Calibration frames and camera settings

[cjdawson]

Hi all.

After the major success on Saturday, I'm finally happy with my setup.  That is the mount, alignment, focusing, control image capture and having all the hardware in place to make it doable and easy.  For me, from here on out, it should be more about tweaking the gear that I have, getting the most from it and all that good stuff.

In saturdays attempt at the horsehead, see this thread... 

 

 

I'm very happy that I can get image data.  So from here on out, it's about the quality of that data.

I'm going to be using a ZWO ASI1600MM-Pro with LRGB filters.  To capture the data.   The image that I captured on saturday was with the camera settings on default, so gain and offset where 10.

I've been looking at other threads and see that the gain would be better at 139 (this is in the tech specs for the camera) and people have been talking about an offset of 25?

 

Firstly, what is this offset?    I don't understand that.

 

For capturing subs, I'm using SGPro, and am intending to capture lights, flats, darks and bias.

The process the data in Astro Pixel Processor.

The image that I captured on saturday wasn't brilliant and I was told on here after posting the horsehead that my flats were not working.

I had another go at capturing the flats again last night and reprocessed.  To me the output was almost the same.

 

So in addition to wanting to know about the offset, I'd also like to know how to capture good quality flats.   I tried the flats wizard in SGPRo and I'm not sure why what I did wrong.

 

Are there any good links that will help me with solving the mystery of the offset and also the seperate question of how to take good, useful flats?

[daz]

Your flats need to be captured without moving the camera and at the same focus point (or near as dammit) as the lights. I tend to take mine in the morning after my lights have captured.

The exposure needs to be long enough so that the histogram is about 2/3rds over to the white end of the graph. For my Atik, I aim for about 25-30,000 ADU (depending on filter) - not sure if this holds for your ASI1600, but this puts the graph about 2/.3rds the way UP the graph.

In SGP, in the wizard, you can define the minimum and maximum ADU count for each filter, and min and maximum exposure times. SGP will then measure each filter and determine the correct duration to reach your required ADU count. You then need to save this data to the profile in SGP!

Running the flats is then as easy as selecting the right image type and SGP will populate the exposure time accordingly.

 

[cjdawson]

Thanks Daz.

I've done exactly that, and yet it seems to have not worked.  AFAIK The focus hasn't been touched since my imaging session, so doing that last night should have been ok.    It's intersting that you says 2/3's of the way up the histogram.   I ran the flats calibration thingy and did the flats wizard too.  That produced images that were 1/2 way up the histogram.   I think this warrants further investigation.

[ollypenrice]

You also need darks for flats. With a CCD camera you can just use a master bias for this but with a CMOS camera your darks for flats need to be taken at the same settings as the flats themselves, whatever that was.

Olly

[gorann]

As for gain and offset, I use 139 and 50 on my ASI1600MMpro. I think that is pretty much standard (at least a good starting point) and for earlier models of the 1600 I read somewhere that offset was factory locked at 50.

[cjdawson]

Yep. I tried doing processing an image yesterday without darks for flats and saw it wasn't working.     I've got a full set of lights, darks, bias, flats and darkflats so will try reprocessing using those to see what happens.

I get where gain of 139 comes from, and I can see that my ASI290MM should have a gain of 111.  Have no idea what the offset is for, but I'm happy to try 50 and see what happens.  From there on out, I'll be tweaking the settings and seeing if I can get improvements.

[cjdawson]

Just had another go a processing the data for my M95 image, 1 hours worth of data on each of the filters. No binning.  processed with Lights, Darks, Flats and DarkFlats - ignored the bias frames.  This became very interesting as the artifacts that I'm seeing are only present in the coloured images.  Might have to give my filters a clean with a lens pen.

I'm getting excited for my next attempt in about 6 months once the cloud take for saturday's session has been paid.