#1  
Old 07-12-2019, 05:18 PM
DiscoDuck's Avatar
DiscoDuck (Paul)
Raider Nation

DiscoDuck is offline
 
Join Date: Apr 2014
Location: Adelaide
Posts: 691
Lines in noise

Hi,
I'm after some ideas on what may be causing the following. I put up an image of NGC 1097 recently and Bart noticed some streaks across the background. I've been trying to investigate this and am a bit perplexed.

The lines are shown in the attached image - stretched drastically. Next to it is a similarly stretched (same ADU difference from black to white) dark image. The dark doesn't show that pattern (nor does a flat). You do see lines in the amp glow area of the dark (other attachment) when extremely stretched. though they are at a different angle and probably a separate thing I'd think.

I stress that this is not a huge problem. The stretch I've used here has a 2 electron difference between black and white, i.e. the outer arms of the galaxy where you can see this background are themselves only about 1 electron above the background (in a 60 second image).

Suspiciously the lines are in the N-S direction, so I don't think it's sensor related but more guiding/tracking. Dithering was used - SGP and PHD2 (scale factor 1 in PHD and I think maybe 2 pixel or so dither chosen in SGP, though it could have been the next one up). And looking at the logs and skipping through with blink dither seems to be doing a good job of moving in BOTH RA and dec - though perhaps I noticed in watching the alignment occurring in PI a slight tendency to movements of the images along that diagonal axis if anything.

Could dither not be being that random?

Could it be a backlash issue? (Though I naively feel that this would show lines in the EW direction as it would tend to fail to move NS then) (the mount has about 400ms backlash).

I think I can reduce the impact of this by taking longer exposures (i.e. much more than my 45-60 second sky limited exposure time) and dithering more drastically, as I think it's to do with fixed pattern noise. Or perhaps I should not dither at all (probably not a good idea) as the lines are more distracting to me than just a generally reasonably uniformly noisy background.

Allan (alpal) and I have been having some really good discussions in trying to understand some issues in the image. But any advice as to what exactly is causing these lines would be much appreciated.
Attached Thumbnails
Click for full-size image (Dark vs Stack small.jpg)
197.3 KB66 views
Click for full-size image (Dark lines small.JPG)
198.7 KB57 views
Reply With Quote
  #2  
Old 08-12-2019, 05:14 AM
alpal's Avatar
alpal
Registered User

alpal is offline
 
Join Date: Jan 2012
Location: Melbourne
Posts: 3,602
Hi Paul,
I'm getting mixed up with your CMOS camera model.
You say it's a ASI1600MM-C. Which is a mono, cooled CMOS camera.

There is one here that I can find:
https://www.astroshop.eu/astronomica...3-mono/p,50241


but it's no longer stocked.
It looks like it's been superseded by this one:
https://optcorp.com/products/zwo-asi1600mm-pro-cooled


that's the ZWO ASI1600MM Pro.


It says theDDR memory buffer is the main difference between ASI “Cool” and “Pro” cameras.

DDR Memory Buffer

The ASI1600MM Pro camera includes a 256MB DDR3 memory buffer to help improve data transfer reliability.
The use of this memory buffer minimizes amp-glow, which is caused by the slow USB 2.0 transfer speeds.


OK - so is the problem caused by not having that latest feature?
Reply With Quote
  #3  
Old 08-12-2019, 06:21 AM
alpal's Avatar
alpal
Registered User

alpal is offline
 
Join Date: Jan 2012
Location: Melbourne
Posts: 3,602
Hi Paul,
I went to your latest FIT files here.
https://drive.google.com/drive/folde...a_DeQ3gURtM4la

I've examined the darks and the flats.
I can see no evidence of diagonal lines in the darks or flats.
I am wondering if the diagonal lines are caused by either:
(1) a gradient in the sky or
(2) some kind of magnetic field in the camera?

Those lines are way down in the noise & I easily processed them out.

I attach a highly stretched version of the master flat for Luminance.
It's processed to show dust doughnuts and I can see a checkered pattern noise.
Attached Thumbnails
Click for full-size image (master flat lum_stretched_small.jpg)
179.8 KB41 views
Reply With Quote
  #4  
Old 08-12-2019, 06:59 AM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is offline
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,980
I would say that it is likely to be fixed pattern noise that isn’t entirely being removed. I’d suggest that it is due to the 2 pixel dither being a bit small when stacking 2000 odd frames.
The checkerboard pattern was something I noticed with both the ASI1600 MM-S V2 and QHY163M but it’s also been chatted about with other CMOS sensors. The other question is whether or not you use bias for calibration? It can be better to skip bias altogether and purely use darks.
Reply With Quote
  #5  
Old 08-12-2019, 07:22 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Its fixed pattern noise. It looks like the architecture of the sensor. I have seen this on a Starlight Express Trius 694.

I found a good bias would subtract it out. I often only used a bias with that camera. If I didn't use a bias I would see that.

You need a good bias with lots of frames and at the exact imaging temperature.

Try that and see if it solves the problem. I used CCDstack.

Greg.
Reply With Quote
  #6  
Old 08-12-2019, 08:54 AM
DiscoDuck's Avatar
DiscoDuck (Paul)
Raider Nation

DiscoDuck is offline
 
Join Date: Apr 2014
Location: Adelaide
Posts: 691
Allan: mine's the older 9now obsolete) ASI1600MM-C. The buffer is there to help reduce slow download issues (amp glow) for USB 2 users. I download over USB3. (Still get amp glow, but it is calibrated out with the darks).


Re the darks/flats, yeah, I can see no lines, except the ones I pointed out in the amp glow, which are at a different orientation and I think even fuirther down in the small scale noise.


Re bias ... I currently don't use bias frames (per Colin). But can try (per Greg) and see if that would help.


There's still the point that these lines are suspiciously running north-south. I don't believe in coincidences!
Reply With Quote
  #7  
Old 08-12-2019, 09:35 AM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is offline
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,980
There has been quite a lot of testing done over on CN and bias frames for the ASI1600 have generally been considered to cause issues. Bias frames are great with CCDs for dark scaling as you can get a true dark frame but the issue is that most CMOS sensors don’t have a linear dark current progression like CCDs do. This gets even worse when amp glow is thrown in as that’s even further from being linear.

A good dark frame is a true representation of the time integrated bias (bias + dark current).
Reply With Quote
  #8  
Old 08-12-2019, 09:38 AM
DiscoDuck's Avatar
DiscoDuck (Paul)
Raider Nation

DiscoDuck is offline
 
Join Date: Apr 2014
Location: Adelaide
Posts: 691
Thanks Colin. It was that logic that led me to not use biases. But just checking now with bias to see if there's any difference.



Quote:
Originally Posted by Atmos View Post
There has been quite a lot of testing done over on CN and bias frames for the ASI1600 have generally been considered to cause issues. Bias frames are great with CCDs for dark scaling as you can get a true dark frame but the issue is that most CMOS sensors don’t have a linear dark current progression like CCDs do. This gets even worse when amp glow is thrown in as that’s even further from being linear.

A good dark frame is a true representation of the time integrated bias (bias + dark current).
Reply With Quote
  #9  
Old 08-12-2019, 09:42 AM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,425
Yeah in SGP I use "Dither: Very High" (in Guiding settings)
Reply With Quote
  #10  
Old 08-12-2019, 09:53 AM
DiscoDuck's Avatar
DiscoDuck (Paul)
Raider Nation

DiscoDuck is offline
 
Join Date: Apr 2014
Location: Adelaide
Posts: 691
So no difference using a bias. As I said, I might try upping the dither next time, like Dunk says he does.
Reply With Quote
  #11  
Old 16-12-2019, 06:35 AM
MoTab (Mo)
Registered User

MoTab is offline
 
Join Date: Nov 2019
Location: Melbourne, Australia
Posts: 6
Bias frames on the Pana

Quote:
Originally Posted by DiscoDuck View Post
So no difference using a bias. As I said, I might try upping the dither next time, like Dunk says he does.
I think the consensus on CMOS sensors is to not use bias at all but to use only darks and dark-flats. You can actually use biases if you lengthen the sub times a little as a lot of the issues people have are with combinations of certain gains and short subs. Dark flats are long enough to avoid the issue and you, in effect, get to take out the bias signal that way.
Reply With Quote
  #12  
Old 16-12-2019, 06:21 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Looks like fixed pattern noise to me Paul. I see this a little on images here. 4 pixels might help stop this happening.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 07:55 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Testar
Advertisement
Bintel
Advertisement