Second tracked image - please critique

[GazK]

So, flushed with the relative success of my first tracked image, and amazed at more actual clear skies, I decided to go for broke and step up to my WO Zenithstar 71, still on the Star Adventurer. Took advice from here and dropped ISO to 400 - too scared of failure to go to 200! - and pushed subs up to 30 secs. 40 mins of data, no darks, no flats (because I'm an idiot who can't read a histogram!)

I mean I'm blown away at this image considering I did my first polar alignment a week ago, and before this month had never imaged a DSO, and this is close to the weight limit of the mount. But as always I'm keen to know what I can do to improve on this.

[GazK]

Is it just me or does the colour temperature need warming up a little?

[Swillis]

I think going up to the WO71 was a good idea, the star shapes look much better. Maybe it was your lens which was the problem with the star shapes previously. Zoom lenses are not great for AP, maybe there are some good ones out there but I have all sorts of weird star shapes with mine. 

I agree, I does look a bit blue.  What are you processing in?  In Siril you can do photometric colour calibration, which works brilliantly.

Great work 👍

[alacant]

11 hours ago, GazK said:

what I can do to improve on this

Hi

Lovely shot.

Minor improvements which are easy to do...

1. 80d: ISO 200. 100 may be even better. 

2. Use flat frames: Without removing the camera, set the 80d on Av and point the telescope at an evenly lit light source so as to give an exposure of around 2s. Use distance/paper/card/t-shirts to control the light. Switch back to M and set the same exposure. Take at least 15 frames. 

3. IIRC, to control the halos on a wo, focus on a white star using a bm. remove the latter then ease focus away from that point until the blue halo begins to turn red. Now compromise between those two focus positions. 

4. Use a field flattener

53 minutes ago, Swillis said:

Siril you can do photometric colour

+1. 

Cheers and HTH

[AstroMuni]

Here is a screen shot of what it could look like. I have applied Photometric calibration in Siril to your uploaded image @GazK

Your shot is really good...And to be fair you have taken on one of the harder DSOs - its easy to capture but hard to process due to the extreme dark & bright shades on it.

Edited March 2, 2022 by AstroMuni

[alacant]

1 hour ago, AstroMuni said:

Photometric calibration

Nice.

@OP: you could apply that to the linear image, then process. It should really shine.

Edited March 2, 2022 by alacant

[Elp]

It's a good image. Orion nebula is so bright I think you can get away with a low iso, even short exposures of 2-4 seconds duration it captures decently without blowing the core out too much, 10 seconds even better. You need to find the balance based on what settings you're likely to use.

[GazK]

9 hours ago, Swillis said:

I think going up to the WO71 was a good idea, the star shapes look much better. Maybe it was your lens which was the problem with the star shapes previously. Zoom lenses are not great for AP, maybe there are some good ones out there but I have all sorts of weird star shapes with mine. 

I agree, I does look a bit blue.  What are you processing in?  In Siril you can do photometric colour calibration, which works brilliantly.

Great work 👍

Thanks very much! Yes, I'm still at the stage of using what I have from visual astro and normal photography, so the tele zoom is far from ideal. Don't get me started on the manual focuser (ugh) and tendency for the zoom to slide at high elevations. At some point I'll probably be dropping cash on a Rokinon.

Stacking was DSS, processing in photoshop. I'll have a play with Siril, thanks

[GazK]

8 hours ago, alacant said:

Hi

Lovely shot.

Minor improvements which are easy to do...

1. 80d: ISO 200. 100 may be even better. 

2. Use flat frames: Without removing the camera, set the 80d on Av and point the telescope at an evenly lit light source so as to give an exposure of around 2s. Use distance/paper/card/t-shirts to control the light. Switch back to M and set the same exposure. Take at least 15 frames. 

3. IIRC, to control the halos on a wo, focus on a white star using a bm. remove the latter then ease focus away from that point until the blue halo begins to turn red. Now compromise between those two focus positions. 

4. Use a field flattener

+1. 

Cheers and HTH

Thanks! Yes will drop to 200 now I know my effort won't be wasted. For some reason I'm struggling to get flats to work, will practise in the warm this cloudy weekend. 

And I'll give that focusing trick a go. In the medium term I want a FF, but its quite a lot of £££ when I'm still working out where I want to go with the hobby.

[GazK]

3 hours ago, AstroMuni said:

Here is a screen shot of what it could look like. I have applied Photometric calibration in Siril to your uploaded image @GazK

Your shot is really good...And to be fair you have taken on one of the harder DSOs - its easy to capture but hard to process due to the extreme dark & bright shades on it.

Thanks! That's really helpful. Just lifting the exposure and moving the black point in the final image has brought me closer to your process.  I'm having a play with Siril now

[glafnazur]

It's a lovely shot, well done 👍

[GazK]

16 minutes ago, glafnazur said:

It's a lovely shot, well done 👍

thank you!

[GazK]

1 hour ago, Elp said:

It's a good image. Orion nebula is so bright I think you can get away with a low iso, even short exposures of 2-4 seconds duration it captures decently without blowing the core out too much, 10 seconds even better. You need to find the balance based on what settings you're likely to use.

Thanks! The 80D has the advantage of being ISO-invariant, which I assume must have helped hold the core down despite 30 second subs.

[Swillis]

26 minutes ago, GazK said:

tendency for the zoom to slide at high elevations

Electrical tape is your friend (or masking tape, but it doesn't stick so well). Same for the focus. 

[GazK]

Taken on the feedback and processed in Siril, with just the lightest of touches at the end in PS. Feels a much more natural image to me, plus the background is a bit more under control. Could have gone brighter but then I lose the core. Comments welcome. 

[AstroMuni]

1 hour ago, GazK said:

Taken on the feedback and processed in Siril, with just the lightest of touches at the end in PS. Feels a much more natural image to me, plus the background is a bit more under control. Could have gone brighter but then I lose the core. Comments welcome. 

Looks great. Try a bit of deconvolution to make the stars more crisp.

[glafnazur]

That looks loads better, great detail and it looks more natural, nice one 👍

[GazK]

16 hours ago, glafnazur said:

That looks loads better, great detail and it looks more natural, nice one 👍

thanks!

[alacant]

On 02/03/2022 at 21:12, GazK said:

its quite a lot of ££

The best way of finding out would be to try one at an astro club. 

[ollypenrice]

I'm surprised nobody's mentioned this, but there is something rather odd going on in this image. I'm afraid I can't offer an explanation for it, though. It's very obvious in the first version but still present, to a much lesser extent, in your excellent rework. (The rework really is excellent.) The thing is that the colour simply drops dead below the line in this screen grab of the original.

 

My own data do show hints of a similar effect but something in your workflow may have exaggerated it. Was there a radical change in the processing between first and second? If so it would be a good idea to identify it because it's going to be important.

Another thing: a key aspect of image processing is managing different parts of the image in different ways. Sometimes this is called The Zone System. In M42 it involves processing, and even capturing, the Trapezium differently from the rest. If you don't have a set of short exposures for the Trapezium it isn't the end of the world. You can do a separate stretch for the Trap, much gentler, and you can boost the saturation to get colour back into it.  Then you can combine the two in an HDR process (High Dynamic Range.)  There are countless ways to do this. You can do it manually in a layers-masks program like Photoshop, or use an automated version of the same (again in Ps or similar), or there are HDR packages like Photomatrix to try.  I dare say the other astro-specific programs have something as well but I don't know them myself. This target absolutely does need different stretches for Trap and the rest. Your data are worth the effort.

Olly

[vlaiv]

41 minutes ago, ollypenrice said:

I'm surprised nobody's mentioned this, but there is something rather odd going on in this image. I'm afraid I can't offer an explanation for it, though. It's very obvious in the first version but still present, to a much lesser extent, in your excellent rework. (The rework really is excellent.) The thing is that the colour simply drops dead below the line in this screen grab of the original.

My guess would be that it is due to synthetic flats / software background removal.

[GazK]

4 hours ago, vlaiv said:

My guess would be that it is due to synthetic flats / software background removal.

First image was stacked in DSS, then followed one of Nico Carver's videos for processing in photoshop, admittedly that video was for an untracked DSLR stack. Second image was using Siril for stacking and most processing, using a different video tutorial.

I just assumed that the lack of colour was down to the limitation of 30 second subs, I had to work quite hard to get anything at all out of the lower portion. But I screwed up the flats and so used a different session's flats, that could be the problem. And yes I did use the Siril background removal tool.

I do have some 15sec subs from that session, so I'll try adding that as suggested for the core, might allow me to push the faint neb further in the 30sec stack.