Melt pool monitoring system and method for detecting errors in an additive manufacturing process

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional melt pool monitoring systems in additive manufacturing machines are ineffective at identifying process faults in real-time, leading to quality issues, scrapped parts, increased material costs, and machine downtime.

Innovation Solution

A method and system for real-time monitoring of additive manufacturing processes using a melt pool monitoring system that analyzes electromagnetic emissions from the powder bed, identifies outlier clusters through spatial proximity analysis, and generates alerts or adjusts the process to correct errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional melt pool monitoring systems are used to monitor the additive manufacturing process, then the build process can be evaluated after completion, but the system is delayed in identifying process issues and is ineffective at detecting errors in real-time

Engineering Contradiction:
Improveerror detection accuracyVSAvoiddetection delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system implements real-time feedback by continuously monitoring electromagnetic emissions during the additive manufacturing process and immediately comparing them against expected patterns. When deviations are detected, the system provides instant feedback to operators, enabling timely intervention before defects propagate through the build process.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis of emission patterns during the build process itself, rather than waiting for completion. By establishing baseline expectations for normal emissions and continuously comparing actual readings against these baselines, the system identifies anomalies before they result in scrapped parts or excessive downtime.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If conventional melt pool monitoring systems perform data analysis after build completion, then complex analysis can be done, but the system is not effective at identifying process faults that result in quality issues and scrapped parts

Engineering Contradiction:
Improveprocess fault identificationVSAvoidmonitoring system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system replaces complex post-build mechanical inspection methods with optical/electromagnetic sensing. By monitoring the electromagnetic emissions naturally produced during the melting and solidification process, the system identifies process faults without requiring physical contact or complex post-processing inspection equipment.

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

Solution Approach 2:

The system uses electromagnetic emissions as an intermediary signal to detect process faults. Rather than directly observing the melt pool or finished part quality, the system monitors the electromagnetic radiation emitted during processing, which serves as a proxy indicator of process health and potential defects.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of substance

If real-time monitoring and correction is implemented, then scrapped parts and material waste are reduced, but the monitoring system complexity increases

Engineering Contradiction:
Improvematerial wasteVSAvoidmonitoring system complexity
Core Design Contradiction:
Loss of substanceVSDevice complexity

Solution Approach 1:

The system enables self-service monitoring where the additive manufacturing process itself generates the monitoring signal through its inherent electromagnetic emissions. The process does not require external tracers, markers, or additional materials - the normal operation of the laser melting metal powder naturally produces detectable electromagnetic radiation that the system captures and analyzes.

Inventive Principle:
Principle #25Self-service

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 immediate detection and correction of print errors, reducing scrapped parts, material waste, and downtime by providing real-time feedback and process adjustments.

Implementation Method 1

an energy source such as an irradiation emission directing device that directs an energy beam, for example, an electron beam or a laser beam, to sinter or melt a powder material

Methodology Applied
Scientific EffectLaser beam irradiation: Laser

Implementation Method 2

melting entails fully melting particles of a powder to form a solid homogeneous mass

Methodology Applied
Scientific EffectLaser melting: Melting

Implementation Method 3

cameras or light sensors for detecting light that is radiated or otherwise emitted from the melt pool generated by the energy beam

Methodology Applied
Scientific EffectLight radiation from melt pool: Thermal Radiation

Data Source

PatentEP3650142B1Melt pool monitoring system and method for detecting errors in an additive manufacturing process
Publication Date: 2025.11.26 GENERAL ELECTRIC CO
  • EP3650142B1 patent drawingFigure 1
  • EP3650142B1 patent drawingFigure 2
  • EP3650142B1 patent drawingFigure 3

AI summary

A system (100) and method (300) of monitoring a powder (142)-bed additive manufacturing process is provided where a layer of additive powder (142) is fused using an energy source (120) and electromagnetic emission signals are measured by a melt pool monitoring system (200) to monitor the print process. The measured emission signals (250) are analyzed to identify outlier emissions (254) and clusters of outliers are identified by assessing the spatial proximity of the outlier emissions (254), e.g., using clustering algorithms, spatial control charts, etc. An alert may be provided or a process adjustment may be made when a cluster is identified or when a magnitude of a cluster exceeds a predetermined cluster threshold.