Programmable voltage-output, floating-gate digital-analog converter

In this paper, we describe implementations of binary-weighted and equal capacitance floating gate digital-analog converters (FGDACs). We improve the accuracy of charge amplifier DAC circuits and reduce their large element spread by utilizing epots in their implementations. We analyze these circuits and present experimental data from the chips that were fabricated in 0.5μm CMOS process available through MOSIS. The characterization results prove that by using the equal capacitance FGDAC structure, we can obtain very accurate results while eliminating the problems caused by the large element spread.