PSI - Issue 8

P. Conti et al. / Procedia Structural Integrity 8 (2018) 410–421 Author name / Structural Integrity Procedia 00 (2017) 000–000

414

x 10 -5

3

35

Bulk Powder

Bulk Powder

30

2.5

25

2

20

1.5

15

1

10

thermal conductivity [W/(m°K)]

0.5 thermal expansion coefficient [1/°K]

5

0

0

10 3

10 3

temperature [°K]

temperature [°K]

Fig. 3. Thermal expansion coefficient vs. temperature

Fig. 4. Thermal conductivity vs. temperature

The resulting dependence of the thermal characteristics with temperature is summarized in fig. 1 to 4. The Young modulus was assumed very small (the value should be virtually zero but a small value was adopted to obtain a positive definite stiffness matrix) for the powder and for the melted material; the other values were interpolated from data reported in K.C. Mills (2002). Fig. 5 summarizes the dependence of the Young’s modulus from temperature.

10 12

Bulk Powder

10 10

10 8

10 6

10 4

Young's modulus [Pa]

10 2

10 0

10 3

temperature [°K]

Fig. 5. Young’s modulus vs. temperature

2.4 Finite element model The Finite Elements (FE) model is intended to reproduce the SLM strategy called “Island Scan Strategy” developed by Concept Laser ™ to reduce residual stresses and deformation. The idea is to randomly scan very small areas across the layer to limit local strains. The layer area is first broken into a square grid and then single squares (“islands”) - randomly chosen - are scanned with a regular pattern, L.N. Carter et al. (2014). Two adjacent islands have always orthogonal scanning directions. The strategy is explained in Fig. 6 (from I. Gibson et al. (2010)). The scope of this work is limited to the simulation of a single island. The iterative FE analysis reproduces the scanning pattern, the square area will be scanned with path snaking from left to right, and right to left at fixed speed until the entire area has been scanned.