Back to forums

Intra-star color problem...

ZWO ASI183MC Tilt Artifacts
The OSC Club (One Shot Camera) •14 replies•234 views
Michael Feigenbaum avatar
Michael Feigenbaum avatar
I use and ASI183mc which I like very much for the most part…

However, I have a problem from time to time where when I process an image, there is a color gradient within the brighter stars.  In these cases, the star might be nice and round but on the right side, for example, it will have a orange tint and then on the left side it might have a bluish tint.

I have seen this on many other images here on AB and I am wondering of anyone has fought this battle and won.

My thought is that this must be caused by a tilt in the image train, or a tilt in the sensor itself, which is causing the different colors to have different focus points.

I'd be interested in the thoughts of others on this subject…
Well Written Insightful Respectful Engaging
Frédéric Auchère avatar
Frédéric Auchère avatar
Hi Michael,

Could you precise with what instrument you see this? Is it only with the brighter stars? And is the red blue gradient always in the same direction? Can you give an example?
With  a refractor or camera lens (or any system that has a refractive element), if you always see blue (or red) fringes further away from the image center (i.e. blue-red (or red-blue) in the left size of the image and red-blue (or blue-red) fringes on the right side), what you see may be chromatic aberration. In addition to a slightly different focus in the blue and the red, the net effect is a slightly different plate scale (angular size of the pixels) in the blue, and in the red. That means that your image is slightly zoomed in (or out) in the blue (or red). The way to fix that is to determine the scaling factor between the channels and scale all of them to the same plate scale. If the effect is not huge a linear approximation (simple zooming factor) may be sufficient, but there can be higher order effects.

CS

Frédéric
Helpful Insightful Respectful Engaging
Rich Sky avatar
Rich Sky avatar
Yes, I think you are correct with a tilt in the image train or with the camera and the scope. Now, This issue can be corrected in post processing by doing a color shift (red/blue shift by 0.1 to 5 pixels ) or pre-acquisition by collimation.
This problem of mis-alignment becomes more important at lower f ratio, I believe, and very common with the Rowe Ackermann Schmidt Astrograph (RASA).

Now, one other thing, if you have a dark spot inside a star, well, the camera is near focus but not at focus.

In the image below, I have manually shifted the red channel by 35 pixel, if I realign the channel, the stars are all 'normal'.

Clear skies.

Helpful Insightful
Elmiko avatar
Elmiko avatar
Very interesting! I learn something new every day on this site!
Well Written Respectful
PrairieSkiesAstro avatar
PrairieSkiesAstro avatar
Greetings! I use a Celestron 9.25" EdgeHD SCT telescope and an ASI294 MC Pro camera. I notice the same behavior as you in the corners of my images (mostly in the upper left corner). I assume this is because I have a tilt in my optical train somewhere (the stars are also a bit elongated in the corners). Can you (Frederic and/or Rich) elaborate a little bit more on how you fix this (say in Pixinsight) by "scaling" or "color shifting" as you describe above?

Thanks!
Well Written
Frédéric Auchère avatar
Frédéric Auchère avatar
Hi Brad,

What I described (color aberration) will manifest itself by an opposite order of the colored fringes on opposite sides (or corners) of the images. And it should be non-existent at the center of the frame. The fix in that case is a rescaling of the channels.
If the ordering of the colors is the same everywhere in the image, then it may be dispersion (the system behaving as a prism). In that case, as described by Rich, the fix is a shifting of the channels.
I don't use PixInsight, but I suppose that you can register the channels against each other after stacking (the way you register the subs for stacking). That would work both for rescaling and shifting.

BTW if it is a shift, it can also be due to atmospheric refraction. This you can check with the orientation in the images. It should be along the vertical of your object.

Cheers,

Frédéric
Well Written Helpful Insightful Engaging
Michael Feigenbaum avatar
Michael Feigenbaum avatar
Great responses, thank you…  Frédéric, I am using a WO GTF81 which is an apochromatic refractor from William Optics.  This supposedly has a 3 element air spaced objective made from FPL-53 glass.  I was using it at its native focal ratio of f/6.5 but have also used it with a 0.8x reducer which yields f/5.3.

Thank you very much for your thoughts on this.

Bons Ciels,
Mike
Well Written Respectful Concise Supportive
Michael Feigenbaum avatar
Michael Feigenbaum avatar
Rich Sky:
Yes, I think you are correct with a tilt in the image train or with the camera and the scope. Now, This issue can be corrected in post processing by doing a color shift (red/blue shift by 0.1 to 5 pixels ) or pre-acquisition by collimation.This problem of mis-alignment becomes more important at lower f ratio, I believe, and very common with the Rowe Ackermann Schmidt Astrograph (RASA).

Now, one other thing, if you have a dark spot inside a star, well, the camera is near focus but not at focus.

In the image below, I have manually shifted the red channel by 35 pixel, if I realign the channel, the stars are all 'normal'.

Clear skies.

Exactly... thank you very much for this information!
Frédéric Auchère avatar
Frédéric Auchère avatar
Michael Feigenbaum:
Great responses, thank you...  Frédéric, I am using a WO GTF81 which is an apochromatic refractor from William Optics.  This supposedly has a 3 element air spaced objective made from FPL-53 glass.  I was using it at its native focal ratio of f/6.5 but have also used it with a 0.8x reducer which yields f/5.3.Thank you very much for your thoughts on this.

Bons Ciels,
Mike


Thanks for the details! On that oneit does not seem to be either a pure shift or a pure scaling. This could be caused by an unperfect collimation.

Frédéric
Michael Feigenbaum avatar
Michael Feigenbaum avatar
That was my fear… I will investigate.

thank you very much!
Rich Sky avatar
Rich Sky avatar
Thank you for sharing your image.

R
Frédéric Auchère avatar
Frédéric Auchère avatar
Hope you find the cause, those can be hard to track down.

Cheers
Well Written Respectful
Colin avatar
Colin avatar
There is two different types of chromatic aberration; lateral and longitudinal.

Lateral CA tends to have red pointing towards the centre and blue pointing outwards, this will get worse as you move further off axis. Lateral CA can also be even across the entire frame which is what you seem to have. I get the same thing with my Mewlon 250CRS. It was also something I saw with a Samyang 14mm F/2.8 I used to have.

Longitudinal CA is what leads to blue or purple halos around stars.
Helpful Concise
Michael Timm avatar
Michael Timm avatar
I am having the same issue like the OP Mike (and like many others). Very interesting discussion, thanks @ll.
Michael Feigenbaum avatar
Michael Feigenbaum avatar
Frédéric Auchère:
Hope you find the cause, those can be hard to track down.Cheers


Thank you very much, you have given me a lot to think about, very much appreciated!
Well Written Respectful Supportive