Interleaved ADC Filtering With Memory Preload for Chopping Settling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analog-to-digital converter (ADC) circuits experience reduced data throughput due to chopping techniques, which increase settling time and reduce effective data output rate, particularly in delta-sigma modulator-based systems with Sinc decimation filters, as a result of periodic chopping phases and the need for full settling of filter outputs.
Innovation Solution
The implementation of a method that preloads selectable memory circuits in the filter with reference data, allowing for the combination of sampled input data with preloaded reference data to generate filter outputs, effectively reducing the settling time and maintaining data throughput at unchopped rates by utilizing past states and preloading techniques across multiple stages of the filter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If chopping is used to reduce DC voltage bias, then voltage offset is cancelled, but settling time of the filter is increased
Solution Approach 1:
The patent applies preliminary action by preloading the filter memory circuits with reference data from previous chop phases before the current chop phase completes. This allows the filter to have data ready for immediate output generation, effectively overlapping the settling process with the chopping operation and reducing the apparent settling time without compromising offset cancellation performance
2Measurement precision
If chopping is used to eliminate DC bias, then offset voltage is reduced, but effective data throughput is reduced by factor N
Solution Approach 1:
The patent implements continuity of useful action by continuously generating filter outputs during the chopping process rather than waiting for complete settling. The filter memory circuits are continuously updated with new samples while simultaneously being read out to generate outputs, maintaining continuous productive operation throughout the chopping cycles
Solution Approach 2:
By preloading the filter memory with reference data from previous chop phases, the system has data prepared in advance for immediate use in output generation. This eliminates idle waiting time and maintains continuous data throughput equivalent to unchopped configurations
Data Source
AI summary
Techniques to increase a data throughput rate of a filter circuit by preloading selectable memory circuits of the filter circuit with reference data, sampling input data at an input of the filter circuit, combining the sampled input data with the preloaded reference data, and generating a filter output based on the combined sampled input data and preloaded reference data.


