Diffraction Distraction

When it was time to pick my college major, I was torn: Outer space or cyber space? I chose cyber for the job prospects, figuring I could always pursue astronomy as a hobby.

Many years later, and here I am, pursuing it as a hobby. My setup is still pretty basic (much to the dismay of my wife who thinks I've spent entirely too much money already). I'm using a Sony a7 IV, and a Rokinon 135mm f2 lens mounted to a Star-Adventurer 2i tracker that I bought used from eBay. I love a good deal.

I've managed to get some nice photos. Nothing APOD worthy, but I'm having fun learning the constellations and how to find things in the sky. 135mm produces some very wide field shots, and unless there's a big ol' nebula filling the frame, they can feel sort of... flat. Not boring, I mean, there's freaking galaxies in there, man, but flat. Nothing to really draw the eye.

I love the Hubble Space Telescope. The Hubble Deep Field images are some of the most incredible images ever captured by humanity. Having grown up on Hubble images, I have its star patterns seared into my memory.

While I'm never going to get anywhere close to that... Look at the stars! So dreamy, so pointy. These are called diffraction spikes. They're an optical side effect of how Hubble captures its images. TLDR: Light hits thing and makes line. Here's a great graphic by NASA:

You can get them with any camera by stopping down the lens. The light hitting the aperture blades will create some spikes. The pattern will depend on the number and shapes of your lens' blades. Here are some cool spikes I capture from Jupiter with my vintage CZJ Flektogon 35mm:

Stopping down has a major drawback though: way less light. It's probably pretty obvious that when shooting objects 50 million light years away, the amount of light that hits the camera sensor is a crucial factor. More light is usually better.

The spike pattern is also very much determined by your lens' construction. I wanted a bit more control. For example, I want to be able to choose the size and shape of the spikes based on what I'm shooting.

Enter: Diffraction Masks. These put a barrier between the light and your camera sensor. So light will hit the thing and create line. I had read online that some people had luck with fishing line or wire. But I have a 3D printer. And when one has a 3D printer, every project suddenly starts requiring 3D printing.

A few minutes of CAD later:

Boom. Baby Hubble. The resemblance is uncanny.

I wasn't expecting much, but the results were actually pretty rad. I shot the Coma Star Cluster in the constellation Coma Berenitis over two nights, totaling about 3 hours of integration. Some stacking and processing in Siril later, and:

A wide field shot with some additional depth! NGC 4565 (the Needle Galaxy) added some more flavor.

Some people don't like these spikey stars, and that's ok. Photography is subjective. I personally love them, especially in wider shots like this. They turn a fairly standard wide field shot into something that "pops".

Plus, as vast as the universe is, we are all looking at the exact same night sky. There's only so many ways we can point our cameras up there to capture shots. Having one additional gear I can turn to tune things helps keep the creative spark alive.

More Gears

I am continuing to try out new patterns to see what results I can get. The goal is to build a library of diffraction masks so I can swap them out based on the look I want. This is ongoing. I may eventually upload these designs to Printables, but they're pretty easy to replicate in any CAD software and will probably need to be resized to fit your lens hood anyway.

I'm not super well versed on the technical aspects here so forgive me if my terminology is off. Just due to how I built this in CAD, I'm referring to a "spoke" as one of the lines from the center of the circle to the outer edge. Like a bike wheel.

135mm Gallery

Here are some test shots with my Rokinon 135mm f2 lens. Click an image to expand.

3 Spokes

Odd numbers double the spikes.

4 Spokes

Even numbers have the same number of spikes.

6 Spokes

Identical to the 3 spoke design. Light is weird.

8 Spokes

Here's an image I attempted of the Leo Triplet. After a few hours on night 1, I decided that I didn't really like how the 8 pointed stars looked near the galaxies. I stopped gathering data and switched to the 4 spoke mask instead. Theta Leonis at the top looks glorious with 8 points, but the 8-pointed star by the galaxies was too distracting. Again, gears.

Hashtag

This looks identical to the 4 spoke design. Light is really weird.

360mm Gallery

Since the original writing of this post, I've acquired a Zenithstar 61 II. The diffraction masks work with this scope! The lens hood on my 135mm lens is fixed so I can't adjust the distance between the lens and the diffraction mask. On the Z61, however, the hood (aka dew shield) is extendable. Will this affect the look of the diffraction spikes?

To find out, I printed 3 designs and tested each with the dew shield extended all the way out, all the way in, and somewhere in the middle.

Here are the masks in the same order as the results below:

I'm naming these The Line, The Wonky T, and The Classic Cross.

Based on the results so far can you guess what pattern each of these will produce?

2 Spoke - The Line

Can you tell which one is closest to the lens and which is farthest? Neither can I. There's maybe a very slight difference in spike length, but it's extremely minor if present at all.

They're ordered by closest to farthest from the lens. Moral of the story: Extend the dew shield, put on the mask. Shoot away.

The Line mask is one of my favorites. The spikes can be aimed by rotating the mask. When I'm happy with the orientation I usually lock things down some with painter's tape so it doesn't shift throughout the night. Here's a quick shot I took of M35 with this diffraction mask:

3 Spokes - The Wonky T

Again, no perceivable difference. The Wonky T produces some nice 4 pointed stars.

4 Spokes - Classic Cross

Same results.

Wonky T vs Classic Cross

You'll notice that the Wonky T and the Classic Cross produce the same 4-pointed diffraction spikes. Is there a difference between the two? On the left is the Wonky T, the right is the Classic Cross, both at the farthest dew shield position:

The result is subtle, but I do notice that the Wonky T produces slightly less bright spikes in one direction. In the above example, the horizontal spikes are slightly less bright than the vertical ones.

What's Next?

I'm still experimenting with different mask types but have settled on my two favorites: The Line and the Classic Cross. More than 4 spikes per star starts to look a bit busy, in my opinion. But it could be useful for shots with only a few bright stars.

I'm going to keep experimenting with different diffraction masks types. Here are a few ideas:

• A "donut", where the spokes are attached to a central circle.
• Short spokes that don't extend all the way across the frame. Do they need to cover the full frame for the effect to happens? No idea!

If you have any other ideas, I'd love to hear time. I'll update this post as I learn more.

Contact

Questions? Comments? Concerns? Just want to chat? My public profiles are on Mastodon and Bluesky.