LRGB - Exposure times on RGB channels?

[Droogie 2001]

I have a question regarding LRGB exposure times.
From what I have read it is recommended that you take x3 times the exposure times with your coloured filters than the Luminance due to these letting less light through.

I also understand that you can reduce this time by binning the coloured subs, however from previous threads I have read and worked out that my Arc Seconds per pixel is 3.18 unbinned which leads me to my dilemma.

The largest Arc Seconds per pixel for my setup is 3.4 (FL 351mm) so binning is not ideal as it will increase my Arc Seconds per pixel to over 6 which could lead to issues with massive star bloat?

I have also read that really I should be aiming to take as long and as many luminance subs as possible.
So if I take a 10 minute luminance sub then I should take 3x10mins i.e. 30mins for each coloured filters? So for an hour of luminance I would need to take 3 hours of each filter? Giving me 9 hours of coloured filters for 1 hours of Luminance, a total of 10 hours?!
What is the best way to approach this? Is it that important to take x3 RGB the exposure time?
How can I reduce this down from 10 hours as I don’t have an Observatory and have to put up with the UK weather.

Could I bin the colour channels and leave the luminance at 1x1? Will the potentially bloated star issue from binning and giving me +6 Arc Seconds PP be an issue or can I get away with this as the colour channels are only used for their colour and not for detail such as stars?
I realise that it will also depend on the target and some of the colour can be adjusted in post processing though at the expense of adding noise. I would be interested to hear other people’s opinions on what, where and if they reduce exposure times down on similar setups.

Thanks

[cfpendock]

My opinion is that your stars could look a bit "blocky" if you binned the colour at your resolution.  I also think that the amount of time you allocate to L compared to RGB depends on the target.  If you don't bin the colour, there are also ways of enhancing the L with a false Luminance made of the RGB subs (after all, L is made of RGB).  

Others more expert than I will hopefully give more information.

Chris

[vlaiv]

I think you can bin your RGB channels without too much worries about having stars bloated.

If you think about how L+RGB work, L giving "shade" and RGB giving color to pixel. If L is recorded at full res, then pixels around star will be black, and when you try to color the black pixel it will remain black. Depending on how many subs you take for each RGB channel, you might try to do regular L+RGB and regular L + x2 drizzle RGB and see what gives you better image.

With exposure times, huh, not sure on that one. Depending on target and your sensor/filter response in R, G & B you will be able to target specific SNR per channel, but I'm not sure what that target SNR should be in comparison to L SNR. This is because human eye/brain is less sensitive to color variation then to light variation. Therefore noise in L is more perceivable in image than noise in R, G and B, and I think our brain even responds differently to hues so SNR of each channel can be different. So I think that one can get away with smaller SNR for R, G and B, but how smaller that I don't know.

[Droogie 2001]

Thanks for the responses. I guess it is going to be a case of testing and seeing what the results are between the different methods.
 

[Skipper Billy]

I have had some success with shootting RGB 1x1 and no luminance layer at all - works for me !

[ollypenrice]

I would not bin colour at your resolution. (One of our rigs runs at 3.5"PP and we never ever bin at that.) Far too coarse. It isn't bloat that is the problem but the blocky look of the stars. 

I have never shot 3x as much colour as L either. I don't know anybody who does this. It isn't necessary or productive. If you stick to parity, so equal amounts of LRGB, you will find it easy to process and retain good colour. If you shoot more L than RGB you can still fill the L with colour but more advanced processing techniques are needed.

L can be either very important indeed, even vital, for an image or it can be a positive disadvantage, depending on the target. Some examples below;

Stars are best wthout L if they are unbinned. They are smaller and more colourful, so clusters are great in RGB.

Anything faint (tidal tails, faint nebulosity) thrives on L and you will simply never get there in RGB (not in the real world, anyway!)

Dusty objects thrive on L, too.

If going for emission nebulae in Ha and OIII you need very little colour. Have a look at Gnomus' Seagull on the DS board. Lots of Ha and OIII but just 4x400 secs for the colour.

I would just shoot equal amounts of LRGB until you get the hang of it, then start doing fancy weightings once you understand how it works.

Olly

 

[Droogie 2001]

Thanks for the advice and its good to know that this x3 exposure time RGB is unnecessary.

Just getting to grips with the CCD so will sticking  with Mono and post processing first. Finally got everything working correctly last night which is a great relief.

Once fine tuned this I will then come back to the LRGB.

[ollypenrice]

50 minutes ago, Droogie 2001 said:

Thanks for the advice and its good to know that this x3 exposure time RGB is unnecessary.

Just getting to grips with the CCD so will sticking  with Mono and post processing first. Finally got everything working correctly last night which is a great relief.

Once fine tuned this I will then come back to the LRGB.

This is a good plan but don't become convinced that RGB is difficult. It is really easy. Personally I use AstroArt for stacking, calibration and colour combining because it is fast, easy, transparent and works. 

Olly