Manufacturing of Ti6Al4V Impeller of Centrifugal Compressor
Avez Shaikh, Ajinkya Shinde, Satish Chinchanikar, and Tanmay Deshpande
Abstract Additive manufacturing (AM) making radical changes in traditional
manufacturing through innovation. The 3D printing modeling and simulation tech- niques are evolving to avoid difficulties and losses in manufacturing. A critical aspect of failure in metal 3D printing is due to design distortion and internal (residual) stresses. Simufact Additive is a simulation tool that has emerged to reduce process development time and produce high-quality parts by having prior information of the part distortion. Simufact Additive simulates and predicts structural behavior (stress, strain, failures, and distortion). This study analyzes the effect of voxel-based surface mesh size on the process simulation for metal additive manufacturing of Ti6Al4V impeller of the centrifugal compressor. The part is analyzed considering different mesh sizes to obtain highly accurate and quality components without design distor- tion. The analysis of simulated results shows that the increase in element size reduces the amount of distortion. Meshing and the simulation time observed as decreasing with the increase in element sizes.
3D printing or Additive Manufacturing’s application in industry is vastly growing
due to the instantaneous and accurate results it provides in product development. But it is costly, sluggish, and needs multiple approaches before the final product is ready. The need for speed and precision is needed in the process which can be achieved through simulation modeling and analysis. Many software has been developed that captures the whole procedure of modeling and gauge the simulation to the needs of the handler. The simulation and analysis carried out in this research paper are done with Simufact Additive [1].
A. Shaikh (B) · A. Shinde · S. Chinchanikar · T. Deshpande
Mechanical Engineering, Vishwakarma Institute of Information Technology, Pune 411048, India e-mail: [email protected]
It is easy to work with and allows fast predictions about failures and stress in parts due to error or improper technology or manufacturing glitches. These predictions and simulations can address the problems in manufacturing through 3D printing. The errors or deterioration can be allowed through ineffectual forbearances or high- cost treatments. Residual stresses that are not required can cause self-distortion of the part during or post-production. Simulation in Simufact Additive allows users to digi- tally analyses and simulate the part with optimization such as required parameters, materials, environmental conditions, and creating or removing direction, support, etc. [2]. The software uses the finite element method to precisely approximate the stresses and distortions on the model or part. The finite elements are formed by the finishing element network. The most important part of the finite element analysis is whether the element size and number of elements used is sufficient for that analysis. If the element network is sufficient, and assuming other model inputs are correct then the outcome of the analysis is satisfactory. Finite element density is used to moderate the precision of the analysis which consist of element type and shape which highly affect the accuracy of the analysis [3]. There is abundant of work done in meshing and anal- ysis on different models and with different materials. From medical applications to constructive approaches in different domains simulation with additive manufacturing to detect anomalies is growing. Introduction to the voxel-based point cloud represen- tations is first-time given. There are key techniques for creating voxel-based point cloud representations to understand behavior and applications using voxel-based depictions [4]. In terms of material science, a new progressive damage model for 3D woven composites has been worked on with a depiction of volume cell with high fidelity is constructed. In this, Voxel and conformal mesh models of 3D woven composites are generated and then simulated on different environments to see the responses of materials and composites with a comprehensive comparison of damage simulations between the voxel and conformal models. It has pointed out that fake stress concen- trations and artificial connections are induced by voxel mesh, which are identical to the real ones that could be induced in the part in real environment conditions [5]. With 3D printing already on the verge of becoming a traditional method for prototyping. Newer approaches such as 4D printing are on their way to disrupt the industry. A conceptual design framework for 4D printing has been proposed. It deals in voxel- based modelling and simulation of non-programmable shape-changing smart mate- rials and composites. Their simulation and analysis can provide a quick evaluation of any material distribution behavior upon exposure to the stimulus and environment change. High accuracy can be obtained in a high-density network only if there is no singularity region in the part. As the complexity of the model and its network elements increases the need for high specifications in the computer and the time required to complete increases. Whenever multiple trials and iterations have required this problem regarding time consumption will be encountered, especially for transient and non-linear analyses. The model used in this study is an impeller from a centrifugal compressor. Also, this study focuses on the voxel mesh and its advantages over Effect of Voxel-Based Surface Mesh Size on Process … 251
other mesh types. The dimensions of the part and finishing elements can be fully represented and worked upon. Different materials with different geometric properties of different objects can be inspected. The boundary conditions can be used simply. Though there are disadvantages like the difficulty of model connection designs, modeling assumptions, estimation of component interactions difficulty, and damping is usually ignored [6]. The voxel mesh is necessary for the disadvantages to being coped with and making the manufacturing of the part possible. In this study, the Simufact Additive simulation tool has been used to predict structural behavior (stress, strain, failures, and distortion) of Ti6Al4V impeller of the centrifugal compressor considering the effect of voxel-based surface mesh size while the process simulation for metal additive manufacturing. The part is analyzed consid- ering different mesh sizes to obtain highly accurate and quality components without design distortion. The analysis of simulated results is presented to reduce process development time and produce high-quality impellers by having prior information of the part distortion.
2 Metal Powder Bed Fusion
In the Metal Powder Bed Fusion process, high power energy sources i.e., lasers are used to produce parts layer by layer by the melting of finely sized metal powders. The first commercialized PBF process for metals was Selective Laser Sintering (SLS) which included sintering or fusing of regions of thin powder layers. Since then, various technological advancements have been done in the PBF process which has enabled the production of components with increased and better mechanical char- acteristics, performances, and properties. Thus, this led to subcategories of various processes under metal powder bed fusion technology [7].
2.1 Selective Laser Melting (SLM)
Selective Laser Melting (SLM) is a technology under the metal powder bed fusion AM process that employs using a laser that unintermittedly melts layers of metal powder under an inert atmosphere [8]. The noteworthy advantages of this AM process include having high flexibility in the manufacturing of components with high complexity both in terms of lattice structures and design. SLM’s material research mostly prevails on Titanium alloys because of their high corrosion resistance, specific strength, and biocompatibility. With such process capabilities and mate- rial properties, SLM finds its applications widely in aerospace; space, biomedical and automotive fields [9–11]. 252 A. Shaikh et al.
2.2 Direct Machine Laser Sintering (DMLS)
Direct Metal Laser Sintering is a technology for the manufacture of metallic compo- nents directly from the 3D CAD Part/model. In DMLS mechanical properties of the components fabricated have been observed and studied by many researchers. The observations have concluded that DMLS can manufacture practical components in a small capacity. However, the properties of the components/parts manufactured through this process are inferior in quality as compared to components manufactured through other traditional manufacturing processes [12].
3 Process Simulation for Metal Additive Manufacturing
Software Process Simulation Modelling (SPSM) on models is a newer concept that
is getting popular due to its multiple benefits and the fact that it is the solution to the disadvantages of 3D printing metal parts. The failures and internal stresses of a model can be analyzed without manufacturing the actual part through simulation. The process starts with a model or part that needs to be analyzed before mass production. First, it is voxel meshed, with different sizes such as 0.5, 1, 1.5, and 2 mm. A voxel signifies a value on a consistent grid in 3-D space. In short, Voxel mesh refers to the volumetric mesh structure. The solution is provided by separating the mesh pieces into three dimensions. The surface is laid after the surface mesh, the voxel mesh covers the entire surface. The residual stresses and distortions are calculated by the end of the analysis. The residual stresses can be reduced most of the orientation of the optimum part that can be determined and the support elements optimized along with editing environment data. The simulation armed with this information aims to analyze and reduce stress and distortions by adjusting voxel meshes along with element sizes which will result in lower distortion and after manufacturing less failure of the part. The software Simufact Additive works smoothly with the simulations and analysis part [13].
3.1 Simufact Additive Software
MSC Simufact Additive is a software specially developed for metal powder bed fusion additive manufacturing processes such as direct metal laser sintering (DMLS) Selective laser melting (SLM), laser beam melting (LBM) and electron beam melting (EBM) methods. The primary benefits of this software are that we can process simu- late the complete AM process, including steps such as determination of various distortions and residual stresses, printing steps simulation, heat treatment, actual support cutting/removal and hot isostatic pressing (HIP). The software also uses effi- cient computational methods to predict: shrinkage, warpage, optimum location of Effect of Voxel-Based Surface Mesh Size on Process … 253
part placement., etc. The software provides the only solution of using a hexahedral element type for a voxel mesh. It enables the part to be built right the first time. For this study, we have used MSC Simufact Additive 1.0.
3.2 Materials and Methods
In this research paper, a case study component i.e., an impeller of the centrifugal compressor was used in the experiment. The main objective of this was to find the main effect of voxel size reduction of 3 predicted properties i.e., distortions, residual stresses, and effective strain in the final component before actually manufacturing it. Also, another main objective was to reduce the distortions in the simulated impeller part and to hence provide better manufacturing of the component. A generalized flowchart for the experimental process of performing the simulation is as shown in Fig. 1 [14]. In the very beginning the case study component i.e., the centrifugal impeller was imported into the software. The imported component must be always a CAD drawing in.stl format. (Fig. 2) The default metal powder fusion AM process was simulated i.e., SLM. The machine used for this process is EOS M 400 of dimensions 400 × 400 × 400 mm. The three process stages include building, simple cutting, and
Fig. 1 Flowchart of simulation model [14]
254 A. Shaikh et al.
Fig. 2 3D drawing of impeller
support removal. The imported part is centered with distance from left and front as 82 and 4.2 mm from the bottom for all four simulations. The next process includes generating supports for the impeller with a support radius of 0.25 mm and a critical surface angle of 45°. A total of seventeen supports were generated as shown in Fig. 3. The base plate material is X155CrVMo12-1_h. The metal powder used is Ti6Al4V, and its material curves, composition, and constants were already predefined and as depicted in Table 1. The next step involves assigning all the supports for the support removal step. The surface mesh for all four analyses is 2 mm. The layer thickness for each increment is 0.03 mm. The properties for laser beam are beam width of 0.00015 mm, the velocity of beam travel as 1 m/s, power as 200 W, and efficiency of 0.5. Impeller with generated supports is as shown in Fig. 3a. The next step involves defining the voxel mesh size and the voxel mesh parameter size is the topic main focus of the research paper. The generated voxel mesh for one of the sizes can be seen in Fig. 3b. As discussed, the software used hexahedral
Fig. 3 a Impeller with generated supports and b voxel meshed impeller
Effect of Voxel-Based Surface Mesh Size on Process … 255
Table 1 Ti6Al4V metal powder properties
Particulars Values Density 4.41 g/cm3 Minimum wall thickness 0.3–0.4 mm Maximum particle size layer thickness d50 (DIN ISO 13320) 30 µm Minimum temperature 293.15 K Maximum temperature 1873.15 K Material composition Al (5.5–6.5%), V (3.5–4.5%), O < 2000 ppm (parts per million), N < 500 ppm, C < 800 ppm, H < 120 ppm, Fe < 2500 ppm, Titanium balance
elements also called voxel elements. The voxel size is defined as the edge length for the hexahedral i.e., voxel element. The size of elements represents the number of metal powder layers comprised. The minimum fraction for the parts and supports is 5 and 1%. The simulations were performed for the following voxel mesh sizes—0.5, 1, 1.5, and 2 mm. The next step involved defining the solver properties i.e., parallelization, output division, and solver. Multiple cores were used i.e., six cores on Intel i7, 10th genera- tion laptop, with features such as 16 GB RAM and 6 GB NVIDIA graphics card for computation. The domain decomposition feature inclines to use of multiple cores for process computation. The meshing time for different voxel meshes was also recorded. Finally, once the simulation was completed the different results for distor- tions, residual stresses, and effective strains were compared to draw conclusions and hence determine the effect of mesh sizes on the results.
4 Results and Discussion
All the simulations were performed to predict 5 important parameters for the metal PBF AM process that are total distortions, effective stress is also known as residual stresses, effective plastic strain, and mean normal stress. • Distortion is the most common problem in additive manufacturing. Improper and Inconsistent shrinkages and accumulations during the additive manufacturing process drive the final distortion of the part. (Refer to Fig. 4 a, b which show the distortion comparison values between the analyzed simulation results for voxel mesh sizes of 1 mm and 1.5 mm, respectively). • Effective stress is defined as stress, which depends mainly on 2 factors i.e., pore pressure and applied tension, which in turn controls the strength and strain behavior of the metal powder (Ti6Al4V). (Refer to Figs. 5a, b which shows the effective stress comparison values between the analyzed simulation results for voxel mesh sizes of 1 mm and 1.5 mm, respectively). 256 A. Shaikh et al.
Fig. 4 Distortion for voxel mesh size of a 1 mm and b 1.5 mm
Fig. 5 Effective stress for voxel mesh size of a 1 mm and b 1.5 mm
• Effective plastic strain—The value of effective plastic strain is the integral of
stepwise increments of plastic deformation for a period t. (Refer to Fig. 6a, b which show the distortion comparison values between the analyzed simulation results for voxel mesh sizes of 1 mm and 1.5 mm, respectively). The predicted results for all four simulations are represented in Figs. 7 and 8. After performing simulations with four different voxel mesh sizes (0.5, 1, 1.5 and 2 mm), it was found that as the voxel mesh size is reduced more precise results were obtained.
Fig. 6 Effective plastic strain for voxel mesh size of a 1 mm and b 1.5 mm Effect of Voxel-Based Surface Mesh Size on Process … 257
Fig. 7 Effect of voxel mesh size on effective stress and meshing time
Fig. 8 Effect of voxel mesh size on distortion and effective plastic strain
Hence, the cause is that the more sensitive solution causes more deformation terms. It was also observed that if voxel mesh has an element size as small as the value, the experiments time will increase. Figure 7 depicts the effect of voxel mesh size on effective stress and meshing time and Fig. 8 depicts the effect of voxel mesh size on distortion and effective plastic strain. It can be seen that voxel mesh element sizes and dimensions affect the numerical computation, analysis, and underlying final results. The simulated results show that varying voxel mesh size and constant surface mesh, i.e., with an increase in the element size, reduce the number of distortions, effective stress, effective strain, meshing, and the simulation time for each particular analysis. 258 A. Shaikh et al.
5 Conclusion
Simulation modeling has multiple applications in pre-processing of additive manu-
facturing. It has the potential for a common standard in manufacturing industries. Distortion of parts through the production process constitutes a significant problem to MNCs, causing them to lose precious time and capital. Software like Simufact Additive is continuously improving to create a better-simulated environment for the user that tackles the disadvantages of Additive manufacturing. This study finds that voxel mesh element sizes and dimensions affect the numerical computation, analysis, and underlying final results. The simulated results from this study show that varying voxel mesh size and constant surface mesh, i.e., with an increase in the element size, reduce the number of distortions for each particular analysis. This study observed that finite element density, i.e., element type and shape are prominent metrics to control the accuracy of the predicted results. Effective stress, effective strain, distor- tion, meshing, and the simulation time observed as decreasing with the increase in element sizes. Finally, this study draws attention to that the future of pre-processing manufacturing lies within the lines of simulation modeling instead of the traditional practice of trial and error.
References
1. Litzkow J, Hwang IH, Mehmert P, Bernhardt R (2017) Simufact additive 1.0–a new era in simulating 3D-printing processes. In: Proceedings of the conference of the Korean society of precision engineering, pp 262 2. Çelebi A, Appavuravther EZ (2018) Analyzing the effect of voxel mesh and surface mesh appli- cation on residual stress by simufact additive software. Düzce Üniversitesi Bilim ve Teknoloji Dergisi 6(4):930–940 3. Jawade SA, Joshi RS, Desai SB (2020) Comparative study of mechanical properties of additively manufactured aluminum alloy. Mater Today: Proc 4. Sossou G, Demoly F, Belkebir H, Qi HJ, Gomes S, Montavon G (2019) Design for 4D printing: a voxel-based modeling and simulation of smart materials. Mater Des 175:107798 5. Beiges J, Chaumont M, Moreira CA, Cervera M, Codina R (2020) An adaptive finite element strategy for the numerical simulation of additive manufacturing processes. Addit Manuf:101650 6. Zheng T, Guo L, Tang Z, Wang T, Li Z (2021) Comparison of progressive damage simulation of 3D woven composites between voxel and conformal discretization models. Mech Mater 158:103860 7. Grasso M, Colosimo BM (2017) Process defects and in situ monitoring methods in metal powder bed fusion: a review. Meas Sci Technol 28(4):044005 8. Kruth JP, Froyen L, Van Vaerenbergh J, Mercelis P, Rombouts M, Lauwers B (2004) Selective laser melting of iron-based powder. J Mater Process Technol 149(1–3):616–622 9. Gong H, Rafi K, Gu H, Ram GJ, Starr T, Stucker B (2015) Influence of defects on mechanical properties of Ti–6Al–4 V components produced by selective laser melting and electron beam melting. Mater Des 86:545–554 10. Li P, Warner DH, Fatemi A, Phan N (2016) Critical assessment of the fatigue performance of additively manufactured Ti–6Al–4V and perspective for future research. Int J Fatigue 85:30–43 11. Sterling AJ, Torries B, Shamsaei N, Thompson SM, Seely DW (2016) Fatigue behavior and failure mechanisms of direct laser deposited Ti–6Al–4V. Mater Sci Eng, A 655:100–112 Effect of Voxel-Based Surface Mesh Size on Process … 259
12. Anton J, Lars P (2016) Characterisation of carbon fibre-reinforced polyamide manufactured
by selective laser sintering. Addit Manuf 9:7–13 13. Sunny S, Gleason G, Bailey K, Mathews R, Malik A (2021) Importance of microstructure modeling for additively manufactured metal post-process simulations. Int J Eng Sci 166:103515 14. Bian P, Shi J, Liu Y, Xie Y (2020) Influence of laser power and scanning strategy on residual stress distribution in additively manufactured 316L steel. Opt Laser Technol 132:106477
