels, which is dithering 100 times. The noise from your 20 darks and flats will be equivalent to 20 x 100 =2000 darks and flats. So your noise from those darks and flats will be very low compared to the noise from your 100 lights. Dithering is extremely powerful, as it effectively multiplies the number of flats or darks you took.  Taking only 20 darks and 20 flats with 100 images and no dithering, will add a lot of noise to your images, 

So keep the number of dithers (the number of shifts) * number of flats (or darks) >> number of subs if you can. This should ensure little noise from your flats and darks creeps into your lights. 
Rick

Regarding where in my processing sequence I use GradientXTerminator:  I am using PhotoShop (not PixINsight).   My first step after converting the image to .tiff format (so I can process it in PS)  I first go to "levels" and adjust the black point right up to the left edge of the peak in the histogram.  Then, I go to "curves", and do a stretch to lighten up the background sufficiently to that the gradients at the corners  show up.  THEN  I run Gradient XTerminator, after which I  go back to "levels" and re-adjust the black point.
Stargazer 66207