CMOS Camera Gain

Hi,

New to posting here so please bear with me.

Has there been any discussion as to the best CMOS camera gains for spectroscopy? I'm going to attach a few graphs to illustrate why I'm asking the question.

The first graph shows the QE of a camera. If you take the "red" cutoff at 800 mn, there is about a factor of 3 difference in the QEs between the min and max. The next picture shows the number of Dynamic Range stops and the read noise as a function of gain. For imaging, the gain of 100 gives the best combo of gain, noise, and dynamic range with about 13.5 stops, or a DR of 11k. 

However, do we need that kind of DR when doing spectroscopy? Do we even need it for Variable Star work?

When doing VS we look at a target and our reference stars are usually within 5 magnitudes of the target unless the target has a huge range, as some do. A factor of 100 is covered by 7 stops so a gain of 400 on this camera would be adequate.

The same argument can be made about spectroscopy work. The dynamic range required between the min/max portions of a spectrum, accounting for the sensor QE, can be handled by a gain higher than what is optimal for imaging.

OK -- I can't attach any images. Maybe because I just joined this group.

Question: when submitting a spectrum, what percentage of the well depth is normal either in terms of electrons or ADU?

Thanks.