Initial pedestal distortion?

Data from a single pixel on APS-3 #2

reset before pixel select

reset following pixel select

Voltage continues to rise for 500 to 700 micro seconds after reset switch is turned off this will cause a pedestal error for the first frame.

Estimate of pedestal error for a 4 ms frame read time:

select region for linear fit

Voltage out as a function of time after reset

Pedestal after the bump for 4 ms, in electrons

current in electrons per second from slope

Pedestal from slope with 4 ms integration

electrons

Pedestal from first two reads 4 ms apart

electron error on the first frame

What is stated here is for the first pixel of the frame. For later pixels both reads will be well into the linear region, so there should be no error in this case. To really compete this measurement the pixel select should be delayed many ms after reset to be sure that this is being interpreted correctly.

Conclusion 1st frame has been bad because of the delay from reset open to full voltage. The pixel selection does not affect the delay, because the delay also occurs for doing reset 100s of ms following pixel select.

Question why does the voltage continue to rise after the reset switch is turned off? Is this the clearing of carriers from the channel?