Semiconductor Furnace Temperature Control via Prediction Model

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current semiconductor manufacturing processes face challenges in optimizing PID parameters for temperature control in furnaces, requiring trial and error and relying on engineer intuition, which can be inefficient and inaccurate, especially when heater temperature characteristics vary or when engineers lack sufficient time.

Innovation Solution

A technique that involves acquiring temperature data and power supply values, creating a prediction model to estimate temperatures, and calculating optimal power supply values to minimize deviations, thereby automating the temperature control process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If PID parameter optimization is performed by trial and error based on engineer intuition, then the temperature control can be adjusted flexibly, but the process becomes time-consuming and unreliable when heater characteristics vary

Engineering Contradiction:

Improvetemperature control reliabilityVSAvoidparameter optimization time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs self-diagnosis and self-optimization of PID parameters by automatically acquiring temperature data, analyzing heater characteristics, and calculating optimal parameters without requiring external engineer intervention or trial-and-error processes

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system continuously monitors temperature data from the furnace and heater, compares actual temperature with target temperature, and uses this feedback to automatically adjust and optimize PID parameters for maintaining reliable temperature control

Inventive Principle:

Principle #23Feedback

2Measurement precision

If PID parameters are optimized manually by engineers, then the process can be adapted to specific situations, but the accuracy decreases when heater temperature characteristics vary greatly

Engineering Contradiction:

ImprovePID parameter accuracyVSAvoidadaptability to heater variations

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system automatically detects and adapts to specific heater characteristics by acquiring temperature data and performing analysis specific to each heater configuration, eliminating the need for manual adaptation while maintaining high accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically determines optimal PID parameters based on actual temperature characteristics data, allowing the parameters to change and adapt automatically according to specific heater properties rather than using fixed manual settings

Inventive Principle:

Principle #35Parameter changes

3Ease of manufacture

If trial and error method is used for PID optimization, then the process can be simple to implement, but the productivity decreases due to repeated adjustments

Engineering Contradiction:

Improveimplementation simplicityVSAvoidtemperature control setup efficiency

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The system replaces the manual mechanical trial-and-error adjustment process with an automated computational system that acquires temperature data, analyzes characteristics, and calculates optimal PID parameters through algorithmic processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary automatic optimization of PID parameters before actual production use by acquiring temperature data and calculating optimal parameters in advance, eliminating the need for repeated adjustments during production

Inventive Principle:

Principle #10Preliminary action

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

This approach maintains high-level performance in temperature control by optimizing power supply to heaters, reducing reliance on intuition and improving efficiency in setting optimal PID parameters, even when temperature characteristics change.

Implementation Method 1

the substrate is accommodated in a furnace of a semiconductor manufacturing apparatus and an inside of the furnace is heated

Methodology Applied

Scientific EffectJoule heating: Joule Heating

Data Source

PatentUS11158529B2Method of manufacturing semiconductor device, method of controlling temperature and non-transitory computer-readable recording medium

Publication Date: 2021.10.26 KOKUSAI DENKI KK

AI summary

There is provided a technique that includes (a) acquiring temperature data of at least one of a heater temperature defined by a temperature of a heater and a furnace temperature defined by an inner temperature of a process chamber, and acquiring a power supply value indicating an electric power supplied to the heater; (b) acquiring a reference temperature of the temperature data; (c) creating a predetermined equation using a prediction model of estimating a predicted temperature of the temperature data; (d) calculating a solution of minimizing a deviation between the reference temperature and the predicted temperature based on the predetermined equation; and (e) outputting a calculated power supply value calculated from the solution, and processing a substrate while controlling heating of the heater based on the calculated power supply value.