Measured gain not matching published value

jmgoldba

The gain I measure using flats from an ASI6200MM, following your suggested procedure here, doesn't come close to matching your published value for the camera. E.g. at Gain 100 (10dB) I measure a value close to 10x greater than the 0.27 e-/ADU that's published. Also, there seems to be a correlation between your published FW(e-) values and e-/ADU gain, such that GAIN(e-/ADU) = FW(e-)/(2¹⁶ - 1).

So my questions are:

How could my measured values be so far off?
How are the published FW(e-) values determined, i.e. are they simply derived from the measured gain or do they come from elsewhere?
Would you provide (say on a cloud drive) a pair of dark corrected flats from a 6200MM camera that provide a measured gain that matches your published value?

Jesse

Tech@ZWO

jmgoldba Could you share your results and fits files with us? thanks.

jmgoldba

Tech@ZWO

crop.zip

184kB

Attached is the cropped centers of a couple flats. The gain I calculate is about 3.1 e-/ADU which isn't close to the 0.8 reported for EGAIN in the FITS header. Note that I obtain the same 3.1 value if I use the whole image and/or I subtract 500 ADU from each image (aka OFFSET x 10) before calculating the gain.

Thanks,
Jesse

jmgoldba

Tech@ZWO
For another example I put a couple of full size binx2 bias frames on Dropbox. When I calculate the read noise using EGAIN in the header (which matches your spec for the GAIN at Gain = 100) I get 0.77 e- which is about half the 1.5 e- (at binx1) that you spec. Why's that?

Jesse

jmgoldba

[crickets] I just measured another set flats and got a gain of about 1 when EGAIN in the header was 0.2. Am I just pissing in the wind with these posts?