Hey all,
does anyone know what can be the cause of these double stars? first thought was pinched optics or tilt but this seems like a bit too much for just that.
Thx for the help guys :)
the starmask is a 7/8H stack lightly stretched
Explore Scientific ED102 triplet with a field flattener and asi mc533📷 starmask_M106_RGB.jpg
Unless it’s on every frame in your data set, it’s tracking errors from your mount. Backlash etc. It basically clunked into place.
Due to the backlash on my mount I get this after almost every time I dither.
I would blink through your frames one by one. To get this kind of effect there would have to be movement, not between subs but in single subs themselves. The stars would be in one position for part of the exposure and suddenly shift to a slightly different position for the rest of the exposure.
What does your control software chain look like? How exactly are you running everything?
This is not pitched optics. That would cause stars to be in a diamond shape. Tilt will cause out of focus in different corners of the chip.
Some corners will betoo far in-focus ans others will be to far out-focus. The area where the star is that was use to focus, will be in focus.
What is odd about this image is part of the star is in focus and the other part is out of focus with a donut ring. It's as if the optics are focusing the stars twice and slightly off center.
Does this happen in every sub frame? Does it happen with every session?
Did you only notice it in the final stack? If so, then it may be caused simply by some bad out of focus subs slightly not aligned property. Was there a meridian flip with focus shift.
If this was only in one session, I would suggest count that session as a lost and move on. If it has happened in other sessions then, remove the reducer and see if the situation continues. If the stars are then normal, the problem is the reducer. After removing the reducer and the situation continues, then rotate the camera and see if the situation continues in the same orintaion. If it remaines the same after rotation, then it is in the imaging train. It the star orintation distortion rotates, then it in the optics.
A possible guess would be the oppisite of pitched optics, loose optics with optican elements foucusing differently. But that is only a wild guess.
Another more likely conclusion is focuer shift caused by the optical train weight, as the mount slews.
Lynn K.
Lynn K · May 5, 2026, 02:42 PM
Another more likely conclusion is focuer shift caused by the optical train weight, as the mount slews.
That’s a great point, there may be some focuser shift due to weight/sag or something being loose in the optical train that’s moving as the mount moves through the night.
I think this is a rejection error / alignment / registration issue. Try a different reference image, maybe just after meridian flip. I have had similar issues before and can almost always be solved by changing reference image and or choosing other rejection methods. I doubt you’ll see this effect in a single sub.
So this is a stack of 7-8h, let’s say around 100-150 images?
If so, this looks like a problem with star registration. Is there any possibility that one group of images has been registered to a different reference frame than another group? For example between days or so?
The easiest way to check is to blink the registered images, while zooming in on a small group of stars. As you click through them, the stars should all be in exactly the same place. If images are registered to different reference frames, you will see stars in subsequent frames flip between two locations.
Your rejection algorithm will have rejected pixels in a very unpredictable way. That’s what could explain the donuts. Also the small stars are normal, the bright ones are the shifted ones. Possibly the rejection algorithm just completely rejected one of the two positions in the small stars, which would indicate that the group ‘wrongly’ registered images is much smaller than the other group.
Hard to tell exactly, but if the above assumptions are correct, I would have a close look at the registration process.
CS, Willem Jan.
bigCatAstro · May 5, 2026, 05:03 PM
Lynn K · May 5, 2026, 02:42 PM
Another more likely conclusion is focuer shift caused by the optical train weight, as the mount slews.
That’s a great point, there may be some focuser shift due to weight/sag or something being loose in the optical train that’s moving as the mount moves through the night.
The thing is, what I see is a sudden shift in the middle of a sub. Sagging would be a gradual shift that would be taken out during stacking alignment.
Tony Gondola · May 5, 2026 at 10:09 PM
bigCatAstro · May 5, 2026, 05:03 PM
Lynn K · May 5, 2026, 02:42 PM
Another more likely conclusion is focuer shift caused by the optical train weight, as the mount slews.
That’s a great point, there may be some focuser shift due to weight/sag or something being loose in the optical train that’s moving as the mount moves through the night.
The thing is, what I see is a sudden shift in the middle of a sub. Sagging would be a gradual shift that would be taken out during stacking alignment.
That’s a good point too. It’s an interesting issue and probably needs more session data to diagnose.
I had a similar issue about 6 months ago. Every star was doubled in about one out of every few hundred frames. Since the doubling was occurring in the original sub-frames it was clear that it wasn't a registration issue. And this was with a 10 Micron gm1000 mount.
The problem disappeared a few months ago as mysteriously as it began and I still do not know what the cause was.