Pipelined ADC Residue Dithering for Better Linearity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analog to digital converters (ADCs) face challenges in achieving linearity, as non-linearity issues lead to differential non-linearity errors, which are exacerbated by the limitations of current dithering techniques that require modifying the input signal or increasing the dynamic range, thereby complicating the design and performance of ADCs.
Innovation Solution
The proposed solution involves a method for an ADC stage that uses a residue forming stage with a sampling digital to analog converter (DAC) to apply and remove dither values without modifying the input signal to the analog to digital converter, allowing for improved linearity and noise performance by distributing the dither across multiple stages, thereby reducing the dynamic range impact and enhancing conversion accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dither is applied to the input signal of the ADC stage, then non-linearity errors are reduced, but the dynamic range increases and design complexity increases
Solution Approach 1:
The patent extracts the dither application from the traditional input signal path and relocates it to the residue forming stage. By taking out the dither function from the main signal path and implementing it separately through the sampling DAC, the solution reduces design complexity while maintaining the linearity improvement benefits.
Solution Approach 2:
The patent introduces the sampling DAC as an intermediary component that applies dither to the residue signal rather than directly to the input signal. This intermediary approach allows dither to be applied in a controlled manner that improves linearity without requiring significant changes to the overall ADC architecture or increasing dynamic range.
2Measurement precision
If dither is applied to improve linearity, then conversion accuracy improves, but dynamic range increases
Solution Approach 1:
The patent applies dither locally at the residue forming stage rather than globally to the entire input signal path. By applying dither only to the residue signal through the sampling DAC and not to the full-scale input signal, the solution improves conversion accuracy while limiting the impact on dynamic range requirements.
Solution Approach 2:
The patent applies partial dither action by introducing dither only at specific points in the conversion process (at the residue stage) rather than throughout the entire signal path. This partial application of dither is sufficient to improve linearity and conversion accuracy without requiring the full dynamic range expansion that would result from applying dither to the complete input signal.
Data Source
AI summary
A dither is an uncorrelated signal, usually pseudo-random noise injected into the input of an ADC such that a given input value of the wanted signal becomes spread over a plurality of codes. This reduces the effect of DNL and also smooths the integral non-linearity (INL) response of the ADC. The advantages of introducing dither could be obtained without having to perturb the signal input to the ADC. This avoids the introduction of additional components in the ADC. The dither can be applied to the components used to form a residue of the ADC stage within a pipelined converter. For example, a dither can be applied solely to a DAC part or different dithers can be applied to a ADC and DAC parts respectively. This allows greater flexibility of linearization of the ADC response and the formation of an analog residue by the DAC.