Noise-Shaping SAR ADC Calibration for Stable Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Noise-shaping SAR ADCs face variations in noise transfer function due to coefficient variations from capacitor matching and amplifier gain variations, leading to reduced noise-shaping effectiveness, accuracy, and input bandwidth limitations.
Innovation Solution
Implementing a technique for testing and calibrating noise-shaping circuitry within SAR ADCs, using programmable/trimmable circuit components and modifications to digital logic to selectively skip conversion processes and utilize VREFH/VREFL switches during testing/calibration, ensuring accurate coefficient calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If noise-shaping circuitry is implemented in SAR ADC to improve SNR, then signal-to-noise ratio is improved, but coefficient variations from capacitor matching and amplifier gain variation reduce noise-shaping effectiveness
Solution Approach 1:
The patent implements trimming circuitry that allows adjustment of the noise-shaping coefficient to compensate for variations caused by capacitor matching and amplifier gain variations. By changing the coefficient parameter through trimming, the system maintains optimal noise-shaping performance despite manufacturing tolerances.
Solution Approach 2:
The patent incorporates feedback mechanisms where the actual coefficient values are measured and used to adjust the trimming controls. This closed-loop approach ensures that the noise-shaping circuitry maintains effectiveness by continuously compensating for parameter variations.
2Measurement precision
If noise-shaping coefficients are trimmed to improve accuracy, then ENOB is improved, but additional trimming circuitry and controls increase device complexity
Solution Approach 1:
The patent implements self-trimming mechanisms where the ADC automatically adjusts its own noise-shaping coefficients without requiring external intervention. The trimming circuitry is integrated into the ADC structure, allowing it to self-calibrate and maintain optimal performance.
Solution Approach 2:
The patent combines the trimming controls and circuitry with the existing ADC structure, integrating the coefficient adjustment functionality into the noise-shaping circuitry itself. This merging approach minimizes additional complexity by reusing existing circuit elements where possible.
3Measurement precision
If oversampling is used to reduce comparator and pre-amplifier noise, then signal-to-noise ratio is improved, but basic oversampling only gains 3 dB per factor of 2 increase in oversampling ratio
Solution Approach 1:
The patent implements dynamic noise-shaping that adapts to different operating conditions and input signal characteristics. By dynamically adjusting the noise-shaping coefficient and adapting the filtering characteristics, the system achieves better than 3 dB per factor of 2 improvement in SNR.
Solution Approach 2:
The patent combines multiple noise-reduction techniques including noise-shaping, oversampling, and digital filtering into a composite approach. This combination of methods achieves superior noise reduction performance compared to basic oversampling alone.
Data Source
AI summary
Testing of the noise-shaping circuitry within a successive approximation register (“SAR”) analog-to-digital converter (“ADC”) (“SAR ADC”) to ensure it will function as expected, while also providing a method for calibrating the coefficients of the noise-shaping circuitry. Programmable/trimmable circuit component(s) can be used to calibrate the coefficient(s) of the SAR ADC. Digital logic within the SAR engine enables it to selectively skip portions of the ADC conversion process and to use voltage references rather than an analog voltage input signal in sample mode during such test/calibration modes.