CCDciel

This is a special tool to analyse automatically your camera read noise, sensor gain, well capacity and linearity. You will just need a flat panel or substitute in front of your telescope and a cap to cover your telescope to make darks. Analysing will go fully automatic. Halfway the process the tool will ask you to remove the flat panel and cap the telescope for darks.

It is important that CCDCiel has access to your camera gain setting in menu Edit, Preferences, tab Camera and check mark option “Manage gain and Offset if available”. After using the tool you can uncheck it.

The recorded data values can be copied to a spreadsheet/word-processor.

In the main windows you can set the gains to test. If the camera allow to set the offset it is very important you set a value as low as possible that not result in any truncation for the range of gains to test.
The linearity test is optional and can be set for one gain (in screenshot at camera gain 120) .
The dark current test is also optional and can be set for one gain ( in screenshot at camera gain 120). To increase the accuracy your could choose to repeat the number of tests from one to three.
At the bottom of the menu there is a setting (400 sec) for the duration of the dark current test. For older cameras you can set it lower at 100 or 200 seconds. For modern cameras with a very low dark current it could be required to increase it to 1000 seconds. The dark current test will abort is the difference recorderd is two or less adu.

Actual gain in e-/adu:

Read noise for each gain. One line is for the pure read noise. The second line is for the combination of read noise and random telegraph noise (RTN), also called salt & pepper noise. The read noise is calculated as follows

read noise including RTN := stdev(bias1-bias2)/sqrt(2)

read noise excluding RTN := stdev2(bias1-bias2)/sqrt(2) where stdev2 is using sigma clipping to remove hot pixels.

Full well capacity:

Charge persistence (cmos image lag). The bars indicate the median dark pixel values. This dark series is taken directly after the last flat. If the first bar is higher then the second bar (both are for the same camera gain) then there is some minor charge persistent. So there is some charge left from the previous flat image. The height & values of the other bars can be influenced by the gain setting.

Dark current. The red line indicates the total measurable noise including the random telegraph noise. For some modern cameras it could be required to increase the default dark exposure from 400 up to 1000 seconds. The dark current test will take two dark so the duration will double the exposure time.

Camera info from the camera driver:

Note: The RTN, Telegraph Noise is measured in the difference of two darks. It is defined as all pixels with a value above 3.0 sigma of the the noise. For a clean Gaussian noise only 0.27% of the pixels exceed this value. Any more then 0.27% is defined as RTN. The noise sigma value is determined using a sigma clip routine.