Research progress in the application of nondestructive testing in additive manufacturing technology

In conventional mechanical manufacturing techniques, casting is often used for the production of complex workpieces, such as precision casting or pressure casting. At the same time, with the development of forging and mold forming technology, its products have been difficult to process or have been impossible to manufacture. Therefore, industrial development urgently requires major improvements in existing technologies or the emergence of new technologies, so the emergence of additive manufacturing technology is of great significance. Laser additive manufacturing technology is the most representative type of additive manufacturing technology. According to its forming principle, the most representative ones are laser selective melting (SLM) and laser metal direct forming (LMDF) technology. Laser selective melting The principle of the technology is shown in Figure 1.

1. Additive manufacturing

At present, additive manufacturing technology has become one of the most dynamic and promising technologies in industrial production. Compared with the traditional mechanical manufacturing technology, the additive manufacturing technology has a series of advantages such as short cycle, no mold, high flexibility, and no restrictions on material and part structure. In automotive, medical, electronics, military, aerospace. Other fields have been widely used. By using honeycomb, crystal lattice or other complex structures, it is also possible to optimize the weight and function parameters of the workpiece and reduce the wall thickness. Quality control is a very critical issue for all new technologies. So far, the research on the quality control of additive manufacturing technology is not deep enough, and non-destructive testing is the key technology to improve the quality control level of additive manufacturing. This paper summarizes the types of defects in the additive manufacturing process, points out the requirements of non-destructive testing technology in the additive manufacturing process, and draws the application potential of laser ultrasonic technology by summarizing the advantages and disadvantages of various non-destructive testing technologies.

2. Defect type of additive manufacturing technology

A typical additive manufacturing workpiece is shown in Figure 2. Additive manufacturing can be divided into four stages according to the process: raw material, preparation process, preparation completion and service process. Different types of defects may exist in each stage, and the contents to be tested are also different. The main contents to be tested in the raw materials include powder size, particle shape and morphology, physicochemical properties and material supply. The main testing contents in the preparation process are stress state, molten state, material properties, part distortion, pores, residual stress (elimination The self-phase-balanced internal stress remaining in the object after external force or uneven temperature field, etc.), over-melting depth and fusion quality. After preparation, the main objects to be detected are geometric deviation, residual stress, product anisotropy. , cracks, bubbles, inclusions, surface defects, pore clusters, deep embedded defects and porosity (refer to the percentage of pore volume in the bulk material and the total volume of the material in the natural state), the defects formed during service mainly have surface Defects, cracks and deformations. Several typical defects in laser additive manufacturing are shown in Figure 3.

Among them, continuous defects such as cracks, pores and clusters of holes are the most dangerous, and these defects are usually present in most workpieces. One of the outstanding features of additive manufacturing workpieces is their higher porosity compared to conventional forged, cast or molded parts. An increase in porosity may reduce the strength of the part. Localized hole clusters may result in the formation of cracks in service, and the presence of micropores usually determines the dynamic properties (such as fatigue) of the workpieces produced by the additive. At the same time, local metal segregation due to quenching in a partial region causes a specific stress-strain state to occur during metal crystallization. Higher residual stresses lead to deformation, geometric dimensional changes and the formation of microcracks, so the stress state is the key component of the additive manufacturing process.

3. Non-destructive testing requirements for additive manufacturing technology

Additive-made workpieces are typically disposable and extremely expensive to manufacture, so traditional destructive testing is often not available for additive manufacturing. At the same time, because the additive manufacturing workpieces are created layer by layer, the properties are more difficult to predict, which poses a challenge to the quality inspection of additive manufacturing workpieces. In a sense, non-destructive testing can perform quality assessment of the workpiece without damaging the integrity of the workpiece and service performance, and can meet the unique inspection requirements for additive manufacturing workpieces.

The whole process inspection requires the non-destructive testing method to be used for the characterization of the metal material in the molten state during the additive manufacturing process, which is far more difficult to detect than the prepared workpiece, and requires the detection process not to interfere with the processing of the additive manufacturing process. Improved non-destructive testing sensors and controllers are needed in additive manufacturing equipment and processes to improve inspection and control capabilities, provide real-time visibility and adjust the manufacturing environment. The real-time detection and material properties in the deposition process need to be able to improve the production of qualified parts, so that parts produced by additive manufacturing can be directly used for installation.

In order to improve the quality of the workpieces produced by additive manufacturing, it may be necessary to perform closed-loop process control of the entire system, such as the ability to monitor parts and control or mitigate the distortion and residual stress of the parts, while providing a detailed record of the production of each additive. Process control can also be extended to raw materials prior to manufacturing and verify the microstructure, geometry and quality of the part. Since the process parameters deviate from their optimum values ​​during the manufacturing process, the service performance of the processed workpieces may deteriorate. Therefore, the additive manufacturing process parameters need to be evaluated by non-destructive testing results. The main parameters of the evaluation are as follows: System deviation and optical sensor to determine the bath depth.

There are five main requirements for non-destructive evaluation of materials: non-destructive testing of raw materials, non-destructive testing of finished parts, monitoring of defect effects, design of product databases and physical parameter reference standards. Non-destructive testing of raw materials, such as metal powder size, particle shape, microstructure, morphology, chemical composition of molecules and atomic composition, these parameters need to be quantified and ultimately evaluated for performance consistency; non-destructive testing of finished parts involves the fabrication of workpieces (without further processing) and Post-processing workpiece (need to be further processed), the detection content includes small-size pores, complex workpiece geometry and complex internal features; defect effects, non-destructive testing method to characterize the defect type, frequency and size of the finished workpiece, easy to understand the product The impact of attributes on product quality and performance; designing a product database, a microstructure database can compile and clarify the relationship between process structure and performance, including images or photos collected during each process, such as input material properties, in-situ process monitoring and After the manufacturing and post-processing, the generated features are completed; the physical parameter reference standard, the current lack of suitable full-size workpieces to evaluate the feasibility of the non-destructive testing method in the additive process, due to the complex geometry of the parts manufactured by the additive, and the embedding deep

In the additive manufacturing process, real-time monitoring of possible defects is required. It is necessary to overcome the influence of surface topography and preparation temperature. It is necessary to integrate non-destructive testing technology with the manufacturing process without affecting the additive manufacturing process; Evaluate during the acceptance phase and lifetime to determine its service performance. In addition, throughout the life of the part, it is necessary to characterize the microstructure and morphology of the material, to make fine measurements of atoms and molecules, to characterize internal stress states, and so on. In short, timely and reliable detection of defects of different nature and monitoring how these defects develop is of great significance for the additive manufacturing process. Therefore, non-destructive testing methods need to meet material, design, and testing requirements, and can be used throughout the life cycle of materials, including optimization during manufacturing, implementation process testing, post-production quality acceptance, and quality monitoring during service. Therefore, there are clear requirements for non-destructive testing at all stages of additive manufacturing.

4. Development of Nondestructive Testing Technology in Additive Manufacturing Technology

At present, the technologies in non-destructive testing mainly include: computed tomography, penetration test, eddy current test, ultrasonic test and infrared camera measurement. As shown in Figure 4, an experimental system for monitoring 3D printing using acoustic emission is shown.

X-ray inspection has a wide range of applications in the industry and can undoubtedly be used to detect porosity, dimensional errors and other defects in additive manufacturing parts. The X-ray incident angle directly affects the size and shape of the detected defect and can show defects of less than 2% of the sample thickness. Computed tomography scans all samples, while ultrasonic and permeation tests are performed on the surface of the workpiece. X-ray computed tomography has the ability to detect internal defects and internal features to detect closed pores and high density inclusions. At the same time, computed tomography detection technology also has certain limitations, such as the volume effect of X-rays is obvious. At the same time, since the crack perpendicular to the X-ray beam cannot be detected, it is impossible to reliably detect the defect. In general, X-ray computed tomography is a powerful technique for non-destructive testing of additive products, making it possible to describe the material's structure, shape distribution and quantitative dimensions of defects.

A prominent feature of additive manufacturing is the higher porosity than conventional forged, cast or molded parts, where irregular rough surfaces are present, making traditional non-destructive testing methods for detecting surface defects difficult to apply. Penetration testing is a surface inspection technique that detects surface and near-surface defects of solid materials and their parts. It is used to detect porous or rough workpieces that are not processed and polished. It is difficult to measure complex internal structures or crystals with deeper positions. The lattice structure needs to update the more sensitive non-contact non-destructive testing method.

Dinwiddie et al. used infrared cameras to reveal defects such as voids, unfused and splashed materials during additive manufacturing. The image processing special algorithm they developed can quantitatively describe the porosity, but does not specify the minimum defect size that can be detected. Gatto and Harris are mounted at a position of 135 mm from the working surface with a CMOS camera with a resolution of 508 pixels/inch. During the synthesis process, the camera takes a layer-by-layer photo and then processes it through a specially developed algorithm to obtain geometric parameters of each layer. The photo can determine the geometry of the pores and calculate the shape deviation of the cross section. The limitation of this method is that on the one hand, only the external surface condition can be analyzed and the inside cannot be detected. On the other hand, the surface roughness can seriously affect the detection result.

Guan et al. evaluated the selective laser-sintered workpiece using the EX1301 Michelson optical coherence tomography system, which can achieve a spatial resolution of about 10 μm (compared to 50 μm for X-ray) and can detect hollow , unbonded and surface roughness, as shown in Figure 5, but this method cannot detect large workpieces. Guan et al. pointed out that the penetration depth of light waves depends on the absorption and reflection characteristics of the material, and the spatial coherence and temporal coherence of back-reflected light waves can affect the measurement accuracy, so this technique can only be applied to non-metallic materials. This method has the same sensitivity as X-ray computed tomography, but optical tomography can be used for process detection of layer-by-layer growth.

Rudlin et al. studied eddy current, laser ultrasonic, and laser imaging methods for the detection of additive manufacturing processes. In fact, none of the above three methods have been used for the inspection of the manufacturing process, and can only be used for the detection of artificial defects after preparation, and to evaluate the near surface defects of the workpieces manufactured by the additive. The principle of laser thermal imaging is to use real-time thermal imaging of the laser-heated part of the sample by infrared camera to reveal the non-uniformity of laser heating of the sample section. The method has low sensitivity when detecting defects below the surface, and only one test is reliably detected in the test. Defects with a diameter of 0.6 mm and a depth of 0.2 mm. At a depth of 0.5 mm or more, the detection sensitivity of the eddy current technique is 0.4 mm, and when detecting near surface defects, the sensitivity of the laser ultrasonic and laser imaging methods is less than 0.2 mm, as shown in Fig. 6. Show.

Laser ultrasonic testing is a non-contact detection method that can be used for fast scanning. The ultrasonic can be used to characterize the anisotropy of ultrasonic propagation in a sample by using different transverse and longitudinal speeds on the metallographic cross section. Parameters can be used to determine the size and depth of near-surface defects, and are often used for defect identification of welds to ensure the integrity of the pipe and track. There are currently few studies using laser to ultrasonically examine metal sample powder deposits. Because laser ultrasound can use a laser source to generate powerful ultrasonic pulses with easy-to-do waveforms and a wide spectral range, its spatial resolution is 3 to 10 times higher than piezoelectric excitation. At the same time, since the laser-induced ultrasonic pulse does not oscillate and the pulse duration is 6-7 times shorter than the PZT, higher resolution and higher sensitivity can be achieved, and the blind spot is small. At present, laser ultrasound has a detection depth of 700 μm for discontinuous defects ranging from 150 to 500 μm, but the sensitivity is significantly reduced when the depth exceeds 300 μm. The porosity and anisotropy detection of laser ultrasonic for additive manufacturing products. There are still few studies. In summary, the use of laser ultrasound for non-destructive testing of additive manufacturing demonstrates greater potential, but it also needs to be integrated with the manufacturing process, considering the use of this method for layer-by-layer real-time monitoring of the additive manufacturing process.

Further, the method of measuring the residual stress can be classified into a physical measurement method and a mechanical measurement method, and the mechanical measurement method is usually a destructive method such as a grooving method. Non-destructive testing methods that can be used for residual stress detection include magnetic methods, X-ray diffraction methods, and ultrasonic methods. Among them, the magnetic method is determined according to the relationship between the stress and the magnetization curve during the saturation process of the ferromagnetic body, and is used within a certain range; the X-ray method is perfected, but there is radiation damage and only the surface stress and the specific position can be measured. Lattice distortion is difficult to measure, so its application is greatly limited; ultrasonic law is the most promising method in non-destructive testing methods, with fast, convenient on-site measurement, both surface measurement and internal residual stress, especially Laser ultrasound technology has greater application potential.