Wafer Special Pattern and Probe Card Defect Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current wafer testing methods rely on manual judgment to detect special patterns and probe card defects, leading to inefficiencies and increased labor costs due to potential false judgments and misuse of equipment.

Innovation Solution

A method that divides the wafer into testing partitions, uses a probe card to test each partition, and applies a chi-square test to determine the presence of special patterns or probe card defects by calculating P-values, allowing for automated detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual judgment is used to detect special patterns and probe card defects, then the testing process can identify abnormal die distributions, but the detection efficiency is low and labor costs are high

Engineering Contradiction:

Improvedetection accuracyVSAvoiddetection efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the manual mechanical judgment system with an automated computer-based analysis system. The test machine automatically acquires testing maps, calculates abnormal die distribution ratios, compares them against thresholds, and identifies special patterns or probe card defects without human intervention, thereby maintaining detection accuracy while dramatically improving efficiency

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The testing system performs self-diagnosis by automatically analyzing its own testing data to detect special patterns and probe card defects. The system uses its acquired testing maps to calculate abnormal die distribution ratios and automatically determines whether defects exist, enabling the system to self-verify its operation quality without external manual inspection

Inventive Principle:

Principle #25Self-service

2Reliability

If manual judgment is used to determine special patterns, then defects can be identified, but the process is time-consuming and labor-intensive

Engineering Contradiction:

Improvedefect identification reliabilityVSAvoiddetection time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces time-consuming manual analysis with automated computer processing. The test machine rapidly calculates abnormal die distribution ratios by comparing actual testing maps against expected distributions, automatically determining defect presence in seconds rather than requiring prolonged manual examination, thus maintaining reliability while minimizing time loss

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary automated analysis of testing maps immediately after data acquisition, calculating abnormal die distribution ratios and comparing them against pre-set thresholds before any manual review would occur. This preliminary automated detection quickly identifies obvious special patterns or probe card defects, reducing the overall time required for defect identification

Inventive Principle:

Principle #10Preliminary action

3Productivity

If automated testing is performed without pattern detection, then testing speed is high, but false judgments occur due to setting errors or probe card damage

Engineering Contradiction:

Improvetesting speedVSAvoidjudgment accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent introduces a feedback mechanism where the test machine automatically analyzes the acquired testing maps to calculate abnormal die distribution ratios and compares them against threshold values. This feedback loop continuously monitors for special patterns and probe card defects, allowing the system to self-correct or alert operators when issues are detected, thereby maintaining high testing speed while improving judgment accuracy through automated verification

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The testing system performs self-verification by automatically analyzing its own testing data for special patterns and probe card defects. The system uses its acquired testing maps to calculate abnormal die distribution ratios and automatically determines whether defects exist, enabling the system to self-verify its operation quality without external manual inspection, thus maintaining both speed and reliability

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9435847B2Method for testing special pattern and probe card defect in wafer testing

Publication Date: 2016.09.06 MACRONIX INTERNATIONAL CO LTD

AI summary

Methods for testing a special pattern and testing a probe card defect in wafer testing are provided. In the method for testing the special pattern, a wafer is divided into multiple testing partitions, in which each of the testing partitions includes multiple dies. The dies in each testing partition of the wafer are respectively tested by multiple sites of the probe card to obtain a testing map. Then, a number of the dies having defects and a number of the dies without defect within each of the testing partitions in the testing map are accumulated to construct chi-square test and calculate a maximum P-value. Finally, it is determined whether a minimum of the maximum P-values of all of the testing partitions is smaller than a certain predetermined threshold. If the minimum is smaller than the threshold, it is determined that the testing map of the wafer contains the special pattern.