Back to forums

[RCC] Strange artifacts in long exposure image

Siril Team Siril Request for feedback Sh2-129 Troubleshooting Artifacts
20 replies•617 views
Götz Golla avatar
Götz Golla avatar
Not sure if this is the right place to post an image which is giving me a headache. I have recently started to take  longer exposures because my object, SH2-129, is very weak in OIII. Here is my equipment:
  • Takahashi Epsilon 130D
  • ASI 6200MM camera with 16bit ADC sensor
  • for this exposure: the MAX FR 12nm OIII narrowband filter of Astronomik
  • colimated, tilt-corrected, in focus
  • I am using a flat field panel by Gerd Neumann utilizing electroluminescence

Here are the images I took for the final image
  • 60 Bias-Frames
  • 60x1sec Short Darks
  • 60x180sec Darks
  • 60x1sec Flats with the OIII filter
  • 70x180sec Lights

Here is my workflow. I am using Siril for Linux. It uses 32bit for the stacked master frames.
  • I calibrate the Short Darks with the Bias-Master before stacking to a master short dark
  • I calibrate the Darks with the Bias-Master before stacking to a master dark
  • I calibrate the Flats with Bias and short dark master before stacking to a master flat
  • I calibrate the Lights with Bias, Dark and Flat before stacking to the final image

The flat works pretty well removing the vignetting of the lights. The single calibrated lights look perfect. However when I stack a lot  of the single images, I get this:

The image is not linear but using Sirils histogram transformation view to enhance the background.

My first idea was that maybe the master flat was somehow not OK. So I redid the flat several times with different exposures, taking care that the min and max are well within the 16bit range of the ASI 6200MM camera. I also moved the flat field panel a little, but the cirular patterns never change. The flats were done directly after observation, or the next night. It doesnt make any difference.

So I am out of ideas right now. Any ideas about that might have gone wrong are welcome.
André Bremer avatar
André Bremer avatar
1/ It’s not entirely clear from your post, but does your master flat show the same artifact? If so, then there’s likely something off with your calibration process, because it should correct out as expected. If it’s not in the flats then it has to be an outside factor. Stray light, glare?
2/ Is it just showing up with your OIII filter (flats, lights)? If so, then it’s likely something on your filter.
3/ If it’s showing up regardless of the filter, rotate the camera. If the pattern stays fixed it’s something on your sensor or the window. If it rotates, it’s something in your OTA.
Helpful Insightful Concise
Götz Golla avatar
Götz Golla avatar
The ring is definetely visible in the master flat as well.

Or maybe only in the master flat !?

I have a baffle sitting above the primary mirror to reduce the reflections from the rim of the mirror. Maybe the bright light of the flat panel is reflected off that baffle. But shouldnt that happend in the same way on the Light, and  be corrected with the flat ??? I am not sure if calibrating with a flat takes out reflections as well.

Here is the image without calibration. Hard to say, but maybe the rings are not visible.
andrea tasselli avatar
andrea tasselli avatar
When was the flat taken taken? Day or night (just curious)? If you can see the ring in the flat but not in the stacked un-flattened frame then it is obviously an artefact of the flat and the way you take them. Try using a dusk flat instead, much better way if you ask me…
Götz Golla avatar
Götz Golla avatar
The flats were taken at night just after the observation.

I currently have three theories
  • the image is slightly overcorrected with the flat
  • the relatively bright light of the flat panel was shining into the optical train from outside
  • internal reflection by the baffle around the primary, or some other internal reflection

I might try to remove the baffle tomorrow and see what happens to the flats....
Götz Golla avatar
Götz Golla avatar
This is an interesting article about the effect of bias on flats:https://siril.org/tutorials/synthetic-biases/
The main point  is that if you have an offset in your flat and you divide it with your light, you might get a wrong curvature of the correction. Maybe thats the effect I am seeing. I will do some experiments with  subtracting a synthetic offset from my flats and see what happens.
André Bremer avatar
André Bremer avatar
I once inadvertently used bias frames with an offset that was different from the flats and it broke the calibration. The FITS header in the individual subs should tell you what offset was used in the acquisition of your bias and flat frames. If they are different that might be the cause.

Another quick sanity check you could try is to stack frames without flats calibration and see if the ring appears. That will tell you for sure if there's an issue with the flats.
Well Written Helpful Insightful Concise Supportive
Mike avatar
Mike avatar
Maybe frost?  I used to see that kind of thing when my dessicant  needed recharging.
Götz Golla avatar
Götz Golla avatar
André Bremer:
I once inadvertently used bias frames with an offset that was different from the flats and it broke the calibration. The FITS header in the individual subs should tell you what offset was used in the acquisition of your bias and flat frames. If they are different that might be the cause.

Another quick sanity check you could try is to stack frames without flats calibration and see if the ring appears. That will tell you for sure if there's an issue with the flats.

The image without flat calibration is the one I posted as the second image in this thread. It hard to say if the rings are there since they are very weak, maybe not.

Today I processed a series of images with varying synthetic biases. The FITS header says that the offset is "50". The real bias image has ADU values of 500 with very small variance, so a factor of 10 higher. I went through the calibration process with synthetic biases of 490, 500, 505, 510 and 520. The results are below:



Its pretty clear that the rings are independent of the Bias values. In the last image the rings are outshined by the undercorrected background. I would say that a problem with the calibration process like overcorrection can be excluded.

In these small images another artifact is visible, a darker vertical bar on the right side of the ring. I am scared .

I will remove the baffle of the primary mirror and create new flats today asap.
Helpful Insightful
Dale Penkala avatar
Dale Penkala avatar
I’m far from an expert here but I have to agree with 2 comments above that mention frost or moisture on the sensor. I’ve had very similar artifacts in my images as well and found that either using the sensor heater (071mc pro) or in the case of my 294mc pro drying out the desiccant tablets inside the camera fixes the issue or the combination of both for the 071mc pro.

Dale
Helpful
andrea tasselli avatar
andrea tasselli avatar
Dale Penkala:
I’m far from an expert here but I have to agree with 2 comments above that mention frost or moisture on the sensor. I’ve had very similar artifacts in my images as well and found that either using the sensor heater (071mc pro) or in the case of my 294mc pro drying out the desiccant tablets inside the camera fixes the issue or the combination of both for the 071mc pro.

Dale

If it were dew/frost the dimming would be immediately visible in the raws and the dimming would proceed from centre toward edge since is the centre of the sensor the coolest place and the walls the warmest (since they are there to dissipate heat).
Götz Golla avatar
Götz Golla avatar
Unfortunatelly my ASI 6200MM died on me today while making flats smile

What I can say is that the raw images look OK. I did my observerations on several nights in the past few weeks. So shouldnt any problems with frost of moisture be variable over a few days ? These rings are at exactly the same position on every night. So I consider it unlikely.
André Bremer avatar
André Bremer avatar
Götz Golla:
Its pretty clear that the rings are independent of the Bias values. In the last image the rings are outshined by the undercorrected background. I would say that a problem with the calibration process like overcorrection can be excluded.

In these small images another artifact is visible, a darker vertical bar on the right side of the ring. I am scared .

I will remove the baffle of the primary mirror and create new flats today asap.

Given all this, I am pretty certain the issue is with your flats. The uncalibrated subs look pretty clean otherwise from what I can see.

Gruss nach Deutschland!
Dale Penkala avatar
Dale Penkala avatar
andrea tasselli:
Dale Penkala:
I’m far from an expert here but I have to agree with 2 comments above that mention frost or moisture on the sensor. I’ve had very similar artifacts in my images as well and found that either using the sensor heater (071mc pro) or in the case of my 294mc pro drying out the desiccant tablets inside the camera fixes the issue or the combination of both for the 071mc pro.

Dale

If it were dew/frost the dimming would be immediately visible in the raws and the dimming would proceed from centre toward edge since is the centre of the sensor the coolest place and the walls the warmest (since they are there to dissipate heat).

Ok I’ll shut up then 😊 It was just an idea is all.
Götz Golla avatar
Götz Golla avatar
Dale Penkala:
andrea tasselli:
Dale Penkala:
I’m far from an expert here but I have to agree with 2 comments above that mention frost or moisture on the sensor. I’ve had very similar artifacts in my images as well and found that either using the sensor heater (071mc pro) or in the case of my 294mc pro drying out the desiccant tablets inside the camera fixes the issue or the combination of both for the 071mc pro.

Dale

If it were dew/frost the dimming would be immediately visible in the raws and the dimming would proceed from centre toward edge since is the centre of the sensor the coolest place and the walls the warmest (since they are there to dissipate heat).

Ok I’ll shut up then 😊 It was just an idea is all.

Hi Dale,
I think its good that everyone here can post his/her ideas. It helps to think in all directions and to finally find a solution. So thanks a lot for your contribution.
The "solution" in the case of this thread will have to wait a little since yesterday my ASI 6200MM stoppped working. I hope to be back in a few weeks.
Götz
Respectful Supportive
Megsgallery avatar
Megsgallery avatar
I have the same camera and also use astronomik narrowband filters. 

had issues with this, too

maybe this thread can help you?

https://www.cloudynights.com/topic/786350-narrowband-users-show-me-your-stretched-flats/
Dale Penkala avatar
Dale Penkala avatar
Götz Golla:
Dale Penkala:
andrea tasselli:
Dale Penkala:
I’m far from an expert here but I have to agree with 2 comments above that mention frost or moisture on the sensor. I’ve had very similar artifacts in my images as well and found that either using the sensor heater (071mc pro) or in the case of my 294mc pro drying out the desiccant tablets inside the camera fixes the issue or the combination of both for the 071mc pro.

Dale

If it were dew/frost the dimming would be immediately visible in the raws and the dimming would proceed from centre toward edge since is the centre of the sensor the coolest place and the walls the warmest (since they are there to dissipate heat).

Ok I’ll shut up then 😊 It was just an idea is all.

Hi Dale,
I think its good that everyone here can post his/her ideas. It helps to think in all directions and to finally find a solution. So thanks a lot for your contribution.
The "solution" in the case of this thread will have to wait a little since yesterday my ASI 6200MM stoppped working. I hope to be back in a few weeks.
Götz

Gotz, I’m really sorry to hear your camera has stopped working! That just sucks! Again I’m no pro at this stuff but it was nothing more than giving  you something else to consider. Like I said I’ve had similar (not the same) and found dew and frost to be the culprit. Andrea is probably correct but just wanted to give you something else to consider is all.

Best of luck!

Dale
Götz Golla avatar
Götz Golla avatar
I have the same camera and also use astronomik narrowband filters. 

had issues with this, too

maybe this thread can help you?

https://www.cloudynights.com/topic/786350-narrowband-users-show-me-your-stretched-flats/

Very interesting thread on CN. Thanks for letting me know.
Well Written Respectful
Megsgallery avatar
Megsgallery avatar
Absolutely!

I just tried last eve my flatmaster 150 recommended by Isaac at Stellarvue in that thread. 

worked like a freakin charm. 

❤️🙏❤️

good luck with the troubleshooting.  You will eventually have your eureka moment.  When your camera is up and running again.  ;-)
Götz Golla avatar
Götz Golla avatar
Hello folks,

after my ASI 6200MM finally came back to me from China, I have just found the cause for the strange artifacts in my image(s). Thats because it happend again this week. Here is what I have done (wrong):
I made 60 flats with my luminosity (L-) filter, then did a few images for testing and adjusting the tilt. The images all had a  perfectly flat background after preprocessing with this flat, with the L-Filter as well as with the narrowband Halpha filter. That is to be expected since the L-filter and my narrowband filters are supposed to be homofocal. However, when using the narrowband OIII-Filter I got those rings again.

Using the OIII-filter itself was not the reason for the rings, but I found that the OIII filter needed a slight adjustment to the focus.  My mistake was not to produce additional flats for the refocused OIII setup.  After taking und using the OIII flats, the rings were gone.

To investigate the difference between the flats, there is a very useful feature in some image processing tools. PI has it, and Siril has intruduced the feature recently as well. Its called "pixel math". With pixel math I simply divided the L-filter by the OIII-flat to see the differences. The result is stunning:

So here you see the rings again, coming from the a slightly different focus of the two flats.
One thing is certain: pixel math is an extremely useful tool to investigate problems in image processing !!!
Helpful Insightful Engaging Supportive
Reg Pratt avatar
Reg Pratt avatar
I experienced this exact ringing problem in the calibrated lights with my RCs. After weeks of troubleshooting I finally found that it happens when my flats are too bright. I started underexposing my flats significantly and the problem went away immediately.
Well Written Helpful Insightful Concise