Electrical Fuse Self-Test and Repair Circuit

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Electrical fuses in integrated circuits are not scalable for deep submicron technologies due to pitch limitations and lack self-test and repair capabilities, leading to increased manufacturing costs and yield loss due to faulty fuses.

Innovation Solution

A self-test and repair module is integrated into the electrical fuse circuit, utilizing multiplexers and an indication bit to compare intended data with stored faulty unit information, allowing for electrical programming and inversion of data to correct faulty bits.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If electrical fuse is tested on ATE, then testing capability is provided, but manufacturing cost increases and test time increases

Engineering Contradiction:
Improvetesting capabilityVSAvoidmanufacturing cost and test time
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements self-test functionality where the electrical fuse circuit tests itself using built-in test circuits and control logic. The system can detect faulty fuses and generate repair information without requiring external ATE equipment, enabling the circuit to serve its own testing needs and eliminating the need for expensive external testing equipment.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs testing and fault detection before the actual programming operation. By pre-testing the electrical fuse circuit and identifying potential faults in advance, the system can prevent defective fuses from being programmed, thereby avoiding manufacturing defects and reducing the need for costly rework or discarding entire chips.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If ATE is used for testing, then electrical testing is performed, but repair capability is lost and yield decreases

Engineering Contradiction:
Improveelectrical testingVSAvoidrepair capability
Core Design Contradiction:
ReliabilityVSEase of repair

Solution Approach 1:

The patent implements a feedback mechanism where the testing results and fault locations are fed back to the control logic. The system identifies which specific fuses are faulty and generates repair information that can be used to correct the defects. This closed-loop feedback enables the system to not only detect faults but also initiate repair processes, transforming the one-way ATE testing into an interactive repair-capable system.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary repair information storage and processing mechanism between the testing phase and the programming phase. This intermediary component stores fault location data and repair instructions, acting as a mediator that enables correction of faulty fuses without requiring complete chip discarding. The intermediary layer facilitates the transition from detection to repair.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of manufacture

If laser-fuse is used, then programming is achieved, but pitch scalability is limited by laser wavelength

Engineering Contradiction:
Improveprogramming capabilityVSAvoidpitch scalability
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent replaces the optical/laser-based programming mechanism with an electrical current-based mechanism. Instead of using laser beams to evaporate metal or polysilicon links, the system applies electrical current through the fuse link to induce resistance changes. This substitution of electrical mechanism for optical mechanism eliminates the wavelength limitation and enables scaling to deeper submicron technologies.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the fundamental programming parameter from optical energy (laser wavelength) to electrical parameters (current density and duration). By controlling electrical current density and application time, the system achieves programming without being constrained by optical wavelength limitations, thereby enabling pitch scaling below the laser wavelength threshold.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables efficient testing and repair of electrical fuses, reducing manufacturing costs and improving product yield by allowing for the identification and correction of faulty bits within the circuit.

Implementation Method 1

When programming the electrical fuse, a high current density, typically 600 mA/um2 for the silicided polysilicon, is applied to the electrical fuse link for a certain period of time. The resistance of the electrical fuse will rise due to the electrical stress in its fuse link.

Methodology Applied
Scientific EffectJoule heating: Joule Heating

Data Source

PatentUS7545666B2Electrical fuse self test and repair
Publication Date: 2009.06.09 TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
  • US7545666B2 patent drawing
  • US7545666B2 patent drawing
  • US7545666B2 patent drawing

AI summary

A circuit for testing and repairing a fuse device having a plurality of fuse units and being able to serially input and output data is disclosed, the circuit comprises a first multiplexer configured to select either a true or an inverted data for being stored in the fuse device, a second multiplexer configured to select either a true or an inverted data being read out from the fuse device, a storage unit configured to store information of faulty fuse units, and an indication bit being programmed to reflect a comparison between the data intended to be stored in the fuse device and the stored faulty unit information, wherein when the indication bit is at a first state, the first and second multiplexers select the true data, and when the indication bit is at a second state, the first and second multiplexers select the inverted data.