Looking at mono CCD for the Pentax (budget ~4-5K)

[NickK]

This is a preliminary discussion around CCDs and possibly Large format CCDs.

I'm open for suggestions - knowing that the Pentax will illuminate a Kodak 11000 chip (Mario Weigand's Horsehead Nebula using the same scope with an STL-11000 is an example - stunning). However I'd also like to image smaller DSOs with the option of wider field if needed (thus the 9uM worries me).

The ATIK 16ic did a good job of showing me what a sensitive 659x494 pixels (7.4x7.4um) can pull out in terms of detail. It also spoils you with a low noise of 7e.

The KAI-11002 chip is:

* 4008x2672 pixels (9.0x9.0uM)

* Mono QE = 50% - low quantum efficiency (especially with Ha),

* RGB QE = 34%@450nm, 37%@530nm, 42%@650nm

* 13e high noise and thus needs good cooling and darks.

* 1-3fps - no so hot for focusing.

* Anti-bloom

The KAI-04022 chip is:

* 2048x2048 pixels (7.4x7.4uM),

* Mono QE = 55% - low quantum efficiency

* 11e higher noise and thus needs good cooling and darks.

* Anti-bloom

KAI-8300M

* 3326 x 2504 pixels (5.4x5.4uM)

* QE max at 540nM (~56%), 48% at 470nM and 656nM.

Measure that against the Sony ICX-285AL (ATIK 314L+):

* 1392x1040 pixels (6.45x6.45uM)

* Sony only give relative QE..

* anti-blooming

But with stats put to one side, with 675mm FL, those chips give:

KAI-11000 -> 2.77 arcsec/pixel, 123.2 x 184.9 arc min FOV

KAI-04022 -> 2.28 arcsec/pixel, 77.7 x 77.7 arc min FOV

KAI-8300M -> 1.66 arcsec/pixel, 69.3 x 92 arcmin FOV

ICX-285AL -> 1.98 arcsec/pixel, 34.4 x 46.0 arc min FOV

And for reference

ATK 16ic -> 2.28 arcsec/pixel, 18.7 x 25 arc min FOV

Now looking at the rosette neb using CCD calc:

KAI-11002 -> 1 shot with space left over either side!! This would do the horsehead neb in one shot!

KAI-04022 -> most likely a 2x2 mosaic to get the full nebulosity.

KAI-8300M -> 1 for about 95% of the nebulosity. You could probably just squeeze the horse head neb in one shot, more like 2.

ICX-285AL -> you'd need a 2x2 mosaic, possibly a 3x3 mosaic

ATIK 16ic -> well lets put it this way.. 6x6 mosaic possibly even a 7x7!

The problem here is the pentax can go wide and reasonably high res - how many objects would warrant the very wide field/lower res 11002 vs the 4022 vs the 8300M?

Squiddy wowed people with the Sony 285 chip (IIRC it was the ATIK 314L+) and his rosette nebula where the sensitivity and noise pulled out really fine detail..

It makes me think there's a need for two camera - one medium resolution and one high resolution?

*ponders*

I'd love to have some discussion around this and hear people's views on this. Although people's situations would vary as their scopes vary.

Edit: added the KAI-8300M

[ollypenrice]

Personally getting excited over sampling rate strikes me as a waste of time. I take pics either side of optimal. The worst sampling rate I ever used produced the most internet-visited pic I've yet taken. http://ollypenrice.smugmug.com/Photography/Widefield-images-including/ORION-FIN-V1BNWEB/1167645022_FqzDg-X2.jpg

The big 11 meg chip is fantastic but there are issues. It needs serious computing and I think the Pentax will probably vignette on it. I have a Pentax owner here now but it may not be just the same model. If you have LP then gradients might be a persistent issue. Quite a bit of image 'cleaning' may be necessary, too. But what a field!!!

As for darks and flats, only the 285 can manage without darks. All others need that or a pixel map and flats are always essential.

The 4022 is a good compromise and being square makes the most of your light cone/flat field.

You can't hava a camera for every scope - or can you???

Olly

[NickK]

The big 11 meg chip is fantastic but there are issues. It needs serious computing and I think the Pentax will probably vignette on it. I have a Pentax owner here now but it may not be just the same model. If you have LP then gradients might be a persistent issue. Quite a bit of image 'cleaning' may be necessary, too. But what a field!!!

Very valid points. Quite what the light cone characteristics are (diameter and quality) is difficult to tell. I'll ping Mario and see what his experiences are.

To quote the marketing blub: "The Pentax 105 SDP has an excellent focuser with a free diameter of 94mm. This allows full illumination of the 6x7cm film format.".

Just for reference, the focuser's first thread is at 84mm (3.3"). I then need to use two Baader adaptors to fit a 2" by replace the russian doll: 84mm (3.3") > M68 then M68 > 2". If I'm using the 1.25" EPs etc then the russian doll supports it.

My concern is the 11002 is like taking a 3" brush to the canvas when a 1/4" is required for the portrait. It's design is for big posters prints. Then I ask myself if all the images were on a distribution bell curve, the 11002 seems to make sense for only the top 10%. Although smaller - the higher resolution chips allow for all but the lower 25% of objects.. then these seem to make more sense.

Afterall - quality of detail within the image is important to me. And if I'm looking at DSOs such as M81 etc then seeing nuances the dust lanes etc is more important than one-shotting an entire nebula.

7.4nM is a nice pixel size and 5.4uM (52% less area per pixel than 7.4uM) is nice but I'd need twice the time to image then add a smaller 25,000e bin size with higher noise makes me concerned with the 5.4uM KAI-8300M.

Edit: just occurred to me that the QE is higher with the 8300M which offsets some of the reduced size. It may be 20% or so thus reducing the exposure times.