Re: [APML] New SBIG Cameras

[Radu Corlan <rcorlan@pcnet.ro>]

On Wed, 9 Apr 2003 Aplanatic@aol.com wrote:

> In a message dated 4/9/2003 1:48:34 PM Eastern Standard Time, rcorlan@pcnet.ro writes:
> 
>  
> > OTOH, one should carefully consider the higher noise and Dark current of 
> > the KAI sensors, which would mean that you generally need an exposure 
> > roughy three times as long as on a KAF3200 to get the same SNR. At 650nm, 
> > the ratio is even larger, maybe 5-6times.
> > 
> > Radu
> 
> This is quite misleading.  In practice, photon shot noise from the sky
> background almost always dominates the total noise, not the dark current
> or readout noise.  In typical situations the exposure needed to achieve
> the same SNR will not change much at all.  There are exceptions.

Dave,

The 5-6 time ratio does not take dark current into account. It's the 
QE ratio (80% / 30%) times the noise ratio (30 e- / 12 e-).

sky level depends critically on f/ratio. Also on filtering. A low-noise 
sensor used without a filter on a fast scope will most likely be 
sky-limited. Not necessarily so in other cases.

As an example, i have some frames taken at f/4 under a LVM 4-5 sky. With 
the green filter (which also leaks some IR, but that's another story), i 
get a sky level of ~800 electrons with a 4-minute exposure. So the sky 
photon shot noise is 28 electrons. My camera is sky-limited (read noise 
~10e). But a KAI11000 camera would just be nearing sky-limiting, and the 
resulting noise of the frame would be 41 electrons. 

The signal would be lower by a factor of, say, 1.4, due to the lower QE. 
So the SNR would be 2.05 times lower; since we are sky-limited, we need to 
increase the exposure time by the square of that to obtain the same SNR, 
so we need a factor of 4 increase in exposure time, even though we are 
nearly sky-limited.

Now for the dark current

Assuming the temperature of the chip is -30C, the dark current is 3e/sec 
(from 800 e/sec at 25C). So the dark current is 720 electrons, and it's
corresponding shot noise is 26e. The noise level of the KAI camera rises 
to 48e. We factor in the loss of QE, and our SNR is now 2.3 time lower, so 
we need 5.3 times more exposure time. 

So, at -30C, f/4, with a green filter, under a so-so sky, to get the same 
SNR you would either take a 4-minute exposure with a KAF3200E, or a 21-minute
exposure with a KAI11000.

At f/8, the ratio would be even larger.

Radu

> 
> -Dave Rowe-
> 
> _______________________________________________
> Astro-Photo mailing list
> Astro-Photo@seds.org
> http://seds.org/mailman/listinfo/astro-photo
> 

-- 

The future isn't what it used to be. 

-------------
Radu Corlan       Snail Mail: Bucuresti sect. 1, 
rcorlan@pcnet.ro  str. Argentina nr. 28, Romania

   You can still escape the "Gates" of Hell!   
                 Use Linux!                       


_______________________________________________
Astro-Photo mailing list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo