MLC NAND Flash Read Method Voltage Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

NAND flash memory devices face read errors due to shifting cell threshold voltages over time, which complicates accurate data retrieval in Multi-Level Cell (MLC) storage systems.

Innovation Solution

A read method that dynamically adjusts read voltages based on the retention characteristic of MLCs by using multiple read commands (A, B, and C) with corresponding read voltage groups to compensate for voltage shifts, allowing for accurate data reading even when error correction becomes difficult.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple read commands with different read voltages are used to compensate for threshold voltage shifts, then data reading accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvedata reading accuracyVSAvoidread operation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic read voltage adjustment by selecting from multiple pre-defined read voltage groups (first, second, third read voltage groups) based on the retention characteristic of the MLC. The controller dynamically determines which voltage group to use by evaluating whether error correction is difficult under current threshold voltage conditions, thereby adapting the read operation to changing storage conditions without requiring complex real-time voltage tuning circuits

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the read voltage parameter by providing multiple discrete read voltage groups corresponding to different retention states. Instead of using a single fixed read voltage, the system switches between different voltage levels (first read voltage, second read voltage, third read voltage) depending on how much the threshold voltage has shifted due to retention, thus resolving the contradiction between accuracy and complexity through parameter discretization

Inventive Principle:
Principle #35Parameter changes

2Reliability

If read voltage is adjusted to compensate for threshold voltage shift, then read error rate is reduced, but operation time increases

Engineering Contradiction:
Improveread error rateVSAvoidread operation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-defining multiple read voltage groups and their corresponding threshold voltage ranges before actual read operations occur. The controller has advance knowledge of the retention characteristics and pre-prepared voltage compensation strategies, allowing it to quickly determine the appropriate voltage group without performing complex real-time analysis, thus reducing the time penalty associated with voltage adjustment

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback through the error correction difficulty evaluation mechanism. After an initial read attempt, the system evaluates whether error correction is difficult based on the read data quality, and if so, switches to a different read voltage group. This feedback loop allows the system to adapt to actual read conditions and minimize unnecessary voltage switches, thereby reducing operational time while maintaining low error rates

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS7751241B2Read method of memory device
Publication Date: 2010.07.06 SK HYNIX INC
  • US7751241B2 patent drawing
  • US7751241B2 patent drawing
  • US7751241B2 patent drawing

AI summary

A read method of a memory device including a MLC includes the steps of performing a data read operation according to a first read command; determining whether error correction of the read data is possible; if, as a result of the determination, error correction is difficult, performing a data read operation according to a second read command; determining whether error correction of read data is possible according to the second read command; and if, as a result of the determination, error correction is difficult, performing a data read operation according to a Nth (N≧3, N is an integer) read command.