Memory Test Method for Identifying Failing Cell Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory device testing methods, such as compression tests, are inefficient in identifying which cell array fails among multiple arrays, leading to increased testing time and costs due to the inability to pinpoint specific failing cell arrays within the memory device.
Innovation Solution
A test method that outputs compressed data from multiple cell arrays and locks the output data of all but one cell array as normal data, allowing for the identification of failing cell arrays by generating secondary compressed data, enabling the detection of which cell array has failed during the testing process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If compression test is performed on multiple cell arrays simultaneously, then test time is reduced, but the ability to identify which cell array has failed is lost
Solution Approach 1:
The patent segments the compression testing process into multiple phases: first performing a global compression test on all cell arrays to detect failures, then performing individual compression tests on each cell array to identify the specific failing array. This segmentation allows both time efficiency and failure location identification.
Solution Approach 2:
The patent performs a preliminary global compression test on all cell arrays before performing individual array tests. This preliminary action quickly identifies whether any failures exist, and only if failures are detected does the system proceed to the more time-consuming individual array identification phase.
2Productivity
If data from all cell arrays is compressed into 1-bit compressed data, then test speed increases, but the specific failing cell array cannot be determined
Solution Approach 1:
The patent dynamically adjusts the compression level based on the testing phase. During the global test phase, all arrays are compressed together for speed. During the identification phase, the system dynamically changes to compress only the target array's data, maintaining high speed while enabling precise failure identification.
Solution Approach 2:
The patent applies partial compression action by compressing data from only the necessary cell arrays at each testing stage. Instead of always compressing all arrays simultaneously, it selectively compresses only the relevant arrays during the identification phase, maintaining efficiency while providing precise failure location information.
Data Source
AI summary
A test method for a memory having first and second cell arrays, first compressed data obtained by compressing output data of the first cell array and output data of the second cell array is outputted. When the first compressed data represents that a fail exists, output data of one of the first and second cell arrays is locked as normal data, and second compressed data obtained by compressing the normal data and output data of the other of the first and second cell arrays is outputted.