newbie ccd astrophotography

[spartan2007]

Hi guys about to take the plunge into photography which i have never done before .I am mainly interested in dso but would not mind doing planets as well but will use guide camera when i get one i have set a budget of 2000-2500 AU for main camera .My scope is a celestron cgem 925 edge hd i have been looking at the 428ex osc or mono would this be a good start point y/n.How hard is the learning curve on ccd photography any input would be great .clear skies

[Geordie mc]

First of all, the learning curve is quite steep. If you haven't already seen it, I recommend the book 'Making every photon count' as a good place to start. Secondly, think long term. A colour camera is easier to work with and needs less time spent imaging and processing usually, but a mono camera with filter wheel and a selection of lrgb and narrowband filters will give better images and allow more flexibility, but they are expensive and you need a lot more time. I started with a dslr for colour work, and still use it occasionally but went mono for ccd. I thought I'd made a mistake as it took awhile to get used to using it, but now I am pleased I persevered. As you learn the software and get used to the routine things get easier. Still have loads to learn, but getting there. If you get lots of clear nights and can have a permanent setup then I'd def go for mono. If you rarely get clear skies or have little time then colour might be a better option. I don't know the 428, but it has good reviews. To be honest I think anything by Atik, Starlight Express, QSI, SGig etc will be a good ccd, but you need to match it to your scope. There a formula somewhere which takes into account the CCDs pixel size and your scopes focal length. Sorry don't have it to hand. One last thing, with your scope you'd probably be better guiding with an OAG than a guide scope. There really is so much to consider. Try to find a local group with some experienced imagers and see what they are using. Good luck.

[swag72]

I totally second the comment above abut getting hold of the book 'Making Every Photon Count' - While it won't specifically cover CCD imaging alone, it's an excellent resource all round as far as astro imaging is concerned. Well worth having on your book case.

I started off using a DSLR, then moved to a mono CCD and also had a dabble in a Colour (OSC) as well. I have to say that, in my opinion, the learning curve is with astro imaging in general, so if you are looking at starting with a CCD then totally go for it. 

I prefer a mono - I found a OSC data really difficult to work with when compared to mono data. It's more versatile all round as you can use narrowband filters as well to help combat light pollution as well as to counter the moon effects a little (but not totally).

So it would be mono for me all the way

[Owmuchonomy]

Imaging DSOs at that focal length is not for the faint hearted or the inexperienced. If you are planet chasing however your scope is perfect. Think about what you want to achieve and work out the correct scope/camera option from there.

[Geordie mc]

Spot on about the focal length Chris. A reducer would help and as I said an OAG. Good mount calibration and guiding will be essential. It's not impossible but a hard start. If you decide to do planetary work I can recommend the ASi120MM. It's terrific for the price. A colour model is also available.

[Physopto]

Just be aware that a colour camera cannot be binned (generally as not 100% sure about all different masks). Binning if you don't know increases the pixel size, I.e. 2x2 is a super pixel of 4 real pixels, 3x3 is a super pixel of 9 real pixels. This makes the camera much more sensitive to photons (get more photons in 3x3 than on one ordinary pixel). This shortens the imaging time needed. As someone has said long focal length astrophotography is very much more difficult than wide field. You will need a very good mount for long focal lengths. Setting up a long focal length scope on a mount can be almost impossible sometimes. I watched a friend once trying at a star party. He was nearly pulling his hair out. Each time he got the polar alignment right and tried to tighten up the mount it moved off. It was a very expensive celestron mount as well. So if you are just starting out try either wide field first to learn the ropes, or try imaging the planets. Still full of pitfalls, read up as much as possible and get specific advice first. Join a club if you can loads of good advice usually freely available.

Derek

[kalasinman]

I am aware that there are members who would disagree that OSC is faster in any case than mono and filters...

[Geordie mc]

I think gathering data using OSC is generally quicker, but processing it is a different ball game. Then again, I have plenty of cloudy nights to process and precious few for collecting photons.

[Physopto]

Yes in Shields, if it ain't raining or cloudy there is always the fog! OSC is quicker and you can get nice images. But there are far more options with mono and filters. Plus the added "now what do I do" factor. An ordinary dslr can get you started fairly cheaply and many love them. I suppose it depends upon what you like or want to do. Big decision spending money! That's where I think a club and talking to members comes in.

Derek

[kalasinman]

As I recall the rebuttal goes; a mono CCD has increased sensitivity so collects faster, an OSC has a Bayer filter so each channel is cut by 33% or more. Then, add in the more difficult processing of OSC, and I hear net that mono and filters is better and faster in total.

[Physopto]

As I recall the rebuttal goes; a mono CCD has increased sensitivity so collects faster, an OSC has a Bayer filter so each channel is cut by 33% or more. Then, add in the more difficult processing of OSC, and I hear net that mono and filters is better and faster in total.

Not quite! Using mono takes longer and is more difficult to get to grips with. By binning, you increase sensitivity. You usually bin when using colour filters and use no binning when imaging using clear or cls filters. The clear/cls filters give the luminance component and detail. You don't need the same detail for the colour components to get a great image, you can but you waste time for something you will not really notice.

Derek

[kalasinman]

Not quite! Using mono takes longer and is more difficult to get to grips with. By binning, you increase sensitivity. You usually bin when using colour filters and use no binning when imaging using clear or cls filters. The clear/cls filters give the luminance component and detail. You don't need the same detail for the colour components to get a great image, you can but you waste time for something you will not really notice.

Derek

I read an analysis by a member here  who claimed that because of the binning, and the lack of a Bayer matrix that mono LRGB data collection was similar in time or faster than for a finished picture with OSC, even with the multiple filters. For example , assuming the same number of photons per same size pixel per unit of time, the OSC takes 4x as long to collect the same number in red or blue, and twice as long in green at best. Adding in 2x2 binning for each color with RGB. and each of those times are further reduced to 1/4 of that.

So for red and blue the equivalent to an hour in OSC could be collected in 4 minutes each, 8 minutes in green,  with RGB binned 2x2 = 16 mins. This leaves roughly 45 minutes to fiddle with focus and collect luminance. The luminance channel is 4x faster than OSC even without binning, as it collects and combines all colors 100%.Of course the total data would still be greater in this example than with the OSC as OSC creates luminance out of Bayered data, and the luminance in LRGB would be additive to data of the color channels already collected.

Well, I apologize if I made a mess of this, just recalling a statement elsewhere on SGL.

[Physopto]

Hopefully some one else will have a link to that data. Sounds interesting. I know that some of the pictures that others have taken with a DSLR were very quick compared to me getting colour via a QSI 583/683 camera. I could not match their colour in the same time. I have not used my DSLR for astro imaging, so don't have a direct comparison for my own times.

Thanks,

Derek

[kalasinman]

I think as I inferred the part about luminance data that it could be wrong. The rest was from a post by Olly.

[spartan2007]

Thank guys lots to think about the reason i looked at the atik 428ex was maybe once i learn i was  try hyperstar. But thats a long way off i think i will need learn the basics first but have been told i need  a focal reducer for my scope with its focal length  and finding one is a challenge ,anyway cheers

[Geordie mc]

The hyper star looks good on first glance but be aware that getting sharp focus on such a fast system is extremely difficult. You would need a motorised focus system in my opinion. The cost of astro imaging can very quickly spiral upwards. As has been said, look into all options carefully. Be clear what you want to achieve and how much you'd be prepared to spend.

[Physopto]

" The cost of astro imaging can very quickly spiral upwards"

Yes, tell my bank manager!

[ollypenrice]

I think gathering data using OSC is generally quicker, but processing it is a different ball game. Then again, I have plenty of cloudy nights to process and precious few for collecting photons.

OSC absolutely isn't quicker, in my view. An OSC is blinded to 2/3 of the light by its colour filters all the time. When shooting the luminance in a mono camera you shoot all three colours at once. How can this fail to be faster? I know it's counter intuitive but I recently posted a couple of ustra quick LRGBs.

Far more importantly for the OP is to go for a camera with large enough pixels for his focal length which, even reduced, is going to be long. By binning mono pixels 2X2 a reasonably sensible pixel scale can be acheived. It would be a would be a terrible shame to make binning impossible by going for OSC.

Olly

[Stub Mandrel]

Consensus above seems to favour mono, yet it's interesting that Making every Photo Count sets out boths sides of the argument, then comes down in favour of colour on the basis of personal preference and 'it does everything I want to do' - with the caveat that he uses mono for narrowband imaging.

<edit - I can spel, I just can't get on with thi$ bl00ming tripewriter>

[Physopto]

I think it is a case of "horses for courses". It is what suits you at the time. Trouble is if you really get into it or change your mind it can costly which ever way you go. Read up as much as possible before spending. I still messed up even after doing just that! It's a part of being human we change our minds all the time.

Derek

[ollypenrice]

Consensus above seems to favour mono, yet it's interesting that Making every Photo Count sets out boths sides of the argument, then comes down in favour of colour on the basis of personal preference and 'it does everything I want to do' - with the caveat that he uses mono for narrowband imaging.

<edit - I can spel, I just can't get on with thi$ bl00ming tripewriter>

But you can't get away from pixel scale. I don't have Steve's book in front of me but I'm sure he'll have something to say about working at a reasonable pixel scale. A Canon 500D, for instance, in the OP's scope would give 0.38 arcsecs per pixel which would be far too fine.  Unbinned, an Atik 428EX would give 0.4, also far too fine, but binned 2X2 it would give 0.8. Add the reducer and it would be about 1.2 arcsecs per pixel and that would be good. However, an OSC version of the camera would be stuck at an unproductive 0.4 arcsecs per pixel or a still unsatisfactory 0.6 with reducer. 0.6 is possible on a really accurate mount under a steady sky. This means more accurate than just passing the old 'round stars' test which tells you nothing about any truly random errors since the random errors give a round distribution anyway.

Olly

PS By the way, I've talked to Steve about this quite a lot and fully accept the fact that OSC can be less exasperating in UK stop-start-stop-scream imaging conditions. My other points about OSC were specifically concerned with speed and binning. It is perfectly possible to prefer OSC while accepting that it's a little slower.

[Stub Mandrel]

Probably a naive question, but why is oversampling bad? I can understand that big pixels collect more light and that, when dealing with very, very faint sources you could lose objects that only just make it above background, but surely the only real danger with oversampling is sharpening it before you downsample it and 'creating'  objects that aren't really there out of quantum errors.

Looking at audio - most audiophiles agree that oversampling creates a 'more natural' sound than applying the nyquist criteria; although it can't introduce detail that isn't there the result is fewer artefacts in the fine detail.

>ducks<

[kalasinman]

So who said it was bad?

[Geordie mc]

On the binning I agree and I did point out the need to check the formula for matching camera to scope. However, I can go out and take 20x2 min exposures, dark frames and crucially light frames in a couple of hours and get a reasonable result with OSC but would usually take longer if I had to take 4 lots of data, even allowing for binning, because of the time taken to refocus between filters, and then take flats for every filter. If you took 5x2 min then binned 2x2 and took 5x1 min for each colour, then all the flats to match and then darks at 2 min and darks at 1 min, it seems to add up to more time to me. In theory, OSC may be slower, but in practice I have not found that to be the case. I suppose to an extent it depends upon the brightness of the object being imaged. If it is very feint then OSC is going to struggle to get enough photons with short exposures and will need more of them.

Obviously Olly has much more experience than me and so is probably correct in his post, but for me, if I'm tight for time I go out and shoot a bright dso using OSC.

[kalasinman]

I get that. Olly only uses a flat from one filter for all, and says it works fine. Auto focus and parfocal filters? I dream of a CCD, but a OSC would cost the same and I'd feel like halfway there. No NB, no binning, no false color pallets.