Dark bands on images

[RedLeader1]

Last night I did some imaging and did my flat frames as usual and used the same dark calibration frames with my ZWO ASI071MC pro cooled to -5 with the brightness set to the same level as well as my gain and exposure times being the same.
With the top image I took last night you can see dark banding on the right about half way up but in the second image you can hardly see any banding using the same settings. Banding suppression is set to 0 for both images so what could be causing this?

[RedLeader1]

I have had a bit of a play around with the banding suppression settings using the help guide on Sharpcap Pro and I might be getting somewhere. I played around with setting the gain high and keeping banding suppression set to 0 and live stacked 4 frames. I then did banding suppression turned on with threshold set to 23 and suppression set to 73 for the other image using the same high gain setting and you can see the difference.
I also used the same banding suppression settings for my master darks which was applied to both captures below with top image without banding suppression on and bottom with it turned on.

Going to be cloudy in the UK for the next few days so won't be able to test properly but not sure if the above test a good pre-test to do before using "live"?

[RedLeader1]

In addition to the above I noticed something odd happen when using Sharpcap to check darkness levels with banding suppression on and master dark calibration frame enabled. Gain was set to 90 and exposure time is 15 seconds. With the camera cap on I noticed noise on a dark live stack done with 15 frames and this was able to repeat it with other live stacks. I then set the dark frames to 0 and banding suppression/threshold to 0 and then reselected the same dark calibration frame and then re-enabled banding suppression/threshold to the same levels and same settings as used with the 15 frame live stack. I then did a live stack of 29 frames and noticed there is no noise in the dark image.

What could have caused the noise in the 15 frame stack?

I attach the images captured along with the settings files used for each and I can't see any difference in the settings files.

[RedLeader1]

In order for the noise to not appear it would appear that testing further I have to deselect what is selected for the subtract dark and then I turn off the cooler for the camera and then deselect the camera from the cameras menu. I then reselect the camera and then turn on the cooler and wait to get to -5 and then reselect the file for subtract dark and then I get can a dark image with no noise.

Would this be possibly an issue with Sharpcap because I can repeat the issue by making the noise appear if I don't deselect the file selected for subtract darks and then close the camera and then reselect the camera and perform a live stack with camera cap on?

[admin]

Hi,

so, starting at the very beginning, the key with the banding suppression is to set the 'threshold' as low as possible, and to set it while capturing dark frames. Basically, you turn down the banding threshold until the banding re-appears, then turn it back up just enough to make it vanish again. That way, SharpCap won't include the brighter parts of the image (stars, nebulae, galaxies, etc) in its measurement of the banding and try to cancel those features out (that leads to the dark band you saw in the first post). The full procedure is here : https://docs.sharpcap.co.uk/4.1/#Banding%20Suppression

I think the difference between the two stacks comes down to an oddity of how SharpCap scales the brightness of the stacked images when building an image from the stack... If the brightest parts of the stack are very dim compared to the maximum possible brightness, SharpCap rescales the whole stack upwards to fill the full brightness range. If the brightest parts are less dim then it doesn't do that. This is really there to give an image of sorts when people have badly underexposed their camera (at least with an image of sorts you can see what's going on and hopefully get your settings right in time - harder with a black screen). I think that your different test runs are just landing on different sides of this threshold, so are sometimes being scaled up in brightness and sometimes not. Annoyingly, I can't find the code that implements this little 'feature' at the moment, so I can't advise on the exact nature of the threshold etc. However, in a stack of 'real' images with stars in them that are at or near a decent percentage of maximum pixel saturation, you will never see this effect.

cheers,

Robin

[RedLeader1]

Thanks Robin for the explanation. In amongst the information that I provided I had a read through of the banding suppression and followed the steps and I think I have found the levels at which it needs to be set at. Will see what happens when clear skies return.

Thanks.