Melt-Crystal Interface Gradient Estimation for Ingot Puller Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ingot pullers for growing single crystal silicon ingots lack the capability to precisely measure thermal gradients in real-time, which are crucial for controlling the growth process and ensuring high-quality ingot production.
Innovation Solution
A model-based lumped parameter estimation system is employed to estimate axial gradients at the melt-crystal interface using a controller and sensors, allowing for real-time control of the ingot pulling process by deriving thermal gradients from measurable variables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If thermal gradients are measured directly, then measurement precision is improved, but device complexity increases due to inability to measure in-run
Solution Approach 1:
The patent uses measurable variables (temperature, growth rate, heater power) as intermediaries to indirectly estimate thermal gradients through a mathematical model, avoiding the need for direct gradient measurement devices that would add complexity and cannot be implemented in-run
Solution Approach 2:
The patent replaces direct physical measurement of thermal gradients with a computational approach using a mathematical model that calculates gradients from measurable process variables, substituting a complex measurement system with a computational estimation system
2Manufacturing precision
If real-time thermal gradient measurement is implemented, then manufacturing precision is improved, but loss of time increases due to measurement limitations
Solution Approach 1:
The patent implements a real-time feedback control system where the mathematical model continuously estimates thermal gradients from current process variables, allowing immediate detection of gradient changes and timely adjustment of heater power or growth rate to maintain ingot quality
Solution Approach 2:
The mathematical model is pre-configured with the relationships between process variables and thermal gradients, allowing instantaneous calculation and prediction of gradient changes before they affect ingot quality, enabling proactive process adjustment
Data Source
AI summary
A computer device includes at least one processor in communication with at least one memory device. The at least one processor is programmed to: a) receive data for simulating a process on the device; b) determine a plurality of gradient variables; c) determine a plurality of fit coefficients for the plurality of gradient variables; d) perform fitting operations on the plurality of fit coefficients and the plurality of gradient variable to determine superpositions for the plurality of fit coefficients; e) transmit the superpositions for the plurality of fit coefficients to a controller of the device; f) retrieve operating parameters of the device; g) determine one or more attributes that are not directly measured; h) determine values for the one or more attributes based on the superpositions of the plurality of fit coefficients; and i) control the device to perform the process.


