Simulation technology unleashes the potential of 3D printing additive manufacturing

2024-02-29

Additive manufacturing, commonly known as 3D printing, is a new manufacturing process that emerged at the end of the last century. Due to various limitations, it has not been widely applied in the industry. In the technological conditions and application environment of the new industrial era, this technology has been re valued by the industry and rapidly applied.

In "You Use Simulation as a Vase, We Use It to Change the World", Tian Feng, Senior Vice President of Anser Asia Pacific, mentioned that the reason for achieving redesign is the emergence of new technologies and processes, among which additive manufacturing is a representative new process. Do not view additive manufacturing solely as a manufacturing technology, and oppose using additive manufacturing technology to print a traditional product. Additive manufacturing can achieve designs that traditional manufacturing methods cannot manufacture, endowing forward design with infinite freedom. Product design only needs to be based on requirements and functions, without considering manufacturing constraints and engaging in disruptive innovation. For the infinite innovation space provided by additive manufacturing, design itself has no norms and standards, so simulation has become an important tool. Therefore, redesign is the first deep-seated application of simulation technology.

While looking forward to the unlimited development space brought by additive manufacturing, metal additive technology also faces enormous challenges. Without simulation, metal additive manufacturing will encounter serious bottlenecks and can only be sealed in low-level application spaces. This article will directly face the simulation of additive technology - the second deep level application of simulation technology.

The challenges faced by metal additive manufacturing

Although the growth rate of metal additive manufacturing has been significant in recent years, there are several major challenges in both direct energy deposition and powder bed melting processes:

The types of printable metal materials are limited, and there is an urgent need to develop more metal types to meet industrial needs;

Restricted by printing speed and efficiency, not suitable for mass production;

Printing costs are too high, including machine costs, powder costs, and additional costs caused by high printing failure rates;

Requires tedious and lengthy post-processing steps for printing;

The quality assurance of printed parts and the difficulty of process debugging are very high.

Among them, quality assurance is a crucial factor in obtaining qualified printed copies. Metal additive manufacturing may encounter problems such as component deformation and cracking. Under the same processing parameters, layers, and materials, the microstructure and properties of the finished product vary with different orientations and positions. For example, the residual stress level of the vertical columnar crystal is low, while the residual stress level of the horizontal martensitic phase is high.

The simulation of additive manufacturing process mainly studies the effects of processing parameters, powder, geometric configuration, and other factors on macroscopic deformation, residual stress, microstructure and properties of micro internal components. Macro control and micro control are two important evaluation indicators in metal additive processes: macro control focuses on issues such as warping deformation, component cracking, scraper collision, or support cracking; Microscopic control requires attention to micro characteristics such as porosity, phase transformation, spheroidization, particle size, primary and secondary dendritic structures, and initial dislocation density, which will determine the mechanical properties and characteristics of metal parts.

Metal additive manufacturing may seem simple, but the actual process is very complex. The success of printing a qualified component is influenced by many factors such as materials, printing machinery and equipment, process design, process parameters and settings, and post-processing. For an actual metal printed piece, relying solely on experience or intuition results in a lower success rate of printing. The trial and error method not only increases costs but also extends the product manufacturing cycle.

The significance of metal additive process simulation

Using simulation technology to obtain the performance characteristics of printed products in advance is an important means and method to solve the quality problems of metal additive processes. By predicting in advance and optimizing the process based on this, the probability of printing failure can be reduced, and printing costs can be greatly reduced. The number of unqualified products and the number of trial and error can also be greatly reduced.

Although additive printing technology has the characteristic of being manufacturable no matter how complex it is, products with the same function can greatly enhance printability by using slightly different designs. This has a significant impact on the success rate of printing and manufacturing compensation, which may increase machine utilization, shorten product printing cycles, and ensure the repeatability and quality of product printing. If the prediction of microstructure and properties can also be achieved through simulation, it will greatly accelerate the development of new materials, new machines, and new process parameter packages, reduce research and development costs and cycles, and make it possible to obtain personalized microstructure and control material properties. The summary of the value of simulation for metal additive printing is shown in the following figure.

The value of metal additive process simulation

Difficulties in simulating metal additive processes

Although the value of additive process simulation is enormous, the difficulty in achieving it is also enormous, mainly including the following aspects:

The discrete scale of space I is huge, and the distance between time and walking is significant, resulting in long computation time

The micro size of the light spot and the huge macroscopic size result in a huge scale of grid discretization due to the size contrast between them. At the same time, the printing time of printed items is generally longer, with small items calculated in hours and large items in days, while the time step of thermal solid coupling simulation needs to be discrete in microseconds or even smaller magnitudes. It is very difficult to simulate printing processes with existing computing hardware resources.

I Multiscale problems with coexistence of macro, micro, and meso scales

Whether it is physical phenomena or the scale of the research object, the dynamic study of rapid cooling and solidification non-equilibrium inside the melt requires the use of material microscopic theory. How to introduce mesoscopic methods to unify microscopic and macroscopic phenomena requires analysis from a multi-scale perspective.

The mechanism of physical processes is complex

Considering only the physical phenomena inside the melt pool, additive metal printing is already very complex, including physical processes such as wetting, capillary, surface tension, Marangoni convection, melt pool dynamics, and phase transition. The accurate mechanism and evolution law of its physical changes need to be verified and summarized through experiments in engineering, and it is difficult to fully predict and generalize using only physical control equations.

I involve multiple factors and stages

The quality of additive metal manufacturing is not only related to the quality and characteristics of metal powder, but also to the printability of additive design, machinery and equipment, printing process, printing parameter package, and post-processing.

I have multiple sources of uncertainty and error

Due to the long process and multiple factors involved, there are also many sources of uncertainty and error.

Simulation of Typical Metal Additive Processes (SLM)

In addition to SLM, EBM, SLS, and DMD, metal additive processes also have derived process methods such as LBW, EBW, RPD, etc. Here, we take the popular and commonly used SLM (Powder Bed Melting Process) as an example to introduce the simulation of metal additive processes.

The simulation of SLM metal additive manufacturing process is a very complex and typical multi-scale and multi physical field analysis process. Multi scale is reflected in the scale spanning from macro to meso and then to micro. Multi physical fields require analysis of the forming temperature field, gas field (protective gas), melt flow field (molten pool fluid), velocity field (powder spreading process), and solid stress and deformation field of the printing structure, which can be applied to every stage of metal additive manufacturing and forming.

Process simulation at macro scale

Macro scale simulation analysis mainly focuses on the process simulation of part forming, predicting the stress-strain, forming temperature field, and potential risks during the forming process. The object of macroscopic analysisthe printed part itself and the supporting object of process design, which may also include substrate and necessary machine equipment information such as laser light sources. According to different process simulation algorithms, there are currently two main methods applied to macro scale metal additive manufacturing process simulation, namely temperature structure coupled (thermo elastoplastic) finite element analysis method and inherent strain finite element analysis method. The simulation analysis results of macro scale process usually include: deformation and residual stress of components and supports (before/after removing support), layer by layer stress and deformation, deformation compensation, scraper collision detection, high strain areas, and stress optimized support.

Analysis of melt pool and powder at mesoscale I

The simulation analysis at the mesoscale mainly focuses on the analysis of the molten pool and powder, including the fluidity of the molten pool, the size and morphology of the molten pool, the fluidity of the powder, heat transfer of the powder, and phenomena such as evaporation and splashing after melting. It is necessary to consider the surface tension, capillary, wetting, and Marangoni convection inside the molten pool. Currently, there are mainly equivalent thermal coupling and CFD methods applied to this analysis. By predicting the melting process and solidification process through melt pool dynamics, the phase transition process, temperature history, temperature gradient history, and solidification cooling rate can be obtained.

There are currently two main methods for simulating metal additive manufacturing processes at the mesoscale, namely the method that does not consider the powder scale inside the melt pool and the method that considers the influence of powder. Mesoscale simulation analysis simulates the flow and heat transfer inside smaller scale melt pools. In addition to predicting temperature, temperature gradient, and cooling rate, it can also predict surface quality, interlayer viscosity, porosity, etc. The simulation analysis at the mesoscale is generally a single scan of the object, and rarely involves multiple scans. However, the conclusions and results obtained can correct the macroscopic simulation results and can also be used as input for subsequent microscale analysis.

Microscopic scale organizational simulation

The temperature gradient or solidification cooling rate obtained from macroscopic or mesoscopic scale analysis can be used to predict the crystal structure morphology, grain size and orientation, defects and properties of products through microscale simulation. The main important methods currently used include Phase Field method, Cellular Automation, etc. Each method has its own characteristics and limitations.

The microstructure obtained during the metal additive manufacturing process will directly affect the performance of the formed parts. Obtaining a high density and crystal structure with good grain orientation and size is an important goal of metal additive manufacturing. The simulation analysis of crystals is also quite challenging due to the complex process of metal additive manufacturing.

After obtaining the temperature field or phase transition results data through macroscopic or mesoscopic analysis, the thermal gradient, curing rate, cooling rate, and morphology factor can be further calculated, which are input parameters for microstructure simulation at the microscopic scale.

Numerical simulation of microstructure typically includes deterministic methods, probabilistic methods, and phase field methods. Deterministic methods usually include frontier tracking methods, while probability rules include Monte Carlo and CA methods. Both deterministic and probabilistic methods for simulating grain growth require tracking the solid-liquid interface to simulate the morphology of dendrites, but there are certain difficulties in simulating three-dimensional morphology. The phase field method is based on the Ginzburg Landau theory, using differential equations to reflect the comprehensive effects of diffusion, ordering potential, and thermodynamic drive. With a unified control equation, it does not need to distinguish between solid-liquid phases and their interfaces, and can directly simulate the formation of microstructures. The phase field method and cellular automaton method are two commonly used numerical simulation methods for microstructure simulation.

The relationship between macro, meso, and micro scale simulation analysis of I metal SLM additive process

The relationship between macro, meso, and micro scale simulation analysis of metal SLM additive process is shown in the figure.

The overall relationship diagram of macro, meso, and micro scale simulation analysis of metal SLM additive process

Other areas of focus for additive process simulation

The current applications of additive process simulation that are of great concern include the following topics, and the details will not be elaborated here:

Special post-treatment (such as hot isostatic pressing), analysis of the impact of heat treatment on macroscopic deformation and residual stress elimination, microscopic simulation (such as density enhancement and metallographic structure improvement simulation), simulation of subsequent machining processes, simulation of smooth surface inside the flow, etc;

The support processing and equivalent simulation in macroscopic simulation, including body support, cone support, and block surface support, will also be considered in macroscopic process simulation for richer support in the future;

Microscopic metallographic simulation will directly support the prediction and evaluation of material mechanical properties.

The trend and development direction of additive process simulation

With the support of simulation technology, additive manufacturing will break through bottlenecks, fully leverage its advantages, and achieve the huge innovation space that people expect. But with the development of additive manufacturing technology, additive process simulation technology will also continue to improve. We believe that its future development trends are mainly in the following directions:

Macro scale simulation of additive processes will become increasingly popular and applied in engineering. The entire cycle of additive design, process, and manufacturing will gradually introduce additive process simulation to ensure the printability of designed products;

Materials - equipment - printed parts - support design and process design - process parameter package - macroscopic characteristics - microscopic characteristics - post-processing - performance prediction, the entire process will be streamlined and platformized;

Mesoscopic and microscopic analysis will gradually move from the research and research stage to engineering applications;

Support design and optimization software driven by physical process simulation will gradually be introduced;

Using test data and simulation data, AI algorithms and multi-scale algorithms will achieve offline prediction of additive processes;

More metal material data will be tested and entered, and more metal additive process methods will be simulated.

Researchers Develop New Micro Biorobots for Drug Delivery and Treatment