Characterization of ZnO-degraded varistors used in high-tension devices

The effects of the degradation process on the structural, microstructural and electrical properties of ZnO-based varistors were analyzed. Rietveld refinement showed that the BiO2-x phase is affected by the degradation process. Besides the changes in the spinel phase, the degradation process also affects the lattice microstrain in the ZnO phase. Scanning electron microscopy analysis showed electrode-melting failure, while wavelength dispersive X-ray spectroscopy qualitative analysis showed deficiency of oxygen species at the grain boundaries in the degraded samples. Atomic force microscopy using electrostatic mode force illustrated a decrease in the charge density at the grain boundaries of the degraded sample. Transmission electron microscopy showed submicrometric spinel grains embedded in a ZnO matrix, but their average grain size is smaller in the degraded sample than in the standard one. Long pulses appeared to be more harmful for the varistors' properties than short ones, causing higher leakage current values. The electrical characteristics of the degraded sample are partially restored after heat treatment in an oxygen-rich atmosphere. (C) 2006 Elsevier Ltd. All rights reserved.

The failure analyses on ZnO varistors used in high tension devices

The main purpose of this work is to evaluate the failure caused by electrical discharge on commercial ZnO varistor doped with oxide of Bi, Sb, Si, Cr, Co utilized in electric transmission systems. In order to observe the effect of electrical discharge over the microstructure and electrical properties of the varistors, two kinds of pulses were applied: long pulse (2000 ms) and short pulse (8/20 mu s). In both cases, a decrease in grain size and increase in micropores and leakage current were observed. The degraded samples present oxygen defficiency mainly in the grain boundary and phase tranformation from the bismuth oxide phase. (c) 2005 Springer Science+ Business Media, Inc.

On the effects of turbulence modelling in design optimisation for high-lift devices

Aerodynamic shape optimisation is being increasingly utilised as a design tool in the aerospace industry. In order to provide accurate results, design optimisation methods rely on the accuracy of the underlying CFD methods applied to obtain aerodynamic forces for a given configuration. Previous studies of the authors have highlighted that the variation of the order of accuracy of the CFD solver with a fixed turbulence model affects the resulting optimised airfoil shape for a single element airfoil. The accuracy of the underlying CFD model is even more relevant in the context of high-lift configurations where an accurate prediction of flow is challenging due to the complex flow physics involving transition and flow separation phenomena. This paper explores the effect of the fidelity of CFD results for a range of turbulence models within the context of the computational design of aircraft configurations. The NLR7301 multi-element airfoil (main wing and flap) is selected as the baseline configuration, because of the wealth of experimental an computational results available for this configuration. An initial validation study is conducted in order to establish optimal mesh parameters. A bi-objective shape optimisation problem is then formulated, by trying to reveal the trade-off between lift and drag coefficients at high angles of attack. Optimisation of the airfoil shape is performed with Spalart-Allmaras, k - ω SST and k - o realisable models. The results indicate that there is consistent and complementary impact to the optimum level achieved from all the three different turbulence models considered in the presented case study. Without identifying particular superiority of any of the turbu- lence models, we can say though that each of them expressed favourable influence towards different optimality routes. These observations lead to the exploration of new avenues for future research. © 2012 AIAA.

On the effects of turbulence modelling in design optimisation for high-lift devices

Aerodynamic shape optimisation is being increasingly utilised as a design tool in the aerospace industry. In order to provide accurate results, design optimisation methods rely on the accuracy of the underlying CFD methods applied to obtain aerodynamic forces for a given configuration. Previous studies of the authors have highlighted that the variation of the order of accuracy of the CFD solver with a fixed turbulence model affects the resulting optimised airfoil shape for a single element airfoil. The accuracy of the underlying CFD model is even more relevant in the context of high-lift configurations where an accurate prediction of flow is challenging due to the complex flow physics involving transition and flow separation phenomena. This paper explores the effect of the fidelity of CFD results for a range of turbulence models within the context of the computational design of aircraft configurations. The NLR7301 multi-element airfoil (main wing and flap) is selected as the baseline configuration, because of the wealth of experimental an computational results available for this configuration. An initial validation study is conducted in order to establish optimal mesh parameters. A bi-objective shape optimisation problem is then formulated, by trying to reveal the trade-off between lift and drag coefficients at high angles of attack. Optimisation of the airfoil shape is performed with Spalart-Allmaras, k - ω SST and k - ε realisable models. The results indicate that there is consistent and complementary impact to the optimum level achieved from all the three different turbulence models considered in the presented case study. Without identifying particular superiority of any of the turbu- lence models, we can say though that each of them expressed favourable influence towards different optimality routes. These observations lead to the exploration of new avenues for future research. © 2012 by the authors.

Comparative electrical behavior at low and high current of SnO(2)- and ZnO-based varistors

The complete I-V characteristics of SnO(2)-based varistors, particularly of the Pianaro system SCNCr consisting in 98.9%SnO(2)+1%CoO+0.05%Nb(2)O(5)+0.05%Cr(2)O(3), all in mol%, have been seldom reported in the literature. A comparative study at low and high currents of the nonohmic behavior of SCNCr- and ZnO-based varistors (modified Matsuoka system) is proposed in this work. The SCNCr system showed higher nonlinearity coefficients in the whole range of measured current. The electrical breakdown field (E(b)) was twice as high for the SCNCr system (5400 V/cm) than for the ZnO varistor (2600 V/cm) due to a smaller average grain size of the former (4.5 mu m) with respect to the latter (8.5 mu m). Nevertheless, we consider that another important factor responsible for the high E(b) in the SCNCr system is the great number of electrically active interfaces (85%) as determined with electrostatic force microscopy (EFM). It was also established that the SCNCr system might be produced in disks of smaller dimensions than that of commercial ZnO-based product, with a 5.0 cm(-1) minimal area-volume (A/V) ratio. The SCNCr reached the saturation current in a short time because of the high resistivity of the grains, which is five times higher than that of the grains in ZnO-based varistors.

Comparative degradation of ZnO- and SnO(2)-based polycrystalline non-ohmic devices by current pulse stress

The degradation behaviour of SnO(2)-based varistors (SCNCr) due to current pulses (8/20 mu s) is reported here for the first time in comparison with the ZnO-based commercial varistors (ZnO). Puncturing and/or cracking failures were observed in ZnO-based varistors possessing inferior thermo-mechanical properties in comparison with that found in a SCNCr system free of failures. Both systems presented electric degradation related to the increase in the leakage current and decrease in the electric breakdown field, non-linear coefficient and average value of the potential barrier height. However, it was found that a more severe degradation occurred in the ZnO-based varistors concerning their non-ohmic behaviour, while in the SCNCr system, a strong non-ohmic behaviour remained after the degradation. These results indicate that the degradation in the metal oxide varistors is controlled by a defect diffusion process whose rate depends on the mobility, the concentration of meta-stable defects and the amount of electrically active interfaces. The improved behaviour of the SCNCr system is then inferred to be associated with the higher amount of electrically active interfaces (85%) and to a higher energy necessary to activate the diffusion of the specific defects.