I too have had my eye on one of these sensors use in an astronomical
application. Other than the moderately shallow well capacity of 20.5Ke-, it
looked pretty decent spec-wise.
One of the reasons why I shot the IC1795 the way I did, using clear
luminance atop emission line data (which doesn’t alter the color balance btw,
unlike Halpha used in that capacity) was to see just how much of a problem the
stars would be in a 15 minute clear exposure.
Well some of them saturated but the antiblooming treated them nicely.
There were no ill effects resulting.
For the emission line data, I saw only a handfull of saturated stars and
then only on the really bright ones...
My conclusion is that well depth is way overrated...
I’ve observed that ‘conventional wisdom’ in the astro world is usually
nonsense propagated by the ignorant that are both ignorant of the technology and
ignorant of the depth of their own ignorance, and after using this camera for a
couple of months I lump the well capacity comments I have heard from these
self-appointed non-technical “experts'” as further examples of the second class
of ignorance.
The sensor is plug compatible with the newly upgraded KAI16000 that is
called the KAI16060 so even if this experiment was a “bust” (It was not),
there’s a nice pathway to switching to a more conventional sensor.
The camera features both a 12MHz and a 4MHz mode of operation and I used
the slower speed mode. It takes a while to download because FLI elected to use
the sensor in single output mode, so all pixels must be read-out via a single
port, unlike the PL39000 that uses two ports.
I think for general astro-market consumption, that was the better
choice because there’s no need to deal with differential offset level shifts and
their impact on calibration that I generally deal with using the PL39000 and my
habit of keeping a library of darks that I use for a year or longer.
In the PL39000 case, the ambient temperature in play has some impact on the
absolute offset levels for the two independent readout systems so I find
that to use my library darks I often need to apply a fixed offset to
one side versus the other to eliminate the seam. If I shoot darks on the night I
take my image data usually that is less of an issue. However I have developed a
good method for avoiding the differential offset issue by careful calibrations
and experience. Quite honestly I don’t like losing the imaging time by shooting
darks, because I typically use no fewer than 20 darks of a given exposure
duration and that can be a substantial amount of time to collect a good
set.
But the ignoramuses that typically populate the hobby would find such a
protocol unacceptable and so I think FLI made a wise decision to trade off
simplicity of use for the masses versus higher speed operation for people like
me that appreciate it and don’t mind a little extra work
being an interline sensor, flat fielding is simple using broadband filters
with the sky as an illumination source (preferred for a host of reasons): you
can open the mechanical shutter and use the electronic global snap-shutter
inherent in the interline architecture to great advantage: taking sub-100
millisecond exposures if necessary.
All in all I think this will be a very popular product if the pricing is
set to a good level and what I have seen is that it will be priced above the
KAI11002-based cameras and below those using the KAF16803.
The only thing I would like to see improved on this prototype camera is a
little better cooling. The sensor in mine has a lot of vertical lines in the
darks that are completely tamed by cooling. During the Dog Days of Summer when I
took this image the ambient temperatures were running high during the day, and
remaining high at night, so I could not quite reach –25C and could see that the
sensor would benefit from a colder operating temperature than the –20C I elected
to use. Now that we are getting into cooler weather this isn’t even a factor.
However using dithering and bad pixel repair algorithms I had no problems
managing the defects at the warmer temperatures. I intend to publish a full
characterization report when I can make the Photon Transfer analysis but at the
moment it is mounted on my telescope and I am using it for imaging when I get
the chance to image: I just returned this past Friday from three weeks in
Asia... I will be headed back just after the elections it appears.
