A B the simulation is the analysis of the mechanical processes in the vicinity of the collision point, the apparent deviation related to the welding process itself (elongation of the flyer part at the free edge) can be neglected. Figure 10 presents images of the simulations of the aluminum flyer tubes in a 15-mm coil at the time of initial impact with the parent part for lw < lcoil, lw = lcoil, and lw > lcoil, respectively. The areas of initial contact between the parent and flyer parts are visible in the numerical simulations. The angles at the contact point for each image were calculated. Figure 10A and C presents clear indications of a one- or two-sided front, respectively, and correlate well with the experimental results presented in Fig. 5. At working lengths ≤ 7 mm, the front is one sided. This is an indication that the relationship between magnetic pressure at the flyer interface and material deformation is such that the flyer contracts first at the edge, and then continues along a single, continuous collision front. The simulation in Fig. 10C shows that the deformation of samples placed over the coil length (lw = 17 mm) occurs in a bowed manner, and the point of impact moves to both directions. The development of the collision angle, however, differs between these two directions. In the direction of the flyer tube (to the right in Fig. 10), the collision angles rise quickly to a range regarded favorable for welding. To the other side, the collision angles WELDING RESEARCH remain small, probably too small for a weld formation. The simulation results shown in Fig. 10B exhibit very small impact angles with an essentially flat tube deformation. A comparison of the impact angles and radial impact velocities for working lengths of 10 and 7 mm are shown in Fig. 11. This exemplifies the notion that by using a smaller working length both the impact velocity and impact angle are increased. Also assessed in this simulation was the orientation of the flyer part in the asymmetric coil (from which side the working length was referenced). Although the impact angles for the two orientations are similar, orientation 1 (working length referenced from the 90-deg edge) used in the current experiments results in systematically higher initial impact velocities compared to configuration 2 (working length referenced from 45-deg edge). Figure 12A, the simulated effect of the working distance on the magnetic field Htangential, provides insight into this phenomenon, showing a comparison between the magnetic field experienced on the inner and outer surface of the deformed tube, as well as the magnetic pressure pmagnetic. The magnetic pressure is a mathematical conversion of the volume Lorentz forces (Ref. 18). It depends on the magnetic fields on the inner and outer surfaces and the permeability (see Equation 2). pmagnetic(t)= 1 2 μ 2 (t) – Htangential,inner ( 2 (t)) (2) Htangential,outer Here it is apparent that the magnetic field at the inner surface of the tube is independent of the flyer-coil configuration; however, the tangential magnetic field in configuration 1 shows a higher external magnetic field, and therefore a higher magnetic pressure. Because the general form of the magnetic field distribution across the flyer surface is mirrored for orientations 1 and 2, the flyer impact velocities differ with only minimal changes in the impact angle. Through simulations as well as viewing the interfaces of samples joined under nonwelding conditions, it can be seen that positioning the edge MARCH 2016 / WELDING JOURNAL 107-s Fig. 12 — Axissymmetric simulation (with FEMM) of the tangential magnetic field Htangential at flyer surfaces for various configurations. A — 10mm coils; and B — working lengths in a 15mm coil. Fig. 13 — Simulated maximum interface shear stress for various working lengths, normalized to the maximum interface shear stress for lw = 4 mm.
Welding Journal | March 2016
To see the actual publication please follow the link above