Neutron Soft Error Rate Calculation for Semiconductor LSI

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for evaluating neutron soft error rates in semiconductor LSI require multiple neutron radiation tests across various energy bands, leading to high costs and long evaluation times.

Innovation Solution

A soft error calculation device that uses low energy neutron radiation data to derive the neutron soft error rate, incorporating a SEU cross section function calculator, error count actual measurement value calculator, and error rate calculator to calculate the soft error rate, reducing the need for extensive radiation testing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple neutron radiation tests are conducted across various energy bands to obtain soft error partial cross-sectional areas, then the accuracy of soft error rate derivation is improved, but the evaluation time and cost increase significantly

Engineering Contradiction:
Improvesoft error rate derivation accuracyVSAvoidevaluation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary neutron radiation tests only at low energy (e.g., 1 MeV) to obtain initial soft error cross-sectional area data. This preliminary action allows the system to calculate equivalent cross-sectional areas for higher energy bands through mathematical conversion, eliminating the need to conduct actual radiation tests at multiple energy levels while maintaining derivation accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the energy parameter of neutron radiation from multiple discrete values to a continuous energy spectrum model. By measuring at one energy point and using energy spectrum conversion formulas, the system derives soft error rates for the entire energy range, transforming a multi-parameter measurement problem into a single-parameter measurement with mathematical transformation

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple neutron radiation tests are conducted across various energy bands to obtain soft error partial cross-sectional areas, then the accuracy of soft error rate derivation is improved, but the operational cost increases

Engineering Contradiction:
Improvesoft error rate derivation accuracyVSAvoidoperational cost
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The patent performs preliminary neutron radiation tests only at low energy (e.g., 1 MeV) to obtain initial soft error cross-sectional area data. This preliminary action allows the system to calculate equivalent cross-sectional areas for higher energy bands through mathematical conversion, eliminating the need to conduct actual radiation tests at multiple energy levels while maintaining derivation accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a mathematical model (energy spectrum conversion formula) that copies the soft error characteristics from low energy measurements to high energy predictions. This virtual copying allows the system to derive high energy soft error rates without physical high energy radiation, significantly reducing operational costs

Inventive Principle:
Principle #26Copying

3Reliability

If neutron radiation is performed multiple times with different energy levels, then comprehensive soft error data is obtained, but the radiation equipment operating cost increases

Engineering Contradiction:
Improvesoft error evaluation completenessVSAvoidaccelerator operating cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent makes low energy neutron radiation data serve multiple purposes: it directly provides low energy soft error cross-sectional areas and simultaneously enables calculation of high energy soft error rates through energy spectrum conversion. This multi-functionality allows a single low energy measurement to replace multiple energy-level measurements, reducing accelerator usage and operating costs

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the energy parameter of neutron radiation from multiple discrete values to a continuous energy spectrum model. By measuring at one energy point and using energy spectrum conversion formulas, the system derives soft error rates for the entire energy range, transforming a multi-parameter measurement problem into a single-parameter measurement with mathematical transformation

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10401424B2Soft error rate calculation device and calculation method for semiconductor large scale integration (LSI)
Publication Date: 2019.09.03 HITACHI LTD
  • US10401424B2 patent drawing
  • US10401424B2 patent drawing
  • US10401424B2 patent drawing

AI summary

Neutron soft error rate derivation is calculated from data at the low energy neutron radiation. An outline value of an SEU cross-section function corresponding to a given neutron energy value is outputted. This outline value and the low energy neutron spectrum data are used to calculate an error count basic value of errors to occur over time. An error count actual measurement value over time is calculated from an error count during radiation of low energy neutrons and low energy neutron radiation time. The error count basic value and the error count actual measurement are used to calculate a proportionality coefficient of the SEU cross-section function. While holding a natural neutron spectrum, an error rate calculator outputs a neutron flux corresponding to a neutron energy value. The neutron soft error rate is calculated by an integration operation of multiplying the SEU cross-section function with the natural neuron spectrum.