PSI - Issue 14

Manojakumar Chimmat et al. / Procedia Structural Integrity 14 (2019) 746–757 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

747

2

1. Introduction Laser additive manufacturing (LAM) has evolved significantly in the last decade or so in a variety of applications ranging from aerospace, space, automotive, healthcare, tool and die and general engineering applications (Bi et al. 2014)(Yang et al. 2016)(Schoinochoritis et al. 2017).The process comprises building a part through layer-by-layer deposition of materials to create complex structures directly from a computer model design (Grunberger and Domrose 2015)(Thompson et al. 2015). A vast number of efforts have gone into process optimization to obtain parts that are dense and pore free to obtain the required geometry as well as to be able to have the required surface roughness conducive to enabling the additive manufactured (AM) part to be used in functional applications without further post processing (Basak and Das 2016). In the direct metal laser sintering (DMLS) process, a laser scanning system scans the material on the powder bed as per the input given by the CAD file. At the laser interaction zone material melts and solidifies (Grunberger and Domrose 2015). More recently, additive technologies have begun to be applied in the parts with complex internal geometries and enabling the consolidation of previously made parts. One of the key challenges with the DMLS process is the build-up of residual stress during a layer wise melting and solidification that takes place, which may result in cracking and distortion of the built part (Protasov et al. 2016)(Spears and Gold 2016). These residual stresses can build-up in the AM parts due to the combination of the high-laser-power combined with low-bed-temperatures, leading to curling of parts (Gibson et al. 2015). LAM processes exhibit complex thermo-mechanical behavior as they involve, melting, solidification and re melting/reheating of the previously deposited layer (Mukherjee et al. 2017) (Zheng et al. 2008). The origin of stresses in a fusion based deposition process are mainly due to mismatch in the thermal expansion co-efficient that occur during solidification, cooling and phase transformations. In single phase materials, the stresses are mainly thermal in origin while in multiphase materials phase transformations can also play a significant role (King et al. 2015). Residual stress can have a significant role on the reliability of the repaired/deposited additively manufactured component in terms of the mechanical properties and microstructure and can lead to unacceptable losses in dimensional tolerance, debonding between deposited layers during the as built condition as well as after heat treatment (Yadroitsava and Yadroitsev 2013) (Monroy et al. 2013). LAM is seen as the only solution for enabling a viable technology to have a high buy to fly ratio and to lead to enhanced productivity, especially in gas turbine and aerospace repair applications (Schoinochoritis et al. 2017). Therefore, it is imperative to establish the residual stress characteristics in the as repaired/deposited condition as well as after the applicable heat treatment to ensure reliable use of the LAM in critical repair applications (Bi et al. 2014). A fundamental understanding of how residual stresses evolve in LAM processes will enable prediction of part reliability to distortion, cracking, as well microstructure aspects such as recrystallization (after heat treatment) and hence the mechanical behaviour. This study attempts to elucidate the residual stresses experimentally via the X-ray diffraction sin 2 ψ technique on as printed, grit blasted and heat treated DMLS alloys, CoCrMo and SS316L.

Nomenclature AM

Additive manufacturing DMLS Direct metal laser sintering LAM Laser additive manufacturing CAD Computer aided design

2. Experimental methods

The alloys, CoCrMo and SS316L were printed using an EOS M280 single laser machine, at INTECH DMLS Pvt Ltd, Bengaluru. The nominal composition of the alloys is listed in the Table 1. The powders were sourced from Praxair and EOS, GmbH, for SS316L and CoCrMo, respectively. Two geometries were printed, a. Cube of 12.7x12.7x13.4 mm and b. flat plate with a dimensions of 115x80 mm and thickness of 4.5 mm at the base, for CoCrMo to study the variation in residual stresses with geometry. For SS316L, only Cube was printed with the same dimensions as that of the CoCrMo Cube as shown in the Fig. 1a. Figure 1 shows a schematic of the two geometries used in this study. These will be referred to as Cube and Coupon respectively through the rest of the paper. The processing parameters used for printing both these alloys are listed in Table 2. The process parameters were