Nonvolatile Memory Data Retention via Margin Detection Cells
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-level threshold voltage algorithms in nonvolatile memory cells are vulnerable to data errors due to charge leakage and margin loss, as the narrowing of threshold voltage states increases sensitivity to read disturbs and charge loss.
Innovation Solution
The implementation of nonvolatile memory integrated circuits with data cells and margin detection cells, where the margin detection cells have a narrower operating margin than data cells, allowing for more precise sensing windows and automatic refresh mechanisms to prevent data errors, using separate or shared sense amplifiers and control circuitry to manage bias arrangements and detect charge loss or gain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multi-level threshold voltage algorithm is used to increase storage density, then storage density is improved, but data retention reliability deteriorates due to narrowed margin between threshold voltage states
Solution Approach 1:
The patent segments the memory array into data cells and margin detection cells. The margin detection cells are further divided into groups, with each group containing multiple margin detection cells corresponding to different data cell groups. This segmentation allows independent monitoring of threshold voltage margins without affecting the primary data storage function.
Solution Approach 2:
Margin detection cells serve as intermediary elements between the data cells and the control circuitry. These intermediary cells monitor the threshold voltage margins by comparing their own threshold voltages against reference values, and they trigger refresh operations when margin degradation is detected, thus protecting the primary data storage without requiring direct intervention in the data cells.
2Productivity
If threshold voltage states are bunched closer together to increase storage density, then storage capacity is improved, but vulnerability to charge leakage and read disturbs increases
Solution Approach 1:
The patent implements preliminary monitoring of threshold voltage margins through margin detection cells before actual data corruption occurs. These detection cells continuously assess the health of threshold voltage states and trigger refresh operations in advance, preventing charge leakage and read disturbs from causing data errors.
Solution Approach 2:
The system establishes a feedback loop where margin detection cells monitor threshold voltage margins and provide information to control circuitry. When the control circuitry detects that margins have degraded below threshold values, it triggers refresh operations on the affected data cells, creating a closed-loop system that actively maintains data integrity despite closely bunched threshold voltage states.
3Measurement precision
If margin detection cells with narrower operating margin are introduced, then sensing precision is improved, but device complexity increases
Solution Approach 1:
The patent designs margin detection cells with the same basic structure and functionality as data cells, allowing them to be integrated into the existing memory array without requiring entirely new cell designs. The same sense amplifier circuitry and control logic are used for both data cells and margin detection cells, reducing the impact of device complexity while enabling precise margin monitoring.
Data Source
AI summary
Disclosed are apparatuses, methods, and manufacturing methods relating to improving data retention in nonvolatile memory. In many embodiments, reference memory corresponding to the data memory is used to determine whether to refresh the data memory.