Has anyone got a reason for this?

[myk68]

Hi All

So this is a zoomed image from my shot of the Cocoon nebula the other night and is from a ZWO ASI533 on a SW 80ED pro. I got about 60 2 minute images. pre-processed and stacked in Pixinsight.

The stars are points so the guiding was fine. It seems to be background noise.

However, 2 other objects imaged the same night dont have this issue and are fine.

Thanks for looking

Mike

[ollypenrice]

23 minutes ago, myk68 said:

Hi All

So this is a zoomed image from my shot of the Cocoon nebula the other night and is from a ZWO ASI533 on a SW 80ED pro. I got about 60 2 minute images. pre-processed and stacked in Pixinsight.

The stars are points so the guiding was fine. It seems to be background noise.

However, 2 other objects imaged the same night dont have this issue and are fine.

Thanks for looking

Mike

The stars are not points to my eye, Mike. They fan out to the right with soft extensions. The smaller ones show this best. This is enlarged and annotated to show the extensions I'm talking about.

If I had to guess, I'd say it was not from guiding error but was optical in origin. This kind of artifact can arise simply from soft focus but it can also be explained by tilt or mis-collimation. Refractors can be out of collimation though we usually think of reflectors needing attention.

The fact that you didn't get it on all images points either to focus having drifted for this image or to the angle of the scope affecting any tilt. Slack in the drawtube, for instance?

I think your main concern is the diagonally aligned colour noise, though? Could the cooler have gone off for some reason? Maybe the chip temp is shown on the subs' file data?

Olly

[myk68]

Hi Olly

Sorry it was the diagonally aligned noise I was meaning.   My fault I never am clear in my explanations.

I have gone though the images again, looking closer and found a few that the guiding bumped on and I think that's the star artifact you have picked up on.

Maybe the focus and temp moved too (electric focuser on order 😉, the boss doesn't know about that yet)

For more info on the images. I did them in 2 sets of 30, one after the other.  The above image is when all stacked together

I have just stacked 1st 30 and the 2nd 30 separately and they look fine. Maybe a bit of cloud crept in the first

going to see what happens when I stack these 2.

Mike

[myk68]

Yup restacking the 2 together gets that strange aligned noise??

[CCD-Freak]

It is called "walking noise"  there are several possible reasons for it such as differential flexure, focuser flexure or polar alignment which causes a slow creep in the guiding. 

This is a fairly common issue....now the fun part....figuring out what is causing it. 🤪   Dithering your guiding typically helps with it too.

 

John
CCD-Freak
WD5IKX

Edited November 5, 2020 by CCD-Freak

[ollypenrice]

Are you stacking two stacks together or stacking all the individual subs in one fresh pass? I read your post as stacking the stacks.

Olly

Edit. Dither is an excellent weapon against walking noise.

Edited November 5, 2020 by ollypenrice

[myk68]

Hi John

That sounds fun to figure out(not).  But if there a reason why it then wouldn't do it in 2 following exposures that night.

Or is that because I moved the scope to look elsewhere and it settled better maybe.

Hi Olly

I cheated a little with this, just to try it. So stacked the final 1st 30 and 2nd 30 together, but as pixinsight needs more than 2 images to stack I doubled up but the result was the same as when I stack all 60 frames

[CCD-Freak]

If you search on "walking noise astrophotography" you will find much information and discussion about it.   There have been many threads here and on Cloudy Nights.

I think at some point we have all suffered from this affliction.

 

John
CCD-Freak
WD5IKX

 

 

Edited November 5, 2020 by CCD-Freak

[ollypenrice]

Have you tried a stack without darks?

Olly

[vlaiv]

I would try two different things here.

First is to figure out if there is drift and what sort of drift it is (is it straight/linear - bad or random - sort of good). To do that, just stack subs without alignment and look at formed star trails.

Second is to try to use different debayering method. Use super pixel mode and see if that changes anything.

What software are you using for stacking? Changing registration alignment method might help as well. Don't use linear if you can help it. Use more advanced resampling method.

[myk68]

Olly and Vlaiv

This image is the 2 sets of 30x2min side by side. directly as saved from the camera in fits. NO processing NO alignment.

I use Pixinsight totally at the moment and there is a tool in there called blinked that just lets you look through your pictures to make sure they are ok and then saves the ones you select to a folder to then work further.  These images are not even debayered yet.

I can see no trailing or anything which to me means my tracking works really well.!!  unless I am missing something

the walking noise would surely only be if your stars move in position on the image frame, which doesn't seem to be happening due to my guiding.

The only thing here I would think is that between the 2 sets of 30 there is a positional difference and in fact there is. 

So does this mean I cant use these two sets together or is there a way to sort out the noise first.  do I clip the background down or run a denoise first on maybe 4 stacks of 15 then stack those.

[vlaiv]

7 minutes ago, myk68 said:

I can see no trailing or anything which to me means my tracking works really well.!!  unless I am missing something

the walking noise would surely only be if your stars move in position on the image frame, which doesn't seem to be happening due to my guiding.

Try to take one set of subs (whole set) and stack them without alignment. Use average stacking method.

Although stars don't move on successive subs - they might move over couple of frames. I had that - perfect guiding over one frame, but difference between first and last frame over few hours could be as much as 15-20px.

[ollypenrice]

I find Pixinsight inordinately complicated and would simply try stacking in a different program.  It does look as if the stacking is introducing the problems including the stellar distortions.

AstroArt is my choice for stacking and calibrating. The free trial version allows you to to stack but not save. Alternatively you could try DSS, which is free, or post your precise workflow on the PI forum to see if anyone can spot an error.

Olly