Diffraction ring pattern around bright stars

[Andyy]

What causes these diffraction ring pattern around bright stars? Visible in single subs but become more prominent in the stacked image.

Equipment used: Skywatcher 130PDS, Baader MPCC MK III, Baader 7nm Ha filter, ZWO ASI1600MM

Edited October 14, 2021 by Andyy

[Andyy]

I'll answer my own question in case others experience the same. After som research I found out that there rings are quite common in obstructed optics and there is nothing wrong. The rings are caused by the secondary mirror. They can even be seen on images taken by the Hubbe Space Telescope. When processing images the rings easily blends in and become more or less invisible in the final image.

Edited October 14, 2021 by Andyy

[tooth_dr]

3 hours ago, Andyy said:

I'll answer my own question in case others experience the same. After som research I found out that there rings are quite common in obstructed optics and there is nothing wrong. The rings are caused by the secondary mirror. They can even be seen on images taken by the Hubbe Space Telescope. When processing images the rings easily blends in and become more or less invisible in the final image.

What scope is it?  Havent seen this myself in any images of my own.

[Adam J]

On 12/10/2021 at 18:08, Andyy said:

What causes these diffraction ring pattern around bright stars? Visible in single subs but become more prominent in the stacked image.

Equipment used: Skywatcher 130PDS, Baader MPCC MK III, Baader 7nm Ha filter, ZWO ASI1600MM

Am guessing that this is a very large aperture reflecting scope ?

If so then it is actually normal and can indicate very good optics you will see this on bright stars in Hubble narrow band images. Find a high resolution image of the pillars of creation and you will see the effect, but now quite so obvious unless you can find a mono Ha only image. 

 

Adam 

Edited October 14, 2021 by Adam J

[vlaiv]

3 hours ago, Andyy said:

I'll answer my own question in case others experience the same. After som research I found out that there rings are quite common in obstructed optics and there is nothing wrong. The rings are caused by the secondary mirror. They can even be seen on images taken by the Hubbe Space Telescope. When processing images the rings easily blends in and become more or less invisible in the final image.

I'm not sure this is the case.

You are referring to Airy pattern, and yes, every scope will under ideal conditions produce Airy pattern - which is most visible in narrow band images as pattern itself depends on wavelength.

Here is pattern produced by laser in lab:

(see https://en.wikipedia.org/wiki/Airy_disk)

Problem is that Airy pattern is much smaller than represented in your image above.

130PDS has Airy disk diameter of 2.54", which means that distance between fringes is half that or 1.27". Further, this scope has FL of 650mm and ASI1600 has pixel size of 3.8µm which together gives sampling rate of 1.21".

This means that single fringe (bright and dark part) is about as large as single pixel.

Fringes in your image cover about 4px:

I don't think this is Airy pattern we are seeing here, and I think that some other sort of diffraction is happening.

[Adam J]

3 hours ago, vlaiv said:

I'm not sure this is the case.

You are referring to Airy pattern, and yes, every scope will under ideal conditions produce Airy pattern - which is most visible in narrow band images as pattern itself depends on wavelength.

Here is pattern produced by laser in lab:

(see https://en.wikipedia.org/wiki/Airy_disk)

Problem is that Airy pattern is much smaller than represented in your image above.

130PDS has Airy disk diameter of 2.54", which means that distance between fringes is half that or 1.27". Further, this scope has FL of 650mm and ASI1600 has pixel size of 3.8µm which together gives sampling rate of 1.21".

This means that single fringe (bright and dark part) is about as large as single pixel.

Fringes in your image cover about 4px:

I don't think this is Airy pattern we are seeing here, and I think that some other sort of diffraction is happening.

I took a look and can confirm your findings. Looking a little further you would expect the observed fringes to be produced by a aperture of between 31mm (3 pixels between fringes) and 46mm (4.5 pixels between fringes) the higher end is about the size of the 130PDS secondary mirror shadow and the bottom end is about the size of a 31mm filter.  I suspect the secondary mirror on that basis. 

Adam 

 

Edited October 14, 2021 by Adam J

[vlaiv]

5 minutes ago, Adam J said:

I took a look and can confirm your findings. Looking a little further you would expect the observed fringes to be produced by a aperture of between 31mm (3 pixels between fringes) and 46mm (4.5 pixels between fringes) the higher end is about the size of the 130PDS secondary mirror shadow and the bottom end is about the size of a 31mm filter.  I suspect the secondary mirror on that basis. 

Adam 

 

I'm not sure that it works like that.

We see airy disk of aperture because we are "magnifying" image after it has hit the aperture.

So 130mm of aperture will cause diffraction with 2.47" airy disk diameter - if we don't have mirror or lens after to magnify that - no way we are ever going to see it.

You need extremely small apertures to be able to see diffraction effects without "magnification" of telescope. If you want to see airy pattern in lab - you need pinhole, laser and screen that is far enough so that pattern will be amplified enough (light is disturbed by small angle and you need enough distance for small angle to project into reasonable size image).

After light beam has passed primary mirror / lens - hitting secondary (that is flat) or filter aperture - is not going to produce anything. Any diffraction will be tiny as we don't have another "magnifying" element after it.

[Adam J]

41 minutes ago, vlaiv said:

I'm not sure that it works like that.

We see airy disk of aperture because we are "magnifying" image after it has hit the aperture.

So 130mm of aperture will cause diffraction with 2.47" airy disk diameter - if we don't have mirror or lens after to magnify that - no way we are ever going to see it.

You need extremely small apertures to be able to see diffraction effects without "magnification" of telescope. If you want to see airy pattern in lab - you need pinhole, laser and screen that is far enough so that pattern will be amplified enough (light is disturbed by small angle and you need enough distance for small angle to project into reasonable size image).

After light beam has passed primary mirror / lens - hitting secondary (that is flat) or filter aperture - is not going to produce anything. Any diffraction will be tiny as we don't have another "magnifying" element after it.

I mean the shadow of the secondary cast prior to hitting the primary mirror, as opposed to the flat mirrored surface itself.  It would be like cutting a circle out of the light incident on the primary of about 46mm diameter.  I believe that you get interference between this secondary pattern and the pattern caused due to the primary mirror diameter that is the essence of why you get poorer contrast with a reflector as opposed to a refractor?

Might be wrong though.  Having said this I did operate this combination myself for a while (130PDS and ASI1600mm pro) and dont remember this effect. However I never imaged a bright star in narrow band and very good seeing conditions either.  

Adam

Edited October 14, 2021 by Adam J

[vlaiv]

20 minutes ago, Adam J said:

I mean the shadow of the secondary cast prior to hitting the primary mirror, as opposed to the flat mirrored surface itself.  It would be like cutting a circle out of the light incident on the primary of about 46mm diameter.  I believe that you get interference between this secondary pattern and the pattern caused due to the primary mirror diameter that is the essence of why you get poorer contrast with a reflector as opposed to a refractor?

Might be wrong though.  Having said this I did operate this combination myself for a while (130PDS and ASI1600mm pro) and dont remember this effect. However I never imaged a bright star in narrow band and very good seeing conditions either.  

Adam

Obstructed aperture does create different distribution of energy in circles - but minima remains where it is - distance between fringes and their size remains the same.

As you can see from graphs above - minima and maxima are roughly aligned, but it is peak intensity that shifts from central disk to fringes. This is causing contrast loss in obstructed aperture.

[Adam J]

46 minutes ago, vlaiv said:

Obstructed aperture does create different distribution of energy in circles - but minima remains where it is - distance between fringes and their size remains the same.

As you can see from graphs above - minima and maxima are roughly aligned, but it is peak intensity that shifts from central disk to fringes. This is causing contrast loss in obstructed aperture.

Yes, but from the 20% example it looks possible to have a situation when every other ring is attenuated relative to unobstructed aperture even if the spacing between rings of differing brightness is the same. Suppose you could get an additional interaction with the pixel sampling of that effect occluding the underlying detail and only picking out the  delta between the brighter and darker rings.  The 130pds has a 36% obstruction though. I think that the reason that you still have the same fringe spacing is that more energy is placed into the rings created by the full aperture in comparison to the smaller amount contributed by the secondary shadow and as a result you will always see the primary rings and the secondary will simply modulate the primary rings as they move in and out of phase with them. If you hit the right ratio you can produce an effect similar to what is observed. 

As the size of the obstruction is increased the beat frequency will decrease and the attenuated rings will become darker / nulls deeper. 

Adam

Edited October 14, 2021 by Adam J

[vlaiv]

14 minutes ago, Adam J said:

Yes, but from the 20% example it looks possible to have a situation when every other ring is attenuated relative to unobstructed aperture even if the spacing between rings of differing brightness is the same. Suppose you could get an additional interaction with the pixel sampling of that effect occludes the underlying detail and only picks out the brighter rings.  The 130pds has a 36% obstruction though. I think that the reason that you still have the same fringe spacing is that more energy is placed into the rings created by the full aperture in comparison to the smaller amount contributed by the secondary shadow and as a result you will always see the primary rings and the secondary will simply modulate the primary rings as they move in and out of phase with them. If you hit the right ratio you can produce an effect similar to what is observed. 

As the size of the obstruction is increased the beat frequency will decrease and the attenuated rings will become darker / nulls deeper. 

Adam

You might be right on that one - but that is under assumption that there is no atmospheric blurring.

Look at thickness of diffraction spike. It is at least 6px across. That is consistent with say 3" FWHM or a bit more. This same blur affects fringes and they will be smoothed out.

Btw, we should be seeing same concentric rings in diffraction spikes as two "mix" like this:

To my eye - they look superimposed rather than "mixed" in above image.

I think that simplest cause of action is to alter slightly CC to sensor distance - maybe just shim it with less than 1mm shim. Any change in distance will alter optical path and interference pattern will change if it is due to reflection from CC (which is likely cause in this case).

 

[Whirlwind]

There may be other causes.  I think this came up in discussion before (on a Chroma filter IIRC) and I postulated that it could be Newtons rings formed from a spherical surface and touching surface next to it.  For example I imagine the coverslip and the microlens of the CMOS chip.  It is wavelength dependent so may only be seen in narrowband (as broadband the effects at each wavelength would merge).  This should be testable as the equation is well known.

The alternative I more recently thought of was perhaps because of how interference filters are designed (Interference filters - Electronic Imaging - Bedford Astronomy Club).  From my understanding narrowband interference filters can have several layers of material in them.  If some of the light reflects then you can get light 'bouncing' back and forth between the layers if they reflect some of the incident light.  This could result in rings of slightly defocused light reaching the CMOS/CCD.  This would be more dependent on the thickness between the material in the filters so will be harder to determine as companies are less likely to release this information. 

[sharkmelley]

The central obstruction causes a periodic modulation of the Airy rings:

Obstructed optics and rings around stars - Beginning Deep Sky Imaging - Cloudy Nights

Mark 

[vlaiv]

5 minutes ago, sharkmelley said:

The central obstruction causes a periodic modulation of the Airy rings:

Obstructed optics and rings around stars - Beginning Deep Sky Imaging - Cloudy Nights

Mark 

Then it would appear that @Adam J was right to imply that it is effect of central obstruction - it is modulation that shows as pattern here since individual rings are blurred by seeing?

[sharkmelley]

6 minutes ago, vlaiv said:

Then it would appear that @Adam J was right to imply that it is effect of central obstruction - it is modulation that shows as pattern here since individual rings are blurred by seeing?

The individual rings are too finely spaced to be resolved by the sensor but the modulated pattern has a wider spacing which can be resolved.

Mark