Distributed Feedback Sigma-Delta Modulator for Metastability Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sigma-delta modulators face challenges in achieving high-speed operation and low-power consumption without suffering from meta stability problems, particularly at high sampling frequencies where parasitic poles and delays can lead to errors and instability.
Innovation Solution
The implementation of a sigma-delta modulator with distributed feedback paths, each configured to provide a different gain value, utilizing multiple gain stages to decouple quantizer gain requirements from clock speed limitations, allowing for stable operation at high speeds or low powers by distributing gain across feedback paths and optimizing time delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single high-gain feedback path is used, then the quantizer gain requirement is met, but meta stability problems occur at high sampling frequencies
Solution Approach 1:
The single high-gain feedback path is segmented into multiple feedback paths with different gain values. The quantizer is divided into multiple gain stages (first gain stage, second gain stage) where each stage provides a portion of the total gain. This segmentation distributes the gain requirement across multiple paths, preventing meta stability while achieving the necessary quantizer gain.
Solution Approach 2:
Different feedback paths are assigned different gain characteristics tailored to their specific functions. The first feedback path provides a first gain value (e.g., gain of 2) while the second feedback path provides a second gain value (e.g., gain of 4), creating local quality variations that optimize both stability and precision in different parts of the system.
2Productivity
If the sampling frequency is increased for high-speed operation, then productivity improves, but parasitic poles and delays cause meta stability
Solution Approach 1:
The feedback system is segmented into multiple paths with different time delay characteristics. The first feedback path has a shorter time delay associated with it, while the second feedback path has a longer time delay. This segmentation allows the system to accommodate high sampling frequencies by distributing the delay burden across multiple paths.
Solution Approach 2:
The gain values of the feedback paths are specifically designed to compensate for time delay effects at high sampling frequencies. By adjusting the gain parameters (first gain value and second gain value) in relation to their respective time delays, the system maintains stability even when operating at high productivity levels.
3Use of energy by moving object
If power consumption is reduced for low-power operation, then energy efficiency improves, but signal integrity and stability deteriorate
Solution Approach 1:
The power consumption is segmented and distributed across multiple gain stages rather than concentrating it in a single high-gain stage. Each gain stage operates at lower power levels, and their combined effect achieves the necessary total gain. This segmentation enables low-power operation while maintaining signal integrity through distributed amplification.
Data Source
Figure 1~2a
Figure 2b~3a
Figure 3b
AI summary
A sigma-delta modulator (300) comprising a first filter stage (304); a second filter stage (306) in series with the first filter stage (304); a first feedback path (311) between the output of the second filter stage (306) and the input to the second filter stage (306), the first feedback (311) comprising a first gain stage (308, 308') such that the first feedback path (311) is configured to provide a first gain value; and a second feedback path (313) between the output of the second filter stage (306) and the input to the first filter stage (304), the second feedback path (313) comprising a second gain stage (309; 310') such that the second feedback path (313) is configured to provide a second gain value. The first gain value is different from the second gain value.