Senvol, which created the most comprehensive database of industrial 3D printers and materials, offers several products to help companies access, generate, and analyze AM data in order to implement the technology. Several American defense agencies, such as the US Air Force, US Army, and the US Navy, have used Senvol’s tools in the past. Now, the company announced that it recently teamed up again with the US Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL) to publish a technical report on on pedigreed additive manufacturing data.
The report, “Collection of High Pedigree AM Data for Data Analysis and Correlation,” is the result of a two-year cooperative research and development agreement, which was all about the generation of—you guessed it—pedigreed AM data.
The abstract of the technical report states, “ORNL worked with Senvol to develop and evaluate a standard operating procedure for collection of pedigree data for Additive Manufacturing using the Concept Laser XLine 1000r using an Al-Si-Mg alloy. ORNL independently evaluated and provided feedback to Senvol regarding the standard operating procedure document. Those edits were incorporated in the document that was then used to fabricate builds on the Concept Laser XLine 1000r to evaluate the efficacy of the document in collecting pedigreed data. The builds were done with varying build parameters, and the samples were subjected to tensile testing. The tensile data was used as an input for an artificial intelligence-based data analytics framework to determine the correlation between the build parameters and resulting tensile strength.”
The two worked together to evaluate and execute Senvol’s proprietary Standard Operating Procedure (SOP) document, so it could be used to collect pedigree AM data specifically for a laser powder bed fusion 3D printer using an aluminum (Al-Si-Mg) alloy.
“This SOP covers topics such as collecting appropriate geometric information, key processing parameters for the AM technology, and any key material testing protocols,” explained Peeyush Nandwana, a powder metals and additive manufacturing researcher at ORNL. “These are critical in terms of understanding the true material response, especially when dealing with multivariate analysis approaches in which several of these variables may be interlinked.”
The overall goal here was for ORNL to evaluate Senvol’s best practices document specifically to check on the adequacy of the written procedures for powder bed AM technology data collection. ORNL conducted an independent evaluation of the SOP document, which, according to Senvol, “details how to generate pedigreed additive manufacturing data,” and then provided its feedback to Senvol. The company added the edits into the technical report, which was then used in the making of 3D prints on a Concept Laser XLine 1000r system in order to see just how successful the report was in collecting the pedigreed data.
A variety of build parameters were used in the test prints, such as laser speed and layer thickness, and the 3D printed parts were subjected to tensile testing. The resulting data was then input into the Senvol ML software suite, which includes a machine learning algorithm that analyzes the relationships between process parameters and material performance. In this case, the data was used to figure out connection between the various print parameters and the resulting tensile strength.
“Oak Ridge National Laboratory has distinguished expertise in additive manufacturing, and so we were very pleased to work with them on this project. Collectively we were able to show that generating the data at the scale in this work and leveraging the use of correlation functions from Senvol’s machine learning software, Senvol ML, can provide the basis for isolating the impacts of different variables on resulting material properties and performance. This can be particularly helpful in developing process parameters for new materials and machines,” explained Senvol President Annie Wang.
In addition to making sure that Senvol is capturing the necessary data to extract good information, the company and ORNL concluded that the collection of pedigreed data is crucial to gaining a full understanding of AM material quality. You can download the technical report here and check out the project’s results, such as the ultimate tensile strength and elongation of the 3D printed Al-10Si-Mg alloy, for yourself.
“The impact of this project will be significant in helping the additive manufacturing industry understand the necessity of producing pedigree data,” stated Ryan Dehoff, secure and digital manufacturing lead at ORNL. “We’ve demonstrated that pedigree data collection is critical to understanding the quality of additive manufacturing materials, and ensured that all of the nuanced data required to accurately extract information is captured.”