M106 multi narrowband
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.
Please share the equipment used and if possible camera settings to help others.
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.
Please share the equipment used and if possible camera settings to help others.
M106 multi narrowband
My M106 project is finished. Not as planned, shooting broadband only gave way to orange/brown.
So it has become a multi narrowband capture: Optolong L-Pro (Ha, Hb, OIII, NII and SII) and L-eNhance filter (Ha, Hb, and OIII). In total 15 and a half hours of data.
And it's clear that the best filter, is dark skies
Below a cropped only of the core of M106 with the L-eNhance. Below that, the complete but scaled down to 2400px complete version.
This one can be viewed here: https://i.ibb.co/8KZXPCM/M106-in-multi-narrowband.jpg
Bortle 7/8
Meade LX200 8" f/10 ACF OTA
Ioptron CEM25EC mount (no guiding)
Optolong L-Pro filter
Optolong L-eNhance filter
Zwo ASI071MC Pro camera
Captured with SharpCap Pro @ -10 Celsius
Optolong L-Pro: 83 x 240 sec / Gain 10 / Offset 10 / White balance R50 B50
Optolong L-Pro: 31 x 300 sec / Gain 200 / Offset 28 / White balance R50 B50
Optonlong L-eNhance: 51 x 535 sec / Gain 175 / Offset 20 / White balance R50 B50
Stacked the 3 sets with DeepSkyStacker
Processed with Siril and Photoshop
Siril: each stack separately with Background Extraction and Histogram, saved as TIF
Photoshop: layered the 3 images with 50% opacity, Camera Raw Filter (blacks, saturation, clarity, noise reduction), scaled down to 2400px.
So it has become a multi narrowband capture: Optolong L-Pro (Ha, Hb, OIII, NII and SII) and L-eNhance filter (Ha, Hb, and OIII). In total 15 and a half hours of data.
And it's clear that the best filter, is dark skies
Below a cropped only of the core of M106 with the L-eNhance. Below that, the complete but scaled down to 2400px complete version.
This one can be viewed here: https://i.ibb.co/8KZXPCM/M106-in-multi-narrowband.jpg
Bortle 7/8
Meade LX200 8" f/10 ACF OTA
Ioptron CEM25EC mount (no guiding)
Optolong L-Pro filter
Optolong L-eNhance filter
Zwo ASI071MC Pro camera
Captured with SharpCap Pro @ -10 Celsius
Optolong L-Pro: 83 x 240 sec / Gain 10 / Offset 10 / White balance R50 B50
Optolong L-Pro: 31 x 300 sec / Gain 200 / Offset 28 / White balance R50 B50
Optonlong L-eNhance: 51 x 535 sec / Gain 175 / Offset 20 / White balance R50 B50
Stacked the 3 sets with DeepSkyStacker
Processed with Siril and Photoshop
Siril: each stack separately with Background Extraction and Histogram, saved as TIF
Photoshop: layered the 3 images with 50% opacity, Camera Raw Filter (blacks, saturation, clarity, noise reduction), scaled down to 2400px.
Re: M106 multi narrowband
Hi Menno,
Although it's not conventional to use the filters on galaxies I'd say that it's given you quite a nice contrasty result on the central region of this particular galaxy probably particularly because of the HA feature? M82 may be another to go for with your set up in a similar way and for the same reason?
TimH
Although it's not conventional to use the filters on galaxies I'd say that it's given you quite a nice contrasty result on the central region of this particular galaxy probably particularly because of the HA feature? M82 may be another to go for with your set up in a similar way and for the same reason?
TimH
Re: M106 multi narrowband
What I have read is that the core of M106 is really active with a supermassive blackhole. My guess is that there are enough lines emanating from the core over the whole spectrum to capture.
And yes, M82 is on my to-do list, as is M101 and M33 (both have some nice narrowband structures in them).
Menno
And yes, M82 is on my to-do list, as is M101 and M33 (both have some nice narrowband structures in them).
Menno
Re: M106 multi narrowband
Started with M82timh wrote: ↑Tue Apr 20, 2021 7:03 pm Hi Menno,
Although it's not conventional to use the filters on galaxies I'd say that it's given you quite a nice contrasty result on the central region of this particular galaxy probably particularly because of the HA feature? M82 may be another to go for with your set up in a similar way and for the same reason?
TimH
This is 4 hours of data and now pushed the unguided mount to 600 second exposures
This is with the Optolong L-eNhance. Planning for at least 10 hours of this and then also a set of broadband and mix that.
Re: M106 multi narrowband
Some good detail in that image.
Dave
Dave
Re: M106 multi narrowband
I (finally) learned the good/better way of adding Ha data to RGB: in Photoshop processing RGB and HA separately, then copying the red channels of both, merge those two (50/50 in this case) and then past that result in the red channel of the RGB and then continue processing if needed.
It makes a world of difference: more blue colors, sharper, more detail. If you compare this to the original, the disk itself now doesn't have that red "haze" anymore.
That's one of the great things of this hobby: there is always something new to learn
It makes a world of difference: more blue colors, sharper, more detail. If you compare this to the original, the disk itself now doesn't have that red "haze" anymore.
That's one of the great things of this hobby: there is always something new to learn
Re: M106 multi narrowband
Yes that is way better!
Just for interest - the 'correct' method in PixInsight is similar but probably a bit better defined. You first separately process the HA and RGB. Extract the red channel from the RGB and then subtract that from the HA image to the point that the stars are about cancelled out. What your then left with is the HA - red difference image. It is this 'pure HA image' that you then add back into the red channel of the original RGB image.
The thing that I like about PI is that it all makes mathematical sense (at least to the level that I can judge) and there isn't anything that seems arbitrary. However despite the conceptual neatness of isolating an HA image that really is just HA and doesn't also include the ~ 656 nm blackbody light of the stars it doesn't quite work in practice because the subtraction itself adds noise and the HA star FWHMs tend to be tighter than in RGB etc. Also you have to be careful about how much of the HA difference signal to add back into the RGB red channel - in practice too much distorts the overall colour balance.
Tim
Just for interest - the 'correct' method in PixInsight is similar but probably a bit better defined. You first separately process the HA and RGB. Extract the red channel from the RGB and then subtract that from the HA image to the point that the stars are about cancelled out. What your then left with is the HA - red difference image. It is this 'pure HA image' that you then add back into the red channel of the original RGB image.
The thing that I like about PI is that it all makes mathematical sense (at least to the level that I can judge) and there isn't anything that seems arbitrary. However despite the conceptual neatness of isolating an HA image that really is just HA and doesn't also include the ~ 656 nm blackbody light of the stars it doesn't quite work in practice because the subtraction itself adds noise and the HA star FWHMs tend to be tighter than in RGB etc. Also you have to be careful about how much of the HA difference signal to add back into the RGB red channel - in practice too much distorts the overall colour balance.
Tim
Re: M106 multi narrowband
Hi Tim
Yes, that last bit got me really confused at first with another capture. It all got very green. Took some experimenting to figure out that 50/50 of RGB and Ha red channel is not a golden rule But in this case it did.
Menno
Yes, that last bit got me really confused at first with another capture. It all got very green. Took some experimenting to figure out that 50/50 of RGB and Ha red channel is not a golden rule But in this case it did.
Menno
Re: M106 multi narrowband
Since there are only clouds and clouds alone (and dense fog like now), I did some rework again. More process experience and new/updated software does a lot. Especially the 2nd version of StarNet.
Like with M106 where the data is way higher in detailed now. The vertical secondary arm is also showing way better now.
This again a "mix" where the Ha is blended in the red channel of the RGB.
Menno
Like with M106 where the data is way higher in detailed now. The vertical secondary arm is also showing way better now.
This again a "mix" where the Ha is blended in the red channel of the RGB.
Menno
Re: M106 multi narrowband
Hi Menno, yes the HA 'arm' certainly shows much more clearly now - it is a difficult balance to decide how much HA to blend into the red channel. Ultimately I don't think that there is a 'right' answer since it just depends upon what you want to show. But for M106 - since the HA near the core is one of its unusual features - I think that it is a good call to increase it to the point that the arm is visible as you have.
It is definitely a feature of this hobby that processing presents so many choices and as your learning and the tools improve it becomes evident that you could probably revisit most of your past data and improve it it in some aspect.
My most recent example has been something that I learned recently about CMOS cameras. When dealing with very low light levels as the case when doing narrow band - the camera inherent noise levels are often so low that after subtraction of darks you can lose significant amounts of information simply through so many of the pixels just defaulting to zero after calibration (just because of the inherent statistical error in the low numbers that have to be subtracted one from the other). There is a setting in PixInsight called 'auto' which alleviates this problem by automatically adding in a 'pedestal' value (e.g. say 200 e-) to the lights when it detects that too many of the pixels would otherwise come out at zero. So now I realize that I have been inconsistent in my use of the pedestal in processing narrow band (sometimes set to auto- sometimes not) -- but am I going to go back through recalibrate and redo all the NB images just to check ?! . No that would be far too much -- but I might just revisit one of the better images perhaps?
The dilemma's created by new advances (Starnet 2) or simply learning new things ....
PS What really amused me Menno was to look back and see that your first sentence in this now quite long thread was to say that your M106 project is now finished
Tim
It is definitely a feature of this hobby that processing presents so many choices and as your learning and the tools improve it becomes evident that you could probably revisit most of your past data and improve it it in some aspect.
My most recent example has been something that I learned recently about CMOS cameras. When dealing with very low light levels as the case when doing narrow band - the camera inherent noise levels are often so low that after subtraction of darks you can lose significant amounts of information simply through so many of the pixels just defaulting to zero after calibration (just because of the inherent statistical error in the low numbers that have to be subtracted one from the other). There is a setting in PixInsight called 'auto' which alleviates this problem by automatically adding in a 'pedestal' value (e.g. say 200 e-) to the lights when it detects that too many of the pixels would otherwise come out at zero. So now I realize that I have been inconsistent in my use of the pedestal in processing narrow band (sometimes set to auto- sometimes not) -- but am I going to go back through recalibrate and redo all the NB images just to check ?! . No that would be far too much -- but I might just revisit one of the better images perhaps?
The dilemma's created by new advances (Starnet 2) or simply learning new things ....
PS What really amused me Menno was to look back and see that your first sentence in this now quite long thread was to say that your M106 project is now finished
Tim