NAND Memory Read Sensing with Fail-Bit Adaptive Dual Pass

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing semiconductor memory devices face inefficiencies in sensing operations, leading to increased current consumption and prolonged processing times, particularly in NAND memory devices with multiple program loops and verify operations.

Innovation Solution

Implementing a dual-sensing technique that includes a fast read process with positive sensing followed by a slower read process with negative sensing if fail bit count exceeds a threshold, utilizing zero Volts during the first sensing and a positive voltage during the second, while reducing sensing timing parameters and maintaining consistent reference voltages.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If a fast read technique is used during sensing operations, then processing speed is improved, but current consumption increases and fail bit count increases

Engineering Contradiction:
Improveprocessing speedVSAvoidcurrent consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The sensing operation is divided into two distinct processes: a first sensing process using fast read technique with positive sensing for initial detection, and a second sensing process using slower read technique with negative sensing for verification when fail bits are detected. This segmentation allows the system to use the faster technique only when necessary while relying on the more energy-efficient technique for routine operations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The sensing technique dynamically switches between fast read and slower read modes based on real-time detection of fail bits. The system adjusts the read speed and sensing polarity adaptively - using fast positive sensing initially, then transitioning to slower negative sensing if needed, optimizing the balance between speed and energy consumption based on actual data conditions.

Inventive Principle:
Principle #15Dynamics

2Speed

If a fast read technique is used during sensing operations, then processing speed is improved, but reliability deteriorates due to increased fail bit count

Engineering Contradiction:
Improveprocessing speedVSAvoiddata sensing reliability
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system implements feedback by monitoring fail bit counts during the first sensing process and using this information to determine whether to initiate a second sensing process. When fail bits exceed a threshold, the system triggers a verification pass using negative sensing, ensuring that unreliable fast reads are corrected while maintaining overall system speed through selective re-sensing.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent prepares for potential reliability issues by having a pre-planned second sensing process ready to execute when fail bits are detected. This cushioning approach ensures that if the fast read produces unreliable results, the system can immediately switch to the more reliable slower read technique without compromising overall data integrity.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

3Speed

If positive sensing technique is used with fast read, then processing speed is improved, but energy efficiency deteriorates

Engineering Contradiction:
Improvesensing speedVSAvoidenergy efficiency
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

The patent applies different sensing qualities to different situations: positive sensing is used locally during the fast first sensing process where speed is prioritized, while negative sensing is applied locally during the second verification process where energy efficiency and reliability are prioritized. This localized application of different sensing qualities optimizes the overall energy-speed tradeoff.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12525300B2Energy efficient fast read in a memory device
Publication Date: 2026.01.13 SANDISK TECHNOLOGIES LLC
  • US12525300B2 patent drawing
  • US12525300B2 patent drawing
  • US12525300B2 patent drawing

AI summary

The memory device includes circuitry which is configured to sense threshold voltages of the memory cells of a selected word line of the plurality of word lines in a sensing operation. During the sensing operation, the circuitry, in a first sensing process, senses the memory cells of the selected word line using both a fast read technique and a positive sensing technique. The circuitry also determines a fail bit count and compares the fail bit count to a threshold. In response to the fail bit count being less than or equal to the threshold, the circuitry completes the sensing operation. In response to the fail bit count exceeding the threshold, then the circuitry performs a second sensing process to sense threshold voltages of the memory cells of the selected word line using a relatively slower read technique.