M 81 & 82 + M42 (old)

[JohnSadlerAstro]

Hi,

My first really successful image with my new P-DS, EQ5, and EOS 1000d:

56 mins of 75 sec subs at ISO 1600 (I could have doubled expo length but didn't want to ruin tracking)

a large number of Darks, Bias, and Flats (which have worked for the first time ever ) 

Stacked in DSS; processed in 16-bit GIMP

There are some stacking borders but I haven't cropped them off yet, the only issue I can see is the coma on the left hand side, I think it's caused by poor(ish) collimation, otherwise, I'm all ready for tonight! (Any suggested targets?)

Also an experimental image of M42 on my first attempt at using the P-DS, I was plagued with issues with the calib files, I think due to not taking enough. This one is only 10 mins total, so it's pretty noisy, I had to use IRIS to iron out the gradient. This one will have to do for this year, as Orion is setting by midnight (when the lights go off) but hopefully in September-October this year I can get some more.

Any comments/suggestions would be gladly accepted!

John

[Knighty2112]

 Very nice results. Not an imager so can't critique it in any way other than to say well done!  

[geordie85]

Glad you've managed to get your calibration frames working. If I was you I'd be inclined to continue adding data to this target. Having several hours of data will help with noise and increase your signal which will make for a better image. But then I'm happy to spend several nights imaging the same target. I think this is because when I first started imaging I'd bounce around the sky trying to image 3 or 4 targets a night and always ended up very unsatisfied with my results. 

[JohnSadlerAstro]

2 hours ago, geordie85 said:

Glad you've managed to get your calibration frames working. If I was you I'd be inclined to continue adding data to this target. Having several hours of data will help with noise and increase your signal which will make for a better image.

Hi,

How much better would the image be if I got, say, an extra hour of data? I can quite easily do this, as I simply carried everything into the garage last night without removing the camera/scope/CC or anything. There are a lot of targets I can't wit to get, an I'm just wondering if it will really make much difference given the short exposures I'm taking?

John  

[alacant]

3 hours ago, JohnSadlerAstro said:

suggestions would be gladly accepted

Hi. Nice shots. Well done. I think you've missed focus in 1 and are tilted in both, so I don't think more data will help. For tonight you could try a Bahtinov mask, 1mm more on your cc, adjust the sccondary by getting the sight tube cross-hairs bang in the centre of the primary donut and make sure the camera is square in the focuser. Or just accept what you have; most of the issues can be sorted out in software anyway. Good luck and HTH.

[Carter Thiel]

Which flat frames did you end up using? 1/50 or 1/100?

[JohnSadlerAstro]

I used 1/50 th sec ones.

John

[Filroden]

2 hours ago, JohnSadlerAstro said:

Hi,

How much better would the image be if I got, say, an extra hour of data? I can quite easily do this, as I simply carried everything into the garage last night without removing the camera/scope/CC or anything. There are a lot of targets I can't wit to get, an I'm just wondering if it will really make much difference given the short exposures I'm taking?

John  

It's surprising how much a difference it makes. I think the consensus we reached in the No-EQ Challenge thread was that you needed a good two hours with a DSLR before the processing became easier (i.e. noise levels started to become manageable). We would continue to see improvements in image quality for every hour added. You don't see 'deeper' but what you do capture in each short exposure becomes easy to bring above the background noise.

However, it's all down to preference. You are taking the images for your own benefit first, not ours  If you're happy with the image and want to try new targets, then try new targets. You can always revisit the targets later and add new exposures to the old ones and re-integrate them. Just make sure to save your master calibration data and use dates in their names so you can use the correct calibration data with each session of light data. I personally calibrate each sub and save the calibrated versions before they are integrated. That way I can just calibrate the latest set, then add in the already calibrated historical data and run a new integration.

As for the image, I love these two targets. There is something about M81 that looks so pristine, almost like it's a bride nebula in its white veil. I can see quite a bit of noise in the background which will improve with more data. You will also find processing the image easier if you crop the stacking artefacts before stretching. It means you will have a cleaner histogram and know where your true data sits.

You have a gradient running left to right that seems to start blue and end red, so your background removal has not fully worked. I'd give it a second run to see if it can completely remove it. I'm guessing you have more light pollution to your east?

And it looks like your PA could be refined a little more. You're seeing some field rotation (hence the stacking artefacts). I've only just got my first EQ mount so I can't help you any further than that. It's a cracking good start with the new kit and I'm glad it's finally starting to come together for you!

[JohnSadlerAstro]

Hi,

Thanks! I'm not quite sure if I will get up again tonight, one night might be enough, I'll see how the weather looks later.

John

[geordie85]

16 minutes ago, Filroden said:

It's surprising how much a difference it makes. I think the consensus we reached in the No-EQ Challenge thread was that you needed a good two hours with a DSLR before the processing became easier (i.e. noise levels started to become manageable). We would continue to see improvements in image quality for every hour added. You don't see 'deeper' but what you do capture in each short exposure becomes easy to bring above the background noise.

However, it's all down to preference. You are taking the images for your own benefit first, not ours  If you're happy with the image and want to try new targets, then try new targets. You can always revisit the targets later and add new exposures to the old ones and re-integrate them. Just make sure to save your master calibration data and use dates in their names so you can use the correct calibration data with each session of light data. I personally calibrate each sub and save the calibrated versions before they are integrated. That way I can just calibrate the latest set, then add in the already calibrated historical data and run a new integration.

As for the image, I love these two targets. There is something about M81 that looks so pristine, almost like it's a bride nebula in its white veil. I can see quite a bit of noise in the background which will improve with more data. You will also find processing the image easier if you crop the stacking artefacts before stretching. It means you will have a cleaner histogram and know where your true data sits.

You have a gradient running left to right that seems to start blue and end red, so your background removal has not fully worked. I'd give it a second run to see if it can completely remove it. I'm guessing you have more light pollution to your east?

And it looks like your PA could be refined a little more. You're seeing some field rotation (hence the stacking artefacts). I've only just got my first EQ mount so I can't help you any further than that. It's a cracking good start with the new kit and I'm glad it's finally starting to come together for you!

 

3 hours ago, JohnSadlerAstro said:

Hi,

How much better would the image be if I got, say, an extra hour of data? I can quite easily do this, as I simply carried everything into the garage last night without removing the camera/scope/CC or anything. There are a lot of targets I can't wit to get, an I'm just wondering if it will really make much difference given the short exposures I'm taking?

John  

Filroden took the words right out my mouth. In short, more data equals a better finished image and an easier time getting there.

[Filroden]

Hope you don't mind but I thought I should include an example. Here's your jpg (not the best starting point good for illustration). By cropping it becomes much easier to remove the gradient (I used ABE in PixInsight for a quick removal). This removed most of the blue gradient and the red from the right border. I also neutralised the background (to minimal effect, ABE did most of the work) and applied some crude noise reduction and desaturation to the background (masking the areas of signal). There is still more that can be done (especially as I did this in a couple of minutes rather than a couple of hours, which the data deserves). You have a good starting point and more can be gained from adding more data (to help control the noise) and through additional processing (to bring out the signal you have captured).