Calibration help [Deep Sky] Processing techniques · Thomas · ... · 19 · 391 · 1

tomtom2245 1.20
...
I'm running into a problem when trying to calibrate my Ha frames in Pixinsight. I'm using an ASI178MM-Cool which I know has some serious amp glow at longer exposures but up until this point, darks have been able to remove it quite successfully. The current problem is that the amp glow from the lower left and lower right is not being removed but the other parts are. This problem is not happening with my Oiii frames and they are actually 300 sec vs my 240 sec for Ha. I've tried image calibration several times and always get the same result. At this point, I think I may need to try and retake all of my dark frames. The current darks were taken at the same exposure (240sec), gain (279), and temp (-15C). In the attached image, the upper left is an uncalibrated light frame, upper right is a calibrated light showing the amp glow still present in the lower corners, and the bottom image is my master dark. Any thoughts as to why it is not being removed? Oh, and during the calibration process, I leave Calibrate and Optimize unchecked in the darks section of the Image Calibration tool in PI.

Like
tomtom2245 1.20
...
I took another set of new darks, this time I did 50 frames versus the 25 I used last time, and still get the exact same result.
Like
gnomus 0.00
...
Hi Thomas.  It’s always a pain trying to track these things down.  That problem looks pretty bad.  Are you certain it’s amp glow?  The two bottom corners might be, but that top right looks awfully like light leakage.  The ‘rays’ across the top of the frame look especially suspicious.  But I should say I’m not a CMOS user so am not any sort of expert .  Are your darks shot in total darkness, perhaps with the camera off the scope in a darkened room?

If this does turn out to be amp glow then I might consider contacting ZWO since there is only so much that you can expect of calibration.
Like
tomtom2245 1.20
...
Steve Milne:
Hi Thomas.  It’s always a pain trying to track these things down.  That problem looks pretty bad.  Are you certain it’s amp glow?  The two bottom corners might be, but that top right looks awfully like light leakage.  The ‘rays’ across the top of the frame look especially suspicious.  But I should say I’m not a CMOS user so am not any sort of expert .  Are your darks shot in total darkness, perhaps with the camera off the scope in a darkened room?

If this does turn out to be amp glow then I might consider contacting ZWO since there is only so much that you can expect of calibration.


I believe it is all amp glow as the pattern is always the same and my darks are taken with the camera removed, cap on, and in my basement in total darkness. After getting the camera, i did some digging around and found several other users with a very similar pattern but all were able to calibrate it out with good darks, as has been my case up until this point.
Like
tomtom2245 1.20
...
I also tried DSS with the same result.
Like
gnomus 0.00
...
OK.  I see that the dark you posted has had the STF applied.  Can you still see the glow in the unstretched master dark?
Like
tomtom2245 1.20
...
Steve Milne:
OK.  I see that the dark you posted has had the STF applied.  Can you still see the glow in the unstretched master dark?


No, there is no glow present in the unstretched master dark.
Like
gnomus 0.00
...
We may need a CMOS expert then.  But have I read somewhere that dark calibration on CMOS cameras might be dependent upon the ambient temperature and that this made dark calibration a bit more awkward?
Like
tomtom2245 1.20
...
Steve Milne:
We may need a CMOS expert then.  But have I read somewhere that dark calibration on CMOS cameras might be dependent upon the ambient temperature and that this made dark calibration a bit more awkward?


If the sensor is cooled to the same temperature, what impact should ambient temp then have?
Like
gnomus 0.00
...
Yes I know - on a CCD it doesn't, but I have a vague memory of reading something about this.  Problem is I cannot remember where.  I'll keep looking.  It is possible, of course, that my memory is incorrect.
Like
tomtom2245 1.20
...
Steve Milne:
Yes I know - on a CCD it doesn't, but I have a vague memory of reading something about this.  Problem is I cannot remember where.  I'll keep looking.  It is possible, of course, that my memory is incorrect.


If that's the case, that would be news to me. Appreciate you looking into it for me!
Like
gnomus 0.00
...
I found one thread of a guy with amp glow that looks exactly like yours.  Quite an interesting discussion about calibration problems (though I've not found the thing I remember seeing yet): https://www.cloudynights.com/topic/533251-asi178mc-cooled-darks-amp-glow/
Like
gnomus 0.00
...
And this from the ZWO site may be relevant:

Most importantly, make sure you disable any dark optimization feature [my emphasis] of your calibration program. DSS, PI, and other programs have dark optimization features, and they are often enabled by default. Optimization will scale the master dark, which will usually result in a mismatch with the glows in the light frames. This can leave behind “remnant glow” as a small amount of slightly brighter pixels in the area where the glow was that eventually sum up to re-form the glow in the final integrated image. Alternatively it may leave behind “inverted glow” or “ghost glow”, which is a small amount of slightly  darker pixels in the area where the glow was that eventually produce a dark, inverted version of the glow in the final integrated image. By disabling dark optimization or dark scaling in your calibration program, you will ensure that the master dark frame properly calibrates each light frame.

https://astronomy-imaging-camera.com/tutorials/what-is-amp-glow.html
Like
gnomus 0.00
...
The third and fourth posts in this thread seems to get into the darks-and-lights-shot-at-the-same-ambient-temperature issue.  Perhaps this is where I saw it before (I really cannot remember).  And, I must repeat, I know nothing about CMOS cameras, so this could all be nonsense.  But, for what it's worth: https://www.cloudynights.com/topic/604214-new-zwo-asi-183-mono-uncorrectable-amp-glow/
Like
0.00
...
(deleted)
Like
tomtom2245 1.20
...
Steve Milne:
I found one thread of a guy with amp glow that looks exactly like yours.  Quite an interesting discussion about calibration problems (though I've not found the thing I remember seeing yet): https://www.cloudynights.com/topic/533251-asi178mc-cooled-darks-amp-glow/


So reading through this thread, particularly post #14, there is the suggestion that this Sony sensor runs an internal calibration when gain value is changed. With that in mind, I did another night of imaging the Helix in Ha and then used the new set of darks for calibration (no gain settings were changed in between) and the Ha lights calibrated perfectly this time. Before, I was using a master dark and in between taking that dark and the previous Ha frames, I had changed the gain setting. This time around, there was no gain change between the darks and lights. I'm now wondering if that is what was going on. Will definitely have to keep that in mind from now on with this camera. Thanks for digging that thread up for me!
Like
gnomus 0.00
...
NP.  Glad you have it working now.
Like
Fatimora 0.00
...
Wow...I have the same camera and will soon begin taking long exposures, this thread may have just saved me a thousand headaches! Thanks for posting!!!
Like
Wilsmaboy 0.00
...
I had problems trying to calibrate using lights/darks captured with a CMOS camera using Pixinsight and Nebulsoity - neither dealt with amp glow.  I tried using Astro Pixel Processor to calibrate and it works a treat - removes all amp glow as long as you check the box referring to 'pedestal'.  I also use darks captured at different times to the lights - essentially a dark library - and haven't found it necessary to captue darks at the same time as lights during an imaging session.
Like
tomtom2245 1.20
...
I had problems trying to calibrate using lights/darks captured with a CMOS camera using Pixinsight and Nebulsoity - neither dealt with amp glow.  I tried using Astro Pixel Processor to calibrate and it works a treat - removes all amp glow as long as you check the box referring to 'pedestal'.  I also use darks captured at different times to the lights - essentially a dark library - and haven't found it necessary to captue darks at the same time as lights during an imaging session.


I haven't had a problem with a dark library with my ASI1600, just the ASI178. PI has always processed out amp glow beautifully but it seems to be the particular Sony sensor in the 178 does some stuff internally to recalibrate after any gain change and this changes the amp glow signature just enough that a dark library is useless if you plan on changing the gain.
Like
 
Register or login to create to post a reply.