The "No EQ" DSO Challenge!

[Nigel G]

 Finally after an epic journey its done

1300D

1200D modified

The main PCB had shorted out, fitting a new one, no problem.....apart from the new board needs focusing to the camera. So an unmodified sensor needs to be fitted to focus the PCB to the camera, then the modified sensor can be re installed and bingo....

Come on clear sky's, I have a new toy to play with.

The modified camera can also be used for daytime photography by setting to jpg taking a photo of a white sheet of paper and doing a white balance if needed.

Nige.

[SteveNickolls]

Good luck Nige in this, can't wait to see the results. Here's to some clear, dark skies.

Steve

[Nigel G]

First light with modded camera.

First impression is I'm amazed at the difference in the subs, the Rosette is clearly visible compared to unmodded hardly visible.

I'm impressed, I managed 3 targets last night, have quickly processed 1 and am taking the wife away for 3 days so they will have to wait till later.

The difference is incredible,

Rosette nebula, modified Canon 1200D with 135mm lens at F3.5, CLS-CCD fliter . 50 x 30s with flats and bias. Star discovery mount.

Looking forward to getting home to take more time over processing.

A quick stretch and colour module produced this.

I also imaged Heart and soul nebula , which looked awesome in the individual 20s subs, and Orion and flame, all with 135mm

Nige

[Stub Mandrel]

Great, I've linked to your post in the thread where someone is asking if modding a camera is worth it.

[Nigel G]

9 minutes ago, Stub Mandrel said:

Great, I've linked to your post in the thread where someone is asking if modding a camera is worth it.

Thanks, I have been following that post  

[Filroden]

50 minutes ago, Nigel G said:

First light with modded camera.

First impression is I'm amazed at the difference in the subs, the Rosette is clearly visible compared to unmodded hardly visible.

I'm impressed, I managed 3 targets last night, have quickly processed 1 and am taking the wife away for 3 days so they will have to wait till later.

The difference is incredible,

Rosette nebula, modified Canon 1200D with 135mm lens at F3.5, CLS-CCD fliter . 50 x 30s with flats and bias. Star discovery mount.

Looking forward to getting home to take more time over processing.

A quick stretch and colour module produced this.

I also imaged Heart and soul nebula , which looked awesome in the individual 20s subs, and Orion and flame, all with 135mm

Nige

Wow. That's a great improvement in the nebula and from only 25 minutes. Can't wait to see the others. 

[parallaxerr]

What's your pixel resolution?

Just wondering what resolution your setups are providing? Mine's at 2.04"/pixel and after looking at some other images of M33, in comparison to my recent attempt, I'm wondering if I'm undersampling and if barlowing x1.5 to bring it down to 1.3"/pixel would help release a little more detail?

Problem is by doing this, I'd be upping the FL to 582mm and FR to 8.8 so not sure if there's any gains to be made?

Edit: I guess this would take my coma corrector that's optimised for f-ratios f3.5 to f6 out of play.

Edited December 12, 2016 by parallaxerr

[Filroden]

I'm  Imaging at 1.96"/pixel as I figure with seeing and tracking anything better wouldn't get any improvement in practice. 

[parallaxerr]

Makes sense Ken, the other images I'm looking at are EQ mounted with 5min subs so perhaps that's where the gains are.

I'm having no joy finding a Nikon astro-modding service in the UK so started looking at ready modded Canons but their pixel sizes are much larger than the Nikon so with my short scope I'd be massively undersampling. The only way to compensate is to go up in FL as far as I can see.

Looks like I'll be missing out on Ha until I can afford a new mount, scope & camera.

Edited December 12, 2016 by parallaxerr

[Stub Mandrel]

34 minutes ago, parallaxerr said:

Makes sense Ken, the other images I'm looking at are EQ mounted with 5min subs so perhaps that's where the gains are.

I'm having no joy finding a Nikon astro-modding service in the UK so started looking at ready modded Canons but their pixel sizes are much larger than the Nikon so with my short scope I'd be massively undersampling. The only way to compensate is to go up in FL as far as I can see.

Looks like I'll be missing out on Ha until I can afford a new mount, scope & camera.

If your pixels are twice as big you can expose for twice as long, choose bigger targets!

Worng.. forgot the issue was rotation..

Edited December 12, 2016 by Stub Mandrel

[The Admiral]

1 hour ago, parallaxerr said:

What's your pixel resolution?

Just wondering what resolution your setups are providing? Mine's at 2.04"/pixel and after looking at some other images of M33, in comparison to my recent attempt, I'm wondering if I'm undersampling and if barlowing x1.5 to bring it down to 1.3"/pixel would help release a little more detail?

Problem is by doing this, I'd be upping the FL to 582mm and FR to 8.8 so not sure if there's any gains to be made?

Edit: I guess this would take my coma corrector that's optimised for f-ratios f3.5 to f6 out of play.

I'm running at about 1.7"/pixel, when I'm using the focal length reducer/flattener, according to astronomy tools FoV calculator. The Dawes limit is supposedly 1.14". And of course, with a OSC camera the whole thing becomes complicated and is not what it seems, especially if the sensor has a low-pass filter in front of it. I'm not sure that one can afford to get too hung up about it. Not only that, but if you up the FL then you'll end up with fewer photons/pixel, which means longer integrated exposure times.

I don't think M33 is an easy target, and to me it always looks a bit fuzzy. I know it's relatively large but it isn't very bright. If you want something as 'clear' as https://stargazerslounge.com/topic/282951-m33-with-10-quattro-and-atik-383l/#comment-3096430, you'll need to remember that he's using a 10" f/4 (800mm FL) and presumably good seeing, rather better than we're accustomed to getting!

I'm still dubious about the 'necessity' of a modified camera, and I know I'm going against the tide here and no doubt there are many who'll argue the case, though it does rather depend on what sort of performance your unmodified camera has. And just what you want to image. Would an modified camera help with M33? Do you want to record exaggerated reds or natural colour?

Just my 2d. worth.

Ian

[parallaxerr]

1 hour ago, Stub Mandrel said:

If your pixels are twice as big you can expose for twice as long, choose bigger targets!

Worng.. forgot the issue was rotation..

And there in lies the rub, rotation! Are you saying, for an eq mounted rig, with less than ideal resolution, you can lengthen exposure to allow photons to spill over to adjacent pixels?

5 minutes ago, The Admiral said:

I'm running at about 1.7"/pixel, when I'm using the focal length reducer/flattener, according to astronomy tools FoV calculator. The Dawes limit is supposedly 1.14". And of course, with a OSC camera the whole thing becomes complicated and is not what it seems, especially if the sensor has a low-pass filter in front of it. I'm not sure that one can afford to get too hung up about it. Not only that, but if you up the FL then you'll end up with fewer photons/pixel, which means longer integrated exposure times.

I don't think M33 is an easy target, and to me it always looks a bit fuzzy. I know it's relatively large but it isn't very bright. If you want something as 'clear' as https://stargazerslounge.com/topic/282951-m33-with-10-quattro-and-atik-383l/#comment-3096430, you'll need to remember that he's using a 10" f/4 (800mm FL) and presumably good seeing, rather better than we're accustomed to getting!

I'm still dubious about the 'necessity' of a modified camera, and I know I'm going against the tide here and no doubt there are many who'll argue the case, though it does rather depend on what sort of performance your unmodified camera has. And just what you want to image. Would an modified camera help with M33? Do you want to record exaggerated reds or natural colour?

Just my 2d. worth.

Ian

Funnily enough, a while ago I compared my camera to yours Ian as I remember you saying you had good red response. Mine is poor in comparison, by about 50% as I recall but I'll be sticking with it as I like the camera.

I have read posts along the same lines of yours stating that modding isn't necessary, but they've all been using circa 300s subs on eq's which helps.

All I know is that my attempt at the HH was poor. It was however at low alt and in polluted skies, which doesn't help. Might be one to save for a dark site session.

[The Admiral]

I think that probably if you are going to go on a nebula safari that you might well wish for a better red response, but obviously your choice in the end.

As for under-sampling, have you zoomed right in to pixel level in ST, and do you find stars comprised of a group of pixels, or solitary square blobs? That'll tell you if you are under-sampling.

Cheers, Ian

[parallaxerr]

This is how the smallest stars look up close...

[The Admiral]

To me I would say that they were over-sampled if anything. Is that before or after software binning?

Ian

[parallaxerr]

Binned 50% in ST

[The Admiral]

Just now, parallaxerr said:

Binned 50% in ST

OK, then definitely NOT under-sampled! I think I've also read that in order for deconvolution to work you need some degree of over-sampling anyway. I agree with Ken, under our typical conditions you are probably not going to get significantly more detail. And with field rotation I think your best resolution is really only going to be in the centre of the FoV, as in the periphery the image will be smeared to a greater or lesser extent, depending on exposure time/location in the sky.

Ian

[parallaxerr]

Another one of my theories turned on it's head!

A little more reading and I agree Ian, oversampled if anything.

So, given that I'm just over 2", which I understand is about as good as we can expect in the UK, am I right in saying the seeing has spread my stars out over a few pixels and there's room for improvement based on conditions?

[The Admiral]

Hmm, now you're asking! . Well my take on it would be that the if we could take the 2.04"/px as gospel, to which we might add the effects of less than perfect focus, Bayer array, low-pass filter, and 'scope diffraction (which I am assuming isn't incorporated), then sky conditions would contribute the rest of the blurring. It is said that you shouldn't have significant over-sampling because I suppose you are spreading the photons more thinly than you need (though deconvolution may require some) and the camera's resolution is being wasted, so I guess that there is room to improve the situation. But, I'm not sure what effect that 'improvement' will actually bring.

You need the input of a seasoned imager here, I think! But the thing NOT to do is chase the numbers.

Ian

Edited December 12, 2016 by The Admiral

[The Admiral]

Also, I use a 0.79x reducer on my 'scope to reduce the focal length from its native 715mm, with the aim of reducing the degree of oversampling and improving the number of photons per pixel. And it gives me a much needed wider FoV. It's still oversampling though. But in the end, there's only so much fine tuning you can do with available optical accessories.

Ian

Edited December 12, 2016 by The Admiral

[Stub Mandrel]

3 hours ago, parallaxerr said:

And there in lies the rub, rotation! Are you saying, for an eq mounted rig, with less than ideal resolution, you can lengthen exposure to allow photons to spill over to adjacent pixels?

No I'm saying that bigger pixels are more forgiving of misalignment.

[Nigel G]

Orion, H/H, flame 40 x 30s Modded and comparison, the original was twice the total exposure time and sub length, Modded far easier to process, but a lot of taming reds.

I think the exposure times can be reduced quite a bit with the modded camera, looking forward to using the scope with it.

All with 135mm lens and 1600 ISO,

Nige.

HnS Modded

 

[happy-kat]

The first image has nice stars, tight.

Is the last image unmodded?

[Stub Mandrel]

9 hours ago, Nigel G said:

Orion, H/H, flame 40 x 30s Modded and comparison, the original was twice the total exposure time and sub length, Modded far easier to process, but a lot of taming reds.

I think the exposure times can be reduced quite a bit with the modded camera, looking forward to using the scope with it.

All with 135mm lens and 1600 ISO,

Nige.

HnS Modded

 

Which is which Nige - the second one has excellent colour, the first is 'sharper'. If the first is modded, try manipulating the histogram or curve for the red channel alone.

[Filroden]

I love the colour and sharpness of the modded Orion. The Ha really comes through and you can see it extend into the region to the right of the Horsehead. And the stars are perfect. I love their subdued feel. There are signs of hollow stars again so I think one of the StarTools processes is eating them. 

Edited December 14, 2016 by Filroden
I meant modded, not unmodded!