Design Ideas: August 15, 1996

To get the maximum performance from an optical instrument that uses a laser source and a CCD linear-array sensor (Figure 1a), an A/D converter should perform over the maximum possible range. You can maximize this range by stretching the upper reference of the flash A/D converter to coincide with the highest lit pixel in the CCD array. Most flash A/D converters allow you to control the extremities of the reference-resistor array. To compensate for optical-path variations, you can use the signal from the peak detector to control the light-source intensity.

For applications in which the shape of the light distribution is more important than the absolute light intensity, the A/D converter's reference must be constant during a single CCD exposure, but can change between exposures. In this case, you can use the scheme in Figure 1b as the light peak detector, if the light doesn't change significantly between two subsequent exposures.

C₁, C₂, and D₁ form a conventional peak detector. When C₁ connects to the diode through switch S_A1, C₂ supplies the output through SB₃. Likewise, when C₂ connects to the diode through SB₁, C₁ supplies the output through SA₃. SA₂ and SB₂ discharge the capacitors before SA₁ and SB₁ connect them to the diode.

Figure 2 illustrates the circuit's timing. The switches close when the corresponding signal is high. The ROG signal relates to the sensors; this signal is low during the exposure and high between two successive exposures. You can easily design a PLD to generate the drive signals for the analog switches. (DI#1909)