Laser Powder Bed Fusion
Laser powder bed fusion (LPBF) process is one of the 7 basic AM methods specified by ASTM. Laser as a heat source in powder bed melting processes is the most used source together with electron beam in the production of metal materials by AM method. Although there are many differences between the two methods; the most obvious differences are the heat source used (laser or electron beam), the preheating requirements of the powder used as raw material and the particle size distribution.
Besides plastics, metals are commonly used in additive manufacturing. The AM subgroups for all materials can be listed as vat photopolymerization, material extrusion, material jetting, binder jetting, powder bed fusion, direct energy deposition and sheet lamination. Powder bed fusion technology is the most widely researched and utilized AM modality up to now . In powder bed fusion AM, the main principle is building the part in a powder bed layer by layer by utilizing the thermal energy of a laser or an electron beam.
Direct Metal Laser Melting (DMLM) is one of the powder bed fusion modalities among others such as Selective Laser Sintering (SLS) or Electron Beam Melting (EBM). DMLM is also known as Selective Laser Melting (SLM), or Direct Metal Laser Sintering (DMLS). Although EBM and DMLM depend on the same joining mechanism, i.e. melting, there are some differences as the heat source, processing environment (vacuum or protective gas) or preheating temperatures. A comparison between DMLM and EBM processes is presented in Table 1.1.
The Laser Powder Bed Fusion process is schematically demonstrated in Figure 1.1 in details.
The Laser Powder Bed Fusion process comprises a bunch of steps from CAD data preparation to taking produced component with the build platform out of the powder bed. In order to transfer CAD data into a Laser Powder Bed Fusion machine, it is converted to Standard Tessellation Language (STL) format and then it is pre-processed by special software, such as Materialise Magics, to orient and locate the part on the base plate as well as to insert support structures for any overhanging areas.
Then, slice data is generated by the same software to specify the laser scan path in every layer. After data transfer to Laser Powder Bed Fusion machine, the build starts with coating a predetermined layer of metal powder on the base plate which is clamped to the build platform. After one layer of powder is coated, a high energy density laser is operated and hits on the coated powder layer and selectively follows the path that is needed to be solidified in this specific layer.
The laser melts and solidifies the metal powder particles and the melted area is fused together. Once the laser completes scanning the whole area that is needed to be melted, the build chamber is lowered by one layer thickness and powder chamber (platform) which stores the powder is raised up by one layer thickness multiplied with a dosing factor in order to compensate the shrinkage or powder loss. Then the new layer is deposited on the already solidified one. The process is then repeated for successive layers of powder until the required components are completely built. This build concept for the DMLM process is demonstrated in Figure 1.2.
Need of Support Structures in Laser Powder Bed Fusion
Support structures, as shown in Figure 1.3 and Figure 1.4 , are one of the main features of a build in metal additive manufacturing because they support the main part as it acts as a “support” to successful build. In the end, they are not a part of main component to be produced. That’s why, they need to be removed after build is completed.
In Laser Powder Bed Fusion, mainly there are two reasons why to utilize support structures:
Eliminating or minimizing residual stress
“Residual stresses” are the stresses that remain in a part and deformed them plastically or elastically in the absence of any external stresses. Residual stresses can be classified in three different types and they are called differently due to they have different mechanisms. Type-I is macro-residual stresses, and they are generated is several grains. Type-II residual stress is micro-residual and are developed in one grain.
For instance, martensitic transformation induces Type-II residual stress. Type-III ones are sub-micro residual stresses and crystalline defects like vacancies, dislocations trigger to formation of this type of stresses. Almost all production technologies induce residual stresses in the product; however, it is known that the larger amount of residual stresses are induced in the Laser Powder Bed Fusion technologies compared to other ones due to high cooling rates encountered in the process. There are two main mechanisms which are attributed as the cause of residual stresses in Laser Powder Bed Fusion.
The first mechanism is called as the “Temperature Gradient Mechanism (TGM)” After the laser beam hits the powder bed, a huge amount of heat is inserted to the powder surface and large thermal gradients occurs around the laser spot due to the rapid heating of the upper surface and rather slow heat conduction.
Owing to the exposure of huge heat input on the top layer, it starts to expand and restrained by the underlying layer, elastic compressive strains are induced. If the material’s yield strength is reached, plastic compression is observed on the top layer. During cooling this plastically compressed upper layers start shrinking and a bending angle towards the laser beam develops. Figure 1.5 portrays the first mechanism which is elaborated above.
The second mechanism that causes residual stresses in the part is considered as the cooling-down of molten upper layer. In this period, the upper layer to be cooled down tends to shrink because of the thermal contradiction. Like the first mechanism, underlying layer tends to restrain this deformation. At the end, tensile stresses are introduced in the added top layer and compressive stresses are introduced below of upper layer.
The residual stresses which accumulated layer by layer during the process can cause one or more layers to be deformed. If this deformation occurs in the z-direction, or build direction, failures such as collision of the re-coater with the part or delamination may occur, and the build may stop. If it occurs along x or y directions, the final product quality is affected, and modelled product cannot be produced with high accuracy. To avoid such problems, overhanging features are propped up with support structures basically for two reasons:
- to restrain the geometry and make a solid connection between part and build platform.
- to dissipate of heat by using solid material instead of powder bed to minimize thermal gradient by knowing the fact that solid support structures have much higher thermal conductivity than powder bed.
Enhancing dimensional accuracy
Figure 1.6 represents a schematic view of Laser Powder Bed Fusion fabrication of inclined surface. As seen, overhang surfaces do not have fully solidified material underneath. Instead, a vast majority of this area is surrounded with metal powder. It is known that solid material has one hundred times higher thermal conductivity than packed powder bed. This causes the melt pool to become too large and to start collapsing into the powder bed as the result of gravity and capillary forces. At the end, dross will be created as demonstrated in Figure 1.7 and Figure 1.8. Meaning that the produced part geometry will deviate from its CAD geometry and results in dimensional inaccuracy.
As it is stated in previous section, support structures are also vital to eliminate dross formation and prevent part from dimensional inaccuracy. They are used as an anchorage to base plate, and they are also in touch with overhang features. Hereby, they prop up the these features physically.
Due to these two reasons, overhang locations need to be supported and then these support structures require to be removed from the surface by applying different post-processing methods such as milling or hand benching. Meaning of applying post-process for removing of support increments additional cost and time as well as obtaining bad surface finish of the overhang surfaces because of redundant.
Therefore, there are many research and studies have been conducted to understand support structures’ behaviors and its effects on either main part or build environment. Besides, reduction of support structures is one of the thirsting research topics in literature. The studies are tried compiled and shared as following.
Gan et. al. investigated different types of support structures to produce same main part. This aim was to understand whether less material usage can be achieved by using different types of supports as well as exploring the heat distribution on main part by changing the distribution and orientation of support structures. Cloots et. al. worked on to find optimum scan strategies with different support structure.
In another work, Hussein et al. Attained that less powder usage can be achievable by using certain cell size and structure type of support profile. Han et al. worked on effect of different support structures on different geometries and compared results from surfaces without support structures. Calignano tried to find optimum support structure geometry which serves the best main part dimensional accuracy at the end of the process with varying materials such as aluminum and titanium.
Obtained results show that the overhanging features having angles up to 30° can be manufactured without supports and there is no observed detrimental effect on the surface roughness. Strano et al. came up with a novel idea which is using cellular support structures substituted for solid ones in AM. Modelling of these cellular structures basically depend on mathematical expressions and results showed that these type of support structures give an opportunity for consuming less material to produce support with support overhang features as much as solid supports.
Langelaar proposed a new method combining support design and layout, build direction and optimization of part geometry simultaneously by running 2D test problems. This gives a huge opportunity to the design team of the part to achieve the most optimum support strategy with lower cost and higher part performance. Craeghs worked in his doctoral dissertation on controlling of process parameters to enhance surface quality by melt pool monitoring system.
Hussein et al. investigated cellular support structures due to the advantages of their low volume, easy removability, and reduction in build time. Krol et al. worked on modelling of support structures by the help of finite element analysis to understand their behaviors in powder bed fusion process. Poyraz et al. studied microstructural analysis of mating surfaces of block support structures and main part by changing parameters of support structures such as hatching distance. Moreover, Jhabvala et al. reported that using pulse laser gives better results than continuous laser in terms of production time because of utilizing high scanning speed with high power during the printing of support structures.
Xiang et al. provided results showing that the melt pool characteristics are very dependent on the oblique angle meaning that the angle between overhang feature and x-axis. The melt pool length is longer there is any overhang feature in part because powder dominant area conducts heat less than solid fabricated parts. Considering this aspect, Cooper et al. investigated that using contact-free support instead of contact support structures for overhangs in the EBM process and reported concrete improvement on distortion of overhangs.
Paggi et al. worked on contact-free supports by putting underneath of overhang structures in the DMLM process. The results show that using contact-free support improves the final fabricated part overhang surface quality in Ti Gr23 powder.
Read more articles about engineering
