Magnetic Pulse Hybrid Joining of Polymer Composites to Metals

Abstract

To lighten their vehicles, car manufacturers are inclined to substitute steel structures with aluminum alloys or composites parts. They are then faced with the constraints inherent to dissimilar (galvanized steel/aluminum) or hybrid (metal/composite) assemblies. Recent developments in magnetic pulse welding seems to offer a viable route. Very fast, this process can be robotized and generates a very localized heating system which limits the formation of intermetallic and damage the composite. Low energy consumption, without filler metal or smoke it is recognized as an environmentally friendly process. In this paper, electromagnetic pulse welding is exploited to assemble polymer composite to metals. Two techniques, a metallic insert in polymer composite or an external patch, have been tested with possible design considerations.

1. Introduction

Nowadays and for mid-term future in automotive industry, steels (60%), aluminum (18%), plastics and fiber reinforced polymer composites (FRPC) (8%) remain main materials of application [1,2]. The latter group is attracting attention due to the advantages it can present from a weight-performance ratio point of view. Therefore, aluminum-FRPC, steel-FRPC or aluminum-steel-FRPC component groups can coexist during the design and fabrication stages.

Polymer composites to metals traditional joining techniques, i.e., mechanical joining and the adhesives, remain widely used. In fact, they are the most well-known historically and their theoretical and technical tools are precisely developed. In mechanical joints such as fastening, hole-clinching, etc., the main limitations are the additional weight and the additional steps during manufacturing as well as their effect on the FRPC [3,4,5,6,7,8]: risk of water intrusion, lamination in the joining area, fibers breakage, and the high stress concentration. In the adhesive bonding case [4,5,9,10,11,12,13], the main limitations remain the long surface preparation coupled with cure times which can increase the production cycle, the safety conditions for the operators, the environmental impact and finally the limitation of the joint mechanical performance.

To avoid these disadvantages, multiple techniques were developed and still undergoing further development. One of these is the combination of the two previous traditional ones which called the hybrid joints processes aiming to combine the advantages especially in improving mechanical performances. To mention some of these techniques: bonded/bolted hybrid joints [14], co-curing techniques [15,16,17,18,19,20,21], bolted/co-cured hybrid joining method [22], advanced pinned hybrid joints [23,24]. The cycle time in the above processes is the main limitation in mass production industry. To circumvent such limitations other alternative routes include: thermoplastics as melt adhesives [3,25,26,27,28,29], ultrasonic spot welding [30,31,32], laser direct joining [33,34,35,36,37], friction spot joining [38,39], friction lap welding [40]. These techniques use heating their main source to create the joints which increase the risk to have thermal damages on the FRPC and are limited to the thermoplastics most of the time.

The work presented here afterwards is focused on the demand for developing processes responding to industrial constraints, i.e., being able to overcome the multi-material dissimilarities with inherent constraints (mechanical, thermal, chemical), and basic fixed equipment that is flexible and adaptable when there is a need to switch from one application to another.

Magnetic Pulse Processes (MPPs) are interesting candidates. MPPs are impulse/high-speed technologies which use pulsed magnetic fields to generate contactless forces to tubular or sheet metal parts. They were applied successfully for electromagnetic and electrohydraulic forming, electromagnetic and electrohydraulic crimping, and welding for similar and dissimilar metals [41,42]. All applications use a pulsed power generator as basic equipment that generates the damped sinusoidal current needed in each case to accomplish the desired phenomena.

In the case of welding applications, i.e., Magnetic Pulse Welding (MPW), the principle of the high-velocity impact processes are based on accelerating one metal—names flyer—at very high velocities towards another fixed metal—named parent—where the local progressive collision creates the bond between the two materials. The nature of the process is hence cold and rapid—some microseconds—which eliminates the heating effects on the joined partners. In brief, the MPW is categorized as a pressure base process where pressure is developed by a high impact. The MPW was most used and developed for tubular geometries [43,44], and since the year 2000, sheet metal applications were developed. Aizawa et al. [45] proposed an original magnetic pulse welding (MPW) method for dissimilar sheet metal joints with a linear I-shaped coil (Figure 1a,b). Manogaran et al. [46] developed the magnetic pulse spot welding (MPSW) (Figure 1c,d) which aimed to facilitate further the automation possibility of the process by overcoming the need of standoff via an insulating system (Figure 1a).

In addition, C. Khalil et al. [47,48], in their recent studies, improved the process capabilities and extension of the applications to automotive dissimilar aluminum–steel alloy welding: 6xxx, 5xxx to DC04, DP advanced steels, and coated steels. The literature reveals scanty investigations into MPW applications for the joining of metals to polymer-based composites.

Objectives of the Study

The investigations reported here are aimed to extend the application of the MPW to the FRPC/metal hybrid joints that will make the process a multi-use-multi-material technique. As part of this study, two design solutions were considered for the Metal/FRPC planar sheets joining with MPW and have led to the filing of two patents [42,43]. For design, constraints of MPW technology must be integrated. The first is that one of the metals—a flyer that provides impacting forces—must be a good conductor and the second is related to the maximum deformation—a free forming cone—for a given set of process parameters that the metal can sustain before necking [49,50,51]. This sets the maximum standoff distance between the flyer and the impacted metal. In fact, for welding, it has to be much smaller than the max. This infers, as will be discussed later, that the standoff distance is the maximum thickness of the composite that can be joined when a welding patch is used. In the case of metallic inserts, the stand-off distance is between the flyer and the insert itself.

The first design thus consists of the introduction of a thin metal insert with small dimensions in the FRPC sheet during the manufacturing phase where the polymer will act as an adhesive agent, embodying this insert inside the composite (Figure 2). This metallic insert, which is positioned at the intended joining zone, will act as an intermediate metal on which the metallic partner can be welded using MPW or MPSW. This first configuration can be used when the metallic partner has a good electrical conductivity and a relatively small yield strength and good formability (aluminum sheets, for example).

The second configurations take into consideration the conditions where the metal partner could be a very high-strength metal or have a high thickness, or is an electrically less-conductive metal; for example, ultra-high strength steel. In this case, the solutions are represented in Figure 3 and the principle is based on the use of a metallic patch which will have the role of a flyer metal that will pass through a managed opening in a composite sheet to produce a weld with other parent metals, trapping the composite in-between and creating the joint.

One common characteristic of the above configurations is their recyclability, as the components can be separated easily at the disposal phase for this purpose. Furthermore, the concept, in short, is based on whether the MPW/MPSW of the sheet metal on a metallic insert is inside the composite or is introduced by an intermediary metallic patch. The next sections provide the details and conclusions of tests conducted with the aforementioned configurations.

2. Materials and Methods

2.1. Materials

As previously stated, the two configurations tested use a metallic insert or a metallic patch. In both cases, the FRPC chosen is a DuPont-Vizilon™ SB75G1, which is a heat-stabilized, 2-2 Twill Weave Glass Fabric-reinforced polyamide-based thermoplastic composite sheet (

Table 1. DuPont-TM Vizilon™ SB75G1 thermoplastic composite sheet.

Property	Test Method	Units	Warp Direction		Weft Direction
Reinforcement
Fabric	-	-	2-2 Twill Weave Glass Fabric
Wrap to weft ratio	-	(%)	50		50
Resin Composition	-	-	PA-GF75
Sheet
Thicknesses	-	(mm)	1.2 to 2
Fiber Mass Fraction	-	(%)	75
Fiber Volume Fraction	-	(%)	57
Density	ISO 1183	(g/cm3)	1.98
Thermal
Melting temperature	ISO 11357-1/-3	(°C)	260
Mechanical (23 °C)			Dry	Cond.	Dry	Cond.
Tensile Modulus	ISO527-4	(GPa)	29	24	27	25
Tensile test at break	ISO527-4	(MPa)	491	399	504	411
Tensile strain at break	ISO527-4	(%)	2.2	1.8	2.2	1.8

). For the first configuration, i.e., with metallic inserts, the flyer metal is an aluminum-5182 (

Table 2. Aluminum alloys chemical compositions (% at.).

	Si Max	Fe Max	Cu Max	Mn Max	Mg Max	Cr Max	Zn Max	Ti Max	Other Max
5182	0.20	0.35	0.15	0.50	4.0–5.00	0.10	0.25	0.10	0.15
5754	0.40	0.40	0.10	0.50	2.6–3.6	0.30	0.20	0.15	-

and

Table 3. Aluminum alloys mechanical properties.

Sheet Material	${\mathit{R}}_{\mathit{p}\mathbf{0.2}} \left(\mathbf{MPa}\right)$	${\mathit{R}}_{\mathit{m}} \left(\mathbf{MPa}\right)$	A % ISO 20 × 80
5754-H111$\left(\mathit{e}\le \mathbf{1.5} \mathbf{mm}\right)$	90–130	200–240	21
5182$\left(\mathit{e}\le \mathbf{1.5} \mathbf{mm}\right)$	120–160	260–310	23

) with a thickness $e_f=1.2$ mm. The metallic partners are made of 5754 aluminum (Table 2 and Table 3) and DC04 steel (

Table 4. Steels chemical compositions (%at.).

Material	C Max	Mn Max	Si Max	P Max	S Max	Al Max	Ti + Nb Max	V Max	Cr Max	Mo Max	B Max	N Max	Ni Max	Nb Max
DP450	0.10	0.16	0.4	0.04	0.015	0.015–0.08	0.05	0.01	0.8	0.3	0.005	0.008	-	-
DC04	0.08	0.40	0.10	0.025	0.025	0.020

and

Table 5. Steels mechanical properties.

Sheet Material	${\mathit{R}}_{\mathit{p}0.2} \left(\mathbf{MPa}\right)$	${\mathit{R}}_{\mathit{m}} \left(\mathbf{MPa}\right)$	A % ISO 20 × 80
DC04$\left(\mathit{e}\le \mathbf{1.47} \mathbf{mm}\right)$	160–200	280–340	37
DP450	290–340	460–560	27

). For the second configuration, i.e., with a metallic patch, the metallic patches were 5754 and 5182 aluminums (Table 2 and Table 3) and the metallic partners tested were DC04 steel with a thickness $e_p=0.8 \mathrm{mm}$ and a DP450 steel with a thickness $e_p=1.17 \mathrm{mm}$ (Table 4 and Table 5).

2.2. Equipment

2.2.1. Pulse Generator

The pulse generator used is a 50 kJ developed at ECN with the following characteristics:

$$C_{Gen}=408 \mu \mathrm{F}, L_{Gen}=0.1 \mu \mathrm{H}; R_{Gen}=3 \mathrm{m}\mathsf{\Omega }; V_{max}=15 \mathrm{kV}; I_{max}=500 \mathrm{kA}; f_{short}=25 \mathrm{kHz}.$$

The highest limit for the discharge energy is hence fixed at 16 kJ so that the discharge current does not exceed 80% of the maximum allowable current for the generator:

$$I_{operatio{n}_{max}}=0.8\times I_{max}=400 \mathrm{kA}$$

2.2.2. Coils

The two coils used in this study are: a linear rectangular cross-section coil and an O-shape rectangular cross-section coil as described in our previous investigations (Figure 4) [53]. The active areas of the two coils are the same: 20 × 8 mm. The O-shape coil used during this experimental feasibility study in its steel version for the patches with higher thickness while the linear copper coil will be used for the patches with smaller thicknesses, since the O-shape has a higher efficiency, as demonstrated by C. Khalil et al. [53]

2.3. Experimental Procedure

2.3.1. First Configuration with Metallic Insert

Metallic Insert

For this first configuration (Figure 2), the insert design is focused on having the thinnest possible insert and a good adhesion with the composite. The materials of the inserts are aluminum 5754 and steel DC04, with thicknesses of 0.5 mm and 0.62 mm, respectively, and are minimized to reduce composite weight.

Two designs are tested:

• The first design is perforated metallic inserts, with perforation diameters of 2 mm (Figure 5a) and the inserts were sandblasted on the side in contact with the composite in order to increase the contact surface and to improve the adhesion. At the same time, the design took into consideration the fact that the area of the insert where the welding will take place needs to be free of any perforation. Perforation is still intended to reduce weight.
• The second design also consists of perforated inserts, with 2 mm diameters holes but with flanged holes with a depth of 1 mm (Figure 5b).

The inserts were embedded in the composite sheets by applying an over-molding thermo-compression process using a SCAMAX press consisting of two heating plates and a pressure cylinder. The process parameters that led to the specimens shown in Figure 5c are pressures of 3.67 MPa and a heating temperature of 275 °C.

Setup for MPSW Application

In case of MPSW (Figure 1c,d), the first step of the process is to create the hump in the flyer metal [46,53]. The general geometry of the hump chosen is a rectangular one. This geometry was chosen based on the previous study done by Arun et al. [54] during which different geometries were tested and it was proven that the rectangular one is the more efficient. The hump, as described earlier [54], is stamped in the flyer metal using a hydraulic press die and the pressure used does not exceed 6 tons.

After cleaning the metal surfaces from any residual oil using acetone solution, the flyer metal is then positioned using a guiding system (plastic insulator and laser) so that the hump is facing and is centered on the coil’s active area. The FRCP specimen with metallic insert (parent metal) is then positioned above the flyer metal with the massive die on and finally the system is clamped using a torque of 35 N·m so that the hump will not be deformed due to clamping (Figure 6).

The mechanisms that lead to the weld are schematically presented in Figure 7a. The magnetic pulse initiates the deformation in the central part of the hump and then progresses towards the outer corners. The central point—the first point of impact—is not welded, as the requirement of the impact angle is not met. The red portion is where the weld zone is localized. The interfacial structure (Figure 7b) shows the result of impact progression around the first point of impact by the progression of the wavy interface in opposing directions. This has been previously discussed by Arun et al. [46]

In order to verify that the shock induced by the MPSW did not damage the composite, we carried out observations using a XRadia XCT-400 microtomograph. No damage to the fibers, matrix or insert/matrix bond was observed in our samples (Figure 8).

2.3.2. Second Configuration with Metallic Patch

Metallic Patches

For this second configuration of Figure 3, the metallic patches were made of aluminum 5754 (thicknesses $e_{f_1}=0.5$ mm and $e_{f_2}=1$ mm) and aluminum 5182 (thickness $e_f=1.2$ mm). They have dimensions of $50\times 50 {\mathrm{mm}}^2$. The use of different thicknesses of patches will be detailed in the Section 3. The openings in the composite were dimensioned to exceed the coil active length by 10 mm, i.e., +5 mm on each side. Concerning the geometries of these openings or holes inside the composite, they will be detailed in the Section 3.

Setup for MPW Application

The metallic patch was first cleaned with acetone solution and centered on the hole in the composite. The parent metal, i.e., DC04, was also cleaned using an acetone solution and then positioned regarding the composite as it is presented in Figure 9. The flyer, which is here the aluminum patch, is positioned facing the coil where a Kapton insulation sheet with a 0.1 mm thickness is used to separate it from the coil. The positioning is controlled by a laser so that the part of the patch to be deformed is centered regarding the coil’s active area and the whole system is clamped.

2.3.3. Joint Strength Evaluation

For the joint’s mechanical strength evaluation, the specimens were tested in lap-shear conditions [53]:

• quasi-static at ${10}^{-2}$ mm/s (with an Instron 5584 mechanical tensile machine, Norwood, MA, USA);
• dynamic at 614 mm/s (with an MTS 819 hydraulic high speed tensile machine, Eden Prairie, MN, USA);
• and fatigue under unidirectional conditions ($R=0$, frequency of 20 Hz, $F_{max}=0.6\times F_{\max quasi-static}$ with an MTS Electropulse E10000 machine, Eden Prairie, MN, USA).

3. Results

3.1. Configuration with Metallic Insert

3.1.1. Configuration with 5754 Aluminum Metallic Insert

The first welding tests were done using the first design of aluminum 5754 perforated inserts (Figure 5a). The discharge energies between 10 kJ and 16 kJ using the O-shape coil led to successful welding. The specimens were tested under quasi-static lap-shear tests and the average maximum load attained was between 1.9 kN and 2.5 kN. Different failure mechanisms were observed during the tests:

• at lower discharge energies, failure occurred in the welding and a small part of the insert was detached from the composite (Figure 10a);
• at higher discharge energies, either the insert was totally detached from the composite (Figure 10b) or tearing occurred in the insert while a part of the insert was detached from the composite (Figure 10c).

The second design of inserts, i.e., flanged hole perforated inserts (Figure 5b), was tested with the same energies. When testing the specimens under lap-shear quasi-static conditions, the failure mode was the tearing of the insert, with a better adherence of the insert with the composite.

These preliminary tests had the aim of defining the best design of the inserts, which stands in the composite. Hence, to test the joining strength under different loads, the flanged hole perforated insert was chosen. Three sets of five specimens were prepared using a discharge energy of 16 kJ. Each set was then used for quasi-static, dynamic and fatigue tests.

The results of the quasi-static and dynamic lap-shear tests are respectively presented in

Table 6. Quasi-static lap-shear tests for metallic insert configuration (E = 16 kJ).

Flyer	Parent	Metallic Insert	${\mathit{F}}_{\mathit{m}\mathit{a}\mathit{x}}{}_{\mathit{a}\mathit{v}\mathit{e}\mathit{r}\mathit{a}\mathit{g}\mathit{e}} \left(\mathbf{kN}\right)$	${\mathit{F}}_{\mathit{m}\mathit{a}\mathit{x}} \left(\mathbf{kN}\right)$
				Test 1	Test 2	Test 3	Test 4	Test 5
5182	FRPC	5754	3650	2467	3253	3305	3818	4224
5182	FRPC	DC04	5526	4445	4702	5687	6084	6710

and

Table 7. Dynamic lap-shear tests for metallic insert configuration (E = 16 kJ).

Flyer	Parent	Metallic Insert	${\mathit{F}}_{\mathit{m}\mathit{a}{\mathit{x}}_{\mathit{a}\mathit{v}\mathit{e}\mathit{r}\mathit{a}\mathit{g}\mathit{e}}} \left(\mathbf{kN}\right)$	${\mathit{F}}_{\mathit{m}\mathit{a}\mathit{x}} \left(\mathbf{kN}\right)$
				Test 1	Test 2	Test 3	Test 4	Test 5
5182	FRPC	5754	4931	3412	3844	4819	5186	5877
5182	FRPC	DC04	6210	4641	5179	5794	8256	7179

. With the average load of 3650 kN for static and 4931 kN for dynamic, the results show a wide spectrum of data ranging from 2467 to 4224 kN for static and 3412 to 5877 kN for dynamic. The specimens with the lowest load in both cases were fractured in the weld zone. When taking a closer look at these 2 specimens, part of the welding coincided with the middle holes of the insert, as represented in Figure 11a. In the dynamic tests, the failure occurred in the insert, and is represented in Figure 11b, where the tearing of the aluminum insert is visible. It is worth suggesting that the failure mechanisms would be determined by the strength of the insert composite interface, that of the weld and the insert sheet itself. This may account for the multiple failure behavior observed in this study. The presence of perforations is another factor that exacerbates the analysis from a sheet-tearing perspective. It may thus be concluded that perorated inserts, though propitious for weight reduction, introduce uncertainty of outcome.

The fatigue tests were further performed under unidirectional conditions (R = 0) and at a frequency of 20 Hz and the maximum load for fatigue tests is for each couple equal to 60% of the highest quasi-static lap-shear failure load. The results are given in

Table 8. Fatigue tests results for metallic insert configuration (E = 16 kJ).

Flyer	Parent	Metallic Insert	${\mathit{F}}_{\mathit{m}\mathit{a}\mathit{x}} \left(\mathbf{kN}\right)$	Number of Cycles
				Test 1	Test 2	Test 3	Test 4	Test 5
5182	FRPC	5754	2	41,000	35,000	36,000	42,000	45,000
5182	FRPC	DC04	2.88	45,000	41,000	24,000	29,000	34,000

. The failures during the fatigue tests were a combination between a tearing in the metallic insert and a detachment of the insert from the composite (Figure 12).

3.1.2. Configuration with DC04 Steel Metallic Insert

As mentioned earlier, here the flyer is aluminum 5182 and the insert embedded into the composite is a perforated flanged sheet of steel (DC04). Herein, the flanged design was preferred, as mentioned earlier, for the aluminum insert, as it produced better results. Preliminary welds were attained with the discharge energies between 10 kJ and 16 kJ. When the specimens were tested, the average maximum loads were between 2.9 kN and 5.3 kN for 10 kJ and 16 kJ respectively and the failure occurred in the welds.

Hence, to test the joining strength under different loads, three sets of five specimens were prepared using a discharge energy of 16 kJ. Each set was then used for quasi-static, dynamic and fatigue tests.

The results of the quasi-static lap-shear tests and dynamic tests are given respectively in Table 6 and Table 7. The fatigue test results are listed in Table 8. The failure loads are exceeding 5 kN in quasi-static and 6 kN in the dynamic condition. The fatigue tests showed some disparities in the steel insert case, but this was due to the fact that a part of the welding line coincided with the holes in the insert, giving a lower number of cycles between 24,000 and 29,000, while the others were higher than 34,000 and up to 45,000 cycles.

The failure most of time occurred, in this case, in the welding itself. The phase diagram between aluminum and steel suggests the formation of intermetallic as previously formulated and observed by the authors [55]. When the failure did not occur in the weld, a tearing was observed in the flyer metal due to the lower mechanical strength of aluminum (Figure 13).

3.1.3. Analysis and Discussion

As a first observation, the application of the MPSW to weld a metallic sheet on a metallic insert embedded inside a composite is possible using either aluminum or steel inserts. These inserts showed better adhesion in the composite in the case of the flanged hole inserts.

Another point that was investigated was related to the eventual equivalence between the welding behavior in the metal to the FRPC joints and the metal to metal sheet with identical processing conditions. The idea was to confirm the carryover possibility of the MPW and MPSW parameters from bimetallic welding applications [48,53] to the FRPC/metal joint application, without having to go through a new R&D cycle of experimentation. Figure 14 and Figure 15 show the typical curves for the quasi-static lap-shear configurations found in both applications: 5182/FRPC joints and 5182/sheet metals of the same inserts’ alloys. The results are coherent.

3.2. Configuration with Metallic Patch: Principle Feasibility Validation

The first patch used was a 5754 aluminum patch of $50\times 40 \mathrm{mm}²$ having a thickness of 0.5 mm. The hole in the composite is a rectangular hole with a width of 25 mm and a height of 50 mm and the coil is a linear copper coil [53]. The configuration was tested on both a DC04 steel and a DP450 steel with discharge energies of 10 kJ, 13 kJ and 16 kJ in both cases. The welding occurred in the three cases and for both types of steels (Figure 16).

For energy higher than 13 kJ the patch is teared at the edges of the rectangular hole (Figure 17). In magnetic pulse welding, the impact speed reaches several hundred meters per second. When the projectile part arrives on a sharp edge, it is possible to cut the projectile part. This is what happened in our case with the 0.5 mm thick patch. To avoid this phenomenon, it is necessary to use a patch of greater thickness.

For the rest of the tests, we successfully used 5754 aluminum 1 mm thick and 5182 aluminum 1.2 mm thick patches. The coil used in this case is the O-shape steel coil, since we have thicker sheets [53]. The welding in both cases occurred when applying energies between 10 and 16 kJ and no tearing at the rectangle edges was observed, showing that it is better to go with patches with a thicknesses higher than 1 mm to be able to stands the sharp edges of the composite’s holes.

To evaluate the quality of assemblies, the specimens were tested in the quasi-static lap shear configuration where the velocity was of ${10}^{-2}$ mm/s.

In cases where the patch did not fill at 100% all the composite holes, given the configuration of the assembly, the loading is directly applied on the patch and the tensile test is more of a peeling test.

The maximum loads attained during the tests of 5754 patches are between 120 N and 680 N for the lower and the higher discharge energy levels, respectively. During the tests of 5182-patches, the maximum loads also varied between 120 N for lower discharge energy levels and 600 N for the higher level. The first remarkable point is the failure in both cases that was due to the patch that jumped from its place, and the second one is the relatively long time before the collapse of the insert, which is about 2 min on average.

Figure 18 presents a typical curve of the load vs. displacement for both 5754- and 5182-patches. During the test, the composite sheet was slipping under the insert and hence the load was tearing the weld. When looking at a cut-section of the joining area (Figure 19), the flow of the patch does not fill the corners and the slopes of the deformed patch observed confirm this. Therefore, the tests can be considered more as peeling tests than lap-shearing.

In order to prevent as much as possible this slipping effect and increase the zone of the welding that undergoes the loading it is possible to optimize the orientation of the hole according to the loading in order to avoid any risk of tearing at the corners of the holes. Tests by orienting the weld at 0°, 45° and 90° have shown that it is thus possible, for the same welded surface, to improve the resistance of the assembly by a factor of 3.

In the configuration outlined in Figure 3b, the composite is taken in sandwich between the two metallic partners. In this application, the window in the composite is rectangular and the flyer metal is a 0.5 mm thick aluminum 5754 sheet. The weld was done using a discharge energy of 16 kJ. In Figure 20, the test result is presented together with the test result of a bimetallic application with the same process parameters. The tearing in both cases happened in the flyer aluminum sheet metal, and the peak load exceeded 3.5 kN.

4. Conclusions and Perspectives

The objective of the study to explore and extend the application of the MPP to the FRPC/metal hybrid joints was achieved, thereby making the process a multi-use-multi-material technique. The new proposed solutions covered all possible applications, taking into consideration different configurations as well as the material properties and thicknesses. The examination of feasibility via introduction of a thin metallic sheet in the composite or using a metallic patch through the composite to create the join were proven experimentally.

The important conclusions relevant to the metallic insert configuration examined in this study can be resumed as:

-

metallic inserts’ material is not a limitation, and it can be either aluminum or steel;

-

the design of the metallic insert using flanged holes gives a high adherence with the composite;

-

the holes in the insert do not prevent the welding from occurring, but they contribute to a significant decrease in the effective welding area, leading to weaker welds. Consequently, the design should take into consideration the spatial distribution of holes to avoid any coincidence between the welding line and the holes inside the inserts;

-

the mechanical tests suggest an insignificant loss in the welding strength when applying the MPSW on the insert when the hole–weld juxtaposition was avoided. The spatial distribution of holes and the precise tuning of the weld remain critical;

-

the steel inserts compared to aluminum inserts showed better resistance. With thin aluminum inserts (0.74 mm), the insert was invariably torn during tests.

For the metallic patch, the main points that can be stated are:

-

a thin metallic patch will have the risk of tearing on the borders of the holes in the composite during the application of the MPW, and hence it is preferred to use patches with thicknesses higher than 1 mm;

-

the hole dimensions should be consistent with the coil’s active area length by exceeding this length of a minimum 5 mm on each side;

-

the flow of the patch through the holes in the composite does not necessarily fill the edges of the holes, and hence create slipping risks under loading conditions. These loading conditions, when possible, should be avoided by the design;

-

the transition from MPW bimetallic parameters can be used for FRPC–metallic welds should the sheet thickness of the composite be lower than that defined by the gap between sheets of bimetallic joints. It is important to mention here that a gap between bi metallic sheets is required for the impact acceleration of the flyer;

-

the configuration where the FRPC is taken as a sandwich between two metallic partners has proven its efficiency and the behavior of the welding is the same as in the bimetallic case applications;

-

and the final joint is easy to disassemble for disposal and recyclability.

The design and results presented here suggest the viability of MPW as a tool for the weld joining of metallic and FRPC sheet components. There is a further scope for improved implementation of MPW as a composite-metal assembly process through optimization of operating process parameters, which includes the optimum patch/insert thickness. For future studies, the focus should be given to optimizing both metallic inserts and patch configurations. Inserts are best suited whenever the thickness of the composite part is more than what flyer sheet metal can reasonably sustain. Magnetic pulse welding involves an impact and to what extent the composite is damaged needs to be apprehended. Examination of the adhesion of metallic inserts to composites by testing more design concepts and conducting microscopical observations to explore the post weld effect on the composite microstructures is required. The feasibility proven of the concepts opens the door for further improvements in the design of inserts, as well as the patches and window of processing parameters.

5. Patents

The work presented in this article has been the subject of two patents applications [49,50].