From: K Michael Merrill <merrill@noao.edu>
Date: 5 May 2008 1:25:58 PM CDT
To: Mark Dickinson <med@noao.edu>
Cc: Buell Jannuzi <jannuzi@noao.edu>, Rob Swaters <swaters@astro.umd.edu>, Frank Valdes <valdes@noao.edu>, Tracy Huard <thuard@astro.umd.edu>, Michael Merrill <mmerrill@noao.edu>, Ron Probst <probst@noao.edu>
Subject: Re: NEWFIRM saturation

The current integration timing (1 digital averages) for NEWFIRM is approximately:

	reset | 0.10 s delay | starting read of row one | 0.70 s | end of first read of row 2048 | programmed delay to accomplish integration time
	      | 0.10 s delay | starting read of row one | 0.70 s | end of second read of row 2048 |

The current integration timing (4 digital averages) for NEWFIRM is approximately:

	reset | 0.10 s delay | starting read of row one | 1.32 s | end of first read of row 2048 | programmed delay to accomplish integration time
	      | 0.10 s delay | starting read of row one | 1.32 s | end of second read of row 2048 |

Since data are read row by row, so there is a ramp in photons detected that runs from a minimum a row 1 to a maximum at row 2048.

So at the very fastest for 4 digital averages:

	row 1 in the first read has 0.10 s of accumulated charge,
	row 2048 has 1.420 s of accumulated charge.
	row 1 in the second read has 1.520 s of accumulated charge,
	row 2048 has 2.840 s of accumulated charge.

Remember the first read is non-destructive and that only the difference between the two reads is recorded.

Michael