Dither more?

[Rudeviewer]

As the title states i just got my new Mach1 GTO mount and this was my first test image with 8x30 min subs with the AT8RC and OAG with the Atik 383l+.  I did just a quick process here to post up and ask this question.  

I was wondering as i have this pattern that kind of seems to go almost up and down in the image, if its due to not dithering enough.  I think i had it turned down a bit with my cgem dx since it moved all over due to backlash anyways.  The image i am posting is a 2x drizzle of the bubble to try and get close up of the noise pattern i am talking about.  Let me know what you think and if dithering more will help that noise pattern.

[wimvb]

What does the non-drizzled image look like. To evaluate issues like this, I would do a straight stack, without drizzling.

To see if it's 'walking noise', have a look at the first and last images in the stack, and the direction of offset. If you use PixInsight, you can use the Blink routine to make an animation of all subs. This should reveal the direction of offset very clearly. In that case, you may also be interested in this

http://pixinsight.com.ar/en/info/processing-examples/19/crux-southern-cross.html

Hope this helps.

[Rudeviewer]

I actually figure out is was dither being set too small.  I turned up the dither and the new subs came out great.  Without darks, if dithering only moved a small amount, regular stacking just wouldn't average it out.  

[ollypenrice]

Vertical banding might be helped by longer subs. How long were these and in what filter?

Olly

[Rudeviewer]

 Baader 7nm HA filter and 30 min subs, 9 of them.

[Rudeviewer]

You can see if my post in the deep sky section that I believe I got it mostly taken care of however.  I just started using sgpro and just didn't dither enough I think.  I turned it up and that look in the background went away.  I also used calib files which I hadn't before and it seemed to have cured it.  

[ollypenrice]

OK, the subs are plenty long enough! 

Olly

[Rudeviewer]

4 hours ago, ollypenrice said:

OK, the subs are plenty long enough! 

Olly

Thank you for the Help Olly.  I really appreciate it.

 

[Stub Mandrel]

I got the same effect last night across three different targets, but in each case the 'grain' was in a different direction. As I was guiding without dithering tas well either of these rules out a camera pattern or 'walking noise'.

Allowing for variations in the wind over time as well as the actual location in the sky of each object, it seems that the 'grain' roughly coincides with the drift of the thin haze that was causing poor transparency.

I have seen this effect several times before and ALWAYS when there is poor transparency.It is not affected by darks, so I am convinced it is some sort of noise effect caused by slow moving thin cloud. It's movement presumably creates a non-random texture as the thin cloud drifts through the frame and as it is in every frame it can only be reduced, not removed, by sigma kappa stacking.

 

[Rudeviewer]

24 minutes ago, Stub Mandrel said:

I got the same effect last night across three different targets, but in each case the 'grain' was in a different direction. As I was guiding without dithering tas well either of these rules out a camera pattern or 'walking noise'.

Allowing for variations in the wind over time as well as the actual location in the sky of each object, it seems that the 'grain' roughly coincides with the drift of the thin haze that was causing poor transparency.

I have seen this effect several times before and ALWAYS when there is poor transparency.It is not affected by darks, so I am convinced it is some sort of noise effect caused by slow moving thin cloud. It's movement presumably creates a non-random texture as the thin cloud drifts through the frame and as it is in every frame it can only be reduced, not removed, by sigma kappa stacking.

 

Interesting.  I never thought about that.  Ill have to pay more attention to seeing.  

[Stub Mandrel]

30 minutes ago, Rudeviewer said:

Interesting.  I never thought about that.  Ill have to pay more attention to seeing.

Although what I get looks like classic 'walking noise' I checked a run of the guided subs and all but one had all the stars in the same place within a pixel or so - I can't see how it could be otherwise when guiding is working - but the streaks are probably 15-20 pixels long.

I can see how this could happen with unguided imaging, but you would have to have an awful lot of fixed sensor noise, not just hot pixels.

In fact, if 'walking noise' was a real effect, anyone using 'dither in RA only' should see these streaks horizontally across their pictures.

[Rudeviewer]

3 minutes ago, Stub Mandrel said:

Although what I get looks like classic 'walking noise' I checked a run of the guided subs and all but one had all the stars in the same place within a pixel or so - I can't see how it could be otherwise when guiding is working - but the streaks are probably 15-20 pixels long.

I can see how this could happen with unguided imaging, but you would have to have an awful lot of fixed sensor noise, not just hot pixels.

In fact, if 'walking noise' was a real effect, anyone using 'dither in RA only' should see these streaks horizontally across their pictures.

Ill have to look because i think i have the box checked dither is RA only now that you mention that.  

[ollypenrice]

For me thin high haze never produces this kind of effect. I see it first in star haloes - a big soft glow around bright stars. Because drifting cloud is coninuously in motion I don't see how it could 'freeze' a pattern like this. I still think the source is electronic.

Olly

[wimvb]

1 hour ago, Stub Mandrel said:

Although what I get looks like classic 'walking noise' I checked a run of the guided subs and all but one had all the stars in the same place within a pixel or so - I can't see how it could be otherwise when guiding is working - but the streaks are probably 15-20 pixels long.

I can see how this could happen with unguided imaging, but you would have to have an awful lot of fixed sensor noise, not just hot pixels.

...

If you have a reflector, you also have to calculate in mirror movement (flexure). Even with guiding, I often have a small movement between subs, but never within a sub. My explanation for this is that the moving mirror (even shutter?) of my dslr causes a small vibration which moves the primary mirror. If the primary mirror is loose in its cell, which it should be ever so slightly, it has a tendency to 'fall' to the lowest point in the cell as the scope tracks a target. But this falling isn't continuous due to friction. A vibration caused by the dslr release, can then  initiate the movement of the primary. If this seems unlikely, consider that we're talking about minute movement of a few arcseconds.

I'm not saying that this IS the cause, but it is a possibilty that one shouldn't rule out.

When I was working in the semiconductor industry, I could measure a 2" silicon wafer sagging under its own weight, and that was just a few grams. The sagging was also just a few arcseconds. (I measured x-ray diffraction angles.)

[Stub Mandrel]

22 minutes ago, wimvb said:

I'm not saying that this IS the cause, but it is a possibilty that one shouldn't rule out.

I've been thinking this through and came to the same conclusion it could be mirror movement or 'differential flexure'.

Is there a way to 'manually dither' in PHD2. I.e. can I stop the intervalometer after a few subs, press a button to start a dither movement, then get a notification when the mount has settled and I can restart the camera. I would prefer NOT to go down the road of yet another cable/interface to control the camera from my laptop. Plus Backyard Eos may be great but I'm not watching a 2 hour 41 minute video to find out what it does...

[wimvb]

I think that guiding and manual duthering are mutually exclusive in practice. You would needto stop guiding, dither, find the guide star again and restart guiding. That would take a big bite out of available imaging time.

[Stub Mandrel]

By 'manual dithering' I only mean that PHD2 does it when I tell it, rather than automatically.

[wimvb]

9 hours ago, Stub Mandrel said:

By 'manual dithering' I only mean that PHD2 does it when I tell it, rather than automatically.

Didn't know that.