Bipolar FIR DAC Layout Using Opposite-Polarity Current Sources
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital-to-analog converters (DACs) with finite impulse response (FIR) filters face challenges in providing a bipolar output without increasing noise, circuit area, and power consumption, particularly in achieving well-matched current sources and stable offset references.
Innovation Solution
The DAC circuit divides the FIR filter into stages with mutually exclusive sets of positive and negative polarity current sources, ensuring the midpoint of the current output is zero without requiring additional current sources or switching elements, thereby achieving a bipolar output without increased noise or power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a fixed current source of opposite polarity is added to set the midpoint of the output near zero current, then a bipolar output is achieved, but noise is added to the output without contributing any signal margin
Solution Approach 1:
The current sources are divided into two separate sets: positive polarity current sources and negative polarity current sources. Each set is controlled by mutually exclusive tap weights, eliminating the need for a fixed opposite-polarity current source and thereby reducing output noise while maintaining bipolar output capability.
Solution Approach 2:
Instead of adding a fixed opposite-polarity current source to create bipolar output, the invention inverts the approach by using mutually exclusive positive and negative current sources that are selectively activated. This inversion eliminates the noise-contributing fixed current source while achieving the same bipolar output effect.
2Reliability
If pairs of current sources are provided for each tap to improve signal-to-noise performance and hold output capacitance constant, then bipolar output with improved performance is achieved, but additional circuit area and power are required
Solution Approach 1:
The circuit uses dynamic switching of current sources based on the digital input bit-stream. The shift register dynamically selects which current sources (positive or negative) are activated at each clock cycle, allowing the same physical current source to serve multiple purposes over time, thereby reducing the total number of current sources needed compared to static dual-polarity designs.
Solution Approach 2:
Each current source can serve multiple functions depending on the digital input state. The same current source infrastructure is used for both positive and negative output polarities by controlling which set is activated, making the current sources universal rather than dedicated to a single polarity, thus reducing overall circuit area.
3Manufacturing precision
If current sources are well-matched to achieve accurate bipolar output, then output accuracy is improved, but matching requirements among individual taps increase circuit complexity
Solution Approach 1:
The current sources are segmented into two distinct groups (positive and negative polarity sets) that are mutually exclusive. This segmentation reduces the matching requirements because each group can be independently designed and matched, rather than requiring all current sources across both polarities to be perfectly matched, thereby reducing overall circuit complexity.
Data Source
AI summary
A digital-to-analog converter (DAC) having filter sections with differing polarity provides a low-noise, low area bipolar output solution in delta-sigma modulator based DACs. A shift register receives an input bit-stream and provides a series of tap outputs that are used to control application of a number of current sources to output summing nodes. The current sources are divided into mutually-exclusive sets of positive polarity and negative polarity current sources, which are not necessarily contiguous. In one embodiment, half of one of the sets of current sources precedes the other set of current sources, and the other half of the divided set of current sources provides the final set of output taps. The number of current sources in each set may be equal, so that the midpoint of the output corresponds to zero current.