DAC Background Calibration Using Single-Bin Error Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Integrated circuits in data converters suffer from imperfections that cause noise and spurs, degrading performance and affecting other signal chain components, necessitating effective calibration methods to minimize errors.
Innovation Solution
A background calibration technique for digital-to-analog converters (DACs) using redundant DAC cells to generate reference and calibration tones with opposite polarities, allowing for efficient error detection and minimization in a single frequency bin, without disrupting normal operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration methods are used to remove noise and spurs, then measurement precision improves, but processing overhead and system complexity increase
Solution Approach 1:
The calibration stimulus is segmented into multiple frequency components, with each component targeting specific error types (static vs. dynamic errors). The redundant DAC cells are divided into subsets that receive different weighted versions of the calibration stimulus, allowing parallel calibration of multiple error sources simultaneously without increasing overall system complexity
Solution Approach 2:
The same redundant DAC cells serve multiple functions: they act as both calibration devices for error measurement and as functional DAC cells for normal operation. The calibration stimulus generator and error detector are integrated into the existing DAC architecture, allowing the system to perform both calibration and data conversion functions without adding separate dedicated calibration hardware
2Productivity
If background calibration is performed during normal operation, then productivity is maintained, but measurement precision may be compromised due to signal interference
Solution Approach 1:
The calibration stimulus is applied periodically at specific frequency bins that are distinct from the normal data signal frequencies. This periodic application allows the error detector to distinguish calibration-induced error tones from normal operational signals, maintaining measurement precision while enabling continuous productivity during background calibration
Solution Approach 2:
The redundant DAC cells act as intermediaries that receive the calibration stimulus and generate error tones that can be detected without interfering with the main data signal path. The error detector uses these intermediary error tones to measure DAC cell mismatches while the main DAC continues normal data conversion operations
3Manufacturing precision
If redundant DAC cells are used for calibration, then manufacturing precision improves, but device complexity increases
Solution Approach 1:
The DAC system uses its own redundant cells to perform self-calibration, measuring and correcting its own errors without requiring external calibration equipment. The calibration process is automated through digital signal processing that identifies error tones and adjusts DAC cell weights accordingly, improving manufacturing precision while the redundancy is already present in the design
Data Source
AI summary
Techniques that enable calibration of digital-to-analog Converters (DACs) with minimal processing overhead. A single frequency bin can be used to calibrate errors between bits. A low frequency feedback path can be included into a low frequency low power ADC to determine the error signal that exists in the calibration bin. The bits are calibrated when this error signal is minimized. The calibration techniques described provide an extremely efficient and optimal calibration at the DAC output of both static and dynamic errors.