Capacitor Rotation in Sigma-Delta ADCs for Gain Error Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional analog-to-digital converters (ADCs) face challenges in achieving high absolute accuracy and low DC offset due to component mismatch errors, particularly in instrumentation and measurement applications, where random and systematic errors in capacitors lead to gain errors that degrade conversion accuracy.
Innovation Solution
A component rotation method is employed, where pairs of physical, electrical component units are selected and rotatably connected to function as a single component, averaging out variations in component values and eliminating systematic and random errors by alternating their association with system parameters, thereby reducing gain errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional matched analog components are used in ADCs, then the device can achieve basic conversion function, but component mismatch errors cause gain errors that degrade accuracy
Solution Approach 1:
The patent applies dynamic element matching by periodically rotating the assignment of capacitor units between the first and second capacitors. This dynamic reconfiguration allows the system to average out component mismatch errors over time, achieving high accuracy without requiring perfectly matched static components. The rotation controller systematically varies which physical capacitor units are assigned to which functional roles, converting a static matching problem into a dynamic averaging process.
Solution Approach 2:
The patent implements periodic rotation of capacitor assignments at different frequencies to eliminate both random and systematic errors. By periodically exchanging the roles of capacitor units at carefully selected frequencies, the system averages out random mismatches and cancels systematic gradient errors. This periodic reconfiguration is the core mechanism that transforms component imperfections into canceling error patterns.
2Measurement precision
If static capacitor assignments are used in ADCs, then the circuit structure is simple, but systematic and random errors due to component mismatch cannot be eliminated
Solution Approach 1:
The patent uses dynamic reconfiguration of capacitor assignments through rotation controllers that periodically swap which physical capacitor units are assigned to which functional roles. This dynamic approach eliminates the need for perfectly matched static components by averaging out mismatches over time, achieving high gain accuracy without requiring overly complex static matching networks.
Solution Approach 2:
The system uses its own capacitor units to serve multiple functional roles through rotation. The same physical capacitor units are reassigned to different positions in the differential structure over time, allowing the hardware to self-correct its own mismatch errors through the rotation process rather than requiring external calibration or additional matching components.
3Measurement precision
If capacitor units are fixed in specific positions, then manufacturing is simpler, but gain errors persist due to fabrication variations
Solution Approach 1:
The patent implements dynamic rotation of capacitor assignments where the mapping between physical capacitor units and functional capacitor positions is continuously varied. This dynamic reassignment allows the system to average out fabrication variations inherent in fixed positions, achieving high gain accuracy without requiring complex manufacturing processes or perfectly matched components.
Solution Approach 2:
The system periodically rotates capacitor assignments at optimized frequencies to eliminate both random and systematic fabrication errors. This periodic reconfiguration transforms static manufacturing imperfections into time-averaged error cancellation, achieving high precision without requiring more complex manufacturing processes.
4Measurement precision
If no error correction is applied, then the ADC structure remains simple, but component mismatch causes degraded conversion accuracy
Solution Approach 1:
The patent employs dynamic element matching through rotation controllers that periodically reassign physical capacitor units to different functional positions. This dynamic approach provides error correction by averaging out component mismatch over time, achieving high accuracy without requiring complex static correction networks or additional calibration hardware.
Solution Approach 2:
The ADC system corrects its own gain errors through the rotation process without requiring external calibration equipment or additional correction circuits. The periodic reconfiguration of capacitor assignments causes the system to self-average out its own mismatch errors, providing built-in error correction with minimal additional complexity.
Data Source
AI summary
A method for removing component mismatch errors for a system parameter being set by a ratio of two or more physical, electrical components (“components”) of the same kind on an integrated circuit including providing an array of component units having the same component value, determining the actual component values of each component unit in the array, selecting component units based on the actual component values to form pairs of component units where the pairs have approximately the same total component values, ordering the component unit pairs, assigning alternate component unit pairs to be associated with each of the two or more components, rotating at a first frequency the assignment of the component unit pairs. At each rotation, the component unit pairs to be associated with each component are shifted so that each component unit pair is associated with a different one of the two or more components in turn.