First Light for the D5300a and Baader Ha Filter (California Nebula)

[Xiga]

Hi all

So after getting my newly-modded D5300 back from JTWAstronomy, and also picking up a Baader 7nm Ha filter, I was itching to put them to use, and luckily enough I had a clear night a couple of weeks back (I've been moving house so have had to wait to process it).

I was shooting from my parents house, which I think is in a yellow zone. There are street lights right out the front of the house, and I was having to point the scope almost directly above one. Thankfully there was no moon to make things worse. It takes me forever to set up, so much so in fact that the longest capture I've ever managed was just under 2 hrs, so I was well chuffed that I managed just over 3 hrs this time. The night stayed clear the whole time, in the end I had to pack up as I had work in the morning, but it nearly killed me to leave! (went into work the next day with just 3 hrs sleep, man that was a long day! lol). I was originally hoping to capture a small amount of RGB (by replacing the Ha filter with my IDAS-D1) but there just wasn't time.

This was my first time using a NB filter, but thankfully I didn't find it any more difficult than regular imaging. I knew to up the ISO and exposure in SGP for the plate solving and focusing (using the Bahtinov mask). Other than that it was all pretty much the same. So in the end I managed 184 mins of data (23 * 8 min subs), all dithered. Thankfully 8 mins was enough to separate the Red Histogram from the left hand side. Not by much I should add, but at least it was clearly separated. I probably should have gone for 12 or even 15 min subs, but having never gone above 8 mins before I didn't want to risk it (although I'm sure my belt-modded HEQ5 Pro could manage it if asked to).

But overall I think I'm in love with NB imaging! It's so good not having to worry about LP (of any kind). The D5300a has also really impressed me. I know OSC camera's are obviously not the right tool for the job when it comes to NB imaging, but the D5300's QE of ~55% seems to be high enough to be able to pick up more than enough data to get the job done (especially for those of us who can't afford a mono camera with all the bells and whistles). I'm now actually seriously considering getting an OIII filter to try some Bi-Colour imaging. Watch this space.

Processing has been a learning curve (to put it mildly) and is still ongoing. I tried a few different ways, normal stacking in DSS and throwing away the G and B channels, Super-Pixel stacking in DSS, and finally a more involved method using IRIS (which is what the image below used). Super Pixel mode in DSS was quite good, but I wasn't happy that it was still mixing in the G and B pixels with the R, so I used IRIS to convert the Raws to FITs, then extract only the Red Pixels, before finally stacking them in DSS (using an x2 Drizzle to restore the resolution). This produced a much better image, with more contrast and less noise, but the downside is I wasn't able to use my Flats or Bias files, so yes, the image below doesn't contain any calibration files at all. I'm going to have another go and see if I can get the files calibrated first in IRIS, before extracting the Red pixels and stacking in DSS. The one thing I'm not sure about though, is whether or not converting all the Lights, Flats, Bias files from .NEF to .FIT in the first place makes calibrating them impossible (I'm hoping that IRIS doesn't debayer them when it converts). Unfortunately I don't have the technical expertise to test this out, so all I can do is try and make a couple of stacks and compare them by eye in PS.

Anyway that's enough rambling from me  

Here's what I've ended up with so far. C&C welcome (you can't offend me!). Apart from stretching I didn't do too much. I didn't do any sharpening, NR or contrast enhancement, so there's definitely more I could do to it. So any NB processing suggestions/workflows (or comments on the pre-processing steps above) will be most welcome indeed.

Clear skies!

https://flic.kr/p/YKmccU

 

 

[tooth_dr]

Ciaran, that is a delightful image, and the FF works well.  My next purchase.

 

Best wishes

Adam.

[Adam J]

The HEQ5 with your setup should manage 20min or even 30 min subs once the belt mod is fitter. Mine does with a 130PDS and a DSLR on it. Noise seems quite low considering the camera is not cooled. 

OIII works fine with my cooled 550D not sure how well it would work with an uncooled D5300 though it looks to be a decent chip if your image is anything to go by.

 

 

 

[Xiga]

Thanks Adam. Yeah the sensor in the D5300 is amazing considering the price of the camera. Good to know that the HEQ5Pro can handle those kind of subs, i'll try some longer ones next time. 

I tried calibrating the files before extracting the Red Pixels and stacking, but it didn't work. The calibration resulted in strange files, probably due to converting them from .nef to .fit. 

I think the bet way around this in future will be to set SGP to save both .nef and .fit files for me (which it will do, although it will double the download time, not such a big deal i suppose when you're dealing with long exposures). Then i should be able to calibrate before stacking just the red pixels. That boat has sailed unfortunately as far as this data set has gone, so i'll have to wait to the next session to try that out. 

[Xiga]

11 hours ago, tooth_dr said:

Ciaran, that is a delightful image, and the FF works well.  My next purchase.

 

Best wishes

Adam.

Thanks Adam!

Here is what i use in my optical chain:

Screwed directly to the drawtube (and kept there permanently):

https://www.firstlightoptics.com/adapters/baader-click-lock-2956256-m56-celestron-skywatcher.html

This adapter then attaches to the scope-side of the FF (and stays on it permanently. i also screw my 2" filters into it as well)

https://www.firstlightoptics.com/adapters/flo-adapter-for-skywatcher-focal-reducers.html

Then, on the other side of the FF i attach this adapter, which allows the connection to the D5300:

https://www.firstlightoptics.com/coma-correctors/skywatcher-dslr-m48-ring-adapter.html

 

Using this setup, i have a rock-solid and safe clamping mechanism to hold everything securely. It also allows the use of 2" filters and, importantly, lets me quickly rotate the camera to any orientation (important for getting the framing right in SGP). I highly recommend this setup for when you get your FF.

[tooth_dr]

7 hours ago, Xiga said:

Thanks Adam!

Here is what i use in my optical chain:

Screwed directly to the drawtube (and kept there permanently):

https://www.firstlightoptics.com/adapters/baader-click-lock-2956256-m56-celestron-skywatcher.html

This adapter then attaches to the scope-side of the FF (and stays on it permanently. i also screw my 2" filters into it as well)

https://www.firstlightoptics.com/adapters/flo-adapter-for-skywatcher-focal-reducers.html

Then, on the other side of the FF i attach this adapter, which allows the connection to the D5300:

https://www.firstlightoptics.com/coma-correctors/skywatcher-dslr-m48-ring-adapter.html

 

Using this setup, i have a rock-solid and safe clamping mechanism to hold everything securely. It also allows the use of 2" filters and, importantly, lets me quickly rotate the camera to any orientation (important for getting the framing right in SGP). I highly recommend this setup for when you get your FF.

 

Would the FF attach directly onto the drawtube without the clicklock and adapter?  I use insert filters for both CLS and Ha, so the ability to hold 2" filters isnt so much of a concern at the minute

[Shibby]

Very nicely done! I've done a fair amount of Ha with my DSLR in the past and it can work out OK with enough data and careful processing - as you've proven! Oiii will work well, too, as you get double the number of green pixels on your sensor.

[Xiga]

4 hours ago, tooth_dr said:

 

Would the FF attach directly onto the drawtube without the clicklock and adapter?  I use insert filters for both CLS and Ha, so the ability to hold 2" filters isnt so much of a concern at the minute

Hi Adam

Yes it does. On it's own the FF screws directly onto the drawtube so it's also very secure. The problem I had with it, was in rotating to get the right framing. The camera and FF are all locked together so the only way to rotate is to loosen the clamping ring on the draw tube and then screw the FF in/out. Also, I have a mark on my draw tube for rough focus (good enough that I can plate solve without needing to focus first), and moving the FF position decreased the accuracy of the marking. Needless to say, after a couple of sessions I knew I needed a better solution, which is how I ended up using the above combination. 

[tooth_dr]

Thanks Ciaran. I’ll get a FF first and see how it works. My T adapter can be rotated with three grub screws so might work ok 

[StargeezerTim]

On 05/10/2017 at 14:02, Xiga said:

Hi all

So after getting my newly-modded D5300 back from JTWAstronomy, and also picking up a Baader 7nm Ha filter, I was itching to put them to use, and luckily enough I had a clear night a couple of weeks back (I've been moving house so have had to wait to process it).

I was shooting from my parents house, which I think is in a yellow zone. There are street lights right out the front of the house, and I was having to point the scope almost directly above one. Thankfully there was no moon to make things worse. It takes me forever to set up, so much so in fact that the longest capture I've ever managed was just under 2 hrs, so I was well chuffed that I managed just over 3 hrs this time. The night stayed clear the whole time, in the end I had to pack up as I had work in the morning, but it nearly killed me to leave! (went into work the next day with just 3 hrs sleep, man that was a long day! lol). I was originally hoping to capture a small amount of RGB (by replacing the Ha filter with my IDAS-D1) but there just wasn't time.

This was my first time using a NB filter, but thankfully I didn't find it any more difficult than regular imaging. I knew to up the ISO and exposure in SGP for the plate solving and focusing (using the Bahtinov mask). Other than that it was all pretty much the same. So in the end I managed 184 mins of data (23 * 8 min subs), all dithered. Thankfully 8 mins was enough to separate the Red Histogram from the left hand side. Not by much I should add, but at least it was clearly separated. I probably should have gone for 12 or even 15 min subs, but having never gone above 8 mins before I didn't want to risk it (although I'm sure my belt-modded HEQ5 Pro could manage it if asked to).

But overall I think I'm in love with NB imaging! It's so good not having to worry about LP (of any kind). The D5300a has also really impressed me. I know OSC camera's are obviously not the right tool for the job when it comes to NB imaging, but the D5300's QE of ~55% seems to be high enough to be able to pick up more than enough data to get the job done (especially for those of us who can't afford a mono camera with all the bells and whistles). I'm now actually seriously considering getting an OIII filter to try some Bi-Colour imaging. Watch this space.

Processing has been a learning curve (to put it mildly) and is still ongoing. I tried a few different ways, normal stacking in DSS and throwing away the G and B channels, Super-Pixel stacking in DSS, and finally a more involved method using IRIS (which is what the image below used). Super Pixel mode in DSS was quite good, but I wasn't happy that it was still mixing in the G and B pixels with the R, so I used IRIS to convert the Raws to FITs, then extract only the Red Pixels, before finally stacking them in DSS (using an x2 Drizzle to restore the resolution). This produced a much better image, with more contrast and less noise, but the downside is I wasn't able to use my Flats or Bias files, so yes, the image below doesn't contain any calibration files at all. I'm going to have another go and see if I can get the files calibrated first in IRIS, before extracting the Red pixels and stacking in DSS. The one thing I'm not sure about though, is whether or not converting all the Lights, Flats, Bias files from .NEF to .FIT in the first place makes calibrating them impossible (I'm hoping that IRIS doesn't debayer them when it converts). Unfortunately I don't have the technical expertise to test this out, so all I can do is try and make a couple of stacks and compare them by eye in PS.

Anyway that's enough rambling from me  

Here's what I've ended up with so far. C&C welcome (you can't offend me!). Apart from stretching I didn't do too much. I didn't do any sharpening, NR or contrast enhancement, so there's definitely more I could do to it. So any NB processing suggestions/workflows (or comments on the pre-processing steps above) will be most welcome indeed.

Clear skies!

https://flic.kr/p/YKmccU

 

 

Nice one Xiga...

Did you mix the Ha with RGB? I also did my first Ha imaging last night and when I extract the red channel the image is mono... unless I am missing something ( highly likely!).

[Xiga]

Hi Tim

I'm not the biggest fan in the world of mono images, so I did indeed do a bit of mixing. As I didn't actually capture any true RGB, I just did a 2nd stack purely for colour and used the mono Ha image for the Luminance. I think I just used AHD debayering method in DSS for the colour. Obviously it's not a proper use of colour but it's surprising what you can still get out of the Ha image when it's stacked as RGB. Even though everything is just red really, messing around with the colour image I was able to set the Nebula to almost any colour I wanted just by stretching the individual channels. I nearly went with a funky golden yellow just for fun, but in the end just went with classic red/pink. 

[Adam J]

If you do use OIII you will need longer subs, if you want to match the H-a exposure as the reason OIII appears in both blue and green is that it sits in the dip right between the two filter and so its the QE is quite poor. I normally increase my sub length to 150% of the H-a sub length so a 10 min H-a sub = 15min OIII sub.