[fitsbits] Appropriate BUNIT for ADUs vs electrons?

[Paul Hirst]

It has to be said, GAIN is a somewhat general term to be used for this, 
there are typically many amplifiers in the system, each having a gain.

I know of systems that have used DEPERDN (Detector electrons per data 
numer) or similar to represent this number.

Cutting more back to the original question - is there any support in the 
fitsbits community for the thought that we should be using "counts" when 
we have the data in electrons - on the grounds that counts implies 
you're counting physical things in some sense (as opposed to ADUs) - ie 
it's something which you shouldn't have to apply a scale factor to if 
you're going to do Poisson statistics on it, for example.

Cheers,
  Paul.



On 03/04/2011 07:56 AM, Steve Allen wrote:
> On Fri 2011-03-04T13:27:10 -0300, James Turner hath writ:
>> What do other observatories do?
>
> Ponder the same question, but for reasons noted at the end it's a bit
> moot for us.
>
> As you note, one common way at the telescope and during intial
> reduction is to leave the array values as DN/adu/count and to include
> the GAIN keyword with a value according to the convention used in
> various IRAF procedures.  If GAIN is reset to 1 when it is applied
> then the conventional IRAF meaning is retained.  This is all, of
> course, dependent on keyword conventions for images from CCDs which
> are not registered with the IAUFWG.
>
> Keck data acquisition systems generally do not include a GAIN keyword.
> Lick data acquisition systems have GAIN with entirely different meaning.
>
> --
> Steve Allen<sla at ucolick.org>                 WGS-84 (GPS)
> UCO/Lick Observatory--ISB   Natural Sciences II, Room 165    Lat  +36.99855
> 1156 High Street            Voice: +1 831 459 3046           Lng -122.06015
> Santa Cruz, CA 95064        http://www.ucolick.org/~sla/     Hgt +250 m
>
> _______________________________________________
> fitsbits mailing list
> fitsbits at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/fitsbits