1. Introduction
The recent rise in the adoption of electric vehicles (EVs) has heightened the need for vehicle structures that are both lightweight and high in strength. The incorporation of high-capacity battery systems results in EVs being approximately 200 to 300 kg heavier than their internal combustion engine counterparts (ICEVs), thereby amplifying inertial forces during impact events and increasing cyclic mechanical loads during regular operation1,2). The additional loading promotes the buildup of cyclic and transient mechanical stresses within the vehicle structure, and under impact conditions, this can induce pronounced stress concentrations, thereby heightening the probability of early crack initiation or structural failure. When EVs are developed on ICE- based platforms, suboptimal battery weight distribution often leads to concentrated collision loads, thereby increasing the risk of localized structural damage or partial body collapse3,4). Accordingly, enhancing both flexural stiffness and point stiffness of the vehicle body has become critical for guaranteeing driving stability and crash safety. Various structural enhancement strategies have been pursued in both industry and academia, such as incorporating lightweight alloys (e.g., aluminum, magnesium), composite materials, ultra-high-strength steels, and optimized structural design. However, these methods are constrained by practical challenges, including intricate design complexities, high manufacturing expenses, and limitations related to platform adaptability1,2). Accordingly, this study focused on a localized reinforcement approach that preserves geometric design flexibility while minimizing manufacturing complexity through the application of conventional arc-based deposition techniques. Arc-based deposition is widely applied to structural components demanding strength and dimensional accuracy, including body frames, cross members, and exterior panels. A typical example is cladding, which enhances corrosion and wear resistance by overlaying dissimilar metals such as stainless steel onto the base metal5). Cladding is considered an effective local reinforcement technique, as it enhances surface properties without necessitating alterations to the underlying structural configuration. Arc- based welding technology has recently gained prominence through its evolution into Wire Arc Additive Manufacturing (WAAM), which enables the fabrication of metal components through layered deposition. WAAM offers cost-effective structural reinforcement with high process flexibility, accommodating diverse materials and geometries without major equipment changes6). The WAAM process effectively enhances quality and production efficiency in metal additive manufacturing and is considered a promising approach, particularly for reducing weight and improving the durability of automotive components6-8). WAAM technology effectively strengthens point stiffness and impact resistance while preserving the original structure, offering practical benefits in design flexibility and production efficiency by enabling direct fabrication of reinforcement beads without redesigning the structure. A previous study (Josten et al., 2020) experimentally validated the enhancement of structural rigidity through the application of grid-patterned reinforcement beads on flat surfaces and gusset-shaped beads at corners via the arc-based deposition technique6). This study employed regularly spaced reinforcement beads to strengthen flat areas and utilized a gusset structure to alleviate stress concentrations at joints and corners. However, the approach was confined to simple geometric shapes, limiting the optimization of reinforcement design, deposition path efficiency, and the verification of structural integrity in real vehicle applications. Therefore, this study newly developed a reinforcement bead geometry to address structural requirements of both flat and corner surfaces, alongside a deposition strategy employing the Cold Metal Transfer (CMT) process, renowned for its exceptional arc stability at low heat input. Furthermore, various experiments were carried out to confirm the structural integrity of the designed reinforcement and its suitability for practical vehicle applications.
2. Experimental Method
This study utilized 2 mm thick SAPH 440 plates as the base material, with ER-70 filler metal of 1.0 mm diameter. The chemical composition is detailed in Table 1. The base material employed in this study was SAPH 440 plate with a thickness of 2 mm9). ER-70 welding wire (1.0 mm in diameter) manufactured by KISWEL was used as the filler material, and the chemical compositions of both the base and filler materials are detailed in Table 1. To assess stiffness based on the reinforcement bead geometry, flat specimens were prepared in accordance with ISO 7438 by bending the plate to dimensions of 200 mm × 60 mm × 2 mm, while corner specimens were fabricated by bending the plate to 150 mm × 60 mm × 2 mm at an angle of 100°10). In the arc welding process used to form the reinforcement bead on the 2 mm thick specimen, the Cold Metal Transfer (CMT) mode was applied to minimize deformation caused by accumulated heat input, as it offers excellent arc stability even under low current conditions. Welding was performed using an automated setup, as shown in Fig. 1, consisting of a Fronius Transpuls Synergic 3200 power source and an industrial robot arm manufactured by Yaskawa. A shielding gas composed of 90% argon and 10% carbon dioxide was delivered at a flow rate of 15 L/min.
Table 1
Chemical composition of base metal and welding wire (wt%)
| Materials | C | Si | Mn | P | S | Al | Fe |
|---|---|---|---|---|---|---|---|
| SAPH 440 | 0.21 | 0.3 | 1.5 | 0.003 | 0.025 | 0.01 | Bal. |
| ER70S-6 (1.0 Ø) | 0.07 | 0.86 | 1.53 | 0.012 | 0.07 | - | Bal. |
Fig. 1
Experimental setup for arc-based deposition using a CMT welding system with robotic control and Ar-CO2 shielding, applied to SAPH 440 sheets for reinforcement bead formation
A three-point bending test was conducted to assess the effectiveness of the localized reinforcement. The test was performed using a KSU-20M universal testing machine, following the guidelines specified in ISO 7438 to determine the appropriate test conditions. The support span length (l) was set to 40 mm based on Equation (1), while the lower die length (D) was configured at 30 mm, and the test was conducted at a speed of 5 mm/min. Fig. 2 illustrates a schematic diagram of the bending test apparatus utilized in this study, with a denoting the specimen thickness11).
The experiment utilized three distinct reinforcement bead geometries, designated as the GRID pattern (GR), the Two-layer pattern (TLR), and the ‘S’ shape pattern (SR). The absence of any reinforcement bead was denoted as NR. These patterns are detailed in Table 2. The structural features and design rationale for each reinforcement bead shape are comprehensively discussed in the Experiment Results and Discussion section.
Table 2
Schematic illustration of reinforcement bead configurations
| Sample ID | Description | Pattern |
|---|---|---|
| NR | No reinforcement | - |
| GR | GRID pattern reinforcement |
|
| TLR | Two layer pattern reinforcement |
|
| SR | ‘S’ shape pattern Reinforcement |
|
To evaluate the suitability of the novel SR-shaped reinforcement bead and gusset design for actual vehicle components, dimensional variations of the front side member, which is a component of the car body, were examined before and after reinforcement. The curved and corner sections were reinforced with the two respective bead shapes, and the external geometry was captured using Creaform’s HandySCAN BLACK Elite 3D scanner. The captured 3D shape data were processed using PolyWorks|Inspector software, with alignment performed via the best-fit method across the entire surface based on the pre-reinforcement geometry. Subsequently, surface-to-surface deviation analysis was conducted utilizing the Measure Deviations from Surfaces function. The results were presented as a color map, facilitating a quantitative assessment of the local thermal deformation distribution induced by the applied reinforcement geometry.
3. Experiment Results and Discussion
3.1 Design and Stiffness Assessment of Flat Rein- forcement Bead Geometry
For the experiment, optimal process parameters were determined by combining welding current and welding speed, and were subsequently categorized into low, moderate, and high heat input conditions. Table 3 presents the sequence of welding current and welding speed values corresponding to each condition, while Fig. 3 illustrates the representative weld bead profiles and cross-sectional views for each heat input category. The normal weld bead displayed a uniform width across the top surface and maintained a consistent cross-sectional geometry. The absence of surface imperfections confirmed that the welding process had been performed under optimal and defect-free conditions. Under low heat input conditions, humping beads formed as the molten metal failed to flow forward smoothly beneath the arc, instead pooling at the rear and solidifying in repeated humps along the weld path. These characteristics are generally attributed to inadequate heat input or unstable arc pressure11). At elevated heat input levels, excessive penetration occurred, causing the molten metal to reach and bulge out from the underside of the lower plate as a result of excessive thermal input. To identify the optimal bead shape among acceptable formation conditions, flexural stiffness was evaluated using Equations (2) and (3), where E denotes the elastic modulus, I is the second moment of area, b represents the bead width, and h signifies the bead height. Under each condition, three reinforcement beads, each 200 mm in length, were fabricated in parallel. The bending test was then performed three times for each sample, and the average flexural stiffness and maximum bending load values derived from these tests are summarized in Table 4.
Table 3
Welding current and speed combinations used to evaluate bead formation quality under various deposition conditions
Fig. 3
Types of bead shape at different welding condition (a) Normal bead, (b) Humping bead, (c) Excessive penetration
Table 4
Comparative analysis of bead geometry and resulting flexural performance under varying welding conditions
The comparison of bending loads across welding conditions (Table 3) revealed superior load-bearing capacity at a current of 80 A and a welding speed of 30 cm/min. The bead formed at the optimal welding parameters (80 A current, 30 cm/min welding speed) demonstrated a greater cross-sectional area than those formed under other conditions, indicating enhanced load distribution performance. Flexural stiffness, K, is generally defined by Equation (2) as the product of the elastic modulus (E) and the second moment of area (I):
where, the second moment of area I is calculated based on the bead width (b) and height (h), assuming a rectangular cross-section, as shown in Equation (3):
Thus, Given constant the elastic modulus (E),[qls771] increasing the bead height (h) or width (b), or expanding the cross-sectional area, leads to a higher second moment of area (I), which in turn enhances the structure’s resistance to bending. Given these geometric benefits, the combination of 80 A welding current and a speed of 30 cm/min is identified as the most effective condition, providing an optimal balance between mechanical strength and process reliability in bead formation12).
This study introduced a redesigned flat reinforcement bead based on the optimal welding parameters, 80 A current and 30 cm/min speed, and conducted a comparative analysis of its structural characteristics against the traditional GR geometry. The GR geometry used for comparison was originally proposed by Josten et al. (2020). As illustrated in Fig. 4(a), it consists of horizontally aligned long reinforcement beads, with shorter vertical beads interspersed between them6). This geometry is limited by the excessive number of reinforcement beads, which complicates the fabrication process and prolongs the welding process. Accordingly, a new two-layer reinforcement bead pattern (TLR) was developed to maintain equivalent structural rigidity while reducing the total number of beads required. Bending stiffness was theoretically shown-via Equations (2) and (3)-to depend on the cross-sectional geometry. Based on this relationship, the TLR shape illustrated in Fig. 4(b) was developed and its performance assessed through bending experiments. The bending test results are presented in Fig. 5(a), showing that the NR specimen, which had no reinforcement, exhibited the lowest maximum load at around 2.6 kN. The GR shape exhibited a maximum bending load of 9.28 kN, representing an increase of 60% over the NR specimen. The TLR shape further improved performance, reaching 10.38 kN, which is-an additional increase of about 11.8% compared to the GR shape. This result is attributed to the differences in load distribution and structural continuity associated with each reinforcement geometry. Fig. 5(b) illustrates the GR-shaped specimen, and Fig. 5(c) shows the TLR-shaped specimen. The distinct differences in structural response are clearly visible through the geometries of the reinforcement beads and the corresponding deformation patterns exhibited during testing. Notably, the TLR shape maintains continuous connections between the reinforcement beads, which facilitates broader load distribution and enhances structural stability by preventing buckling under stress. Conversely, the GR geometry exhibited limited spacing between vertical beads and inadequate continuity at the joints, which resulted in stress localization and premature buckling under increased loading. Thus, the TLR shape introduced in this study is considered a highly effective reinforcement strategy, offering enhanced bending stiffness with fewer bead applications, thereby improving both structural performance and manufacturing efficiency.
Fig. 4
Schematic diagrams of reinforcement bead patterns, (a) GR pattern with alternating long horizontal and short vertical beads5), (b) Proposed TLR pattern designed for enhanced stiffness with reduced bead count
Fig. 5
(a) Bending test results of NR, GR, and TLR patterns, (b) Reinforced sample with GR pattern, (c) Reinforced sample with TLR pattern
The TLR design boosts flexural stiffness by enlarging the second moment of area under bending loads applied orthogonally to the weld orientation. Conversely, when the bending load aligns with the welding direction, the beads function independently with minimal structural integration. This results in localized stress accumulation and a decrease in the overall effective bending stiffness. To overcome these limitations and enhance both stress distribution and load transfer efficiency across multiple directions, the SR shape was developed with a curved path and a symmetrical structure on both sides. The design enables effective stress distribution along a curved path and enhances structural cohesion by maintaining uninterrupted connectivity through the central region. Fig. 6 presents the development and production process of the SR-shaped reinforcement bead, where images (a) to (c) show the step-by-step deposition of beads following a curved path. Under optimal welding conditions (80 A current and 30 cm/min speed), three parallel reinforcement beads were first deposited as shown in Fig. 6(a). Then, curved “S”-shaped beads were deposited on both sides, as demonstrated in Fig. 6(b) and (c). Subsequently, the final shape was completed by enhancing the continuity between adjacent beads, achieved by increasing the current to 87 A in the central area to fuse the joints, as shown in Fig. 6(d). Fig. 7 presents a comparison of the bending test outcomes for the TLR and SR reinforcement patterns. The TLR pattern demonstrates a maximum bending load of 10.4 kN, with effective load distribution and enhanced structural rigidity achieved through strong cross-sectional connectivity between the beads. In contrast, the SR pattern achieved the highest load-bearing capacity between the two designs, reaching a maximum bending load of 13.9 kN. The applied load continued to rise consistently up to a displacement of about 33 mm, indicating stable and sustained load transfer throughout the process. This outcome is attributed to the smoothly curved and continuous bead path, which effectively eliminates abrupt geometric transitions along the stress flow path and ensures even distribution of stress across the entire structure. Such a characteristic proves the structural efficiency of the bead design based on a curved geometry.
3.2 Revised Gusset Design for Improved Corner Reinforcement
The gusset shape employed in the previous study (Josten et al., 2020) involves depositing a total of 16 layers of reinforcement beads, making the manufacturing process less practical because of the high number of layers involved6). Therefore, this study focused on refining the shape to minimize production time and enhance the ease of the manufacturing process. To reduce irregularities in bead shape and discontinuities between layers that occur during multi-layer deposition, a flattening treatment was applied to the edge area prior to the deposition of reinforcement beads. The surface control method is recognized as an essential technique for improving shape accuracy in the WAAM process, while a multi-bead overlapping approach that minimizes interference between beads has also been shown to achieve similar improvements13,14). To examine the influence of deposition direction on shape stability, Bead- On-Plate (BOP) welding was conducted prior to the deposition of the gusset structure. As shown in Fig. 8, continuous deposition in a single direction led to a progressive increase in bead shape inconsistency and interlayer discontinuity as the number of layers grew. The underlying cause is the repeated accumulation of arc strikes at the starting point of each layer, where the intense heat and localized current from arc ignition are concentrated on the same spot. This leads to shape instability, such as excessive melting, bead widening, and the development of step-like formations. A prior study (Shukla et al., 2020) also demonstrated that in the CMT-based WAAM process, depositing multiple layers along the same path results in a gradual increase in bead width during the initial stages, along with a decline in shape uniformity. These findings further highlight the detrimental effects of arc strike accumulation on the overall shape quality15). In contrast, applying a cross-directional deposition approach dispersed the starting points, thereby mitigating the accumulation effect and leading to a more consistent bead shape and enhanced deposition stability.
Fig. 8
Bead shape according to stacking direction, (a) repeated stacking from the same start point (Fixed-start stacking) (b) stacking with alternating start points from left and right sides (Alternating-start stacking)
A weaving deposition technique was applied in the edge region to enhance initial deposition stability. The weaving technique generates broad, low-profile beads and has proven effective in minimizing surface irregularities resulting from gravitational effects and heat accumulation, particularly within the WAAM process. Earlier studies (Bultman & Saldaña, 2023; Ji-Young Shin, 2024) demonstrated that adjusting the amplitude and frequency of the weaving path leads to improved geometric accuracy and smoother surfaces. Likewise, Aldalur et al. (2020) highlighted that layering methods using triangular or vibrational weaving contribute to stabilizing the initial base shape and ensuring consistent shape alignment in the following layers15-17). To fine- tune the weaving parameters, amplitude (1.5 to 3.0 mm) and frequency (2.0 and 5.0 Hz) were varied, and BOP tests were carried out with a welding current of 80 A and a welding speed of 30 cm/min. Fig. 9 presents the cross-sectional profile and a conceptual illustration of the reinforcement bead under different weaving conditions. The definitions of valley depth and pitch distance are provided together, accompanied by a visual comparison of shape variations resulting from different amplitude and frequency combinations. The experiment revealed a notable reduction in pitch distance with increasing frequency, alongside a tendency for the bead surface area to grow as the amplitude was raised. This trend is quantitatively detailed in Table 5 as well as Fig. 10(a) and 10(b). As frequency rises, valley depth remains relatively stable, whereas pitch distance decreases substantially, resulting in an overall increase in the valley depth-to-pitch distance ratio. After thoroughly evaluating cross-sectional shape stability and deposition efficiency, an amplitude of 1.5 mm and a frequency of 5 Hz were identified as the optimal conditions. Using the determined weaving parameters, a redesigned gusset was created to improve deposition efficiency over the previous gusset configuration. The new design incorporates a combination of the cross- deposition method and the weaving process, following a path where the left side is deposited first, then connected and welded to the right side. During the initial four layers, flattening is achieved through weaving, after which deposition continues using the 3-point line method. Fig. 11(a) illustrates the stacking sequence and a summary of the process. A total of eight deposition layers were used to successfully form a geometric structure resembling the original gusset design. The final fabricated sample is presented in Fig. 11(b), and the total production time was found to be reduced by around 66.6% compared to the previous design.
Fig. 9
Process condition window for weaving deposition: effects of amplitude and frequency on bead surface morphology, including valley depth and pitch distance
Table 5
Results of cross-section observation according to weaving variable
3.3 Deformation Analysis of SR-Shaped Reinforce- ment Beads and New Gusset Using Surface Profile Measurements
The prior deposition experiments on the new SR- shaped reinforcement bead and new gusset served as preliminary tests to optimize process conditions and understand the mechanisms ensuring shape stability using flat specimens. The next step involves assessing potential local deformation when applied to actual vehicle components and quantitatively evaluating their practical applicability. In this study, the two reinforcement designs were implemented on the front side member, which is an actual automotive component, and 3D surface measurements were conducted. Using PolyWorks software, the shapes before and after bead deposition were aligned, allowing for a detailed quantitative analysis of the resulting deformation (Fig. 12). According to a previous study (Ninshu Ma et al., 2020), metal expands when heated during welding and contracts upon cooling. When this expansion and contraction are unevenly constrained, it results in residual stresses and localized deformations18). Specifically, when jig restraints are applied, overall deformation is limited; however, stress tends to concentrate in certain areas, potentially causing localized adverse deformation. Indeed, the evaluation showed that the right SR-shaped bead welded under jig restraint experienced suppressed cooling due to thermal expansion, resulting in an internal shrinkage strain of -1.587 mm. In contrast, the unconstrained SR-shaped bead on the upper left area of the component, where clamping was not possible, exhibited unrestricted expansion and contraction, leading to an outward deformation measuring +1.552 mm. It indicates that the shape expands outward freely, without any restriction on thermal expansion. Despite being based on the same design, the deformations under contrasting constraint conditions were nearly symmetrical at +1.552 mm and -1.587 mm, indicating that the new SR-shaped reinforcement bead can be precisely and uniformly implemented on real curved parts. Addition- ally, only slight deformations of +0.002 mm and -0.005 mm were observed in areas beyond the weld zone, verifying that the base material of the new gusset underwent minimal distortion due to the heat input reduction provided by the weaving process.
4. Conclusions
This study presents an innovative reinforcement bead geometry based on arc deposition, aimed at overcoming the shortcomings of conventional reinforcement methods and enabling more efficient enhancement of local stiffness, along with the optimization of suitable process conditions.
1) The new SR-shaped reinforcement bead achieves high structural rigidity with fewer beads by employing a curvature-driven continuous path design. Bending test results showed that the maximum load capacity of the TLR shape increased by 11.8%, reaching 10.4 kN compared to the conventional GR shape, while the SR shape exhibited a 49.6% improvement, achieving 13.9 kN. Notably, the SR shape minimized interruptions in the stress transfer path by incorporating a continuous bead trajectory and a central joint, effectively dispersing stress over a larger area. This design enabled stable load-bearing behavior without signs of buckling. This highlights the ability of the new geometry to simultaneously reduce stress concentrations, improve structural performance, and lower the bead count required for reinforcement.
2) To improve the manufacturing efficiency of the traditional gusset design, a redesigned gusset was introduced, combining a weaving process with a cross- deposition approach to ensure initial surface flattening. Among the tested weaving conditions, an amplitude of 1.5 mm and a frequency of 5 Hz delivered superior shape stability and deposition efficiency. Using these parameters, the newly developed gusset successfully replicated the geometry of the original design with only eight layers. As a result, the total production time was reduced by 66.6%.
3) To validate the suitability of the new reinforcement designs for real-world components, the SR-shaped reinforcement beads and gusset were applied to the front side member, and deformation was examined through 3D surface scanning and analysis. The SR shape displayed local deformations of -1.587 mm in the constrained region and +1.552 mm in the unconstrained area. The similarity in deformation magnitude under opposite conditions demonstrates the reliability and repeatability of the shape’s implementation. The new gusset similarly showed minimal distortion of the base material, confirming the effectiveness of the weaving process in reducing heat input.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
