PSI - Issue 68

Aleksa Milovanović et al. / Procedia Structural Integrity 68 (2025) 922 – 928 A. Milovanović et al. / Structural Integrity Procedia 00 (2025) 000–000

925

4

The data analysis is performed using the ‘’Matlab’’ (MathWorks, Natick, MA, USA) software. The K IC is calculated according to the formula (1), and the first calculated K IC value is a conditional value (K Q ) until proven validity:

"

(1)

!

$ %

#

$ B

=

!

!"#

$

%

&'

Where B is a specimen thickness, W is its width, and f(x) is a function of ɑ/W . The crack length symbol is ‘’ ɑ ’’. The formula for the calculation of f(x) is shown in Annex 1.4, and all possible values are listed in Table 1, see the ASTM D5045-14 standard. Except for the P Q validity estimation, the plane-strain condition around the crack tip is evaluated according to the formula (2):

!

!  " # %% && ' ( " #

(2)

" "# $ B & B

$ !%&

$ >

Where σ y is the yield stress value. Separate tensile tests were performed in Milovanović et al. (2024) research with the same AM process parameters to supply this research with the necessary data. Next, the J-integral is calculated according to the recommendations from the ASTM D6068-10 standard, with the first J 0 value calculated according to the following formula (3): (3) Where η is a constant (η=2, for SENB geometry), ɑ 0 is an initial crack length. U is the total absorbed energy subtracted by the indentation energy ( U= U t - U i , see Fig. 2-Right) . The following formula (4) takes into account the crack extension ‘’ Δɑ ’’: Namely, from this formula the J max is calculated using only one SENB specimen. For example, the J-R curve construction needs more specimens. To avoid the J-R curve in this early stage of the research, which requires a lot of time and effort to obtain, the J max serves here as a parameter to compare the results. In general, the J max value is a valuable parameter since it can at least indicate the magnitude of the J-R curve. The J-R curves for all tested SENBs will be presented in the extended paper. Also, the DIC cameras were employed here for more comprehensive data collection, similar to Milovanovic et al. (2022c) research. 3. Results and Discussion Load-deflection responses for all four infill cases are displayed in Figs. 3 and 4. Higher infill cases (i.e., 100% and 70%) are shown in Fig. 3, and the other two lower infill cases (40% and 10%) in Fig. 4. Linear response is dominant for 100% and 70% specimens, and a large non-linear domain is evident on lower infill SENBs. All specimens meet the size criterion for the plane-strain condition from the formula (2) and the calculated values are significantly below the 10 mm limit (see Fig. 5-Left). Similar to the findings from Milovanović et al. (2022b), the lower infill cases fulfill the size criterion with a higher difference from the limit value, and Fig. 5-Left misses the calculated values for 10% infill case, which are well below 1 mm. However, only the higher infill SENBs meet the P max / P Q < 1.1 requirements (see Fig. 5-Right). Also, a noticeably higher scatter of P max / P Q ratio is present in the 10% infill group, mainly due to the high difference in P max data (see Fig. 4-Right). ( ) ! ! ! " # $ B = " ( ) ! ! !"#$ % % ! " " # ! ! " # $ % = & "  " ( ) (4)