Grid Pattern in one channel

[Rodd]

Does anyone have any idea what cuased the grid pattern visible in the lower right hand corner?  Its in the SII stack.  Very odd.  Only had this issue once.  The prominence of the lines makes nme think it is not one or two subs-but all of tghe SII subs.

Rodd

[AKB]

Is it on the flats?

[vlaiv]

What software are you using for stacking and what interpolation algorithms does it offer?

Above is artifact related to a small angle rotation when aligning subs and using certain forms of interpolation.

[Rodd]

17 minutes ago, vlaiv said:

What software are you using for stacking and what interpolation algorithms does it offer?

Above is artifact related to a small angle rotation when aligning subs and using certain forms of interpolation.

I use PI.  I have not seen this before or since. I use either Windsor sigma clipping or linear fit stacking. These are the 2 out of about 10 the software indicates are best.  Does it mean there was rotation of the camera or tube or some other part of the system?

[vlaiv]

2 minutes ago, Rodd said:

Does it mean there was rotation of the camera or tube or some other part of the system?

No, it is due to field rotation because of slightly off polar alignment.

It can also be that your subs are all aligned but there is small rotation between channels so one of the channels needed to be aligned to others.

Try restacking - but use different reference frame.

[Rodd]

I’ll have to see if I still have the original subs. I usually delete subs once I register and keep the registered subs. I guess I can reregister the registered subs

but I only saw this this time. All other channels were collected over the same time and my PA was the same. I typically register all subs to the same sub. 

[Rodd]

1 hour ago, AKB said:

Is it on the flats?

No. Only on SII subs

[Rodd]

51 minutes ago, vlaiv said:

No, it is due to field rotation because of slightly off polar alignment.

It can also be that your subs are all aligned but there is small rotation between channels so one of the channels needed to be aligned to others.

Try restacking - but use different reference frame.

Out of luck. I checked and only have the registered SII subs and they all have the grid pattern.  

[vlaiv]

3 minutes ago, Rodd said:

All other channels were collected over the same time and my PA was the same. I typically register all subs to the same sub. 

I guess that you don't cycle filters, but rather take some time to do the Ha, then OIII and in the end SII, not necessarily in that order?

Was this a single night?

In any case, depending where target is in the sky - slight PA error + guiding can lead to different levels of field rotation.

You can have PA error in altitude and/or azimuth - meaning your mount needs to be turned "left/right" to reach true NCP or it needs to be tilted "up/down", or some combination of these. Left/right error differs from Up/Down error in sense that it causes DEC drift in different part of the sky.

One is near meridian and other is when the mount is pointing east or west.

Depending where you are on "a circle", and depending how big your DEC error in that moment - guiding will cause more or less field rotation (field rotation is created when there is DEC drift and guide system compensates. In fact - I think that it would be there even without guiding, guiding just helps to stay on target).

In any case - if you shoot Ha in one part of the sky and SII in other - this can cause one set of subs to have very small field rotation and other to have somewhat bigger.

[vlaiv]

1 minute ago, Rodd said:

Out of luck. I checked and only have the registered SII subs and they all have the grid pattern.  

I guess that process of registration caused SII subs to have grid pattern. You no longer have raws?

You can "cure" grid pattern by not trying to stretch so much. It is really just noise and if you can see it - you are hitting the noise floor. Pattern forms out of noise because some pixels get more averaged out than others and this improves their SNR. These pixels lie on rectangular grid - and after denoising - grid is shown as darker and lighter zones.

[Rodd]

23 minutes ago, vlaiv said:

I guess that process of registration caused SII subs to have grid pattern. You no longer have raws?

You can "cure" grid pattern by not trying to stretch so much. It is really just noise and if you can see it - you are hitting the noise floor. Pattern forms out of noise because some pixels get more averaged out than others and this improves their SNR. These pixels lie on rectangular grid - and after denoising - grid is shown as darker and lighter zones.

I found an alternate stack registered with Windsor sigma clipping that doesn’t have the grid. But the FOV is smaller. I think I must have cropped out the right hand area and registered to a cropped stack as the ha, SII snd oiii combine well. 
 

thd other issue I have is my amp glow does not calibrate out. Others don’t have this issue. It is only apparent in 300 sec subs (NB). I took new darks recently to see if the darks were the problem, but the same glow is visible. The amp glow is supposed to calibrate out. The best answer is to get a 2600, but I can’t 

[Rodd]

1 hour ago, vlaiv said:

I guess that process of registration caused SII subs to have grid pattern. You no longer have raws?

You can "cure" grid pattern by not trying to stretch so much. It is really just noise and if you can see it - you are hitting the noise floor. Pattern forms out of noise because some pixels get more averaged out than others and this improves their SNR. These pixels lie on rectangular grid - and after denoising - grid is shown as darker and lighter zones.

I fouynd an alternatlyt registered SII stack that I used Windsor sigma clipping to register.  Itr was smaller in dimensions, but teh same scale.  I guess I must have registered to a cropped master.  Dont remember.  Seems to have worked.  Thanks!

C11 .7x ASI 1600 about 13 hours SHO

[tomato]

3 hours ago, Rodd said:

I found an alternate stack registered with Windsor sigma clipping that doesn’t have the grid. But the FOV is smaller. I think I must have cropped out the right hand area and registered to a cropped stack as the ha, SII snd oiii combine well. 
 

thd other issue I have is my amp glow does not calibrate out. Others don’t have this issue. It is only apparent in 300 sec subs (NB). I took new darks recently to see if the darks were the problem, but the same glow is visible. The amp glow is supposed to calibrate out. The best answer is to get a 2600, but I can’t 

Glad you sorted out the grid problem. I can't understand why a suitable dark would not remove the amp glow. I use ASI 178 cameras which have horrendous amp glow at 120s exposures but this calibrates out completely with an equivalent dark. The cameras have been retrofitted with an external Peltier cooler so I have to be a bit careful to ensure the darks are taken at  the same sensor temperature as the lights, but this should be straightforward with the internally cooled camera.

Have you tried other software to calibrate your lights? I have had issues in the past where the darks have been rejected due to some difference in the subframe parameters, but no warning has been displayed, the software just doesn't use the darks in the calibration.

[Rodd]

49 minutes ago, tomato said:

Glad you sorted out the grid problem. I can't understand why a suitable dark would not remove the amp glow. I use ASI 178 cameras which have horrendous amp glow at 120s exposures but this calibrates out completely with an equivalent dark. The cameras have been retrofitted with an external Peltier cooler so I have to be a bit careful to ensure the darks are taken at  the same sensor temperature as the lights, but this should be straightforward with the internally cooled camera.

Have you tried other software to calibrate your lights? I have had issues in the past where the darks have been rejected due to some difference in the subframe parameters, but no warning has been displayed, the software just doesn't use the darks in the calibration.

I have nog tried other TV software. I thought it might be due to old darks, but new ones are no different. I guess I could try maxim DL.  Good idea. 

[Elp]

I get this sometimes, I always suspected it was down to stray light sources affecting the lights as my flats usually don't show an issue, once calibrated and stacked the pattern goes away though.

Edited December 26, 2023 by Elp

[Rodd]

On 26/12/2023 at 10:29, vlaiv said:

I guess that you don't cycle filters, but rather take some time to do the Ha, then OIII and in the end SII, not necessarily in that order?

Was this a single night?

In any case, depending where target is in the sky - slight PA error + guiding can lead to different levels of field rotation.

You can have PA error in altitude and/or azimuth - meaning your mount needs to be turned "left/right" to reach true NCP or it needs to be tilted "up/down", or some combination of these. Left/right error differs from Up/Down error in sense that it causes DEC drift in different part of the sky.

One is near meridian and other is when the mount is pointing east or west.

Depending where you are on "a circle", and depending how big your DEC error in that moment - guiding will cause more or less field rotation (field rotation is created when there is DEC drift and guide system compensates. In fact - I think that it would be there even without guiding, guiding just helps to stay on target).

In any case - if you shoot Ha in one part of the sky and SII in other - this can cause one set of subs to have very small field rotation and other to have somewhat bigger.

Regarding filter cycling. I don’t think there would be much difference in my case because I collect allot of data. At least 2 nights per channel. So for each channel my target progresses from about 30 degrees to an hour (sometimes 2) past the meridian.  So each channel has data from the various declinations—just at different times. I check PA nightly, but often I make no adjustments