SAR ADC DAC Mismatch Calibration Without Conversion Interrupts

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

High-resolution analog-to-digital converters, such as two-stage pipelined SAR ADCs, require calibration to address DAC mismatch, amplifier offset, and gain errors, but existing calibration methods interrupt normal operation and do not account for time-varying environmental changes.

Innovation Solution

A method for digital-to-analog converter (DAC) mismatch calibration in a successive approximation register analog-to-digital converter (SAR ADC) that allows continuous operation by performing calibration in the background with minimal additional comparisons, using a control module to determine the presence and direction of DAC mismatch through additional bit comparisons and adjusting capacitance values accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If calibration is performed using known methods, then DAC mismatch, amplifier offset, and gain errors are corrected, but normal SAR ADC operation is interrupted

Engineering Contradiction:
Improveconversion precisionVSAvoidconversion throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The calibration process is performed in advance during idle periods or between normal conversion operations. The control module executes calibration sequences that determine calibration bits and adjust DAC capacitor values before normal operation resumes, ensuring precision is maintained without interrupting the main conversion throughput.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The calibration process is segmented into discrete steps that can be executed independently: generating calibration codes, performing comparisons, determining calibration bits, and adjusting DAC values. This segmentation allows calibration to be interleaved with normal operation without blocking the entire conversion pipeline.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If calibration is performed frequently to account for time-varying environmental changes, then conversion precision is maintained, but operational interruptions increase

Engineering Contradiction:
Improveconversion precisionVSAvoidoperational downtime
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The calibration process is designed to continue operating alongside normal conversion functions rather than stopping them. The control module manages both calibration and conversion operations continuously, using background calibration techniques that maintain precision while minimizing interruptions to the useful conversion action.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

Calibration is performed periodically at optimized intervals rather than continuously or on every conversion. The control module monitors environmental conditions and operational status to determine when calibration is necessary, executing it at periodic intervals that maintain precision without excessive operational downtime.

Inventive Principle:
Principle #19Periodic action

3Measurement precision

If traditional calibration methods are used, then DAC mismatch is corrected, but the calibration process is complex and requires multiple calibration steps

Engineering Contradiction:
ImproveDAC matchingVSAvoidcalibration process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The calibration method extracts and addresses the most critical mismatch errors through a simplified process. Instead of performing exhaustive calibration on all DAC components, the control module identifies and corrects the dominant mismatch sources using streamlined comparison and adjustment procedures, reducing overall calibration complexity while maintaining effectiveness.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The calibration process performs partial calibration focused on the most significant error sources rather than complete calibration of all parameters. By concentrating calibration efforts on the dominant mismatch components and using efficient comparison techniques, the process achieves sufficient precision with fewer steps than traditional comprehensive calibration methods.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3334049B1A method of digital-to-analog converter mismatch calibration in a successive approximation register analog-to-digital converter and a successive approximation register analog-to-digital converter
Publication Date: 2021.04.21 STICHTING IMEC NEDERLAND
  • EP3334049B1 patent drawingFigure 1
  • EP3334049B1 patent drawingFigure 2A
  • EP3334049B1 patent drawingFigure 2B

AI summary

A method of DAC mismatch calibration in a SAR ADC (600) comprising the steps of: determining a number of bits of an analog input signal (VIN); detecting if a binary code determined from the analog input signal (VIN) matches at least one trigger code; using at least one setting code to determine a calibration residue signal (V*RES) and a calibration bit (B*LSB); analyzing a least significant bit of the digital signal (COUT) and the calibration bit (B*LSB); determining an indication of a presence of DAC mismatch; and calibrating the DAC mismatch. As the determination of the calibration bit (B*LSB) requires only one additional comparison, when compared to the normal operation, the normal operation does not need to be interrupted. Therefore, the calibration can be done in the background and, as such, can be performed frequently thereby taking into account time-varying changes due to environmental effects.