Page 37 - MSAM-4-3

Materials Science in Additive Manufacturing Numerical simulation of plasma WAAM for Ti-6Al-4V

Table 4. Melt pool ratio to the overestimated depth of the pre-heating pass, which
significantly influences the results.
Case Preheating Single bead Single bead+pre‑heating
Experiment 4.06 3.31 3.03 The trial-and-error approach used in this study, which
16
Simulation 2.22 3.67 2.45 was also used by Aarbogh et al., involves iterative tuning of
input parameters to fit experimental data. While common,
this method is time-consuming, computationally intensive,
temperature overestimation in WAAM simulations due to and requires considerable expertise. To address these
thermocouple placement and data acquisition limitations. challenges, Gu et al. proposed a technique to extract weld
18
For pre-heating (Figure 6B), lower energy input per pool shape parameters directly from simulations. However,
unit of length resulted in more moderate temperature this technique primarily calculates the depth and width,
peaks. Again, the peak temperatures are overestimated by with the length of the front and rear ellipsoids estimated as
the simulation, but the cooling behavior of the simulation multiples of the measured width (a = 0.5W and a = 2W ).
r
m
m
f
16
matches the experimental trends within 100 s of reaching Aarbogh et al. estimated the front and rear ellipsoid
the peak temperatures, confirming the thermal boundary lengths with a = W/2 and a = W. Unlike width and depth,
r
f
conditions used. which can be determined directly from micrographs of the
weld cross section, the exact determination of the length of
Figure 6C shows the thermal response for single bead
deposition with pre-heating, where the simulated and the front and rear ellipsoid is more difficult to determine
experimentally and are, thus, often approximated by
experimental profiles are in good agreement. The pre- equations. Chujutalli et al. concluded that the parameters
39
heating pass raises the temperature of the weld region to of the heat source influence the weld bead size collectively
~400°C, reducing thermal gradients and enabling stable rather than independently.
fusion at the weld interface. The highest measured peak
temperature at TC1 (~1,000°C) exceeds the β-transus The numerical model developed in this study for
temperature of Ti6Al4V, resulting in microstructural simulating the thermo-mechanical behavior of single-
transformation, particularly within the HAZ. In a single track deposits is distinguished by its use of a fully transient
weld bead, the final HAZ occurs at the location where Goldak double-ellipsoid heat source formulation calibrated
the peak temperature rise is approximately equal to directly against experimental thermal profiles and melt
the β-transus temperature. This was confirmed by pool geometries. This contrasts with other modeling
36
microstructural analysis, where the HAZ half-width of strategies presented in the literature that focus on
8.8 mm closely matched the location of TC1 (8 mm from computational efficiency and parametric generalization.
the centerline). The relationship between temperature Yang et al. proposed a semi-analytical model with
40
history and phase transformation is critical in Ti6Al4V temporal acceleration to efficiently simulate residual
processing. 11,37,38 stress and deformation by accelerating heat source
movement and adjusting diffusion time, which allowed
4. Discussion significant reductions in simulation time while preserving
4.1. Heat source calibration predictive accuracy for large-scale WAAM parts. Strobl
et al. employed reduced order modeling using proper
41
In structural welding simulations, an accurate generalized decomposition to enable fast calibration and
representation of the heat source is paramount. The shape parameter estimation through efficient representation of
and the heat flux distribution of the heat source must be temperature evolution in bead-on-plate welds. This method
calibrated against experimental measurements, since lays the groundwork for a future digital twin architecture to
the heat source is as a boundary condition rather than a support real-time process control in WAAM. In contrast,
result of the simulation. Accurate heat source definition is Wang et al. developed a purely geometric model for
42
essential for reliable simulation results. 18 directed energy deposition based on the in-order stacking
Table 5 shows the deviations between measured and of primitives. This model approximates the buildup of
simulated fusion zone parameters. The largest deviations deposition geometry by stacking idealized bead segments
occur during pre-heating, with deviations of up to 26% defined by simple mathematical rules. While highly efficient
in width and 42% in depth. In contrast, the weld pool computationally, the model does not simulate heat transfer
shape of the single bead closely matches the experimental or mechanical responses, and instead is tailored for rapid
measurements. The calculated fusion zone for the single predictions of deposition shape and volume, particularly
bead with pre-heating matches well in width but deviates useful for path planning and early-stage process evaluation.
by up to 28% in depth. This deviation is primarily attributed Recent studies have highlighted the importance of dynamic

Volume 4 Issue 3 (2025) 10 doi: 10.36922/MSAM025140021