作者
Kadiri Pedda Reddy,Nikunj Srivastav,Kamlesh Kukreti,Alankrita Joshi
摘要
DACs, or digital-to-analog converters, are crucial devices that connect digital systems to the physical world. A variety of resistor-based DACs are examined in this review, such as R-2R ladder designs, binary-weighted networks (such as W2W current-mirror versions), and DACs that employ digital calibration in R-2R structures. The emphasis is on how well they perform at high resolution or speed, how much space they require, how complex they are, and how well they function in terms of linearity (measured as INL and DNL). We also examine how these DACs function in various CMOS technologies (such as 45 nm, 90 nm, and 180 nm) to observe how performance and power consumption are impacted by technology shrinkage. Since they only require two types of resistors and have a constant output resistance, R-2R ladders are simple to scale. However, unless they are calibrated, they may have linearity issues because of mismatches. [1], [5], [7] Binary-weighted DACs are fast and simple to scale, but as the number of bits increases, they require more power and space and require numerous components. By using matched MOSFET pairs, W-2W current-mirror designs minimize the number of sections, assist in matching, and minimize power and area. Digitally calibrated R-2R DACs allow for very good linearity without requiring extremely precise components by adding small, adjustable parts that help correct for mismatches. DACs usually need less space, utilize less power, and have better linearity for the same design as technology shrinks from 180 nm to 45 nm. For instance, power consumption decreases by about 30%, and the linearity error of a 4-bit DAC decreases from approximately 2.0 LSB in 180 nm to approximately 1.0 LSB in 45 nm. [1], [4] In conclusion, R-2R and binary DACs sacrifice linearity and space for quick and easy conversions. Accuracy and space efficiency can be enhanced by W-2W designs and calibration methodologies. Large DACs, even smaller technology sizes, and more sophisticated on-chip calibration techniques could all be explored in further research.