Statistical and Data-Driven Methods for Additive Manufacturing Qualification: Proceedings of a Workshop (2024)
Chapter: 9 Overview of Measurement and Metrology
9
Overview of Measurement and Metrology
Bradley Jared, University of Tennessee, Knoxville, offered an overview of measurement and metrology. He first covered terminology, beginning with definitions of metrology (“the scientific study of measurement”) and inspection (“evaluation or scrutiny of an object or individual against an established standard”). He then defined accuracy as the qualitative agreement between a measurement and the true value, precision as a measure of variability, and uncertainty as a quantitative distribution that can be reasonably attributed to a measurement.
Turning specifically to additive manufacturing (AM), he said that it provides ever-expanding design freedom and value but requires increased verification costs, larger design margins, and expensive testing. Advanced metrology is required to address the uncertainties inherent in AM, such as the quantification of critical defects.
From there, Jared discussed how metrology is applied in AM. The different aspects include dimensional surface metrology (done, for example, with optical scanning), volumetric metrology (internal structures, defects, etc.), surface finish (various measures of texture), and machine metrology (measurements of different aspects of the machines used in AM processing).
Metrology is also used to measure various characteristics of the materials in AM, both the feedstock materials and the printed parts. Numerous techniques and standards have been established, with the measurements being used for such things as optimization, simulations, and uncertainty quantification. Both destructive sampling and nondestructive evaluation are options.
Measurements are also an important part of process monitoring, with the goal of being able to predict the performance of parts and materials from the set of measurements taken before, during, and after processing. There are many sensing modes available, including optical, thermal, acoustic, pressure, and electrical, and the measurements tend to generate large datasets.
In conclusion, Jared said that metrology is critical to AM and that many tools for it exist, but it is important to invest in the area to lower costs, decrease uncertainties, and increase throughput. Challenges include better volumetric measures and assessments of anisotropy. Last, he said, the digital thread should be used, with metrology connected to simulations and design tools.
In the question-and-answer period following the discussion, Jared first said that it would be useful to combine the computed tomography images of a finished AM product with data accumulated on the part’s form during the AM process. To a second question, he said that one of the metrology challenges for AM is that while one can measure various aspects of the AM process—the melt pool temperature, perhaps—it is much more difficult to collect the desired information about the AM part itself, such as its tensile strength, while it is being printed. He also spoke about doing measurements on the machines used for the AM process, such as measuring the movement of a laser in three dimensions to determine how accurate and precise those movements are.
Speaking about why investments in metrology and measurement tend to lag those in equipment and processes, Jared said that it is human nature, similar to how many people do not visit a doctor until they notice something is wrong. “You don’t actually go measure something until suddenly your customers are saying, hey, this doesn’t work.”
Concerning some of the most important variables to measure in AM, Jared mentioned laser parameters, layer thickness, and the energy density per unit volume of the material being melted.
