No abstract

In laser powder bed fusion (LPBF) the surface layer temperature is continually changing throughout the build process. Variations in part geometry, scanned cross-section and number of parts all inuence the thermal eld within a build. Process parameters do not take these variations into account and this can result in increased porosity and dierences in local microstructure and mechanical properties, undermining condence in the structural integrity of a part. In this paper a wide-eld in-situ infra-red imaging system is developed and calibrated to enable measurement of both solid and powder surface temperatures across the full powder bed. The inuence of inter-layer cooling time is investigated using a build scenario with cylindrical components of diering heights. In-situ surface temperature data are acquired throughout the build process and are compared to results from porosity, microstructure and mechanical property investigations. Changes in surface temperature of up to 200°C are attributed to variation in inter-layer cooling time and this is found to correlate with density and grain structure changes in the part. This work shows that these changes are signicant and must be accounted for to improve the consistency and structural integrity of LPBF components.

Selective laser melting (SLM) is a relatively new manufacturing technique which offers many benefits. However the utilisation of SLM manufactured components depends on the assurance of their integrity during operation. Fracture toughness testing (JIC) has been performed on as-built compact tension fracture mechanics samples manufactured in three orthogonal directions. When the crack growth plane was transverse to the interface of the build layers, the fracture toughness values were found to be similar to those manufactured using conventional techniques. However, the fracture toughness is significantly reduced when the crack plane is parallel to the interface of the build layers. Simple heat treatments have been performed on Charpy fracture samples and the resulting impact energy values indicate that the fracture toughness of a component may be improved by heat treatment.

Laser powder bed fusion (LPBF) is an additive manufacture technique which builds components up in layers from a powder feedstock, using a scanning laser to selectively melt the powder into the required shape. The process of LPBF can often introduce defects into the structure of a part, since the powder may not fully melt and leave holes, or pores, in the sample. Excessive laser power may also cause the powder to vaporise and create pores. In whatever manner these pores are formed, they can significantly impact the properties of the finished component. Since pores and small defects already exist in LPBF components, the void growth and ductile fracture behaviour of LPBF components under multiaxial stress conditions needs to be characterised and predicted. In this work, notched bar tensile tests have been performed on samples with a range of notch acuities and hence multiaxial stress states. These tests have enabled ductile damage models to be calibrated and finite element (FE) simulations of the notched bar tests performed. The model was validated by comparison to the experimental results. The model agrees well with the results in many cases assessed in this work, but sometimes suffers from discrepancies and premature failure due to variability in material tensile properties, emphasising the need for sensitivity studies.

Uniaxial samples have been manufactured for tension/compression testing from 316L stainless steel by laser powder bed fusion (LPBF). Samples manufactured by LPBF are known to contain high levels of residual stresses. These uniaxial samples were built from a solid cylindrical rod and subsequently machined to reduce the central cross section of the sample to the required gauge diameter and improve the surface finish. Finite element (FE) models have been developed to simulate the LPBF process of the rods, their removal from the build plate and subsequent machining into the tension/compression samples. High tensile residual stresses were predicted at the surface of the samples, balances by similar magnitude compressive stresses along their axis. Post machining however, these stresses were reduced by around 80% or more. Residual stress measurements were performed on the samples post machining using the neutron diffraction techniques. These measurements confirmed that negligible residual stresses remained in the samples post removal from the build plate and machining.