of the flyer piece inside the coil/field shaper width does not guarantee a clean one-sided front process. Figure 12B presents the simulated effect of the working length lw on the magnetic field Htangential for a 15-mm coil width. Immediately apparent is a decrease in magnetic field at the outer flyer surface with an increase in working length, leading to lower initial impact velocities — Fig. 11. This shows that the magnetic intensity on the workpiece is inversely related to the working distance, and for working distances lw ≤ lcoil, the field intensity has a peak at the flyer edge. A higher magnetic field intensity also leads to higher Joule heating. This heating is concentrated on a smaller portion of the flyer, but due to the very short process time of a few microseconds, thermal conduction between the heated and cold areas of the workpiece can be neglected. Figure 13 shows that the deformation behavior of the flyer, which was analyzed by visioplastic investigations (Fig. 8), can directly be related to the interface shear stresses. For the experiments with an intermediate working length between 7 and 13 mm, the observed increase in elongation at 7 mm from the flyer edge is represented in the simulations by an increase of the shear stresses. Discussion Important details about the impact process during MPW can be extracted by pulse forming flyer tubes onto parent parts at energies slightly lower than those required for welding, then cutting away the flyer part. This has advantages over other analysis techniques. For example, a push or tension test of welded samples reveals the joint area, but can smear the part interface during removal, and metallographic images of the interfaces show only a very linear view of the interface as well as being very time consuming. As can be seen in Fig. 5, three different regimes exist: 1) A one-sided front, starting at the flyer edge and propagating away from it. Jet material is able to be cleanly ejected from the propagation front. 2) A transition regime, in which the flyer impacts the parent under the correct conditions to potentially weld at the onset and end of the flyer impact zone, but the contact area contains a gray residue. 3) A two-front process, in which a bowed impact at the center of the coil occurs and two fronts run in opposite directions from this region. Each of the presented regimes exhibits distinct interface characteristics. Regime A (one front) leads to a short region of interaction at the flyer edge, which is clearly seen in Fig. 5A and B for samples with working lengths of 4 to 7 mm. The analyzed orientation 1 is also thought to be capable of creating more efficient welding conditions, as less current is needed to reach a given velocity compared to orientation 2 — Fig. 11. The increase in magnetic intensity at the coil edge seen in Fig. 12B in tandem with the decrease in rigidity at the flyer edge aid in ensuring a clean, singlefront process for deformations in regime A. On the other hand, a higher current density in the workpiece leads to stronger Joule heating. Regime B is present for samples pulsed at working lengths from about 8 mm to a working length corresponding to the coil width lcoil and displays gray matter deposited in the intended welding zone. For this “transition” regime, the case is such that the flyer edge impacts the sample first, but the deformation occurs in a chiefly flat manner, leading to low impact angles, which hinder the jetting and “selfcleaning” nature of the MPW process while the flyer impact perturbs the parents’ surface. Entrapment of the jet is possible at these low angles, as projectile particles may impact the flyer or parent surfaces before escaping the joining pair, becoming trapped between the flyer and parent. This would hinder further jetting and welding processes and also heat up the joining partners. Aluminum has a density of 2700 kg/m3 and a melting point of 660°C. Steel has a density of 7870 kg/m3 and a melting point of 1497°C. Because of the large difference in density and melting point between aluminum and steel, the majority of the jetted material is believed to originate from the aluminum flyer. The brittle characteristics of the deposited material match well with the described jet material for Al-Al welds investigated by Stern et al. (Ref. 5). In regime C, it would be expected that the uniform magnetic field along the workpiece would also result in a flat impact; however, it is known that deformation occurs in a bowed manner and welding is generally possible under this configuration. Impact angles are thought to be higher for coils with a smaller maximum working zone lcoil, as seen by the reduction in “undisturbed” area in the center of the two sided front when deforming with the 10-mm coil. Here, the initial impact occurred perpendicular to the surface, and the initial pressure of the running fronts was not sufficient to deform the material interface. The sample surface indicates the flyer only propagates at high enough impact angles and velocities to deform the parent surface as it reaches the coil edge, which is away from the flyer edge. If the bow-shape of the flyer is too flat (as witnessed in the 15-mm coil), it may be that the correct conditions to cause visible damage to the parent are only reached near the coil edge where sufficient impact angles are formed in agreement with Fig. 5A. To the direction toward the flyer edge, the collision angles remain small, probably too small for a weld formation — Fig. 10C. This indicates that the applied coils with a width of 10 and 15 mm might be inappropriate for two-front welding processes. The observation that specimens joined using a regime C configuration showed indications of material failure can be attributed to the deformation mechanism. This mechanism is characterized by increased shearing and elongation (see Figs. 8 and 13), probably exceeding the formability of the flyer material. Excessive heating due to the contact of jetting particles might also contribute. Experimental and simulated trial results show that a longer working length is not necessarily beneficial for increased welding probability/efficiency. Although a higher portion of the charging energy may be harnessed for deformation, said deformation may occur in a manner that is unfavorable to welding. Smaller maximum coil working lengths, to an extent, are more effective in creating favorable welding conditions for one- and twosided fronts. Results presented here WELDING RESEARCH 108-s WELDING JOURNAL / MARCH 2016, VOL. 95
Welding Journal | March 2016
To see the actual publication please follow the link above