Scan Cell Internal Defect Diagnosis Using Truth Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional fault models are insufficient to diagnose scan cell internal defects in VLSI circuits, as these defects exhibit complex behaviors that cannot be modeled by simple stuck-at or transition faults, leading to incomplete defect detection and diagnosis during scan chain testing.
Innovation Solution
A software-based method for scan cell internal defect detection and diagnosis using a defect truth table to describe the behavior of static scan cell internal defects, which identifies the defective scan cell by simulating scan shift cycles and analyzing output values based on input combinations, and an enhanced fault model to improve diagnostic resolution by masking defect impacts during scan load and unload operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fault models (stuck-at, transition faults) are used to diagnose scan chain defects, then the diagnosis process is simple and fast, but the defect detection coverage is incomplete for complex internal defects
Solution Approach 1:
The patent changes the parameter representation from simple stuck-at values to comprehensive defect truth tables that capture complex input-output relationships. The defect truth table includes multiple input combinations and their corresponding output values, allowing the model to represent complex internal defects while maintaining a systematic diagnostic approach.
Solution Approach 2:
The patent segments the complex defect diagnosis problem into manageable components: (1) generating defect truth tables for candidate faults, (2) simulating scan shift cycles with these tables, (3) comparing simulated outputs with actual test results, and (4) identifying the defective scan cell. This segmentation makes the complex diagnosis process systematic and implementable.
2Measurement precision
If advanced defect models with defect truth tables are used, then diagnostic accuracy for complex defects improves, but computational complexity and diagnosis time increase
Solution Approach 1:
The patent performs preliminary actions by pre-generating defect truth tables for all possible scan cell faults before actual diagnosis. These truth tables are prepared in advance and stored, so during actual defect diagnosis, the system only needs to retrieve and compare against pre-computed results, significantly reducing diagnosis time while maintaining high accuracy.
Solution Approach 2:
The patent creates simplified copies of the complex defect behavior through defect truth tables. Instead of simulating the entire complex internal logic of a defective scan cell, the system uses compact truth table representations that capture the essential input-output relationships, reducing computational complexity while preserving diagnostic accuracy.
3Reliability
If scan chain test patterns are applied to detect defects, then the integrity of scan chains can be checked, but complex internal defects inside library cells cannot be detected
Solution Approach 1:
The patent introduces defect truth tables as an intermediary representation between the physical scan cell and the test pattern application. These truth tables serve as a bridge that translates complex internal defect behaviors into observable input-output relationships that can be detected through standard scan chain test patterns, making internal defects measurable without modifying the physical test infrastructure.
Data Source
AI summary
A diagnosis technique to improve scan cell internal defect diagnostic resolution using scan cell internal fault models.