Carina Nebula with DSLR on Star Adventurer

[jman17]

Hi,


I took this shot of the Carina Nebula at 150mm with a DSLR on a SkyWatcher Star Adventurer and am wondering if anyone can give me some feedback on how it could be improved? It was a series of mostly 90 second shots at 800ISO, for a total of about 40 minutes, with a Canon 450D. I shot it from my backyard in a regional city and processed it in Pixinsight after stacking in Deep Sky Stacker with bias frames and some flats and had to crop it quite a bit. I know darks would have helped reduce the noise more but for this shot I didn't manage to get them. I am still learning Pixinsight and have probably missed out on a lot of processing steps. Is it OK to post a link to my FITS file to see what others with more experience can get out of this shot, and explain the steps they took? It would help me a lot to get a workflow to follow to learn how to use Pixinsight better.



Thanks,
jman17

[OzEclipse]

yes that's fine. Other forums restrict external linking particularly to other astro forums but it's fine here on IIS.

Lots of people post links external to IIS.

I'm not a PI user and can't help you but something looks strange with the stacking. Hopefully someone can help you.

Joe

[kosborn]

I'm only a newbie myself but I'm happy to pass on what I have learned so far. Looks like there is some good data there and I think PixInsight is a good way to bring it out. It is a complex program though and takes a bit of learning. I bought 'Inside PixInsight' by Warren Keller and it is an excellent read with info on dslr, one shot colour and mono/filter imaging. The other really good sources of info I've found is Light Vortex Astronomy tutorials (http://www.lightvortexastronomy.com/tutorials.html) and Harry's Astroshed (http://www.harrysastroshed.com/pixinsighthomenewbie.html).


I'd be happy to play with the FITS file (or better still zip up the RAW files) especially since I've been clouded out for the last 3 weeks and can't get any of my own data!!


Kevin

[rcheshire]

I use a 450D and have found with these older models that good acquisition and proper image calibration is important. If you are particular in these areas, the rest just falls into place. Dithering between shots makes a huge difference to image quality as well.

Dark frames can be a blessing and a curse. Dithering helps. If you use darks in PI take longer darks than lights and use the scaling function. That way you work around the problematic dark / light mismatch.

If you include darks in your workflow without scaling, it's better not to subtract the bias from the darks and lights. Just subtract the master dark from the lights and use a master bias to calibrate the flats.

This is my view of the world. No doubt others may see things differently.

EDIT: In some cases median combining bias darks and flats improves results by reducing random noise in the calibration frames.

[sil]

Great start jman, keep going with PixInsight! The Light Vortex tutorials I think are fantastic and they have a very good workflow, not far from my own(not saying mine is great just I arrived at a similar process to theirs on my own). As a starting point try the Batch Pre Processing (BPP) script in PixInsight. Its sort of a one click process, dump in whatever lights darks Flats you have and it will run through registration (aligning) and integration (stacking) using the files you gave it. Even if its only a handful of lights only. It'll output a master integration file which is your starting point for stretching etc and an autostretch in screentranfser function will show you what you've got. It uses a basic workflow to align and stack your images and serves most people perfectly well.

For me a master integration frame (one file 64bit depth, containing ALL my data aligned and stacked) is my processing starting point. Stacking has reduced noice and removed plane/satellite trails etc. Its a huge file and I makes a copy to start working off from there, as I experiment and learn and make mistakes and find tutorials to follow its this file I recopy to start fresh with. As I learn my workflow changes and with that master integration file I can compare easily how my current workflow compares against my previous since its the same starting point. Registration and Integration I've only adjusted my methods with a couple of times with very little noticable differences at the other end.

Something PI introduced "recently" (months ago, its been ages since I've used it) is a photometry colour correction (forget what its called but its standard in the package, Rick might chime in) that uses plate solving to look up reference data and make colour correction adjustment to get your star colours and nebulosity more accurate. Its an awesome tool and alone I have to redo my entire workflow as it shortcuts some steps I take to preserve colour changes while I process. If you plate solve and save your master integration file it'll use this data (you might have to figure out how to add the date/time/location FITS metadata into the file too).

PI is a steep learning curve at first. But its easy to use to take your data and give you an image from it that'll be good, but its also go immense power if you are willing to take the time to explore the options at each step to maximise the processing at all stages to end up with a fantastic image. Start with BPP script and you can take darks and flats another night if you didnt when you got the lights, just make sure same settings etc. Or you can process with Lights alone.

[raymo]

Nobody so far has mentioned it, but your focus is way off. If the focus had been right that would be quite a nice image. Focus is the single biggest
maker or breaker of an image. Did you use your LiveView magnified
5x?
raymo

[xelasnave]

Good effort.
Mask yourself a focuser mask if you can ...they are a great help.
Also I use a little blue tack to hold the focus.
Alex

[jman17]

Hi,


Thanks for the helpful advice. I wasn't even aware the shot was badly out of focus so thanks for letting me know. Looking at it now it is obvious. I will use the 5X magnification from now on and use some blue tack to lock the focus. I know that that time I didn't do proper flats, they are pretty blue and were underexposed. I had to finish as it was extremely cold and didn't have another attempt at them. I wasn't going to include them but thought I got a better shot with some of the better ones included. Some probably weren't evenly illuminated too which probably caused more problems.

Here is the link to the data, both the FITS file and RAW files- https://drive.google.com/drive/folders/1_ZW_wKQY-FllySRJ4MVOYRzVPqARek58?usp=sharing

Thanks very much if someone can have a process of them in Pixinsight and maybe even include a workflow if possible so I can have an idea of how to follow what they have done. A process in any other processing software would be helpful too if anyone doesn't have Pixinsight as I am interested to see what it is possible to get out of data like this. Please let me know if there is any problems with the link.

Thanks,
jman17

[RickS]

It's called PhotometricColorCalibration. Lots of useful advice in this thread...

Cheers,
Rick.