Image Noise

[TrevorW]

Recently trying to capture NGC2035 in the LMC which is relatively low in the sky at this time of the year doesn't get above 58 degrees - started imaging just before 8:00 for Lum but image seems fairly noisy not processing with flats (need to do some) - what is the best way of checking the Lum frames for noise - the other channels seem to be fairly OK as they where captured later in the evening- feedback :welcome:

[Startrek]

Trev,
Lum is always going to be noisy under B6 to B8 skies , it’s like opening the floodgates to everything , Signal and unfortunately loads of Noise.
Flats are not going to do diddly squat to mitigate noise , only correct your background for vignetting and dust motes etc… Flats may even add a bit of noise to the overall noise floor.
With Lum the noise will reduce per sub as the night goes on , particularly if you started just after Dusk, your camera sensor with Lum is super sensitive to light ( photons )
Under my B8 skies and no moon , my Lum subs using the 2600MM ( which has a higher QE than your 294MM ) are limited to 20 to 30 sec on dim and medium brightness targets and 45 to 60 sec on super bright objects. Even then the histogram is off the Richter scale.
Down at my B3 dark site I can push 3 min Lum subs ( no moon around ) and noise significantly less and manageable with processing
Your RGB subs generally will have less noise than Lum but lack fine detail of the Lum filter.
Any Moon glare is also going to play havoc with your Lum subs !!!
Ways to limit noise on your Lum data -
Shorter subs 30 sec to 60 sec ( as long as the read noise is swamped which under B6 I’m sure it would be )
Set your Gain to lowest read noise , I think 120 ?
Cool your camera down a bit more
Take loads and loads of subs, like 100’s , more if the target is dim ( SNR is king )
Use Darks ( matching lights ) to mitigate , dark noise , hot pixels , cold pixels etc…
Dither each sub to mitigate fixed pattern noise
Use noise mitigation tools in your post processing software ( there’s an obvious limit to that )

Hope the above helps
Martin

[TrevorW]

Thanks Martin gain offset setting currently 139 50 took very short subs of m42 same night 10s. I noticed there was not much noise if those Lum subs :)

[Startrek]

Trev,
NGC 2035 brightness magnitude of 10.99
M42 brightness magnitude of 4.0
Huge difference in signal ( brightness) and size of object between the two
Moons ( 25 to 30% ) up early as well on your Lum subs with NGC 2035
Lum takes a lot imaging time to nail optimum SNR with sub length , Gain and quality/quantity of data
Lum is similar to Oiii NB where the moons glare plays havoc with SNR. For NB imaging I try to time my Oiii subs with minimal Moon glare ( if practicality possible)
There’s no rule on which order or on which night you capture your data. Lum ( Broadband ) and Oiii ( Narrowband) are the only exception to that rule.

Cheers
Martin