Superconducting rf (SRF) cavities are traditionally fabricated from superconducting material sheets or made of copper coated with superconducting material, followed by trim machining and electron-beam welding. An alternative technique to traditional shaping methods, such as deep-drawing and spinning, is electrohydraulic forming (EHF). In EHF, half-cells are obtained through ultrahigh-speed deformation of blank sheets, using shockwaves induced in water by a pulsed electrical discharge. With respect to traditional methods, such a highly dynamic process can yield interesting results in terms of effectiveness, repeatability, final shape precision, higher formability, and reduced springback. In this paper, the first results of EHF on high purity niobium are presented and discussed. The simulations performed in order to master the multiphysics phenomena of EHF and to adjust its process parameters are presented. The microstructures of niobium halfcells produced by EHF and by spinning have been compared in terms of damage created in the material during the forming operation. The damage was assessed through hardness measurements, residual resistivity ratio (RRR) measurements, and electron backscattered diffraction analyses. It was found that EHF does not worsen the damage of the material during forming and instead, some areas of the half-cell have shown lower damage compared to spinning. Moreover, EHF is particularly advantageous to reduce the forming time, preserve roughness, and to meet the final required shape accuracy.

No abstract

International audienceElectromagnetic Forming is a very promising high-speed forming process. However, designing these processes remains quite intricate as it leads to deal with strongly coupled multiphysics process and thus requires the use of computational models. We present here the main features of the numerical model which we are currently developing to model this process. Accurate knowledge of constitutive law parameters for material at high strain rates remains quite difficult to access. We thus introduce here a procedure which has been developed in order to deal with identification of these parameters

The design of processes like magnetic pulse forming and electrohydraulic forming involves multiphysical couplings that require numerical simulation, and knowledge on dynamic behaviour of metals. The forming process is completed in about 100 μs, so that the workpiece material deforms at strain-rates between 100 and 10 000 s-1. In this range, the mechanical behaviour can be significantly different than that in quasi-static conditions. It is often noticed that the strength and the formability are higher. The main goal of this study is to use an electromagnetically driven test on tubes or sheets to identify the constitutive behaviour of the workpiece material. In the case of tube, an industrial helix coil is used as inductor. Simulations with the code LS-Dyna® permit to find a configuration where the tube deforms homogeneously enough to allow axisymmetric modelling of the setup. The coil current is measured and used as an input for the simulations. The radial expansion velocity is measured with a Photon Doppler Velocimeter. The parameter identification is lead with the optimization software LS-Opt®. LS-Dyna axisymmetric simulations are launched which different set of parameters for the constitutive behaviour, until the computed expansion velocity fits the experimental velocity. The optimization algorithm couples a gradient method and a global method to avoid local minima. Numerical studies show that for the Johnson-Cook constitutive model, two or three experiments at different energies are required to identify the expected parameters. The method is applied to Al1050 tubes, as received and annealed. The parameters for the Johnson-Cook and Zerilli-Armstrong models are identified. The dynamic constitutive behaviour is compared to that measured on quasi-static tensile tests, and exhibits a strong sensitivity to strain-rate. The final strains are also significantly higher at high velocity, which is one of the major advantages of this kind of processes.

No abstract