Toggle-Mode NAND Link Skew Compensation for SSO and Crosstalk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for valid window maximization in Toggle Mode (TM) or Open NAND Flash Interface (ONFI) links are inadequate in compensating for simultaneous switching outputs (SSO) and cross talk, leading to timing failures and impractical training times due to assumptions about uniform valid windows and neglecting interdependent crosstalk impacts.
Innovation Solution
A method for calculating and skewing critical DQ signals to maximize valid windows by identifying and skewing critical signals in a Critical Timing bucket, while storing others in a Relaxed Timing bucket, and deriving DQS tap settings based on these signals, without requiring a ninth bit for status indication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods are used for valid window maximization in Toggle Mode links, then the implementation is simpler, but timing failures occur due to inadequate compensation for SSO and cross talk
Solution Approach 1:
The patent segments DQ signals into two categories: critical signals (with smaller valid windows) and non-critical signals (with larger valid windows). By applying skew compensation selectively to critical signals through a Critical Timing bucket, the system achieves reliable timing without the complexity of processing all signals uniformly.
Solution Approach 2:
The patent applies different skew compensation strategies to different signal groups based on their specific timing requirements. Critical signals receive systematic skew addition to maximize their valid windows, while non-critical signals use conventional approaches, optimizing reliability without unnecessary complexity.
2Reliability
If uniform valid window assumptions are made for all DQ signals, then the analysis is simpler, but timing failures occur due to neglecting interdependent crosstalk impacts
Solution Approach 1:
The patent applies skew compensation excessively to critical signals by adding systematic skew beyond what uniform methods provide. This partial over-compensation ensures that even signals affected by interdependent crosstalk meet timing requirements, achieving reliability without measuring each signal's crosstalk individually.
3Productivity
If all DQ signals are processed with the same timing methodology, then the process is simpler, but training time increases due to inadequate optimization
Solution Approach 1:
The patent segments DQ signals into Critical Timing and Relaxed Timing buckets, enabling differential processing. Critical signals receive intensive skew optimization while non-critical signals use standard processing, significantly improving training efficiency by avoiding unnecessary optimization of signals that already meet timing requirements.
Solution Approach 2:
The patent changes the timing parameters (valid window calculations and skew amounts) based on signal category. By adjusting skew tap settings differently for critical versus non-critical signals, the system optimizes training time while ensuring all signals meet their specific timing requirements.
Data Source
AI summary
Embodiments of the disclosed technology relate to the operation of devices including memory devices, and more particularly to valid window maximization in a toggle mode (TM) or Open NAND Flash Interface (ONFI) link using systematic skew while compensating for simultaneous switching outputs (SSO) and cross talk.