Multichannel SAR ADC Noise Shaping With Quantization Error Hold
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Successive approximation type A/D converter circuits face challenges in increasing accuracy and supporting multichannel input while maintaining low power consumption, as hybrid type A/D converter circuits with delta sigma configurations do not currently enable both high accuracy and multichannel input.
Innovation Solution
A circuit device incorporating a charge redistribution type D/A converter circuit and quantization error hold circuits that perform noise-shaping A/D conversion, allowing for multichannel input by using quantization errors from previous conversions to enhance accuracy and reduce noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a hybrid type A/D converter circuit with delta sigma configuration is used to increase accuracy, then A/D conversion accuracy is improved, but multichannel input capability is lost
Solution Approach 1:
The invention divides the A/D converter into multiple independent conversion units, each capable of processing one channel. Each unit includes a successive approximation type A/D converter and a quantization error hold circuit. This segmentation allows each channel to be processed independently while maintaining the noise-shaping capability of the hybrid architecture, thereby achieving both high accuracy and multichannel input support.
Solution Approach 2:
The invention creates a universal A/D conversion architecture where each conversion unit can handle multiple functions: successive approximation conversion and quantization error holding. The quantization error hold circuit serves multiple purposes by storing errors from previous conversions and feeding them back to subsequent conversions, enabling noise-shaping across multiple channels simultaneously.
2Adaptability or versatility
If time division manner is used for multichannel input in successive approximation type A/D converter, then multichannel input is achieved, but A/D conversion accuracy remains limited
Solution Approach 1:
The invention introduces feedback through quantization error hold circuits that capture the quantization error from each A/D conversion and feed it back to the next conversion of the same channel. This feedback mechanism implements noise-shaping, which pushes quantization noise to higher frequencies, thereby improving the effective accuracy of the A/D conversion while maintaining multichannel input capability through time-division multiplexing.
3Measurement precision
If hybrid type A/D converter circuit is implemented for high accuracy, then quantization noise is reduced by noise shaping, but circuit complexity increases
Solution Approach 1:
The invention merges the successive approximation type A/D converter and the quantization error hold circuit into an integrated hybrid architecture. By combining these two functional blocks and sharing common resources such as the D/A converter and comparator across multiple channels through time-division multiplexing, the circuit achieves noise-shaping capability with reduced overall complexity compared to implementing separate hybrid converters for each channel.
Data Source
AI summary
A circuit device includes a selector to which first to n-th voltages are input, an A/D converter circuit to which output voltages of the selector are input as input voltages, and first to n-th quantization error hold circuits that hold charges corresponding to quantization errors in A/D conversion of the first to n-th voltages. The A/D converter circuit performs A/D conversion of an input voltage by a successive approximation operation using a charge redistribution type D/A converter circuit and performs k-th A/D conversion on an i-th voltage by using a charge held in an i-th quantization error hold circuit in (k−1)th A/D conversion of the i-th voltage to output A/D conversion result data DOUT in which the quantization error is noise-shaped.