bias fits vs data fits frames with cmos camera

[cjacobson]

I have started to use an ASI camera zwo 1071 with a sony chip. I produced bias, flats, darks and image frames. The bias frames are all read out by Pixinsight as an adu of 8000 when they are really 280. PI says this due to the AA6 fits data for the bias were not read out as unsigned. I do not exactly know what this means but why would everything but the bias be readable as expected? I checked the fits header for the bias with aa6 and it does not have the bzero or b scale. It does for the darks. This business is new to me. Their previous software acted appropriately.
The bias frames are 1 sec and the darks 7minutes the latter with an adu near 280 as well.
This is a 14 bit camera.
I have used a dslr 14 bit and not had this problem.
charles jacobson

[fabdev]

Hello, are you sure about the 8000? Usually when there's a signed/unsigned problem in supporting FITS the background is wrongly read as 32000. By the way: unsigned data does not exist in the FITS standard, if an image has a range 0..65535 (like your light images) then the ADUs must be shifted into the range -32k 32k and the "BZERO=32768" keyword is added to the header. If instead the range of the image is 0..32k (probably your bias images) then the FITS file contains real ADU values, "BZERO=32768" is not used and BZERO has an implicit value of 0.0 as specified by the FITS File Format 2.1b used by AA.
I guess that you could verify the correctness of your bias frames importing them into a reliable FITS viewer:
https://fits.gsfc.nasa.gov/fits_viewer.html

For example:
SaoImage DS9: https://sites.google.com/cfa.harvard.edu/saoimageds9
Nasa's FV: https://heasarc.gsfc.nasa.gov/docs/software/ftools/fv/
Online viewer: http://www.clearskyinstitute.com/fits/fits.html (select "Square root" for more contrast)

Greetings,
Fabio.

[cjacobson]

tha;nks for you lucid and quick reply. Very helpful. cjacobson