*2.2. Weldability Assessment Trials*

Three different weldability assessment trials were carried out with the 5 investigated alloy 718 casting heats after completing heat treatments described in the previous section, thus, in the solution annealing state. LBW Varestraint tests were carried out in a testing device (Figure 2) that was fully designed and manufactured at LORTEK [15,16]. Performance of the test bench complied with general requirements of ISO/TR 17643-1 "Destructive tests on welds in metallic materials—Hot cracking tests for weldments—Arc welding processes—Part 3: Externally loaded tests" [26]. LBW Varestraint tests were performed on 3.2 mm thickness samples that were electric discharge machined (EDM) from 10 mm thickness casting plates. External surfaces were milled before testing. LBW was applied on the surface of testing samples without adding any filler metal by TRUDISK 6002 disk laser from TRUMPF company, Ditzingen, Germany. The laser beam was guided through

400 μm diameter fibre to a TRUMPF BEO D70 laser welding head (200 mm focus length and 200 mm collimation length). LBW Varestraint tests were completed at 0.5 m/min welding speed, 2300 W continuous mode power, and 0.8 mm diameter spot size. Tests were performed in a closed chamber filled by argon gas to avoid surface oxidation and provide good shielding conditions.

**Figure 2.** (**a**) LBW Varestraint testing device and (**b**) measurement of total crack lengths (TCL). Labels of individual cracks in fusion zone (FZ) and heat-affected zone (HAZ) are displayed.

Different augmented strains (*ε*) were applied during LBW by bending the samples along their length at 150 mm/s stroke rate and employing several interchangeable die blocks with radii varying from 20 to 320 mm. The induced augmented strains which are calculated by the following equation were in the range from 0.5 to 8%.

$$
\varepsilon = \frac{t}{2 \cdot R} \times 100 \,\text{\AA} \tag{1}
$$

where *ε* is the resulting augmented strain as a percentage, *t* is the thickness of the sample in millimetres, and *R* is the radius of the die block in millimetres. Two expendable support plates of 304 stainless steel were positioned in both sides of the testing samples to avoid kinking. Both FZ and HAZ cracking susceptibility were studied by determining total crack lengths (TCL) in these two zones of the welds [2,13,14].

Note that in this case, cracks were measured on the surface of the testing samples using magnification lenses (up to 150×) and after cleaning the surface of the welds by soft manual polishing and oxalic acid electroetching to avoid reflections.

Additionally, hot ductility tests of 5 casting heats were carried out in Gleeble 3800D thermomechanical simulator (DYNAMIC SYSTEMS INC., Austin, USA) owned by West University in Sweden. Here, 6 mm (−0.025 mm, +0.01 mm) diameter cylindrical shape samples were finely turned from 10 mm thickness as-cast plates. Hot ductility testing setup and guidelines included in Gleeble Users Training 2010 handbook were applied. These are comparable to the testing specifications included in procedure B of [26], with minimum differences in samples length. NST temperature was only determined in mould E, concentrating the overall weldability assessment of 5 alloys on on-heating and oncooling tests. A heating rate of 111 ◦C/s from room to testing temperatures was employed in on-heating tests, whereas samples were heated up to 1195 ◦C at the same heating rate and subsequently cooled down at 50 ◦C/s to each testing temperature in on-cooling trials. Temperature profile was recorded with K-type thermocouple welded to the surface of testing samples in the area between clamps. The samples were pulled to fracture at 55 mm/s stroke rate. Percentages of area reduction from initial 28.3 mm2 (i.e., 6 mm diameter) were measured to determine ductility at the different testing temperatures.

Finally, self-restrained bead-on-plate LBW trials were carried out in casting plates. In these representative tests, the parent materials were remelted by scanning the surface with a laser beam with the same energy distribution, shielding conditions and parameters

employed in LBW Varestraint tests. In this case, trials were carried out in samples with less than 3 mm thickness (between 2.6 and 2.9 mm) and 9 mm thickness, the latter resulting from the surface grinding of casting samples. Cross-sections of bead-on-plate samples were metallographically characterised to detect cracks in FZ and HAZ.
