Strange stacking artifact

[hector (Andrew)]

Hi Everyone.
I have recently branched out into Deep Sky photography and I upgraded from an Orion 80ED to the Skywatcher 150ED. The first night out shooing with it on the 29th July and when I stacked my 100 images of CenA I noticed an artefact in the image above the galaxy. I can only describe it as a wobbly line. It almost looks to be like a ghost reflection of the brighter star but the reflection has moved during the exposure. The line almost looks like a repeating pattern, much like a plot of the periodic error of the mount. If you look carefully you can see the same wobbly line across the entire image, just the others are quite faint. The image was taken with a Nikon Z6II at 800ISO and a 30second exposure, 100 images were taken. Any suggestions would be appreciated and better still a solution if people know what it is and how to deal with it. It is also in my shot of 4945 from last night as well, just not as obvious. I used Sequator to stack the images.

[The_bluester (Paul)]

Given the common shape to the lines I would suspect hot pixels not being rejected by the stacking software (Which can depend on various stacking settings) I am not familiar with Sequator though as to how you might set it to better reject hot pixels.


It would be worth having a look at individual subs to see if there are visible hot pixels in the right places. Remembering that the stacking aligns the stars so actual objects stay in the same place and anything fixed with regard to the sensor (Hot/cold pixels etc) will appear to move in the final stack.

[multiweb (Marc)]

+1 Just hot pixels. Some softwares like CCD Stack have a bad pixel map you can build from a dark frame or you can use PixInsight cosmetics process to paint them prior to stacking.

[hector (Andrew)]

I have considered Hot pixels and rejected the idea, there are over 50 of these lines visible in the full image. The camera has been used quite a bit and has never shown any hint of this issue in the past. The lines recorded for the next object NGC4945 are in completely different spots. I will check with another camera tonight though.

[JA]

Originally Posted by hector

Hi Everyone.
I have recently branched out into Deep Sky photography and I upgraded from an Orion 80ED to the Skywatcher 150ED. The first night out shooing with it on the 29th July and when I stacked my 100 images of CenA I noticed an artefact in the image above the galaxy. I can only describe it as a wobbly line. It almost looks to be like a ghost reflection of the brighter star but the reflection has moved during the exposure. The line almost looks like a repeating pattern, much like a plot of the periodic error of the mount. If you look carefully you can see the same wobbly line across the entire image, just the others are quite faint. The image was taken with a Nikon Z6II at 800ISO and a 30second exposure, 100 images were taken. Any suggestions would be appreciated and better still a solution if people know what it is and how to deal with it. It is also in my shot of 4945 from last night as well, just not as obvious. I used Sequator to stack the images.

Hi Hector,

3 questions/suggestions to narrow things down or try.....

1. Is the wavy line pattern evident in the subexposures?
2. Is the wavy line pattern evident if stacked using DeepSkyStacker or other software besides Sequator?
3. Did you use Dark Frames? If not, or even if you did, try using the Reduce Dynamic Noises option in Sequator

Best
JA

[hector (Andrew)]

No the lines are not visible in the individual exposures.
I have started a stack on Deepsky stacker but that takes time
I will try in photshop as well
Reduced Dynamic Noise was active

[The_bluester (Paul)]

It would not be uncommon for any particular sensor to have hundreds of hot pixels (Remembering that this does not mean they will be at maximum value, just that they may end up brighter than the average pixel for the same exposure time) so 50 odd lines from hot pixels would not be out of the question.

They will not always show up in the same place on each image after stacking but if it is hot pixels they should be in the same places on the individual subs. The difference really is that from sub to sub the stars will wander around through tracking errors or dithering, but in the stack, the stars end up in the fixed locations and any hot pixels not rejected by the stacking software will wander around in the final image and make streaks, or dotted lines etc.

30 second exposures, are you guiding or imaging unguided? And do the stars move around on the individual subs in about the same way as these streaks do in the stacked image?

[hector (Andrew)]

Most sensors in that size will have hot pixels that can show. The Images are guided on the Sirius EQ-G mount with an accurate polar alignment. I wasn't using the guide scope as I need a mounting plate to place it properly. The same camera has been used for months shooting Astro without any sign of this issue. The only variable has been the change from the Orion 80ED to the Skywatcher 150ED. The mount is a little light for the scope and I have an EQ-6 coming for that scope. I have inspected images taken prior to the 150ED and seen hint of the anomaly. Every image shot last time with the 150 has shown these wavy lines.
I will check by shooting with the 80 first and then the 150 to see if there is a difference. I will also change cameras to see the difference there as well.

[multiweb (Marc)]

Originally Posted by hector

I have considered Hot pixels and rejected the idea, there are over 50 of these lines visible in the full image. The camera has been used quite a bit and has never shown any hint of this issue in the past. The lines recorded for the next object NGC4945 are in completely different spots. I will check with another camera tonight though.

The patterns are exactly the same so yes. 100% hot pixels.

[Sunfish (Ray)]

Have you used the camera setup to detect hot pixels to update the hot pixel list recently. ?Easy to do with a DSLR. And set the stacking software to remove any “remaining” hot pixels.?

100 shots would generate a lot of heat in the sensor.

If you look at the uncropped stacked images you may see that the movement of corners of the overlapping stacked images matches the movement of your lines in direction.