TLP Voltage Prediction for Noisy ESD Parameter Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electrostatic discharge (ESD) testing methods, such as transmission line pulse (TLP) testing, struggle with inefficiencies in manually interpreting key parameters like trigger voltage, 2nd breakdown voltage, and holding voltage due to noise interference, leading to potential misjudgment and inaccuracies.
Innovation Solution
A method and system using a neural network model to convert TLP measurement data into characteristic curves, predict trigger and 2nd breakdown voltages, and derive holding voltage, employing data pre-processing, model training, and optimization to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual interpretation of TLP data is used to extract key parameters, then human judgment can be applied, but it is inefficient and prone to human misjudgment due to noise in measurement data
Solution Approach 1:
The patent replaces the mechanical manual interpretation process with an automated computer-based system that uses signal processing algorithms and empirical rules to extract key parameters from TLP measurement data, thereby eliminating human error and improving both accuracy and efficiency
Solution Approach 2:
The patent introduces an intermediary automated interpretation system that acts as a mediator between the raw measurement data and the final parameter extraction, using noise filtering techniques and structured analysis methods to bridge the gap while maintaining measurement precision
2Productivity
If empirical rules are written into programs to automate parameter interpretation, then interpretation efficiency is improved, but the program cannot effectively filter out noise in measurement data, leading to potential wrong key parameters
Solution Approach 1:
The patent applies preliminary noise filtering and data preprocessing actions before the main parameter extraction process, preparing the measurement data by removing interference and highlighting key features, which enables subsequent automated interpretation to achieve both high efficiency and high accuracy
Solution Approach 2:
The patent replaces simple programmatic rule-based interpretation with a more sophisticated automated system that incorporates signal processing capabilities, enabling the system to both filter noise and extract parameters automatically, thus achieving both efficiency and precision improvements
3Measurement precision
If more detailed analysis methods are used to filter noise and improve accuracy, then measurement precision is improved, but the complexity of the interpretation system increases
Solution Approach 1:
The patent segments the interpretation system into distinct functional modules including noise filtering, signal processing, and parameter extraction components, each handling specific aspects of the analysis, which maintains measurement precision while organizing system complexity into manageable segments
Data Source
AI summary
Disclosed is a method and system for predicting TLP trigger voltage, 2nd breakdown voltage and holding voltage. The method includes: converting a plurality of measurement data into characteristic curves based on time series. The characteristic curves include a voltage-time curve, a current-time curve and a leakage current-time curve. After inputting the characteristic curves into a trigger voltage prediction model and a 2nd breakdown voltage prediction model respectively, the trigger voltage and 2nd breakdown voltage can be predicted respectively. The trigger voltage and 2nd breakdown voltage are used to calculate a holding voltage. Finally, marking points of the trigger voltage, 2nd breakdown voltage and holding voltage are marked on the current-voltage characteristic curve (I-V curve), and displayed on a display device.