Memory Block Retry Pools for Neighbor Plane Disturb Errors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Semiconductor memory systems face issues with neighbor plane disturb errors during multiplane erase operations, leading to premature retirement of functional blocks due to misclassification as faulty, which negatively impacts production efficiency.

Innovation Solution

A fault assessment system that assigns potentially faulty memory blocks to a retry pool, performs a second fault assessment, and relinks functional blocks to form new metablocks, thereby reducing neighbor plane disturb errors and improving efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If multiplane erase pass operations are used to perform erase functions more quickly, then processing speed is improved, but neighbor plane disturb errors occur causing functional blocks to be misclassified as faulty

Engineering Contradiction:
Improveprocessing speedVSAvoidblock classification accuracy
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The patent applies preliminary action by performing a first fault assessment using multiplane erase pass operations to quickly identify potentially faulty blocks, then conducting a second fault assessment using single plane erase operations to confirm the actual faulty blocks before retirement. This two-stage approach allows fast initial screening followed by precise verification, resolving the contradiction between speed and accuracy in block classification.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If single plane erase operations are performed on each individual plane, then block classification accuracy is improved, but erase time increases

Engineering Contradiction:
Improveblock classification accuracyVSAvoiderase time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the fault assessment process into two distinct stages: a first fault assessment using multiplane erase pass operations for rapid screening, and a second fault assessment using single plane erase operations for precise verification. This segmentation allows the system to leverage the speed advantage of multiplane operations for initial screening while using single plane operations only when necessary for confirmation, thereby reducing overall erase time while maintaining high classification accuracy.

Inventive Principle:
Principle #1Segmentation

3Reliability

If functional blocks are retired due to neighbor plane disturb errors, then faulty blocks are removed from service, but production output efficiency decreases due to premature retirement of functional blocks

Engineering Contradiction:
Improvememory system reliabilityVSAvoidproduction output efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements feedback by using the results of the first fault assessment to determine which blocks require second fault assessment, and then using the second fault assessment results to make final retirement decisions. This feedback mechanism ensures that only blocks that truly fail the second assessment are retired, preventing premature retirement of functional blocks caused by neighbor plane disturb errors, thereby maintaining both reliability and productivity.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS12451213B2Optimized handling of neighbor plane disturb issues
Publication Date: 2025.10.21 SANDISK TECHNOLOGIES LLC
  • US12451213B2 patent drawing
  • US12451213B2 patent drawing
  • US12451213B2 patent drawing

AI summary

Memory blocks of a metablock are analyzed and tested to determine whether multiple memory blocks fail a first fault assessment. If more than one memory block fails the first fault assessment, the memory blocks are assigned to a retry pool. When in the retry pool, a second fault assessment is performed on the memory blocks. If one or more of the memory blocks pass the second fault assessment, the memory blocks are placed in a memory block pool and are relinked to other memory blocks to form a new metablock. However, if one or more of the memory blocks fail the second fault assessment, the memory blocks are retired.