Troubleshooting drift graphs

Discussions of Electronically Assisted Astronomy using the Live Stacking feature.
Post Reply
saguaro
Posts: 68
Joined: Tue Feb 14, 2017 9:04 pm

Troubleshooting drift graphs

#1

Post by saguaro »

I'm using SharpCap Pro v. 3.1.5193 on a Windows 10 laptop, a Celestron Evolution alt-az mount, and ASI290mm mini mono camera, 8-inch SCT with a .5x focal reducer for F/5.5 (1100mm focal length). I think there's a problem with the Evolution mount tracking, and I'd like to know more specifically what the tracking issue is so I can try to fix it in the mount. I think I should be able to get up to 15sec exposures (based on comments from others who use this mount for EAA) without issues, but I found that even with much shorter exposures in the 4-5 sec range I am getting lots of drift.

When I attempt 5sec exposures, I see what I can only describe as a pattern of noise in the stacked image, and the stars appear to be slightly out of focus even though I had just adjusted focus beforehand using a bhatinov mask. Is this due to poor mount tracking? When I look at the drift graphs during live-stacking, I see what appears to be lots of drift. What specifically do the graphs show? If I know how to interpret the graphs it will help me troubleshoot any issues with the mount's tracking.
M81, 113x5sec
M81, 113x5sec
M81 113x5sec.jpg (275.19 KiB) Viewed 4324 times
M5 drift graphs
M5 drift graphs
M5 drift graph.JPG (549.81 KiB) Viewed 4324 times
M51 drift graphs
M51 drift graphs
M51 drift graph.JPG (595.19 KiB) Viewed 4324 times
Thanks for any help.
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Troubleshooting drift graphs

#2

Post by oopfan »

Hi saguaro,

That is "raining noise". I suffer from it too. It's caused by tracking/guiding problems. My mount is 50 years old. It has a stepper motor on the RA axis but nothing on the DEC so active guiding is out of the question. Fortunately I have Periodic Error Correction. I can get good 60-second subs and have even gone up to 100 seconds with only a slight amount of star elongation. The problem is that over the course of 1 hour of imaging the tracking drifts 30 to 40 pixels. Because the pixel matrix is not uniform the noise creates trails on the stacked image.

Thankfully I stumbled across a solution to raining noise: increased exposure. If you use The Brain the recommended exposure is set to "overcome" read noise (that's Robin's choice of word -- I like it!) In our case we need to "slay" it. I tend to use long exposures anyhow so I can't offer a multiple. You need to experiment. If the Brain recommends 5 seconds then I would try doubling it to begin with. Of course, if you can solve the tracking problems then you can revert back to the Brain's exposure setting.

Someone on the forum suggested that I use "dithering" so that I can keep the original exposure.

Brian
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Troubleshooting drift graphs

#3

Post by admin »

Hi,

just to explain the drift graphs - the left hand graph is showing how far offset each frame in the stack is compared to the first frame - the axis measurements are in pixels and the two axes represent X and Y offset. In the first screenshot it looks like you have had a lot of drift, in the second only a few pixels.

The right hand graph is a track of rotation relative to the first frame - you can see here the field rotation due to the alt/az mount building up. The vertical scale here is degrees rotation, the horizontal just tracks frames with the oldest at the left.

In both charts the blue cross is the most recent frame, the green crosses are older frames that were used in the stack and the red are frames that were not used in the stack. In the left hand graph the cross size ranges from largest (recent frames) to smallest (oldest frames).

My original thoughts on the raining noise phenomena was that it was caused by under/over corrected pixels (when using a dark that wasn't quite perfect) and as the frame drifted then the pixel would appear as a streak. I think this ties in with your second image where I think I can see much bigger streaks on the left hand side than the right. I think I can see a circular pattern in this stretch of your second image, centered somewhere on the right...
Untitled.jpg
Untitled.jpg (363.9 KiB) Viewed 4307 times
Even if your mount was trackign perfectly you will still get the rotation effect which may lead to the raining noise. Imaging objects further away from the zenith will give you a slower field rotation rate and improve matters, otherwise it's certainly worth trying out Brian's suggestions.

hope this helps,

Robin
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Troubleshooting drift graphs

#4

Post by oopfan »

I too have the 290M camera. Dark subtraction is problematic with any uncooled camera. I had the same problem with raining noise last year with my ASI120MC. You could also try this:

Take half of your dark frames upfront and the other half at the end of your lights. Average them together to make one Master Dark. I started doing that recently but I haven't performed an exhaustive study of its efficacy.

This technique was suggested to me by the instructor of the CCD Photometry course.

Brian

PS: I noticed in your M51 screenshot that you were not doing dark frame subtraction during Live Stack. I don't either; I post-process with other software. Do you still see raining noise after you post-process?
saguaro
Posts: 68
Joined: Tue Feb 14, 2017 9:04 pm

Re: Troubleshooting drift graphs

#5

Post by saguaro »

Oopfan and Robin,

Thanks for the explanations of the "raining noise" effect with my alt-az mount. I have taken it to my local astronomy shop Starizona, where they are going to make some mechanical adjustments to reduce any backlash. Hopefully that will be enough to minimize the issue.

Robin: Thanks so much for describing the drift graphs; that's really helpful. I'm a bit confused though by your explanation that the raining noise is due to use of a master dark. I didn't use a master dark during live stacking.

I am only interested in live-stacking and not in post-processing. I will investigate using the settings recommended by the Brain, but if the Brain recommends longer exposures, I'm afraid the issue will just be magnified.

I was not using a cooled camera. Would using a cooled camera help to mitigate the raining noise effect? The evening was quite warm (upper 80's). The camera registered 32.1C (about 90-degrees F) for the M81 captures.

Thanks.
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Troubleshooting drift graphs

#6

Post by admin »

Ah, interesting that you were not using a dark frame - the same effect can arise without one though - basically an area of the image that is darker than normal due to camera imperfections gets spread out by the streaking effect of the frame drift/rotation.

A correct dark frame may help with the problem (ie one taken at the same temperature). A poorly matched dark can make the problem worse.

cheers,

Robin
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Troubleshooting drift graphs

#7

Post by admin »

Hi,

you don't happen to still have files for the M81 do you? If you happen to have the raw frames and the final stack and could share them with me it might prove interesting to try to understand what the root cause of the raining noise is and whether anything can be done about it.

cheers,

Robin
saguaro
Posts: 68
Joined: Tue Feb 14, 2017 9:04 pm

Re: Troubleshooting drift graphs

#8

Post by saguaro »

Hi Robin,

I only did "saved as viewed" so I don't have the raw frames for any of the images. I think your explanation that it's a combination of poor tracking and not using a dark frame taken at the same temp as the camera (which was at 32-degrees C at the time of the M81 captures) makes sense.

I took my 3-year old Evolution mount into Starizona and they completely adjusted the gears and helped me to get the best balance for my setup (8-inch SCT @F/5 using 290mm mini camera). I'm hoping to try out my newly tuned-up mount as soon as the weather cooperates. I'll then update this thread with the results. If I'm still having the "raining noise" issue, I will be sure to save the raw files for your examination. Thanks!
saguaro
Posts: 68
Joined: Tue Feb 14, 2017 9:04 pm

Re: Troubleshooting drift graphs

#9

Post by saguaro »

Thanks to Dean at Starizona, my Evolution mount performed much better last night. The adjustments Dean made to the mount's Alt and Az gears allowed me to take 10sec and 15sec exposures with my C8 @F/5.2 (1050mm focal length) using my ASI 290mm mini.

Here is my album of images: https://www.icloud.com/sharedalbum/#B0k5Uzl7VGs1RRh

SharpCap did a pretty good job of aligning each subexposure to greatly reduce field rotation effects. I used a matching master dark which helps reduce noise due to the higher temps we get this time of year in Tucson. During cooler months, I may be able to get away without using a master dark.

The images are not as sharp as I would like, due to the tracking limits of the mount and the focal length I was working at (1050mm). Next time I will use more focal reduction to get to around F/4, which I think will be a more forgiving focal length (807mm) that will hopefully allow me to get even better tracking results. Also, next time I will try using the FWHM filter to ignore subexposures that are less sharp. I will also try using longer exposures at lower gain to see if the mount can track well enough at exposures longer than 15sec.
Post Reply