Sigma-Delta Modulator Feedback Delays for Metastability Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sigma-delta modulators face challenges with metastability, which introduces timing variability in feedback due to the finite time required for quantization elements to determine output levels, leading to potential instability and reduced performance when addressing this with multiple quantizers in series.
Innovation Solution
Incorporating quantizers in the control loop to differentially delay feedback to successive integrators, thereby suppressing metastability errors without compromising the stability of the control loop, by ensuring consistent hold times for feedback signals and using additional quantizers strategically to reduce phase margin impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple DFFs are used in series to reduce metastability, then metastability is reduced, but the phase margin decreases and loop bandwidth must be reduced
Solution Approach 1:
The feedback path is segmented into multiple parallel quantizer channels (first quantizer path and second quantizer path) with different delay characteristics. Each path processes feedback signals with distinct timing, allowing the system to benefit from metastability reduction while maintaining phase margin through differential delay compensation.
Solution Approach 2:
A delay element is introduced as an intermediary component in the second quantizer path to match the delay of the first quantizer path. This intermediary delay element compensates for the inherent delay differences between parallel quantizer channels, ensuring consistent timing without requiring additional cascaded DFFs that would reduce phase margin.
2Reliability
If multiple DFFs are used in series to reduce metastability, then metastability is reduced, but loop bandwidth is reduced
Solution Approach 1:
The feedback path is segmented into multiple parallel quantizer channels (first quantizer path and second quantizer path) with different delay characteristics. Each path processes feedback signals with distinct timing, allowing the system to benefit from metastability reduction while maintaining phase margin through differential delay compensation.
Solution Approach 2:
A delay element is introduced as an intermediary component in the second quantizer path to match the delay of the first quantizer path. This intermediary delay element compensates for the inherent delay differences between parallel quantizer channels, ensuring consistent timing without requiring additional cascaded DFFs that would reduce phase margin.
3Reliability
If quantizers are added to the control loop with different delays, then metastability errors are suppressed, but device complexity increases
Solution Approach 1:
The feedback path is segmented into multiple parallel quantizer channels (first quantizer path and second quantizer path) with different delay characteristics. Each path processes feedback signals with distinct timing, allowing the system to benefit from metastability reduction while maintaining phase margin through differential delay compensation.
Solution Approach 2:
The system changes the timing parameter of feedback signals by introducing controlled delays in different quantizer paths. By adjusting the delay parameter in the second quantizer path to match the first, the system achieves metastability suppression while maintaining consistent hold times without adding excessive structural complexity.
Data Source
AI summary
Described herein is a ΣΔ modulator with improved metastability in which the control loop remains stable. In one embodiment, the ΣΔ modulator utilizes differently delayed feedback to successive integrators of the control loop to suppress metastability errors without compromising the stability of the control loop. This is accomplished by including one or more quantizers in the control loop. This technique may be applied to control loops of at least second order, i.e., having two or more integrator stages, where at least one feedback term after the first is non-zero.