Memory Test Apparatus Fail Information Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional test apparatuses for memory devices are inefficient in storage capacity and usage due to the need to store all possible error types, leading to increased size and decreased efficiency in error management and repair processes.
Innovation Solution
A test apparatus that stores fail information in block-units and includes a fail detecting section to identify defects based on predetermined conditions, updating fail information and disabling defective blocks to optimize storage and testing processes, with a mask processing section to cancel testing on defective blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all possible error types are stored in the test apparatus, then complete error detection capability is achieved, but storage capacity requirements increase greatly
Solution Approach 1:
The patent extracts only the essential error type information needed for repair processes from the complete set of all possible error types. The fail information storage is designed to store only the types of errors that actually occur and are relevant for subsequent repair operations, rather than pre-storing all conceivable error types. This extraction principle reduces storage requirements while maintaining complete error detection capability for relevant errors.
Solution Approach 2:
The patent applies local quality by making the error information storage adaptive to actual error patterns. Different blocks of memory may have different types of errors, and the system stores fail information specifically for the error types present in each block rather than uniformly storing all error types everywhere. This localized approach optimizes storage usage by matching the stored information to the actual error characteristics of each memory block.
2Reliability
If all possible error types are stored in the test apparatus, then comprehensive error management is enabled, but the scale and complexity of the test apparatus increase
Solution Approach 1:
The patent extracts only the necessary error type information for repair processes from the complete set of all possible error types. The fail information storage is designed to store only the types of errors that actually occur and are relevant for subsequent repair operations, rather than pre-storing all conceivable error types. This extraction principle reduces storage requirements while maintaining complete error detection capability for relevant errors.
Solution Approach 2:
Instead of pre-storing all possible error types and then filtering for relevant ones, the patent inverts the approach by initially storing only the error types that are actually detected and relevant for repair. The system builds the error information database dynamically based on actual test results, rather than starting with a complete but unnecessary database of all possible errors.
3Reliability
If fail information is stored for all blocks, then complete error tracking is achieved, but storage efficiency decreases due to unused information
Solution Approach 1:
The patent applies local quality by making the error information storage adaptive to actual error patterns. Different blocks of memory may have different types of errors, and the system stores fail information specifically for the error types present in each block rather than uniformly storing all error types everywhere. This localized approach optimizes storage usage by matching the stored information to the actual error characteristics of each memory block.
Solution Approach 2:
The patent implements a mechanism to discard irrelevant error information and retain only the useful fail information needed for repair processes. The system continuously updates the fail information storage based on actual test results, discarding error type information that does not correspond to actual defects and recovering only the essential error patterns that require repair attention.
Data Source
AI summary
Provided is a test apparatus having a bad block memory for storing a plurality of pieces of fail information in association with blocks of a memory under test, each piece of fail information indicating whether there is a defect in the associated block. The test apparatus writes a test data sequence to a page under test of the memory under test, reads the test data sequence written to the page under test, and compares the read data sequence to the written data sequence. The test apparatus includes an allocation register that stores allocation information for setting which of the plurality of fail conditions for judging whether there is a defect in the page under test are allocated to the plurality of pieces of fail information. The test apparatus detects whether there is a defect corresponding to each of a plurality of fail conditions, outputs the detection result as a fail signal, and updates a plurality of pieces of fail information associated with the block including the page under test using the fail signal corresponding to the allocated fail conditions.