Bit-Adjacency Switched CDAC for Low-Distortion Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Serial charge-redistribution cyclic digital-to-analog converters (CDACs) suffer from distortion due to capacitor mismatch errors and charge injection errors, with existing solutions either arbitrarily swapping capacitor configurations or performing cumulative swap decisions that slow down conversion and increase complexity.

Innovation Solution

A CDAC design that selectively connects first and second capacitors in two configurations based on adjacent bits of the input word, using a decision rule to determine capacitor roles for each bit and swapping capacitors only when necessary, thereby reducing distortion and simplifying the architecture.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If capacitor configurations are swapped based on cumulative calculation from MSB to LSB, then distortion is reduced, but conversion speed decreases and complexity increases

Engineering Contradiction:
Improvedistortion reductionVSAvoidconversion speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent inverts the conventional MSB-to-LSB processing order by implementing LSB-to-MSB conversion. This inversion allows the most significant bit decisions to be made last, enabling faster convergence and reducing the number of iterative calculations needed, thereby improving conversion speed while maintaining distortion reduction through selective capacitor swapping

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The patent performs preliminary actions by pre-determining capacitor configuration decisions based on LSB adjacency patterns before full conversion completes. The controller examines adjacent LSB values and pre-configures capacitor connections, reducing the computational burden during active conversion and improving overall conversion speed

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If capacitor configurations are swapped based on cumulative calculation, then distortion is reduced, but device complexity increases

Engineering Contradiction:
Improvedistortion reductionVSAvoidarchitecture complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies local quality by implementing capacitor swapping only when adjacent bit values indicate a need for correction, rather than uniformly swapping for all bits. The controller examines specific adjacent bit pairs and selectively configures capacitors based on local bit patterns, reducing overall system complexity while maintaining distortion reduction where needed

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent segments the conversion process into independent bit-processing stages that can operate in parallel or sequential fashion. By dividing the N-bit conversion into smaller computational units that process adjacent bit pairs independently, the patent reduces the complexity of cumulative calculations while maintaining accuracy through localized capacitor configuration decisions

Inventive Principle:
Principle #1Segmentation

3Device complexity

If arbitrary capacitor configuration is used, then architecture is simplified, but distortion increases

Engineering Contradiction:
Improvearchitecture simplicityVSAvoiddistortion
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent dynamically changes capacitor configuration parameters based on input bit patterns. The controller monitors adjacent bit values and adjusts capacitor connections in real-time, transitioning between different configuration states to minimize distortion. This parameter adaptation maintains architectural simplicity while achieving low distortion through data-driven configuration selection

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach reduces Integral Non-Linearity (INL) and Differential Non-Linearity (DNL) errors, improves conversion speed, and simplifies the CDAC architecture, achieving about 13.4% improvement in RMS INL and 99.6% improvement in RMS DNL errors compared to traditional methods.

Implementation Method 1

Charge redistribution in CDAC 700 is achieved by selective control of MOS-type switches SB0-SB3 and SBINIT

Methodology Applied
Scientific EffectCharge redistribution: Capacitance

Data Source

PatentUS7425913B2Bit-adjacency capacitor-switched DAC, method, driver and display device
Publication Date: 2008.09.16 SAMSUNG ELECTRONICS CO LTD
  • US7425913B2 patent drawing
  • US7425913B2 patent drawing
  • US7425913B2 patent drawing

AI summary

A cyclic digital to analog converter (CDAC) includes: first and second capacitors C1 and C2; an arrangement of switches selectively controllable to connect C1 and C2 in one of at least two charging-and-sharing configurations, the first configuration having the first capacitor C1 as a charging capacitor CCH and the second capacitor C2 as a sharing capacitor CSH, namely CCH=C1 and CSH=C2, and the second configuration having CCH=C2 and CSH=C1; and a controller to toggle the arrangement of switches between the first and second configurations based upon adjacent bits of a given input word.