PrintRite3D® Machine Learning Models and IPQM® Prediction Identifies Common Anomalies
In this paper, we discuss some acquired and developed knowledge from applying machine learning models using PrintRite3D® IPQM® metrics as features that are trained to recognize four common anomaly types, accurately predict each specific anomaly’s presence with IPQMP metrics, and discuss a great deal of future work that will be required to make these techniques useful in volume production.
Sigma Labs - ANSYS: Combination of Modeling and Thermal Sensing to Understand Additive Manufacturing Processes
For this study, Sigma Labs teamed with ANSYS to compare data collected by Sigma Labs’ coaxial thermal sensor response with ANSYS modeling as methods of monitoring additive manufacturing quality. Thermal effects of scan strategy in critical locations were predicted by ANSYS modeling; these thermal effects were experimentally validated using PrintRite3D® in-situ thermal process monitoring data. Validated thermal models in combination with thermal measurement can offer valuable insight to part design and manufacturing. Disagreement between modeling and monitoring can highlight important changes in the process that affect part quality.
PrintRite3D® Alerts for Anomaly Detection
Sigma Labs’ PrintRite3D® software provides additive manufacturers a vital tool to allow them to better understand production issues when they arise immediately and in real time. This allows rapid and detailed investigation and adjustment by process engineers, thus saving time, money and minimizing material waste.
Data Registration and Machine Learning for Anomaly Detection
Sigma’s PrintRite3D® software, using machine learning, can accurately predict the size and position of anomalies previously identified by CT scanning, as well as others not previously identified. This means PrintRite3D can be used as a new, near-real time quality measurement to supplement other measurements, like CT scanning.
Support Structure Optimization Using PrintRite3D®
Additive manufacturing companies can use less metal powder, reduce machine time and material waste, and remove the part more easily when they use Sigma Labs’ PrintRite3D® software to optimize the design of support structures upon which parts are built. They can minimize material waste, save time and money using PrintRite3D®.
DARPA Study Validates PrintRite3D® Quality Control Process for Certification of Metal Parts
A six-year project with DARPA that relied on Sigma’s PrintRite3D® technology has drawn three key conclusions. First, PrintRite3D® demonstrates and ensures process consistency and product quality in metal additive manufacturing. Second, this technology can also monitor additive manufacturing equipment health. Finally, PrintRite3D® can be used to certify quality and certify components without destructive testing or CT scanning – saving time, money and materials.
The relationship between In-Process Quality Metrics & Computational Tomography (CT) in Additive Manufacturing of Metal Parts
What if you could see and analyze the structure of a 3D-printed part while it was being made? You could then have high confidence in the manufacturing process. You’d also be able to adjust the process in real time to further assures quality. Sigma Labs has developed a method to do just that, with results comparable and complementary to CT testing, as this case study demonstrates.
In-situ Melt Pool "Thermal Signature" Defect Detection of Recoater Failure Using Co-Axial Planck Thermometry
Sigma Labs’ proprietary Thermal Energy Planck (TEP™) metric identifies variances in the production of 3D metal parts in real time. Thus, using TEP™ allows for intervention and adjustment of the manufacturing process in real time, making the process more efficient and less costly. This in turn saves time, money and can ensure product quality in metal additive manufacturing.
Evaluation of Quality Signatures™ using In-Situ Process Control during Additive Manufacturing with Aluminum Alloy AlSi10Mg Part 2
This document reports on the second in a series of experiments to demonstrate the capabilities of Sigma Labs, Inc. (Sigma) PrintRite3D® software to determine the effect of intentional changes in an independent process input variable (laser power) on dependent or response data mined in-situ on a layer to layer and part to part basis.
The Relationship Between Melt Pool Monitoring Metrics and Archimedes Density
Metal additive manufacturing today leaves skilled practitioners with little to no insight into the process. Nor does it provide in-process feedback about part quality. Until now. Sigma has demonstrated that its quality metric TED™ exhibits a strong correlation to both global energy density and Archimedes’ density, both established and trusted measures of part quality in the metal additive manufacturing industry.
In-Situ Process Mapping using Thermal Quality Signatures™ during Additive Manufacturing with Titanium Alloy Ti-6Al-4V
Companies that want to use 3D manufacturing face quality and reliability challenges, as well as cost factors in certifying their parts. Sigma’s proprietary software, as demonstrated in this case study, addresses these challenges. Further, the software can be used to eliminate the traditional trial and error approach to process mapping and validation, shortening the time and reducing the cost associated with today’s “make-and-break” approach to certifying AM part quality.
Evaluation of Quality Signatures™ using In-Situ Process Control during Additive Manufacturing with Aluminum Alloy AlSi10Mg - Part 1
This build was designed to establish a correlation between in-process dependent data mined from in-situ sensor raw traces signals, independent process input variables for example laser power, and post-process dependent data measured during destructive metallographic testing for porosity of as-built specimens.