Atik 490EX

[Earl]

Thanks to FLO for posting up the spec, and I thought it best to start a discussion thread outside of the Sponsors section.

These are very small pixels, I curious as to why go so small, is it really aimed at a Bin'ed market?

[Ibbo!]

a bit too small for me as well

what I would like though is a nice quiet mono Sony chip with lets say 5/6 um and about 15 x 10mm for 100 quid and a bag of sweeties

OK 2 to 21/2 grand

i don't want much do I

meantime I might have to get the 694 chipped ccd

Steve

[Black Knight]

It certainly looks an interesting prospect though. I shall look forward to seeing images when they start coming in ...

[RAC]

It should be a real winner with short FL high quality refracters or lenses.

[ollypenrice]

It should be a real winner with short FL high quality refracters or lenses.

This is the whole point, I think. Use it in very short FL setups and still get good final resolution and fairly wide field. If the F ratio is fast the small pixels probably won't be too lacking in sensitivity.

Small portable mount, small optics, get to a dark site...

Olly

[Ibbo!]

It should be a real winner with short FL high quality refracters or lenses.

yes it should

but I want one for the a bit longer f/l please

Steve

[x6gas]

a bit too small for me as well

what I would like though is a nice quiet mono Sony chip with lets say 5/6 um and about 15 x 10mm for 100 quid and a bag of sweeties

OK 2 to 21/2 grand

i don't want much do I

meantime I might have to get the 694 chipped ccd

Steve

Well binned 2x2 you've got 2.28 million 7um pixels in a 1690x1352 array packaged in the chip size you're asking for. Not close enough?

[Ibbo!]

I prefer not to bin

Steve

[r3i]

I prefer not to bin

Is there a downside to binning?

e.g. hypothetically speaking would a sensor with say 1000*1000*7um square pixels perform better than a 2000*2000*3.5um square pixel array binned at 2x2? (I think I've got the arithmetic right but if I haven't I'm sure you know what I'm trying to ask )

[Ibbo!]

I use a SXH9 and have not got good images with 2x2 binning

More than likely my fault rather than the kit

Steve

[ollypenrice]

I use a SXH9 and have not got good images with 2x2 binning

More than likely my fault rather than the kit

Steve

But at what focal length? If your combined pixels are too big then the stars will look blocky.

Olly

[Ibbo!]

1.8 m Olly

Steve

[jackrussell0232]

If the 490 proves popular, I wonder if the guys at Atik will be offering chip swaps to 460 owners? Looks like a one for one ccd replacement which would only need a firmware squirt? Maybe they'd store your 694 chip for you too, for when you want to swap back.... Sony have been pretty smart here I think.

[ollypenrice]

1.8 m Olly

Steve

Long. Then I'm very surprized. What was the issue?

Olly

[Ibbo!]

Odd star shapes

and I tell a lie 2m with the SCT as I have not tried with the RC as yet

could be down to a bit of mirror movement

now I have the OAG repair(fingers crossed ) I will put the RC back on the mount and give it another whirl or two

Steve

[AlistairHowie]

It took some searching, but here are the absolute, comparative QE figures for the Sony chips in the 490EX, 460EX and 314L+ CCDs. This data comes from Allied Vision Technologies, so I'm making the assumption that the QE is entirely chip dependent and not influenced by accompanying CCD electronics. If that is a false assumption then you should ignore the charts. But I think we can use them. Particularly interesting the difference between the 490 and 460 ...

[ollypenrice]

Yes, I guess this says, Don't go for the tiny pixels if you don't need them at your focal length.

How much the electronics come into play is anybody's guess but I'm quite sure they matter.

Olly

[IanL]

It took some searching, but here are the absolute, comparative QE figures for the Sony chips in the 490EX, 460EX and 314L+ CCDs. This data comes from Allied Vision Technologies, so I'm making the assumption that the QE is entirely chip dependent and not influenced by accompanying CCD electronics. If that is a false assumption then you should ignore the charts. But I think we can use them. Particularly interesting the difference between the 490 and 460 ...

As per my post in the other topic, but for anyone who didn't see it:

Bear in mind that this is from data sheets from a manufacturer of industrial cameras. The black QE charts are for the monochrome versions of their industrial cameras and may be somewhat useful for comparing astro CCDS with the same chips, but the Sony datasheets would be better. The black QE charts for the 490EX and the 460EX are as close to identical as makes no difference, the 314L+ chip is approx 10% worse.

The RGB charts are showing the response for the colour versions of their cameras, i.e. with a bayer matrix of some sort. They are showing the net response for the red, green and blue filtered sensor elements. Clearly each of the filters has a pretty wide bandwidth. These charts are no use for the purpose you are trying to use them for, since the Atik responses would depend on the QE of the mono chip plus the bandwidth of your own filters.

[AlistairHowie]

Thanks, Ian. That's very helpful. Below I've added in the Sony datasheet charts for the 460 and 490 chips. Unfortunately, their QE numbers are relative not absolute, but their charts closely follow the pattern of the other ones, so potentially still valid information if we're talking about comparing their OSC CCDs?

It would certainly be a lot easier if they published absolute numbers rather than relative ones. For such expensive items it's a surprise that basic info on QE isn't easily available. And then there's poor Olly having to remember Pythagorus to calculate chip diagonals!

[IanL]

Have a look at their datasheet and see if they give a peak/maximum QE at a given wavelength, that would allow you to calibrate the relative charts and verify whether the non-Sony ones are valid. The RGB charts are similarly useful if you are comparing OSC cameras and can get the peak/max QE for each filter.

[AlistairHowie]

Have a look at their datasheet and see if they give a peak/maximum QE at a given wavelength, that would allow you to calibrate the relative charts and verify whether the non-Sony ones are valid. The RGB charts are similarly useful if you are comparing OSC cameras and can get the peak/max QE for each filter.

That information absolutely doesn't exist, Sony just doesn't publish it. A Google search will turn up many people complaining about that fact, and having to rely on third party specs. I've searched extensively and there's nothing with Sony's name on it re absolute QE numbers. Which is kind of ironic. The Atik website also generally doesn't give QE numbers for their CCDs, so they've been forced to search for suitable adjectives such as "good" QE for the 450, up to "outstanding" for the 490EX. Only the 4000 and 11000 have stated QEs, which is probably because Kodak publish those numbers in their datasheets.

As I said, strange (frustrating, annoying, etc) that these numbers are not readily available.

[x6gas]

That information absolutely doesn't exist, Sony just doesn't publish it. A Google search will turn up many people complaining about that fact, and having to rely on third party specs. I've searched extensively and there's nothing with Sony's name on it re absolute QE numbers. Which is kind of ironic. The Atik website also generally doesn't give QE numbers for their CCDs, so they've been forced to search for suitable adjectives such as "good" QE for the 450, up to "outstanding" for the 490EX. Only the 4000 and 11000 have stated QEs, which is probably because Kodak publish those numbers in their datasheets.

As I said, strange (frustrating, annoying, etc) that these numbers are not readily available.

Hmmm interesting. Haven't I seen peak efficiency figures quoted which effectively give a reference to make the relative graphs absolute?

[x6gas]

Hmmm interesting. Haven't I seen peak efficiency figures quoted which effectively give a reference to make the relative graphs absolute?

No I haven't, apparently. 77% peak QE seems to be oft quoted on the web but a brief search doesn't reveal an official Sony figure.

The diagonals are quoted, though, so Olly can let Pythagoras rest in peace...

[AlistairHowie]

You're correct on both counts! Sony are silent on QE ... which seems strange given how sensitive their chips are.

It's a minor point re diagonal size which is on spec sheets and is of course easily figured, but I do agree with Olly that Atik and others should be quoting it.

[x6gas]

Wait a second... something strange going on here. FLO delivered my 490ex today and yet it is clear outside. OK, there is a moon, but even so! Don't think I've ever managed first light on anything - even the smallest adapter - on the day it was delivered before!