M42 LHaRHaGB, M44 and NGC 2244

[Guest]

A lucky 4 day stretch of clear cold skies

Welford Observatory springs into life for one of the only times this season

The forecast in the run up to the 9th of February was promising clear skies but frankly I didn't pay much attention to it as the weather has been so changeable this winter.

 

The forecast was right though and I managed to not only get many hours in on the 9th but also the 12th, 13th and 15th.

 

The good weather was not the only surprise though. 

 

I haven't imaged the Great Nebula in Orion since 2012 as I remembered it being very difficult to track from my observatory as it moved along the very edge of the southern wall at the limit of my pier/scope configuration. Also, in 2012 I shot it with a one shot colour camera making life very easy. If I were to image it again I'd be using my ASI1600MM with the need to capture luminance, red, green and blue channels - and if I was very ambitious - H-alpha as well.

 

It turned out I was feeling very ambitious. And it worked. I think I've captured my best astrophoto to date. Also, I pushed my processing skills (limited as they are) to include the H-alpha in both the luminance and red channels. I'm finding Pixinsight so much easier to use than other programs I've tried, mostly because the standard approach to most of the processes takes the guess work out of some of the more creative damage I can do to the image. :-) 

 

Also, the observatory was ticking over perfectly with very few errors made by me in re-configuring it each night for the next round of imaging.

 

And so here are the results of one of my most enjoyable series of evenings in the observatory.

 

Details
Object name	The Great Nebula in Orion
Object ID	M42
Date(s)	9, 12, 13 and 15 February, 2018
Telescope	Teleskop Service 65mm
Camera	ASI1600MM
Luminance	60*1min = 60min
Red	25*2min = 50min
Green	25*2min = 50min
Blue	24*2min = 48min
Ha	19*5min = 95min
Oiii	0
Sii	0
Total time	5.05hrs
Frames	0
Processing	PixInsight / Bias, Flats, Darks / Ha to Lum / Ha to red channel / LHaRHaGB /
Notes	I waited this entire winter for an opportunity to get back into the observatory to do some imaging. The weather and my schedule have been appalling. I guess I can really stop saying that now as it’s always the case…. I didn’t expect to image this object – the last time I did was the 14th of January, 2012. It’s very low to the top of my observatory wall and imaging it always seems unlikely. But on this occasion, I persevered over the few clear nights we had and managed to capture 5 hours of data in H-alpha, red, green and blue channels. This is the most data I’ve captured on an object and it paid off. I’ve processed this image combing H-alpha with both the luminance and red channels – then recombining them all for the finished shot. I may try re-processing in different combinations but this certainly seems to have worked ok.

 

NGC 2244

 

 

Details
Object name	The Rosette Nebula
Object ID	NGC 2244
Date(s)	9, 13 February, 2018
Telescope	65mm
Camera	ASI1600MM
Luminance	0
Red	0
Green	0
Blue	0
Ha	15*5 min = 75 min
Oiii	4*5 min = 20 min
Sii	8*5 min = 40 min
Total time	2.25 hr
Frames	0
Processing	PixInsight / Bias, Flats, Darks /
Notes	Not enough data to make this a good image – but it’s a start and all the weather would allow for this season.

 

 

M44

 

Details
Object name	The Beehive Cluster
Object ID	M44
Date(s)	9 February, 2018
Telescope	Teleskop Service 65mm astrograph
Camera	ASI1600MM
Luminance	0
Red	25 x 2 min = 50 min
Green	19 x 2 min = 38 min
Blue	19 x 2 min = 38 min
Ha	0
Oiii	0
Sii	0
Total time	2.1 hrs
Frames	0
Processing	PixInsight / Bias, Flats, Darks
Notes	The second of the objects I imaged during this session. The Beehive cluster is the first object I (accidently) observed while trying to find Saturn out on the downs one night back in 2002 when I took up this hobby again. I remember being absolutely stunned by the sight of all these stars blazing in such a small patch of sky. The scope I was using had been bought from a work colleague for £50 (I later found out that it was considered ..”toy grade” LOL!) but it was good enough to provide decent images of brighter objects such as this one.

I hope the rest of you were able to take advantage of what was a great clear spell this winter :-)

David

[Allinthehead]

Nice collection 

[Demonperformer]

Nice set images. Particularly like the Rosette image. 

[iansmith]

Great set of images. Lovely star colours in M44 and a great sense of depth in M42.

Cheers, Ian

[Guest]

11 hours ago, Allinthehead said:

Nice collection 

Thanks - it's not much to show for 6 months of darker skies but I was happy to get at least these :-)

33 minutes ago, Demonperformer said:

Nice set images. Particularly like the Rosette image. 

Thank you - I hope to add to this image as I'd like to bring the blue of the centre out more

13 minutes ago, iansmith said:

Great set of images. Lovely star colours in M44 and a great sense of depth in M42.

Cheers, Ian

Thanks Ian - M42 was a big project for me learning all the different PixelMath combinations in Pixinsight. Once I thought I was finished I walked away from it for a couple of days and when I returned to it I noticed how flat it looked - just needed to do the last set of curves on it and it brought out the depth - so much to be said for processing over many days instead of hours :-)

 

[Guest]

Looking at my results above I noticed a massive error in both M42 and NGC 2244

They both appeared light on one side of the image - in M42 it's down the right side where it looks way to red (I thought that was heavy nebulosity coming out in H-alpha)

On the Rosette it's across the top

So the same part of the image in both

I'm using a Gerd Neumann panel 

It turns out it's my h-alpha flat that's causing this. It's got a dark band across the top of the frame - I had seen that before but thought...a flat is a flat...and continued processing with it.

Then, reading on forums I've learned that many don't take narrow band flats as they wind up looking like mine does or having fringing/streaks across them (like my oiii and sii do by the way)

The advice was to use the red filter flat for h-alpha and oiii and the green for sii

As a test I processed the h-alpha data without a flat and compared it to the image processed with a flat (see below)

The difference is obvious - and I feel a bit stupid to have accepted (and not immediately noticing) the obvious error you see in the images above - and I'm now reprocessing everything :-) 

I'd be interested to hear from anyone that has had a similar problem with their h-alpha (in particular). My sii and oiii flats show obvious errors seen by many but the h-alpha dark band is interesting.

Thanks for any advice

David

Bad H-alpha flat

 

h-alpha M42 processed with bad flat

H-alpha processed WITHOUT bad flat

Pretty big difference !

David

 

[tooth_dr]

Much much nicer! That new flat made a big difference. 

[Guest]

2 minutes ago, tooth_dr said:

Much much nicer! That new flat made a big difference. 

Thanks :-)

The second image actually had no flat at all so I'm just trying to reprocess everything without a flat on the h-alpha

Amazing how the fog lifted on the second image lol! And also amazing how I could look at my original image and convince myself it was just "heavy nebulosity" on the right side of the M42 image LOL

David

 

[tooth_dr]

Yes indeed no flat! Lol

[Guest]

I've reprocessed this 5.5hrs of data on M42 without the dodgy Ha flats and.....amazing...it actually looks similar to all the other M42's I've ever seen.

I explained how I felt about seeing this new image as ".....like looking at pictures from the 90s when we all had yellow walls...and thought that was "cool"....and you look at it now with your rooms painted natural colours and go bleeechhhh!!!" LOLOL!

Here it is. Lesson learned....shame I can't delete the original post, quite the badge of shame har har!

(this new one probably isn't very good by most standards either but at least it "looks" like M42 and not the inside of a 50 year old candy floss machine...)

David

[carastro]

I think that looks heaps better David.  Lots of nice dust around.  However I think the Core was better controlled first time around. 

Carole 

[Guest]

37 minutes ago, carastro said:

I think that looks heaps better David.  Lots of nice dust around.  However I think the Core was better controlled first time around. 

Carole 

Thank  you Carole - I agree with you too about the core - I'll have another go  :-)

David

[Guest]

2 hours ago, carastro said:

I think that looks heaps better David.  Lots of nice dust around.  However I think the Core was better controlled first time around. 

Carole 

One thing I've noticed about taming the core of M42 is that the more you try to tame it the more it looks like a postage stamp :-) 

#drivingMeNuts lol

David

 

[Guest]

I reprocessed M42 to include using the red filter flats from the evening on the H-alpha channel.

Appears one more step closer to the norm :-)

I used DBE on this version as well so not all the field correction is down to the flat being applied to the H-alpha - but there was a noticeable difference in pre-processing.

[carastro]

Looking heaps better, that magenta cast has also gone.

Carole 

[Guest]

2 minutes ago, carastro said:

Looking heaps better, that magenta cast has also gone.

Carole 

Thanks! I have to stop misinterpreting every field illumination error I see as nebulosity LOLOL!

David

 

[simmo39]

V nice!

[Guest]

I've finished processing the Rosette that I started above - this is 4.6hrs and I had enough data to make it worth processing now :-)

You'll see I've also got rid of the gradient by ditching the flats all together as I hadn't taken any rgb that night to use with the nb filters - DBE in Pixinsight helped a lot :-)

[Demonperformer]

Well, I said I particularly liked your Rosette pic and it has just got better. Excellent result.

[Guest]

As ever ....processing evolves :-)

In my last iteration of the processing of M42 I used RangeMask in PixInsight to focus on the central nebulosity in order to brighten the region.

I had used this facility in a number of areas during the processing and it struck me this ran against my preference of doing simply histogram and curves stretches to minimize my "painting" of the data (LocalHistogramEqualisation excepted of course lol!) 

In this iteration I also limited myself to one saturation boost 

In short I tried avoid manipulating the data as much as I could to remain close to what I captured - rather than, what can happen with me, "paint" a technicolor monstrosity that while pleasing to the eye, is somewhat too fake for my interest in this hobby.

Not content with my opinion on the matter I showed the most recent image process above and the one below to friends for their opinion.

To my surprise they preferred the one below - normally people like the snazzy, bright, decal look - something that is really striking in detail - and thankfully, this time they preferred the softer lighter touch I've used on the data in the image below.

I notice too that while the image is slightly more noisy it has brought out more of the nebulosity throughout the image.

I'm going to produce a GIF (don't think I can post it here) of the three takes on this (I actually like the first iteration I did too - it's a bit bright but looks more natural I think).

When you blink between the three this latest version really stands out as a bit more detailed and natural.

I've made loads of notes on how I got to this - important to be able to replicate the process, as I no doubt will have to wait a year to try again LOL!

David

 

[Guest]

Here's the GIF I mentioned - looks like I can upload it :-)

David

 

[Guest]

On 06/03/2018 at 09:19, David_L said:

As ever ....processing evolves :-)

In my last iteration of the processing of M42 I used RangeMask in PixInsight to focus on the central nebulosity in order to brighten the region.

I had used this facility in a number of areas during the processing and it struck me this ran against my preference of doing simply histogram and curves stretches to minimize my "painting" of the data (LocalHistogramEqualisation excepted of course lol!) 

In this iteration I also limited myself to one saturation boost 

In short I tried avoid manipulating the data as much as I could to remain close to what I captured - rather than, what can happen with me, "paint" a technicolor monstrosity that while pleasing to the eye, is somewhat too fake for my interest in this hobby.

Not content with my opinion on the matter I showed the most recent image process above and the one below to friends for their opinion.

To my surprise they preferred the one below - normally people like the snazzy, bright, decal look - something that is really striking in detail - and thankfully, this time they preferred the softer lighter touch I've used on the data in the image below.

I notice too that while the image is slightly more noisy it has brought out more of the nebulosity throughout the image.

I'm going to produce a GIF (don't think I can post it here) of the three takes on this (I actually like the first iteration I did too - it's a bit bright but looks more natural I think).

When you blink between the three this latest version really stands out as a bit more detailed and natural.

I've made loads of notes on how I got to this - important to be able to replicate the process, as I no doubt will have to wait a year to try again LOL!

David

 

As seen in the May 2018 "Hot Shots" section of Sky at Night magazine :-) 

[carastro]

Jolly well done David. 

My goodness that has really come on since that first version you sent me by Skype. 

Carole

[Guest]

53 minutes ago, carastro said:

Jolly well done David. 

My goodness that has really come on since that first version you sent me by Skype. 

Carole

Thank you very much Carole - I learned a lot processing this one :-)