Canon 200D - with Irix Firefly 15mm F/2.4 - strange star shapes

[brenski]

I'm really hoping someone can help with this one.

A little bit of background: Initially, I thought the issue was build quality of Samyang 14mm lens. However, I've now purchased a better lens and been to a dark sky site to take his latest batch of Milky Way images.

Equipment: Canon 200d, Irix Firefly 15mm F/2.4 lens, tripod, shutter release cable.
Settings tried: ISO 800-3200, shutter 20-25secs, aperture F/2.4-and various stops down.

Issue: Stars are dashes (even in centre image). This was the same issue with the Samyang (previously). I'm now struggling with why the Irix is producing the same results. Could it be the Canon 200d has a problem?

Note: I've also tried OFF/ON settings for each of the following (in different combinations):

• Lens Aberration
• High ISO NR
• Mirror Lock

Results do not change.

Does anyone, have any ideas that may help me resolve this?

thanks

Edited September 9, 2021 by brenski

[Cornelius Varley]

That will be caused by the stars moving at 15 deg/hour. Take a few dozen or so and create a startrails image.

 

[michael8554]

Hi brenski

If those star shapes were only in the corners, and the centre stars round, this would be a lens feature called Coma.

Since all the stars are elongated, that's due to the stars drifting across the sky during the exposure, exaggerated in Peter's image.

Try faster exposures until you find one that doesn't show drift.

Neither of your lenses is the cause of the problem.

Michael

[brenski]

thanks guys. One of the other images I'd take was around 12secs - it still has some coma - nowhere near as much, but it is there nonetheless.

Attached image is  shortest length exposure taken: F/2.5, 12.5sec, ISO6400

Do you think it's any better?

Much of what I've read suggests 30secs untracked for Milky way shots, but that's appears to be not possible. So, how with <15secs do I capture enough to be able to get decent Milky Way detail? what exposure time can I reduce to and still obtain reasonably-detailed Milky Way shots?

What about processing? What do you do to bring out the best of an image?

 

[michael8554]

2 hours ago, brenski said:

One of the other images I'd take was around 12secs - it still has some coma - nowhere near as much, but it is there nonetheless.

As I said before, this isn't coma, it's star drift.

The exposure that works depends on the Focal Length of the lens, the longer the focal length, the more the star drift is "magnified".

If you're happy with the 12 seconds results, that's all that matters.

To improve requires taking say 20 exposures in quick succession and "Stacking" them together to make one improved image.

Then using Photoshop or equivalent to enhance the contrast and colour to your satisfaction.

You might need to merge one image of the landscape into the Stacked image.

Michael

[brenski]

8 minutes ago, michael8554 said:

As I said before, this isn't coma, it's star drift.

The exposure that works depends on the Focal Length of the lens, the longer the focal length, the more the star drift is "magnified".

If you're happy with the 12 seconds results, that's all that matters.

To improve requires taking say 20 exposures in quick succession and "Stacking" them together to make one improved image.

Then using Photoshop or equivalent to enhance the contrast and colour to your satisfaction.

You might need to merge one image of the landscape into the Stacked image.

Michael

thanks Michael. there's a little bit of a mind-bend happening here. i've read so many articles/tutorials/books etc on this and they all claim the same thing:

500/focal length = max exp (before drift appears). therefore with a 15mm lens, this should give around 20-22secs before drift becomes noticeable. In truth, I've just been looking at two different books: (Night Sky - Jennifer Wu / Landscape Astrophotography - Mike Shaw), both confirm untracked imaging with a cropsensorand 15mm lens, should be fine around 21/22 seconds. 

so, this has me utterly bewildered.

[Cornelius Varley]

1 hour ago, brenski said:

500/focal length = max exp (before drift appears). therefore with a 15mm lens, this should give around 20-22secs before drift becomes noticeable.

That would depend on how you define "noticable".