First LRGB image

[adamsp123]

M33 taken with Atik 383L+ on a 190MN guided with a QHY5v on a finderscope with PHD, over 2 nights one good one a bit murky.

L = 40x8min , RGB = 15x4min 2x2 bin each channel

Stacked in DSS, aligned + LRGB combined in Maxim DL, tweaked in PS.

What a game of soldiers I had with this, complete new experience!

Really not like the DSLR process.

My processing skills are in need of serious improvement as the result suggests I think, I'm sure I could do better with this data.

[Deneb]

Still not a bad image, I think it's not a hard object to image, but a lot harder to process then other galaxies.. But as you say it's a different ball game when imaging with CCD LRGB.

I done a little re edit, hope you don't mind..

Cheers

Nadeem.

[MartinB]

Fantastic detail and a vibrant image but it lacks a little colour. One problem I think is the length of your binned subs, 4 mins is too long in my view and quite a few stars will be loosing colour as a result. 2 mis max is probably sensible if you have much light pollution. The other thing that helps saturation is to give each colour channnel at least the same equivalent overall exposure time as the luminence which in this case would be about 40 mins

[adamsp123]

Thanks for the kind comments

Deneb, I don't mind at all, I will have another play myself.

Martin, thanks for the advice, I thought 4 mins wasn't long enough!! little that I know, and the equivalancy of exposure is something I hadn't even considered.

Oh well lot to learn with Mono CCD - shake off the DSLR mindset I suppose.

[Tim]

With LRGB Pete, your subs will have to be shorter than "usual" because they will just blow out the colour and the detail else.

Remember that great talk that Peter Shah gave at Lucksall, might be an idea to assess how long to go for each filter to get the true colour using a G2V star. Damned if I can remember the details though!

As Martin says, tons of detail in there, you must be well pleased

[Roger B]

That's way better than my ever first LRGB,

Your right it's a different ball game between a DSLR and Lrgb imaging. I tried it for a while but didn't really have the time for LRGB i wanted it all in one go i suppose, so i went back to DSLR. I'm now thinking of taking it up again coz i want to better my hobby.

Roger

[RobH]

Pretty good Pete, and a tricky target too.

The trick with maintaining colour is to keep a close eye on your histogram. If anything is clipped then whatever that is will lose its colour detail....watch the small spikes at the top end of the histogram. These will be your brightest stars. Make sure there is a small gap between them and the white point at the far right.

If you've managed that, then you have kept the colour information in your intitial data. What you then need to do is make absolutely sure that you don't lose it in the processing by clipping the stars.

You can always do a second proecess just for the brighter stars, and drop them into the image.

Regarding binning, although it allows you to get data quicker, you lose resolution, and this does show up where stars and fine detail in galaxies and the like are concerned.

These days, the only time I bin anything is when I'm trying to grab very faint data that doesn't contain much detail, such as faint outer planetary nebula halos etc. The rest of the time, I tend to shoot 8 or 10 minute 1x1 subs for my colour and accept that this means more time imaging.

Cheers

Rob

[MartinB]

The binning debate will never go away Pete! With that set up you are starting to be over sampled except in the very best seeing conditions so a good case can be made for binning.

This is the rationale for binning. Binning reduces read noise by a factor of 4. This increases your dynamic range by a factor of 4 and also quarters the exposure time you need to overcome the read noise. Your camera already has very low read noise and you have a fast scope. If you are able to measure your sky background, by the time it reaches 500ADUs the read noise is making an insignificant contribution. You can expose your luminence for longer if you like and a lot of people on here are convinced doing so pulls out more detail. However it is a very good idea with the RGB data to stick with this minimum exposure time in order to avoid over saturation. The exposure calculators I've seen don't include info for binned subs but I would guess that a 500ADU background will do the trick.

When I use a 500ADU background limit, exposures end up coming right down, typically under 1 minute and usually somewhere between 30 and 50 seconds ( I use a ratio of RGB 1:1:1.4) With these short exposures the stars don't bloat and I think you would get excellent results. If you go for 2 minute exposures your stars will suffer especially when binning. I have seen spectacular images using exposures around the 20-30 second mark. You still need to get the overall exposure of each channel up to 1/4 of the luminence imaging time but you will end up with a significant benefit in overall s/n ratio for a given imaging time by binning.

I think binning is best avoided when images are undersampled but in your situation is will be well worth experimenting and deciding for yourself...but make sure you stick with very short binned colour exposures when you are experimenting!

[RobH]

Interesting info about the 500ADU background Martin.

Also, I agree with what you say about binning with regard to seeing conditions/sampling.

In fact, I've recently bought an SX H18, which has 5.4 micron pixels.

I also have a large SCT ordered, and when this arrives, I'll be using the H18 on it and will run it 2x2 binned, as this will still give me 10.8 micron pixels, but with all the advantages of better S/N etc, and 10.8 micron pixels will be much better suited to long FL imaging in our skies than 5.4 micron pixels.

Cheers

Rob

[MartinB]

Interesting info about the 500ADU background Martin.

Also, I agree with what you say about binning with regard to seeing conditions/sampling.

In fact, I've recently bought an SX H18, which has 5.4 micron pixels.

I also have a large SCT ordered, and when this arrives, I'll be using the H18 on it and will run it 2x2 binned, as this will still give me 10.8 micron pixels, but with all the advantages of better S/N etc, and 10.8 micron pixels will be much better suited to long FL imaging in our skies than 5.4 micron pixels.

Cheers

Rob

That will be one hot combination!

[adamsp123]

Good stuff, I am all for doing some experiments given some clear skies, all points all well taken guys, my camera also has 5.4 micron pixels.

Another ask is regarding Narrow Band what sort of exposures should I play with for these filters? And bin??

[adamsp123]

I had another play with the data and reduced by half the number of Luminance subs.

In doing this I have found a way of stacking and merging the files I feel happy with.....

Stack each of the L R G B separately in DSS, register and merge the RGB in Registar - then register and merge L with RGB (merged) in registar.

Tweak in PS

Result below, I have also added a Canon DSLR image I took a while ago of about 3 hours data IIRC, I just cannot get any colour from it except that blue!