CCD vs DSLR

[Tenacious Del]

Hi, I have been wondering about this for a few days.

Why do DSLR camera's capture all three colours at once, but SBIG CCD cameras need coloured filters to make colour images?

And what is more sensitive and faster in recording light?

Thanks
Del

[Radar]

That is a very good question mate. I'm not 100% sure of the answer, but I imagine that a dedicated CCD has higher resolution. I'm sure there are others here who can shed more light on this subject.

[AstroTasmania]

Traditionally, cooled astro cams are B&W for maximum resolution. All the pixels are concerned with collecting luminance detail info. A digital single lens reflex camera abbreviated to (DSLR) has a colour matrix over the pixels so three out of four pixels are capturing low res colour information.

The most accurate way of getting astro colour information is to use a B&W CCD and image through red, green & blue narrow band filters to isolate the colour info, which is then combined with the B&W detail image during later processing. There are also one shot colour cooled astro CCD cams, but generally cost more than an entry level DSLR (digital single lens reflex)

However, for many amateurs this can be a costly and long learning curve to travel in terms of the cost of a cooled CCD astro cam, filter wheel, etc. plus all the extra time it takes to get a colour image this way.

So many newcomers choose a DSLR as a low cost entry with a camera they may already have for their normal day to day photography. (easier to convince the family treasurer if it can take pics of the kids growing up). It is not practical to use a computer controlled cooled CCD can for family happy snaps - which a digital SLR can do without being connected to a computer.

In round figures an entry level 6 million pixel colour DSLR may cost a thousand dollars and a cooled astro camera will cost 5 to 10 times that figure even more with top line filters and filter wheel. Such a cost can be a stumbling block to getting into astro photography, especially if you are not sure if this is really what you want to do.

Essentially, there are horses for courses and the level that you want to go to.

Just my thumbnail sketch, I have waited to see what the experts had to say before jumping in with my simplified comments. Hope this helps a bit?

I decided last year on the non & modified DSLR approach with a Nikon D200 (not modified), a Nikon D50 with total filter removal & clear glass replacement (www.lifepixel.com) so I can do scientific/medical infra red as well as astro (results are excellent) and my Canon 350D with a Baader enhanced filter replacement and cooler modification, from Mr Yun Lee at http://www.centralds.net/, which I believe is becoming quite popular and still cheaper than a dedicated astro cooled CCD system.

The mods were easy & no trouble to get done, however, I need to know where to get regular cloud-free skies from, anyone know of a supplier??

Clear skies...

[h0ughy]

Traditionally, cooled astro cams are B&W for maximum resolution. All the pixels are concerned with collecting luminance detail info. A digital single lens reflex camera abbreviated to (DSLR) has a colour matrix over the pixels so three out of four pixels are capturing low res colour information.

The most accurate way of getting astro colour information is to use a B&W CCD and image through red, green & blue narrow band filters to isolate the colour info, which is then combined with the B&W detail image during later processing. There are also one shot colour cooled astro CCD cams, but generally cost more than an entry level DSLR (digital single lens reflex)

However, for many amateurs this can be a costly and long learning curve to travel in terms of the cost of a cooled CCD astro cam, filter wheel, etc. plus all the extra time it takes to get a colour image this way.

So many newcomers choose a DSLR as a low cost entry with a camera they may already have for their normal day to day photography. (easier to convince the family treasurer if it can take pics of the kids growing up). It is not practical to use a computer controlled cooled CCD can for family happy snaps - which a digital SLR can do without being connected to a computer.

In round figures an entry level 6 million pixel colour DSLR may cost a thousand dollars and a cooled astro camera will cost 5 to 10 times that figure even more with top line filters and filter wheel. Such a cost can be a stumbling block to getting into astro photography, especially if you are not sure if this is really what you want to do.

Essentially, there are horses for courses and the level that you want to go to.I decided last year on the non & modified DSLR approach with a Nikon D200 (not modified), a Nikon D50 with total filter removal & clear glass replacement (www.lifepixel.com) so I can do scientific/medical infra red as well as astro (results are excellent) and my Canon 350D with a Baader enhanced filter replacement and cooler modification, from Mr Yun Lee at http://www.centralds.net/, which I believe is becoming quite popular and still cheaper than a dedicated astro cooled CCD system.

The mods were easy & no trouble to get done, however, I need to know where to get regular cloud-free skies from, anyone know of a supplier??

Just my thumbnail sketch, I have waited to see what the experts had to say before jumping in with my simplified comments. Hope this helps a bit?

Clear skies...Shevill Mathers


or you can buy a 350D or 400D second hand, send it over to Korea to www.centralds.net and get YunLee to modify the hell out of it and you end up with a cooled DSLR Just like Mine:eartoear: :eartoear: where is the one you ordered? Hope it arrives soon

[rogerg]

To answer a different part of your question:



And what is more sensitive and faster in recording light?


This depends on the specific camera chip. Each chip has a rated QE (Quantum Efficiency). The QE is the efficiency at which the chip converts light in to digital signal. Bad is 0% good is 100%.

Typically a high class grayscale CCD from someone like SBIG might have a QE of between 65% and 90%. Back Illuminated CCD's are becoming more widely used which make the 90% more achievable.

Tyipcally a colour chip like those used in DSLR's and single shot colour cameras from people like SBIG have a QE of between 20% and 65%. The highest I have seen is 65%.

So, colour will give you nice colour but mean less of the faint detail, for a given exposure time.

You can compensate for lacking QE by increasing exposure time, unless your exposure times are already at their limit and it's not practical to make them any longer.

Roger.

[AstroTasmania]

Hi Dave,

I sent mine off quite a while ago but said no rush, whenever, I am busy with other projects and would not have time to use it for a while. I have finally made the base for my new "OctaDome", a lot more work than it looks.

It is 2.2metre across the flats and will accommodate the Alhena twin, triple & finally a quadruple system. The frames, cladding and roof petals are finished. it's assembly now and building the floor level with the stainless disc plate. It is over my vehicle service pit between the upper & lower driveways and also has to clear the top of the pergola. With its hinged 8 pointed roof petals, it will be a intro talking point when visitors arrive.

Until this is finished I will not have much time, still working on the 16" as well and doing some mods in the main dome. Hope to be in good camera shape for winter skies!

Sometimes, buying the camera after the telescope is just the start of a very long and interesting journey!

Clear skies...Shevill Mathers

[Radar]

So what does Yun Lee actually do to these DSLR's?

Puts a fan on? Any chance that could cause vibrations?

What other mods are there?

[AstroTasmania]

Hi Radar,

Log on to his site, which gives details a pictorial breakdown of what he does. http://www.centralds.net/en/index.htm

He removes & replaces the standard blocking filter and puts in a Baader high transmission filter which increases the H-alpha transmission from 27% to 98% but still blocks UV & IR.

There is a regulated Peltier cooler plate behind the image sensor, the cooling fan is part of the Peltier cooling system. There is also a system to prevent fogging of the sensor cover, quite an elaborate proceedure.

It can still be used as a normal camera, which is a big plus.

The whole mod is very extensive, well worth you having a look at his website.

Clear skies...

[Noel Carboni]

It's been said above, but I'll repeat it just for clarity: The difference between a mono camera and a full-color camera is that the full-color camera adds microscopic colored filters above the photosites in an arrangement called a Bayer pattern, which ensures all three colors (red, green, and blue) light are being measured all across the imager.

The operational difference is that with a mono imager, to take a color photo you need to swing the colored filters in front of the imager one by one and take exposures through them, and possibly also capture luminance information without any filter. By contrast the color imager can only capture both luminance and color data at the same time.

As most things in the sky don't change much during an exposure session, either approach works pretty well for astroimaging.

I'd like to say a few things about full-color imagers vs. mono imagers, though... Most folks will say the mono imager is overall "better" than a color imager. However, I'm becoming convinced this is an oversimplification, and both approches have definite advantages.

I've been collaborating with a guy in the UK who has a full-color astro camera, and I think I can dispel some of the myths about mono cameras producing sharper images. The key is stacking... If your mount moves a pixel (or several) this way and that during the imaging set (and who's doesn't?), you will actually be stacking red, green, and blue information in every single pixel in the final full-sized image. With enough exposures (subs) you can actually achieve an astroimage that's detailed to the very pixel, without the trouble of moving filters around. Advantages:

1. The three colors are all exposed at the same time, so the exposure levels are mathematically calibrated to one another. You needn't worry about exposing blue exposures for longer than green, that sort of thing. And your color conversion software, which knows about the imager's characteristics, takes care of the math to make the colors all come out balanced properly.

2. There's less likelihood of different gradients in red, green, and blue channels (and thus complex color gradients in the final image) if all are exposed at the same time (vs. the object moving through the sky and being in a different place for red exposures, green exposures, and blue exposures).

3. The red, green, and blue channels are by definition aligned with one another in each sub, regardless of the telescope's optical distortions. The job of stacking becomes rather easier, though most stacking software does a pretty good job with either approach. However, no stacking software is perfect, and in my experience the chance of color fringing is smaller with the full-color imager due to misalignment of the three channels during the stacking process.

Disadvantages:

1. The full color imager cannot have its filters removed to achieve maximum possible light sensitivity. This can mean longer exposure times to capture very dim objects. The mono imager can be used, by contrast, to create an LRGB image, in which the luminance information for all colors of light is captured simultaneously.

In summary, the simplification of the processing of the images one gets through the use of a full-color imager can offset the light sensitivity advantage of the mono imager. So there is no one "better" approach. In short, don't discount the possibility of a full-color dedicated astro camera.

By the way, this is an unresized crop from an image constructed from 34 subs with a 6 megapixel full color SXVF-M25C imager; I see no loss of resolution here: http://www.ourdarkskies.com/gallery2/d/992-1/M81_M82_Area_Center_Crop_Greg_Noel.jpg

-Noel