which best CCD ?

[richardc]

the sensor has a photoresponse non-uniformity that speaks to the different sensitivity of different pixels to a flat field illumination. the optical system is typically not capable of providing completely uniform illumination due to non-uniformities in the light intensity across the illuminated circle (hot centers are common) and due to things like dust motes and the like.

Take collectively this is a system level Fixed Pattern Noise. That includes all sources, be it sensor nonuniform response to vignetting etc...

Flats correct all these problems in a single operation.

You can use Photon transfer methods to quantify how well your flats are behaving by using a variant called a FLat Field Photon Transfer Curve (FFPTC).

in my talk on Photon transfer methods I gave at the 2009 winter star party in florida, I covered the FFPTC starting on page 16

http://www.narrowbandimaging.com/ptc_method_wsp2009_page.htm

additionally if you like to know the "how" of how flat fielding works to correct the response of an image I call your attention to this:

http://www.narrowbandimaging.com/mathematics_of_flat_fielding_page.htm

the last two pages contain very useful graphs: they show you how many electrons you need in your set of flats to successfully remove FPN as a function of average signal level in the image.

the more signal level in the image, the more total number of flats need to be combined together to remove the FPN. I show how to quantify that....

[NickH]

Richard

Bang on the money, thank you...and Paul, we were talking on crossed wires...glad we've arrived at the same destination

[narrowbandpaul]

indeed nick.....indeed