Process Development and Analysis of Electromagnetic Multi-Spot Joining of AA5052 Sheets

Abstract

In this study, two sheets of AA5052 are joined with the high-strain-rate multi-spot joining process using an electromagnetic system. While producing a single spot joint by electromagnetic joining (EMJ) is common, the distribution and application of the pressure can be modified by the design of the coil and spacers to make multiple joints at once. When a preformed dimple is used to provide the standoff distance, it can eliminate the need for spacers and provide good aesthetics for the final product. In the current study, a joint design is developed to provide three spot joints coincidentally by a single discharge of a capacitor bank. For the experiment, four distinctive discharge energies were used for joining: 7, 8, 9, and 10 kJ. The most successful joint sample was made by 8 kJ and was tested for mechanical properties. The cross-section was observed in order for us to understand the joint quality produced by the process. It was found that the ”I”-shaped rectangular coil produces a variable magnetic flux, leading to different flyer deformation variations in the joint geometry. At the centre of the ”I” coil, the minimum flux was predicted, leading to lesser sheet forming, hence a weaker centre-spot joint strength. Further, a numerical study is performed to find the Von Mises stresses, equivalent plastic strain, impact velocity, and impact pressure on the sheets. This manuscript provides new information regarding coil designing and the changes that could be further made to improve the electromagnetic sheet multi-spot joining process.

1. Introduction

The use of aluminium alloys is continuously increasing because of their high strength-to-weight ratio and versatility. In the automotive industry, in particular, using light metals such as aluminium for structural applications helps reduce weight and increase fuel efficiency [1]. Long et al. [2] demonstrate the formability of high strength aluminium alloys depending on the working temperature and suggest aluminium for most automotive closures and body-in-white applications.

Over the decades, fusion-based welding technologies such as arc welding and resistance welding have been two major techniques used for producing permanent joints between steel sheets for their versatility and inexpensive operation costs. However, the joining of aluminium by conventional welding processes is often challenging due to its relatively high electrical and thermal conductivity. For example, resistance spot welding of aluminium alloys requires a higher electrical input which in turn requires costly instrumentation and frequent replacement of the electrodes. Some of the challenges of gas metal arc welding of aluminium alloys are discussed by Praveen and Yarlagadda [3]. These fusion-based techniques also significantly modify the weld interface and microstructure of the metals, thus controlling the heat-affected zones (HAZ), while solidification of cracks is constantly required.

Other alternative techniques for joining high-strength aluminium alloys include friction stir welding (FSW). Patel et al. [4] explain the major process parameters that need to be considered for weldability when joining high-alloyed aluminium alloys such as the 5xxx and 7xxx series. According to their study, the weldability of the alloys can be largely affected by the amount of heat generation and structure modification which can be controlled by the process setups (tool geometry, offset, etc.) and parameters (rotation and welding speeds). While high-strength joints can be made by FSW, it still faces challenges of structural modification by generating heat-affected zones in the base metals. The control of these regions needs to be considered when the product requires high structural integrity.

Electromagnetic pulse spot welding (EPSW) for the single spot joints has been explored by many researchers. Deng F. et al. [5] performed EPSW of AA1060 and SS304 sheets with a field shaper. Wu H. [6] successfully joined Cu/Al alloy using two-step EPSW with an I-shaped copper coil. Further different mechanical joining processes such as clinching [7], self-piercing riveting [7,8], and screwing can also be used. But these processes have limitations as they require extra steps in the production cycle, additional weight of the rivets, and stress concentration at the joining edges.

Electromagnetic joining (EMJ) is an innovative process that uses a high-speed forming of sheet metal of one side, called the flyer, towards another sheet metal to produce a high-strength joint. The flyer material experiences the high-strain rate forming and clinching by the lightning-speed discharge of a capacitor bank through a coil. A thorough review of the electromagnetic forming and its variations is provided by Psyk et al. [9]. Electromagnetic joining is a preferred method for joining thin conductive metals due to its high speed, cleanliness, environmental friendliness, reproducibility, durability, automation capabilities, longer tool life, and low operating costs. Recent development in the EMJ process has been made through which a non-conductive material can also be joined by providing the impact velocity through the conductive driver [10].

The rapid forming and clinching of the metals not only reduce the cycle time but also induce the local strain hardening of the joint interface and its periphery. At the time of the discharge, the stored electrical current in the capacitor is released during tens of microseconds (µs) through the coil or inductor, as shown in Figure 1. The excited coil generates a Lorentz force to drive one sheet metal (flyer) towards the stationary material (target). The joint provides excellent mechanical properties as the material is usually strengthened by local plastic deformation of the base metals without generating HAZ [11,12].

Another advantage of EMJ is that the type and shape of the pressure can be modified by the coil design; it can focus the pressure in a relatively small region to mimic the size of the typical resistance spot weld, but it can also distribute the energy over an elongated region to make a seam weld. Li et al. [13] provide both empirical and numerical analyses on magnetic pulse welding (electromagnetic joining with metallurgical bonding) of aluminium to titanium by an innovative design of the multi-seams coil. Psyk V. et al. have performed destructive and non-destructive testing of Cu-DHP, S355J2, and EN AW 6082 material joints made using the magnetic pulse welding (MPW) process [14].

The versatility of the coil tool can also be used to make multiple spot welds at a time, wherever the adequate standoff distance is allowed to accelerate the flyer for joining. If the pressure generated by a single capacitor discharge is enough to produce multiple joints, it can reduce the cycle time and increase the structural integrity at the same time. While the single-spot joining by the magnetic pulse has been reported by many [5,15,16], the idea of the multi-spot joining has been much less explored by researchers. The simultaneous joining of multiple spots can improve the uniformity of the structure and property of the weld interface by the small joint design, rather than the enlarged or elongated one.

In the present study, we designed the joint between two sheets of AA5052 such that three spot joints can be obtained in a single capacitor discharge. AA5052 is a common automotive aluminium alloy that is usually processed by plastic working. The discharge energy is varied for making the joints to find out the optimal input energy to achieve high joint strength and avoid the tearing of the flyer by excessive energy. The mechanical properties are analysed by standard testing procedures and the structure of the joints is observed by optical microscopy. Further numerical analysis using the ABAQUS explicit module is performed to find the Von Mises stress, equivalent plastic strain in the flyer sheet, impact velocity, and impact pressure at the interface to understand the physics behind the high-strain-rate multi-spot joining process.

2. Materials and Method

2.1. Experimental Study

AA5052 sheets of 1 and 1.5 mm thickness were used for the flyer and target materials, respectively. The properties of the material were examined experimentally within the laboratory: the tensile strength of the sheet is 232 MPa, with an elastic modulus of 70.3 GPa and 17% elongation at the break. Dimensions of the workpieces are shown in Figure 2. On each target, three slots of a 6 mm diameter and 1 mm depth were made using the CNC milling machine, to provide the standoff distance for the flyer travel. The distance between the slots was 6 mm as per the gauge length of the coil, and both sheets were clamped up to a 10 mm edge distance from the edges.

The joining tests were performed by a customized electro-magnetic forming instrument having a total energy-storing capacity of 10.5 kJ. A schematic diagram of the setup is presented in Figure 3. It consists of a single I-shaped coil with a rectangular cross-section. Figure 4 shows the design and dimensions of the coil used for the joining tests in the EMJ system. The coil was fabricated using a wire electrical discharge machine (W-EDM) for precision. Different input energies ranging from 6 to 10 kJ were used for joining.

As EMJ uses an intense pressure to generate an impact, an intense magnetic pressure must be created by the copper coil when the electrical current flows through it. By using a circular slot which is pre-machined, it eliminates the need for removing the spacers after the process and assists in the consistency of the appearance. A circular slot was preferred over a rectangular shape due to leak resistance and uniform stress distribution during deformation [17]. In the current study, the target plate was pre-machined with three slots of 1 mm depth in order that the joint regions of the flyer sheet could accelerate, as shown in Figure 5. The space in the target plate acts as a standoff distance for the metal to achieve the desired impulse forming of the flyer. Figure 6 shows the discharge current sinusoidal waveform for the 8 kJ discharge energy, as monitored using a Rogowski coil.

2.2. Numerical Study

The numerical study to calculate the impact velocity of the flyer sheet, Von Mises stresses after deformation, equivalent plastic strain, and impact pressure at the interface was performed using ABAQUS software. The geometry of the target plate was drawn with partitions to accommodate variable mashing in different parts.

After drawing the flyer sheet and target plate in ABAQUS as shown in Figure 7, the elastic property such as Young’s modulus of 70.3 GPa, Poisson’s ratio of 0.3 and density of 2680 kg/m³ were assigned to the material. The following constitutive equation for the Johnson–Cook damage model is used to define the plastic behaviour of the material [18].

$$\mathsf{\sigma }=(\mathrm{A}+\mathrm{B} {\in }^n)(1+\mathrm{C}\ln \dot{{\in }_p})(1-T^{\ast m})$$

(1)

where σ is the true stress, $\in$ is the true strain; A, B, C, n, m are material constants in which A is the yield stress, B and n denote the strain hardening effect, C is the strain rate constant, and m is the softening exponent. ${\in }^n$ = equivalent plastic strain, $\dot{{\in }_p}$ = plastic strain rate which is the ratio of the strain rate to the quasistatic strain rate for the strain rate ($\dot{\in }$) = 0.01/s. In this paper, the temperature effects are neglected due to the very small impact time i.e., less than 64 µs so only the strain and strain hardening effects are considered.

$$T^{\ast m}=\frac{T-T_{room}}{T_{melt}-T_{room}}$$

(2)

where T_room is the room temperature and T_melt is the melting point temperature of AA5052.

The following rate-dependent Johnson–Cook model was used for the plastic material property of AA5052 (

Table 1. Johnson–Cook model parameters for AA5052 for plasticity.

Parameter	Density (kg/m3)	E (MPa)	G (MPa)	Poisson Ratio	A (MPa)	B (MPa)	n	C	m	$\dot{{\in }_0}$
Value	2680	64,000	24,600	0.3	143.1	215.7	0.54	0.0046	0.9	0.01

) and the Johnson–Cook damage model for material damage evaluation (

Table 2. Johnson–Cook damage model parameters for AA5052 for damage evolution.

Parameter	D1	D2	D3	D4	D5	Displacement at Failure
Value	0.306	0.0446	−1.72	0.0056	0	0.012

) [19].

The ABAQUS (dynamic explicit) step was selected for simulation due to the high velocity, nonlinearity, and transient nature of the electromagnetic sheet-metal joining process [20]. The simulation was performed in 2D to save iteration time and energy. Dynamic explicit was used with a step size of 64 µs according to the first positive half of the damped sinusoidal current. The non-linear geometry (Nlgeom) option was used to accommodate the geometrical nonlinearity of the process. To include the corner effect and easy convergence, the penalty contact method with finite sliding in surface-to-surface contact was kept between the flyer sheet and the target plate. The target plate was kept as the master surface and the flyer sheet was kept as the slave surface because the flyer sheet is supposed to go through more deformation.

The impact pressure is assumed to be applied according to the first positive half of the damped sinusoidal current profile on the flyer sheet. The input pressure with sinusoidal periodic amplitude was applied on the flyer sheet for a step time of 64 µs and a circular frequency of 49,087 Hertz [21]. While performing the simulation, to replicate the mechanical deformation results observed using optical microscopy, non-uniform pressure is applied.

To replicate the boundary condition of the flyer sheet and the target plate during experiments, the bottom part and sides of the target plate were fixed from the bottom using an encastre boundary condition. The horizontal motion of the flyer sheet was restricted as shown in Figure 8a.

A total of 10,470 linear quadrilateral elements of type CPS4R elements were used for the meshing. The CPS4R element is a four-node bilinear plane-stress quadrilateral, reduced integration, hourglass control element. The reduced integration uses a lesser number of Gaussian coordinates and produces a less stiff element which results in reduced computation time. Enhanced hourglass control was used to control the hourglassing of the element caused due to one integration point at the centre of the element during the use of reduced integration.

3. Results and Discussions

3.1. Mechanical Properties Evaluation of Joints

3.1.1. Energy Optimization for Joints

The joints were generated by the EMJ process with variable discharge energies of 7, 8, 9, and 10 kJ. A photograph of the generated joints is shown in Figure 9. The characteristics of each joint were recorded and are shown in

Table 3. Visual inspection of the joint produced by EMJ at different discharge energies.

Discharge Energy	Outcome (Visual Inspection)	Remarks
7 kJ	Centre spot portion of the joint remains undeformed.	No proper joining due to lesser dissipated energy.
8 kJ	All three spots deformed properly.	Proper joining is due to sufficient energy for deformation.
9 kJ	Shearing in the adjacent spots was observed.	Excessive energy leads to boundary shear of two spot joints.
10 kJ	Shearing across all the spot joints could be observed.	Excessive energy leads to the shearing of all three spot joints.

. It is found that the optimal discharge energy of 8 kJ is required for making all three joints successfully. Discharge energy less than 8 kJ resulted in insufficient forming of the flyer which led to no joining at the central region of the spot. When an energy greater than 8 kJ was used, excessive forming of the flyer occurred, which led to shearing of the joint boundaries.

3.1.2. Peel Test and Lap Shear Tests for Joints

The successful joint samples made using 8 kJ discharge energy were used to test the two different mechanical loading tests: the peel and lap-shear tests. In the peel test, there are two modes by which the joint samples failed. One is the separation failure mode, and the other is the crack failure mode. The joint samples are pulled out after applying the loading condition in the separation failure mode. On the other hand, the base metal and the joint regions are fractured in the crack failure mode. The schematics of the two modes of failure in the peel test are shown in Figure 10.

Two metal plates were used as a clamp to hold the samples of peel tests. The clamp was used to hold the target with the help of the bolted joint. The schematic of the arrangement made for the peel tests is shown in Figure 11. A photograph of the failed sample with the crack failure mode is shown in Figure 12.

The peel test was carried out and the load-displacement curve of the test is shown in Figure 13. After it has reached about 400 N, there is a sudden drop in value and a further increase in the load value to the maximum of about 950 N. The maximum value identifies a proper joint leading to higher plastic deformation, which usually occurs in a high strain rate forming by joining processes. It was also seen that the sample tested at 8 kJ failed on the base metal, leaving behind the unfailed spot joints, as shown in Figure 12.

The schematic of the lap-shear test arrangement is shown in Figure 14. Figure 15 depicts the load versus displacement data during the lap-shear test. For statistical accuracy, three samples were generated and tested. The load-displacement curves show decent reproducibility of the samples in terms of the peak load and elongation until the failure. The first peak is seen at about a 0.22 mm extension with a load value of 3950 N, and the second peak is observed at about 0.27 mm and 3600 N before the load drops. This is attributed to the fact that, during the second peak, the centre spot being relatively weak, started failing, and the total failure was delayed by the adjacent spots ”A and C”.

3.1.3. Optical Microscopic Evaluation of the Joints

The microscopic image of the sample joined by the 8-kJ discharge energy is shown in Figure 16. Three different sections are shown, sections A, B (centre spot), and C, respectively. It can be seen that the central region of the sheet did not form as much as spots A and C. It is assumed that the intensity of the magnetic force is less in the centre of the coil due to the end effect. Spots A and C are expected to contribute most of the mechanical properties of the multi-spot joint produced by EMJ. Hence, the overall joint geometry can be improved by changing the coil shape such that the intensity of the magnetic field becomes higher in the centre region of the coil [22]. Future work on the quantification of the magnetic field may be helpful to optimize the coil design. Adjusting the depth of forming between the spot sections according to the magnetic field intensity may also be another potential solution.

3.2. Numerical Simulation Results

Deformed and undeformed shapes of respective spot joints obtained after the structural simulation in ABAQUS explicit software are shown in Figure 17. The deformed shapes of all three spots are quite similar to images obtained from optical microscopy.

The results of the simulation are discussed below:

3.2.1. Von Mises Stresses after Deformation

Al 5052 sheets are ductile; therefore, the Von Mises failure criterion will give better results.

The Von Mises stress contour is recorded at the 64 µs time step according to the total time of the first positive half of the current profile. The first part of the positive increasing pressure curve is responsible for impact and joining of the flyer sheet with the target plate. The later part of the pressure curve is used to hold the material in the deformed position, which reduces the spring-back effect of the material. The Von Mises stresses contour of spot B and spot C suggests that more stress is at the neck of the flyer sheet and near the milled edge of the target plate due to stress concentration during the deformation of the neck. Among the three spots, the Von Mises stress magnitude for spot A, B and C was 160, 84, and 148 MPa, respectively, at the spot centre at the interface of two sheets. Spot B has the lowest stress corresponding to the stress contour of the joint due to relatively less impact pressure, as shown in Figure 18.

3.2.2. Equivalent Plastic Strain (${\in }^n$)

As the induced stress exceeds the yield point of the material and enters the elasto-plastic region, elastic and plastic strains play a significant role. Elastic strain is responsible for material deformation recovery and plastic strain increases the strength of the material by strain hardening. Equivalent plastic strain for the highlighted lower mid element of the flyer sheet is calculated for all three spots as shown in the small window of Figure 19. In the graph, the maximum equivalent plastic strain for the material is between 0.50 and 0.69 during 13 to 17 µs, and later, after the joining of plates, the equivalent plastic strain becomes constant. The maximum values of the equivalent plastic strain for spot A, B, and C were 0.69, 0.50, and 0.66, respectively. Spot B gives less equivalent plastic strain due to relatively less stress than spots A and C. The equivalent plastic strain and impact velocity are plotted up to 30 µs as there is no change in the magnitude after this time.

3.2.3. Impact Pressure at the Interface

Impact pressure plays a very important role in the joining of materials, and it can be calculated by the following analytical relation [23]

$$\mathrm{Impact} \mathrm{pressure}: \mathrm{P}=\frac{{\mathrm{B}}^2}{2{\mathbf{\mu }}_0} [1-e^{(-\frac{2t}{\mathsf{\delta }})}]$$

(3)

where δ is the skin depth in meters and can be calculated as

$$\mathsf{\delta }=\frac{1}{\sqrt{(\mathsf{\pi }\mathsf{\sigma }\mathrm{\mu }f)}}$$

(4)

and B is the magnetic flux density in Tesla, given by

$$\mathrm{B}=\frac{{\mathbf{\mu }}_0\mathrm{I}[{\tan }^{-1}(\frac{b}{2d})]}{\mathsf{\pi }\mathrm{b}}$$

(5)

where ”I” is the discharge damped sinusoidal current (A), b is the width of the middle web portions of the copper coil (m), and d is the distance of the flyer sheet from the inner surfaces of the web of the copper coil (m); t is time in seconds, σ is the conductivity of the workpiece (m Ω⁻¹), ${\mathsf{\mu }}_0$ is the magnetic permeability of the free space or vacuum (Hm⁻¹), µ is the magnetic permeability of the workpiece (Hm⁻¹), and f is the frequency of the transient current (Hz). For our AA5052 workpiece, the calculated skin depth as 0.5074 mm and the maximum magnetic flux density within the space between the upper and lower parts of the coil was found to be around 10 Tesla at d = 0.

Further, the minimum Magnetic pressure at the given impact velocity for joining can be calculated as [24]

P_min. = ρVC

(6)

where $\mathsf{\rho }$ is the workpiece density kg/m³ and C is the bulk sound velocity of the material.

Figure 20 shows the impact pressure at the interface when the flyer sheet is just striking with the target plate. Different spots show different impact pressure. The impact pressure obtained at the interface of spots A, B, and C was 150, 128 and 142 MPa, respectively. Spot B has less impact pressure due to less impact velocity than the adjacent spots.

3.2.4. Velocity of Flyer Sheet before Impact

Due to the variable nature of the electromagnetic pressure generated by the coil, the impact velocity of the flyer sheet is different at different sections. The impact velocity of the flyer sheet can be calculated by the following formula m [5]:

$$\mathrm{The} \mathrm{impact} \mathrm{velocity} \mathrm{of} \mathrm{flyer} \mathrm{sheet} \mathrm{V}=\frac{{\mathrm{B}}^{2}t}{2{\mathrm{\mu }}_0\mathsf{\rho }\mathrm{s}}$$

(7)

where s is the sheet thickness in m.

Additionally, the minimum impact velocity to join two sheets of similar material is given by [20]:

$${\mathrm{V}}_{\min .}=\sqrt{{\mathsf{\sigma }}_{TU}/C}$$

(8)

where ${\mathsf{\sigma }}_{TU}$ is the ultimate tensile stress.

As shown in Figure 21, the maximum velocity obtained from the numerical study for spot A is 247 m/s, spot B is 202 m/s, and spot C is 222 m/s. This shows that less magnetic flux density is acting on spot B than that on spot A and spot C.

4. Conclusions

The multi-spot electromagnetic joining (EMJ) of AA5052 sheets was introduced in this paper. The magnetic pulse system is utilized for extending the conventional applications of forming and single spot joining. The research contents include the mechanical property evaluation of joints and coupled numerical simulation. Mechanical property evaluation of joints provides valuable information such as the peel test, lap shear test, and optical microscopy; furthermore, the coupled numerical simulation results provide valuable information about Von Mises stresses, equivalent plastic strain, impact pressure, and impact velocity.

The important conclusions of the paper can be summarized as follows:

• The AA5052 sheets, with different thicknesses of 1.5 mm and 1 mm, were multi-spot joined by electromagnetic joining using the 5-mm thick I-shaped rectangular coil.
• Multiple discharge energy varying from 7 kJ to 10 kJ was used for performing the experiments. The 8 kJ input energy is found to be a suitable discharge energy for the joining, where 950 N of the load-bearing capacity for the peel test and 3950 N for the lap-shear test is achieved.
• When the discharge energy was lower than 8 kJ, an improper joint was obtained due to insufficient forming of the flyer sheet. When it was increased beyond 9 kJ discharge energy, shear cracks were observed at all three spots, indicating the excessive use of the energy.
• Upon the peel test, the failure modes of the joints were divided into two categories, i.e., fracture failure and separation failure. In 8 kJ discharge energy, the sample was observed to have fracture failure, justifying the strength of the joints as fracture failure can withstand the higher load.
• The experimental and numerical studies suggest that less deformation of spot B is due to less impact pressure and impact velocity, where impact velocity is directly proportional to the square of the magnetic flux density. The magnetic flux density depends on the dimensions of the coil; therefore, proper deformation can be achieved by optimizing the coil design for the increased magnetic flux density.

The overall objective of the study is to explore and extend the possible application of electromagnetic multi-spot joining in the sheet-joining process. The new proposed works consider the circular spots created using the milling machine. The further development of the process needs to consider the pressure distribution produced by the coil. The three spot joints show variations in the structure and property due to the difference in the intensity of the magnetic force. Either the varied depth of slots depending on the intensity, or the modified coil design can solve the problem and produce joints with higher performance. Only structural analysis was carried out in this study and there is further scope for improvement that can be done by using tightly coupled simulation for electro-mechanical analysis.