20  Prakt. Met. Sonderband 50 (2016) 
Additive manufacturing (AM) techniques have been getting much attention recently. Even the Depending ox 
general public has been taking note of these novel production technologies. 3D printing shall be thermal expat 
mentioned exemplarily. Even though the title of this paper suggests otherwise, AM has been around resistant steel 
in the field of manufacturing for some time. The general interest in AM has been gaining impetus room temper: 
during the past decade or so. However, the shift from employing AM techniques for prototyping other sources 
applications only (rapid prototyping), towards their use in the manufacture of small series of develop upon 
production parts or the making of customized individual parts being made of actual design more than the 
materials is a relatively new trend. build up. Thi 
after having § 
Many different terms were coined over the years which more or less mean the same thing, such as alternating str 
rapid prototyping, selective laser melting (SLM), additive manufacturing (AM), laser sintering 
(LS), direct laser forming (DLF), laser consolidation (LC), or layer manufacturing (LM), to name While the mi 
just a few. The shift from prototyping component models using AM with plastics or other materials factor only, tl 
not intended by design for the final part to be made, just to make a geometric representation of the What is more 
latter for miscellaneous testing purposes, towards employing AM with real structural materials has contributed t 
led to a paradigm shift for the designer. Now it is possible to use the almost unlimited geometrical processes are 
freedom of AM techniques for making the most intricate components in real materials, giving have been re! 
engineers new degrees of freedom for component design. One example in the turbomachinery pointless sinc 
industry is the re-design of cooling configurations in components for better thermal efficiency. For SLM process 
a detailed description of AM methods, such as SLM where “successive layers of metal powder are 
fully molten and consolidated on top of each other by the energy of a high intensity laser beam”, the The subject | 
reader is referred to the literature. along both ¢ 
thickness. Th 
The use of AM processes with actual structural materials like steel and non-ferrous alloys, such as penetrant test 
nickel-base superalloys or aluminium, was driven by the aim of achieving material properties close fuel gas on tl 
to those of bulk material conventionally processed. However, designers will have to take into switchovers, 
account some microstructural peculiarities resulting from AM processing. In particular, SLM pipe surface 
produces extremly fine-grained microstructures due to high thermal gradients and rapid 
solidification. On the one hand, this might be advantageous for some applications requiring high If it is desire 
static and dynamic strength. On the other hand, components requiring high creep strength may not very low C 
be well suited to being made by SLM. Another problem could be the high residual stresses that restraint as | 
remain in the part after AM. If they are not relaxed by stress relief annealing, the superposition of particular ap 
residual and applied stresses can lead to cracking in service, as the following case study will candidate co 
demonstrate. rolling and 
conventional 
Two main cracks are visible which were later forced open. It is surprising how microscopically additive man 
brittle the fracture surface appears. The inner pipe wall shows its characteristic rough as-built SLM. Itis q 
surface. The resulting notch effect might very well have contributed to the failure. pipe by deve 
factors as far 
The subject Y pipe failed in service. In the as-received condition, it exhibited a long and gaping 
longitudinal crack. After forcing this crack open in the laboratory, it became apparent that the Y 
pipe failed in a fatigue mode. There are macro- and microfractographic fracture features that are 
consistent with such a failure mode, namely beach marks, and fatigue striations. Microstructural 
evaluations, using metallographic sections, revealed evidence of imperfections, such as 
microporosity, that are normal for SLM, but nonetheless might have contributed to the failure. An 
end-crater solidification crack found in one of the circumferential welds is unrelated to the failure. 
There is a second crack in the smaller of the two pipe cavities, that is visible only from the inside 
and did not penetrate the wall, hence was not detectable from the outside.