Darks are not removing amp glow

[GroupJ]

https://drive.google.com/drive/folders/ ... sp=sharing

Robin,
I redid my Dark library (see link). I was able to take one 31 frame set of the IRIS nebula with the 3nm Ha filter, 180 sec exposures. Unfortunately, I applied a 120s dark. I still got a lot of amp glo.

1. Could you look at my darks and lights - do those look like a decent set with reasonable offsets?
2. Other than applying the correct dark, is there anything else you could suggest to improve the capture?

I'm headed to dark skies in 2 weeks so the environment should be much friendlier

Thanks for the help.

Al

[admin]

Hi Al,

firstly, the darks look fine with no clipping at the black point in the sample images I checked from the dark library. You can see this easily by opening each FITS in FITSLiberator - it shows the minimum pixel values, and if that is greater than zero then you have no black clipping.

Both flat frames look very noisy - how many individual flats did you ask SharpCap to capture and average when making those? Increasing that count will reduce the noise, as will taking the flats at lower gain and either brighter light or more exposure time.

In theory you can reprocess all the images with the right dark frame and new flat frames - you use the 'Folder Monitor Camera' in SharpCap to do this - it can read back image files from a folder as if they had been captured from a camera and then you can activate live stacking (or some other SharpCap feature) while doing this. See https://docs.sharpcap.co.uk/4.1/#Virtual%20Cameras.

One thing I did note was that your raw image frames were captured with a different black level (offset) setting to the 180s dark frame - You would need to take a new 180s master dark with a matching offset level value for the dark subtraction to work properly (in particular, without doing this you will get poor flat correction).

cheers,

Robin

[GroupJ]

Thanks Robin,

Ok this is good info. I did not use the correct dark file so I will try the post process with the correct one.

I'll retake the flats- for some reason I can not get Sharpcap to complete the flat capture process - it tells me it can not find a suitable exposure. So I end up manually exposing them. Any hints on how to set up for flat capture? I've tried it with both a 102mm f5.6 and 60mm f6 scope. White tee shirt on bright sunny day. I'll do a screen capture next time I try that.

Do I need to do separate flats for the two scopes?

Thanks for all the help

Al

[admin]

Hi Al,

the most likely reason for failing to find a suitable exposure for flats is that the difference in brightness between the center and corners is large (lots of vignetting). SharpCap is happy if it can get all the pixel values into the range of about 20 to 85%, but with heavy vignetting or dark dust spots then that may be impossible and the auto exposure will fail. If that happens, you need to manually adjust to try to balance the bulk of the pixels into the good range (20 to 85/90%) and a few beyond that range on each side.

You need new flats every time you change the optical setup - changing scope, detaching and re-attaching the camera, adding or removing filters or barlow/reducers, rotating the camera to change the orientation of the image, etc.

cheers,

Robin

[GroupJ]

Thank you - Yes I reread the help notes and flats are optical train dependent - I appreciate the explanation on why it is failing to find an exposure. I took a new sets of flats and manually exposed, however I get a warning of more than 10% are under exposed or too many pixels are saturated. I elected to set exposure at point where the 10% under exposed just disappears. Is that the best strategy? I had offset at 50 - Does auto exposure also shift offset to find a solution or could I get some help with manual adjustment?

Al

[admin]

Hi,

the autoexposure will not shift the offset, since it only has a small effect on the total range of values that you can access, and an unexpected offset change can leave you with mismatched darks and lights later, which is a significant problem.

Generally the best bet in a difficult situation is to arrange for the right hand side of the histogram graph to get to about 95% maximum (no pixels above 95% in any colour). If any hit 100% then the flat will not be correcting properly for those pixels, and they are likely to be in the brightest area of the image. If that makes the corners too dark (below 10%), so be it - those are not the focus of the image and will just have to cope as best they can (perhaps crop a little if there is a real problem). The fact that you are seeing a brightness variation of 10 to 1 or so in the flats indicates that the vignetting (or dust) is pretty bad anyway...

cheers,

Robin

[GroupJ]

Robin,
Thanks again - I did post process my stack with the correct Dark and a better Flat and the amp glow almost completely diappeared - I feel like I am getting somewhere.

Al