which best CCD ?

[FranckiM06]

Hello,

I would like to buy a CCD around 2000,00 Euros, but I don't know which one to choice between ATIK 314L+ & Starlight SXVR-H9 !

Now, if some one use already the SXVR-H9 CCD it would be nice to tell me about it.

Thank you in advance for more information about it.

Franck

[NickH]

Both are excellent CCD's, as they both use the Sony ICX285 chipset. The issues are

1: The Atik 314L+ costs less, the cooling is now set point, and two stage (fan and peltier), and it comes with out of the box software which a 3 year old could use. If you're not imaging first night with it, then seriously give up...it's that good and that easy to use.

2: The noise floor on both is very low, some say the H9 slightly better due to case/circuit design. I have the 314L and can run sub frames of 15 minutes with absolutely no requirement to do darks, I suspect the H9 is the same at least. This chipset is almost legendary for low noise.

3: H9 has the option of the SXV guide head linking in to it. Nice solution.. The 314L can run in it's own software with another Atik camera (16IC for example) as a guider. I use PHD with the Meade DSI-C as a guide camera, and have never ever had a problem with it finding guide stars

So...if you are not cost constrained, both are superb. The H9, you WILL have to buy extra software to get any real sense out of it (Maxim DL, Astroart or Nebulosity), but either of these superb cameras will do you proud

[Ben Ritchie]

The SXV guide head for the -H9 is nice, very light and keeps cables to a minimum. But they're both really nice CCDs, you won't be disappointed by either.

I would probably work out a price for Atik 314L + Atik 16ic (guider), then one for SXV-H9 + SXV guidehead + AstroArt, and buy whatever was cheaper

[NickH]

16IC as a guider, hmm...DSI-C or QYH5 in PHD will do the trick for less money probably..

but Ben's spot on.. it's all down to cost, as both of these are superb performers

[ollypenrice]

I use four Atik cameras here, a 16HR (now 314L) and the larger 4000. If you can stretch to the 4000 the big chip is a glory, it really is. SO many targets! On the other hand, as the others have said, the Sony 285 chip is a peach. No darks and really no big need for flats in most cases, or so I find. This is a controversial point but one of my guests, whose images are world class, feels the same.

I use the 16ics for guiding and as I said in my piece in Astronomy Now this month, my excuse is the bombproof certainty of getting a star fast and holding it through the odd bad patch. It is a bit of a luxury solution but if you ain't guiding you ain't imaging!

I have no doubt that the Starlight SXVH9 is also a cracker but the Atik software is not only great but getting ever better. The others have all hit the nail on the head in their replies, I would say. Just adding my penn'orth for good measure.

There are an assortment of large format colour cameras which look tempting on paper but I have never been at all impressed by those I have seen in action here. I feel you have the right shortlist.

Olly

[Ben Ritchie]

16IC as a guider, hmm...DSI-C or QYH5 in PHD will do the trick for less money probably..

16ic was the nearest direct equivalent to the SXV guider in sensitivity etc. that I could think of, although it's obviously not like-for-like (the 16ic is cooled, but also requires a power supply ... so good and bad).

The DSI-C and QHY5 are both good choices and significantly cheaper, but give away some sensitivity too.

[Aza]

As others have said you cant go wrong with the 285 chip.

Recently I bought the Atik 314L+ and I love it.

I have never used darks or flats with the camera.

[Milamber]

I am *so* tempted to jump in here since I could be seen as being biased... but I will anyway

I loved the idea of the dedicated SX guider - I always have - but that's *all* it is, a guider. Unless you cool it of course but the mod is not that practical and poufs your warranty. (No, I won't!). Once you get past purely physical issues (like which camera will fit best in a Hyperstar (SX) or which will fit best the other end of a fork (Atik)), it really is a just question of price and how easy to use they are straight out of the box ... of course there's also the extras (like software) you need to buy as well but there's a fair chance you have stuff that will drive the SX cameras already.

There are many images from both cameras out there - it's a subjective thing and advice is almost certain to be a little partisan, be prepared for a big debate here!

Arthur

[NickH]

It may also be worth readind Dr Craig Stark's report on th 314L in Astronomy Technology Today. The guy who wrote PHD Guide thinks it's one of the best cameras he's ever tested

[FranckiM06]

Hello every body,

First, I want to said to you all thank you for everything and all yours informations. It's great to get lot information about these CCD !

So, after to think and to see your answer I will buy the 314L+ because I think the software is very easy to use like it is write on the top.

So, thank you one more time.

Franck

[narrowbandpaul]

the sony 285 is a great sensor.

low PRNU, DSNU and dark current.

these 3 things coupled together make darks and flats unneccessary....well flats not required from a sensor point of view. If you have vignetting (probably wont over such a small sensor) then you will need flats.

Finger Lakes have just started putting the 285 in their microline package. I am not sure how much they cost, but its probably very realisable. The ML8300 8.3MP is a snip at $3500. and this is a great camera.....

the import tax might kill the price, but that aint the fault of FLI....

anyway the 285 will serve you well.

so would the KAF8300 sensor.

paul

[NickH]

yes, good call Paul, I think for a wide format, the FLI's look impressive too..

[FranckiM06]

Hello Paul & Nick,

Also, I look this moment the ATIK 4000 M with the sensor 4022 wide : 15x15 !

The price is around 2800,00 Euros.

What do you think about this one too ?

Franck

[narrowbandpaul]

also a good sensor.

its KAI4022 so it is kodak, has higher PRNU DSNU and dark current

[steve51450]

i have all 3 sensors and out of the 3 i would choose the 8300 chip everytime. that said you will need a shortish focal length refractor to utilize its pixel size & count to its full extent

then again if we are on a tighter budget than that the 314l is a fabulous camera and you will not need to buy the + version as set point cooling is not needed on that chip a standard 314l was on ebay yesterday for £700 and did not sell so i would email him and save some dosh

steve

[RobH]

285 chips can be found in Atik 16HR's and Artemis 285's also.

Both great cameras. The only real difference between them and the 314 is the shape and the fact they are USB1, so download times are slower at about 15 seconds.

No set point cooling, but that doesn't matter with this chip as has been said before.

Flats though, you need whatever camera you use, as they aren't anything to do with the camera, but are a record of the optical train, and there will always be some pesky dust bunny to remove somewhere!

To apply flats correctly you will also need to take a set of bias frames, but I only take one every 6 months or so, as the camera only slowly changes over time.

Cheers

Rob

[narrowbandpaul]

rob, flats remove fixed pattern noise which is the variation in charge collection efficiency of the pixels. So definately is a property of the sensor

all flats do is remove fixed pattern noise.

vignetting and dust are types of fixed pattern noise

despite darks not being required, set point cooling is a nice feature, especially if you want to make dark transfer curves...

paul

[steve51450]

paul

i only take bias frames with that chip as i have found darks unnecessary and that was with both the 16hr and the 314L, the same cannot be said for the kodak chips

i think what i am trying to say is the 285 is a very easy to live with low maintenance chip that is fitted to a wide variety of ccd cameras by many suppliers that will give great results in almost anyones hand

and to sum it up its only flaw that i can find is i wish it was 4 times larger

[NickH]

Paul, I think you have your flats and bias mixed up..(how is a dust bunny fixed pattern noise?, they move with changes in the sensor position/slip etc)...I guess your terminology may have some technical merit, but to all intents and purposes, most people treat flats as ways to remove optical train defects..

.Rob is right (and flats should be taken at all times I find). Darks...with the 314L I have almost never used them, no need..

I also have an Atik 4021 (4000) in the observatory project in Spain Excellent wide field camera, but not as sensitive as the 314L and also noisier.. still, a joy to use as it uses the same software... I had it for review some time back and was imaging the Horsehead with all the faint streamers first night...

314L is a no brainer...(or the H9 with some decent software) though... and personally I only sold my ART285 to get the 314L (faster DL time makes a difference when wide field imaging the Sun), and the form factor (I have a C11 and am mulling at some stage a Hyperstar)

[NickH]

Webcam Astronomy Dar\k Frames

:-)

[narrowbandpaul]

i dont have it mixed up Nick.

the dust bunny doesnt move around so it is a form of fixed pattern noise, hence it will flat field out.

how would a bias work? it is a zero length exposure to try and capture no light.....?

my terminology is the exact same as used by CCD experts.

optical train defects are fixed pattern noise sources, as is sensor FPN caused by charge collection efficiency variations between pixels......

[RobH]

rob, flats remove fixed pattern noise which is the variation in charge collection efficiency of the pixels. So definately is a property of the sensor

all flats do is remove fixed pattern noise.

vignetting and dust are types of fixed pattern noise

paul

I can take bias and darks without the camera attached to the optical train.

I can't take flats without the optical train though. Therefore flats are directly related to the optical train, not the sensor.

Anything else is just different terminology for the same thing.

Cheers

Rob

[narrowbandpaul]

i can assure you that flats remove FPN associated with the sensor.

it certainly removes optical train effects, but if you have a perfect even illumination but high pixel response non uniformity you still need flats, which could be taken with no lens to focus the light.

paul

[NickH]

????

Paul...

I too know some pretty good CCD people...try the guy who co developed the Mt Palomar CCD's)

Your terminology in this domain is not commonplace. Whilst it may be a function of the CCD to flat field it, what Rob, I and almost everyone else here is referring to is aberrations in the optical train. This can be dust on the filters (which does change as you change filters/wheels, remove camera to change wheels etc) dust on the sensor itself (which would match your fixed pattrn analogy), dust on the sensor window cover (this can/doeas and WILL change, as you move the camera from scope to scope etc)

I think it's just a case of crossed wires here, we're talking about fundamentally different things, probably with different terminology, but for the purposes of 99% of the people here, a flat field to which Rob and I refer, is specific to the optical train. Flat map fielding on stars (which I was doing on the Hale 200" with a team from Oxford University in May this year, is closer to what you are referring to, and not commonplace in the amateur domain...whereas this is

Basic CCD image processing

I guess we're just coming to the same conclusion from different angles. yes, variations on the CCD light collection area etc, do need to be "flattened" out, as does the bias read noise.

No harm done..

[NickH]

And dust bunnies do move...night to night when you are removing cameras :-)

(which is why I tend not to remove mine...)