Non-Destructive Testing Qualification of Complex Parts with Digital Image Correlation and Digital Signatures


Qualification of final parts is a highly discussed topic in additive manufacturing (AM). Process immaturity and the associated uncertainty of part quality in metals AM has led many to seek new methods for qualifying manufactured parts, especially those of high complexity and low-volume production. Manufacturers have defaulted to costly destructive testing methods, often misrepresenting final part properties by creating and testing dog bone-shaped witness specimens. It is becoming progressively understood that due to variations in processing and thermal history that witness specimens do not accurately represent the final characteristics of the manufactured part.

As an alternative to the often costly, destructive testing of AM parts, Non-Destructive Testing (NTD) methods are increasingly part of AM part qualification. The most common methods focus on high-resolution scanning, both internal and external, using methods such as X-ray Computed Tomography, or X-ray CT. While these methods support detailed, fine-grained inspections, they also generate large amounts of data that are not readily interpreted. It is not always known how data should be interpreted to qualify a part against its intended functionality, or how part might perform under intended loading conditions. This subtopic focuses on qualifying parts by leveraging these functional requirements and utilizing NDT techniques to develop repeatable “signatures” of parts. These signatures will be based on the mechanical response of a part’s critical geometry under designed loading conditions, therefore qualifying a part against its intended use.

This subtopic addresses gaps in current qualification methods, specifically the need to address performance uncertainty introduced when manufacturing a part with AM processes. New methods are needed for the NDT qualification of complex parts to complement current predictive modeling, destructive, and non-destructive qualification methods. These methods should be robust enough to apply to any complex geometry while being mature enough to qualify against a part’s intended functionality.

The subtopic seeks to leverage high resolution Digital Image Correlation (DIC) technologies [1] and digital signatures [2,3] to evaluate mechanical responses under critical loading conditions at critical geometries of complex parts. By establishing acceptable loading thresholds and performance expectations, benchmark measurements taken from qualified parts can be compared with those of parts yet to be qualified under simulated loading conditions. The digital signatures of the qualified parts can then be used to verify and validate untested parts through DIC technologies and signature verification. Applications of such qualification techniques include NDT testing for deliberately introduced defects in AM manufactured parts, a major concern in AM security logistics [3].

The topics addressed under this subtopic directly relate to NIST’s goal of supporting U.S. commerce through measurement science. Successful development of the techniques will provide industry an affordable means of qualifying complex AM parts through advanced imaging and modeling technologies while maintaining manageable amounts of data generation.

The goal of this project is to develop and validate novel NDT methods that are well suited for the low-volume, high complexity, high cost production requirements in additive manufacturing. Methods should include novel, high-resolution digital image correlation techniques that can be applied to any complex geometry (including lattice). Based on predictive analytics and modeling, methods will identify functionally-critical locations and establish correlations between critical geometries in the design and surface points from which deformations will be measured. The stress and strain measurements, and analyses results, around these critical point locations will then be used to form a repeatable, function-based, response “signature.”

This project will develop new methods for qualifying functional parts based on DIC technologies and functional loading requirements. Goals include establishing the following methods:

• Based on actual loading conditions, a method will be developed for identifying critical surfaces and volumes on a functional part.

• A method will be developed to correlate actual loading conditions with simulated loading. The method will use simplified loading conditions to simulate stresses observed at critical surfaces and volumes under complex loading conditions. The method will demonstrate that loading experienced critical geometries of complex parts/shapes in actual loading conditions can be simplified so that the same loading conditions can be simulated in a lab environment. The method will be validated by demonstrating consistent mechanical testing results, and alignment between simulated and actual loading conditions.

• A method will be developed that utilizes high-resolution DIC to observe and analyze previously identified critical surfaces/ volumes. Results of the DIC related to stress and strain will then be mapped to the 3D geometry, providing a “DIC signature” that can be benchmarked to establish expected performance of a given material and geometry for given loading conditions.

• Methods will be established that correlate known responses of a qualified parts with the “DIC signatures” of parts yet to be qualified. Methods and thresholds will be established that will then allow for simulated loading conditions and corresponding DIC to attain “DIC signatures” of unqualified parts. These signatures will then be verified and validated against the signature of the qualified part. The “signature verification” method will be validated by demonstrating consistent results across “DIC signatures” for a given geometry, loading, and uncertainty threshold.

In concert, these methods will support the establishment of performance benchmarks for a given complex part. A benchmark signature will then be used to qualify remaining parts by comparing the established digital signatures with those obtained from other parts of the same material and geometry using high resolution DIC technologies.

Phase I expected results:
1. Validated method that induces the response of critical geometries (surfaces/volumes) experienced under actual loading conditions in a simplified, simulated lab environment. 2. Development and demonstration of a high-resolution DIC method that can be used to attain the stress maps around critical part geometries. 3. Validated method that demonstrates repeatability in high-resolution DIC on the critical geometries of complex shapes. 4. Validated method that matches benchmark “DIC signatures” consistently against parts with same material, geometry and loading. 5. Validated method that demonstrates that parts qualified using “DIC signatures” perform within acceptable thresholds and uncertainty, as predicted when compared against the performance of the benchmark part.

Phase II expected results: 1. Identified conditions on which the DIC method can be applied, demonstration of the limitations of the approach. 2. Demonstrated DIC using embedded markers and other imaging technologies to overcome some of the identified limitations. 3. Development of a prototype software package that supports the integration of the developed methods, including DIC, obtaining DIC signatures, and DIC signature validation.

NIST may be available to work collaboratively and provide technical direction and consultation.




Agency Micro-sites

US Flag An Official Website of the United States Government