Astigmatism affect on imaging
Posted 02 May 2013 - 11:05 AM
Posted 02 May 2013 - 02:58 PM
How did you determine that your scope has astigmatism?
Astigmatism is a condition where there are two points of focus, and most people will focus somewhere between these to points.
With this condition, the focal plane is folded like a Pringle potato chip or a saddle seat.
A little astigmatism is not the end of the world.
But again, the question is how did you determine that the scope has astigmatism to begin with. There are some conditions that can have a somewhat similar effect.
Attached is a simulation of .5 waves of astigmatism.
It is very very very unusual to see anywhere near this much astigmatism in anything but department store telescopes.
The red line in the MTF plot shows the contrast loss for a perfect aperture, and the dotted line shows the additional contrast that would be lost because of the amount of astigmatism shown.
The star images so in focus, how an equal magnitude double would look, and how this much astigmatism would look with 2 waves of defocus.
Are you sure you have astigmatism? If you are, it would take this much to do meaningful damage to the image.
.2 Or even .3 waves is not exactly cosmetic, but not enough to be easily seen in focus on extended targets. It would show on an in-focus star though.
Posted 02 May 2013 - 03:01 PM
Posted 02 May 2013 - 06:59 PM
Posted 02 May 2013 - 07:12 PM
If it is not easily perceptible that the ring is not round, then the astigmatism is likely minor.
On the other hand, if the first ring starts to look more like a pronounced oval, then there could be an issue.
If it is a Celestron, has the secondary mirror even been removed and replaced? From time to time, Celestron would touch up a secondary mirror to null out some astigmatism in the primary, and if this was the case and someone rotated the secondary mirror, it could be that perhaps the situation could be improved.
But if the error is not as severe as is shown in my Aberrator simulation, and it is not a Celestron where the secondary has perhaps been rotated, then I guess there is little to do about it.
Astigmatism is not a good error to have if it is more than a little, but even a little will show on the first diffraction ring if you are a careful observer.
But there is this... If you do not see an oval when you defocus, then the other likely candidate is some surface roughness in the mirror.
Dog Biscuit error (unusual in SCTs) leaves regular low spots in the mirror. When in focus, the rings can show this as uneven intensity.
If you defocus and don't see an oval, but the light pattern around the rings is irregular, that could be some surface roughness.
Posted 02 May 2013 - 08:08 PM
The diffraction rings when out of focus seem round, so I guess its best to just leave it alone.?
Posted 02 May 2013 - 08:38 PM
And again, you could be seeing some minor surface roughness which can have the effect of making the first ring unevenly illuminated.
If you see the first ring emerge as more or less round on both sides of focus, even if there is any astigmatism, it would in this case be minor.
Again, see the picture above. If you defocus a small amount and the first ring looks round, and you go though focus and it looks round on the other side, my guess is that if there is astigmatism it is quite minor. If it looks like the picture above, that would be astigmatism.
I don't think I have ever seen any meaningful astigmatism in a Celestron SCT, so that is why I was curious about it.
Posted 04 May 2013 - 10:06 PM
Posted 04 May 2013 - 11:17 PM
Both of the C11s I've owned have shown astigmatism. And just like yours, they produced very fine planetary images:
Scope 1 Jupiter
Scope 2 Jupiter
I only noticed Scope 1's astigmatism during deep sky imaging, when stars were sometimes elongated. I spent six months chasing this down as a tracking error before Glenn and Eddgie helped me recognize it for what it was. If you want to follow the whole ordeal, it's all linked in this thread.
In short, I noticed it because as temps changed - and focus shifted - during the night, stars would elongate. And even when focused, they could show diamond shapes in poor seeing (due to turbulence rapidly shifting focus points of individual stars, IIUC.) Celestron replaced Scope 1 with Scope 2. Scope 2 has less astigmatism than Scope 1, but it still shows it. Hey, I know what to look for now!
Should you care? As Eddgie suggested, it depends how much astigmatism you have. With Scope 2, I've learned to stop worrying and love astigmatism. As long as I stay well-focused, I have no problems (Scope 2 shows no diamond shapes). And because star elongation rotates 90 degrees as I pass through focus, when focusing I can tell which direction I need to move, not just that I'm out of focus.
Would my images be better without astigmatism? I kind of doubt it. I'm pretty happy with how my planetary images look vs other 11 inch scopes. And I'm pretty satisfied with my recent deep sky images. Having seen it in both of my C11s, my hunch is that most mass-produced SCTs have some degree of astigmatism, though most users would never notice it. Either that or I'm the Unluckiest Astronomer in the Universe (which I have sometimes thought!).
Hope that provides encouragement.
Posted 06 May 2013 - 11:16 AM
Here is a simulation of Mars with different amounts of astigmatism yielding a Strehl from approx 0.9 (~0.25 wv PTV) to 0.7 (~0.47 wv PTV). The original image was extracted from WinJupos for the time noted and the planet was then at 11.77 arc sec in diameter. The semi-ideal image (it is more contrasty than a real Mars) was then convolved with a point spread function aberrated by the various amounts of astigmatism. The scale is 0.04 arcsec / pixel.
Doubtful you have as much astigmatism (AST) as the 0.7 strehl case as it would be very obvious visually in the out of focus images of a star with an oval changing orientation by 90 degrees.
Likely the astigmatism is a result of a combination of things. I have seen it from time to time in scopes with little if any astigmatism. Probably a combination of thermal and seeing effects. Its also possible to misdiagnose AST from captured star videos if they are not oversampled enough as the stacking software produces inaccurate up/down and left/right movements.
A decentered corrector in the amount of ~1-2mm would result in detectable astigmatism.
If you are convinced the AST is due to a decentered corrector, and you know by its orientation the correspondence with points on the corrector as they relate to a sensor image you could do the following at your own risk. Move the corrector up and down or left and right with respect to the orientation of the astigmatism. In one direction it will get worse and the other improve. I've lost in all my notes the relation AST from decentering has with respect to an intra vs the extra focal image. I do recall I got the info from Bob Piekiel.
It is true one can partially correct for AST in deconvolution, you need IP software that allows for reading in a PSF (or star image) with the same orientation and aberration. Seldom does one have an image and PSF with enough signal to noise for successful image restoration. Capturing an accurate PSF is not easy for a larger scope unless seeing is quite good. Its vitally important not to overexpose the central peak as it will destroy the linearity of the PSF. In addition the PSF should be oversampled more than is usually the case for high resolution imaging.
Posted 06 May 2013 - 11:30 AM