Non-destructive testing of metal-based additively manufactured parts and processes: a review

Figures & data

Table 1. Advantages and disadvantages of different AM technologies [38].

AM technologies	Advantages	Disadvantage
DED	Large processing range	Complicated phase transitions
	Fast deposition rate	High residual stresses
MEX	Preparative functionally graded properties	Unable to produce small complex parts
	Wide variety of ingredients	Poor repeatability
		Poor surface quality
PBF	Lower surface roughness	Low forming efficiency
	Higher dimensional accuracy	High residual stresses
	Fabrication of internal and complex features
SHL	High hardness and good wear resistance	Used on sheet metal
	High compressive and tensile strength	Low dimensional accuracy
	High efficiency for large parts
	Low cost
MJT	High dimensional accuracy	Lack of stability
	Homogeneous mechanical properties	Limited types of raw materials
BJT	Low machine costs	Not strong enough
	A large variety of material options	Relatively low density
	No explicit support structures required
	Ability to prepare complex structures
VPP	Superior print resolution	Toxicity
	High efficiency	Lack of stability
	Good surface finish
	A wide selection of materials

Figure 1. Schematic diagram of the processing principle of DED.

Figure 2. Classification of the major DED technologies. Adapted from Dass and Moridi [45].

Figure 3. Schematic diagram of SLM processing principle.

Figure 4. Schematic diagram of EBM processing principle.

Figure 5. Classification of the major PBF technologies.

Figure 6. Some typical defects in a metal AM part [15].

Figure 7. Establishing the correlation between microstructure, mechanical properties and the performance of the component through NDT in AM.

Table 2. Analysis of state-of-the-art NDT techniques [22, 66].

Various NDT techniques	Advantages	Limitations
Image-based optical method	Non-contact testing	Surface detection
	Fast acquisition	Dependence on algorithms and tools
	Easy implementation
X-ray computed tomography	Detection of interior defects	High cost
	Versatile applications	Limited object size
	Inspection of a wide range of materials	Unsuitable for online detection
Acoustic emission testing	Cost-efficiency	Limitations in identifying static anomalies
	Real-time monitoring	Sensitivity to weak signals and noise
	Early and rapid damage inspection
Ultrasonic methods
• conventional testing	Relatively high resolution	Unsuitable for high temperature operations
	Quick detection	Dependency on surface treatment
• phased array	Penetration of thicker parts	Require coupling
	No safety hazard	Limited range of detection
• laser ultrasonics	Non-contact testing	More expensive than conventional testing
	Suitable for complex geometries	Relatively low SNR
	Applicable at high temperature operation	Less sensitivity
Infrared thermography
• passive	Non-contact testing	Unsuitable for detecting deep defects
	No external heat sources	Potentially low SNR of the thermal signature
	Suitable for cyclic loading	Susceptible to environmental conditions
	Fast scan
• active	Non-contact testing	Require external heat sources
	Detection of larger areas	Unsuitable for detecting deep defects
	Fast scan	Relatively high cost
Machine learning based	Capacity for uncovering implicit knowledge	Supervised algorithms require large amounts of data
	Handling of large datasets	Lack of standard datasets in real world
		Complexity
Other NDT techniques
• eddy current	Improved sensitivity	Lower penetration depth
	High-speed detection	Surface finish affects defect detection accuracy
	Non-contact testing
• penetrant testing	Low cost	Offline detection
	High sensitivity	Detection of defects open to the surface
	Few material limitations	Dependence on precleaning

Figure 8. Schematic of selective laser melting (SLM) process monitoring system [67].

Figure 9. The elements of X-ray CT system, including X-ray source, detectors, test object, and reconstruction algorithms. Adapted from Biswal et al. [74] and Ramírez et al. [15].

Figure 10. Synchrotron X-ray CT of the L-PBF 316L stainless steel cylinders (grey) and defects (red) using different laser power: (a) 380 W, (b) 320 W, (c) 260 W, and (d) 200 W [90].

Figure 11. Schematic illustration of acoustic emission (AE) testing procedure for internal defect detection.

Figure 12. Test bench setup for AE during L-PBF [96].

Figure 13. (a) Arc-DED manufacturing process and (b) ultrasonic phased array inspection using (c) the total focusing method (TFM). Adapted from Javadi et al. [117].

Figure 14. (a) Experimental setup for ultrasonic phased array testing. Elastic least-squares RTM images of (b) longitudinal wave velocity, (c) shear wave velocity, and (d) density in the multi-material part of the experiment. Adapted from Rao et al. [119].

Figure 15. (a) Laser ultrasonic imaging system and (b) schematics of 3D imaging of subsurface defects [121]. Note that DAQ and LDV stand for the high-speed data acquisition and laser Doppler vibrometer, respectively.

Figure 16. Schematic illustration of a typical reflection test setup of active infrared thermography.

Figure 17. Thermograms of the PBF specimen made of the Inconel 718. (a) Front view and (b) rear view [145]. Note that black circles indicate defect outlines.

Figure 18. Types of machine learning based on input data used to train the algorithm. Note that MLP, CNN and RNN stand for multilayer perceptrons, convolutional neural network and recurrent neural network, respectively.

Figure 19. (a–d) Optical images of laser metal deposition build metal parts and (e–h) corresponding attention maps [149]. Note that defects are marked in red circles and rectangles.

Figure 20. Principle of eddy current testing.

Wong KV, Hernandez A. A review of additive manufacturing. Int Sch Res Not. 2012;2012:208760.

 Google Scholar

Dass A, Moridi A. State of the art in directed energy deposition: from additive manufacturing to materials design. Coatings. 2019;9(7):418. doi: 10.3390/coatings9070418.

 Web of Science ®Google Scholar

Ramírez IS, Márquez FPG, Papaelias M. Review on additive manufacturing and non-destructive testing. J Manuf Syst. 2023;66:260–286. doi: 10.1016/j.jmsy.2022.12.005.

 Web of Science ®Google Scholar

Lu QY, Wong CH. Additive manufacturing process monitoring and control by non-destructive testing techniques: challenges and in-process monitoring. Virtual Phys Prototyp. 2018;13(2):39–48. doi: 10.1080/17452759.2017.1351201.

 Web of Science ®Google Scholar

Dwivedi SK, Vishwakarma M, Soni A. Advances and researches on non destructive testing: a review. Mater Today Proc. 2018;5(2):3690–3698. doi: 10.1016/j.matpr.2017.11.620.

 Google Scholar

Zhang Y, Fuh JYH, Ye D, et al. In-situ monitoring of laser-based PBF via off-axis vision and image processing approaches. J Appl Psychol. 2019;25:263–274.

 Google Scholar

Biswal R, Zhang X, Shamir M, et al. Interrupted fatigue testing with periodic tomography to monitor porosity defects in wire+ arc additive manufactured Ti–6Al–4V. Addit Manuf. 2019;28:517–527.

 Web of Science ®Google Scholar

Choo H, Sham K, Bohling J, et al. Effect of laser power on defect, texture, and microstructure of a laser powder bed fusion processed 316L stainless steel. Mater Des. 2019;164:107534. doi: 10.1016/j.matdes.2018.12.006.

 Web of Science ®Google Scholar

Eschner N, Weiser L, Häfner B, et al. Classification of specimen density in laser powder bed fusion (L-PBF) using in-process structure-borne acoustic process emissions. Addit Manuf. 2020;34:101324.

 Web of Science ®Google Scholar

Javadi Y, MacLeod CN, Pierce SG, et al. Ultrasonic phased array inspection of a Wire+ Arc Additive Manufactured (WAAM) sample with intentionally embedded defects. Addit Manuf. 2019;29:100806.

 Web of Science ®Google Scholar

Rao J, Yang J, He J, et al. Elastic least-squares reverse-time migration with density variation for flaw imaging in heterogeneous structures. Smart Mater Struct. 2020;29(3):035017. doi: 10.1088/1361-665X/ab6ba4.

 Web of Science ®Google Scholar

Lv G, Yao Z, Chen D, et al. Fast and high-resolution laser-ultrasonic imaging for visualizing subsurface defects in additive manufacturing components. Mater Des. 2023;225:111454. doi: 10.1016/j.matdes.2022.111454.

 Web of Science ®Google Scholar

Kolb CG, Zier K, Grager J, et al. An investigation on the suitability of modern nondestructive testing methods for the inspection of specimens manufactured by laser powder bed fusion. SN Appl Sci. 2021;3:1–16. doi: 10.1007/s42452-021-04685-3.

 Web of Science ®Google Scholar

Cui W, Zhang Y, Zhang X, et al. Metal additive manufacturing parts inspection using convolutional neural network. Appl Sci. 2020;10(2):545. doi: 10.3390/app10020545.

 Google Scholar