Multi-Bit DAC Data Weight Averaging to Prevent Tone Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-bit delta-sigma data converters experience noise due to mismatched current sources in the internal multi-bit DAC, leading to tone generation issues, especially when specific data are repeatedly inputted, which can decrease the signal-to-noise ratio (SNR) and affect system stability.
Innovation Solution
A dynamic elements matching circuit using data weight averaging (DWA) technique is employed, where the pointer position and circulation direction are dynamically adjusted based on comparisons of present and previous data and pointer positions to prevent tone generation by randomly selecting unit elements and dispersing mismatching errors across a wider bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a multi-bit delta-sigma modulator and internal multi-bit DAC are used, then system stability is improved and high SNR is achieved with lower oversampling ratio, but noise is generated due to mismatched current sources in the internal multi-bit DAC
Solution Approach 1:
The patent applies dynamic element matching (DEM) technique that dynamically switches between multiple current sources based on input data patterns. Instead of using fixed current sources, the system dynamically selects and combines different current sources to generate the required output current, thereby distributing and averaging the mismatch errors across multiple sources over time.
Solution Approach 2:
The patent changes the operational parameters of the DAC by varying the selection and combination of current sources based on the input data. The system adjusts which current sources are activated and how they are combined, transforming the static current source configuration into a dynamic one that adapts to different input conditions to minimize the impact of mismatch errors.
2Manufacturing precision
If conventional DWA technique is used to select unit elements, then mismatching errors are averaged, but tone generation occurs when specific data are repeatedly inputted
Solution Approach 1:
The patent introduces a modified DWA technique that uses periodic circulation through all unit elements. The system maintains a circulation counter that increments with each input sample and uses this counter to systematically cycle through different unit element combinations. This periodic action ensures that even when the same data pattern is repeated, the system eventually cycles through all possible unit element selections, preventing sustained tone generation while maintaining error averaging benefits.
Data Source
AI summary
A method for converting data includes matching dynamic elements by repeatedly selecting a portion of unit elements among N unit elements according to data and a circulation direction. An existence of a tone generation possibility is determined by comparing a present pointer position with a previous pointer position and by comparing present data with previous data. At least one of the present pointer position and the circulation direction is changed based on the existence of the tone generation possibility. The present pointer position and the present data are stored or the changed pointer position and the present data are stored. Unit elements are sequentially selected by the present data from the stored pointer position in the circulation direction or the changed circulation direction. The present pointer position is moved by the present data in the circulation direction or the changed circulation direction.