Pascal Venet

Accelerated Ageing of Metallized Film Capacitors Under High Ripple Currents Combined With a DC Voltage

In view of their potential for high dielectric breakdown strengths, low dissipation factors, and good dielectric stability over a wide range of frequencies and temperature, metallized films capacitors are very used components for aeronautic applications. Nevertheless, capacitors remain unreliable components; a good acquaintance of their deterioration over time would enable us to perform a predictive maintenance on the component and thus improve the availability of the whole system. This operation requires the knowledge of the capacitor ageing law and their failure mechanisms associated with the application. In a standard six pulses rectifier-fed pulse width modulation inverter, dc-link capacitors are generally subjected to gradients of temperature due to the encountered electrical stresses. Since traditional floating ageing tests, which consist on applying constant voltages and temperature across the component, do not reflect the normal ageing of the component, we propose in this paper to study the degradation kinetics of metallized films capacitors under high ripple currents, alone or combined with a dc voltage across the devices terminals. It will therefore reproduce with a more representative way the ageing of the component in comparison with the traditional floating ageing tests. Based on the resonant circuit properties, the ageing test bench especially developed for our study provides high currents and voltages from a low power source. Through the study and analysis of the capacitors parameters evolution under high ripple currents, an original ageing law is proposed to model the capacitance decrease with time based on the electrochemical corrosion of the capacitors electrodes. The adequacy between the experimental points and the model proved the validity of the proposed law.

Metallized polymer film capacitors ageing law based on capacitance degradation

Abstract Passive components, particularly capacitors, are very used devices in power electronics applications providing key function on board. Nevertheless, capacitors breakdowns can have catastrophic consequences on the financial and human scale; a good acquaintance of their deterioration over time would contribute in the improvement of the availability of the whole system by performing a predictive maintenance on the component. This operation requires the knowledge of the capacitor ageing law and their failure mechanisms associated to the application. Capacitance loss can be mainly attributed to the self-healing process occurring in metallized film capacitors when used under high steady electrical and thermal stresses. In this paper, a capacitance ageing law is proposed based on the identification of voltage and temperature degradation kinetics from three experimental floating ageing tests performed at different voltage and temperature constraints. A total of 34 capacitors provided from different manufacturers using polyester film as dielectric have been studied and compared to validate the proposed law.

Comparison of EDLC impedance models used for ageing monitoring

Electrochemical double layer capacitors (EDLC) electrodes are made with pores of different geometric characteristics. This particularity is called pore size distribution. During EDLC ageing monitoring, it is necessary to know precisely the evolution of impedance. Firstly, this article deals with classical impedance models like single pore or constant phase element (CPE) for EDLC and exposes their strengths and their drawbacks. Then a more recent model grouping pores of the same geometric capacities is presented. Each group can be modeled by an electrical branch and has its own impedance expression. The resulting electrode impedance is obtained by putting n branches in parallel. This new model, called multi-pores model (MP), has been recently presented by A. Hammar [3] and is highly hoped to be used for more precise health monitoring of EDLC. The physical advantage is to take into account the pore heterogeneity. The models precision is finally compared using experimental results.

Balancing circuit control for supercapacitor state estimation

In this paper, an original technique for monitoring the aging of supercapacitor energy storage system is proposed. The aging of supercapacitor is represented by the degradation of its internal parameters as the loss of capacitance and the increase of the Equivalent Series Resistance (ESR). This degradation is continuous until reaching the end of life criteria. The proposed approach is based on estimating the current ESR, of each storage system element with a fast control of the balancing circuit present at the terminal of each storage element. The fast control is applied for a short moment and could be applied at any storage system state with no incidence on the system. The information about this current parameter makes the state of health monitoring possible. The analysis is performed using simulation with realistic storage system models in order to demonstrate and validate the effectiveness and the accuracy of the proposed technique. The results presented covers storage system such as supercapacitor but could be also applied to batteries.

Online lithium-ion batteries health monitoring using balancing circuits

In this paper, an original online technique for the state of health monitoring of lithium-ion batteries energy storage systems is proposed. The aging of lithium-ion batteries is represented by the degradation of their internal parameters as the decrease of the capacitance and the increase of the Equivalent Series Resistance (ESR). This degradation is continuous over aging until reaching the storage system failure criteria. Thus, the original online technique is based on tracking the ESR representing the battery aging. The proposed approach uses a suitable balancing circuit connected to the terminal of the storage element to extract its actual parameter. The information about this actual parameter makes the state of health monitoring possible. The analysis is performed using simulation with realistic electrical model to demonstrate and validate the effectiveness and the accuracy of the proposed new approach.

Luenberger observer for SoC determination of lithium-ion cells in mild hybrid vehicles, compared to a Kalman filter

Li-ion batteries are continually developed for a wide range of use. New variants appear on the market, including Li-ion technologies that allow supplying very high electrical power. By consequence, the battery can be used as an energy buffer in vehicles. The autonomy is limited and it is necessary to determine the State of Charge of each element that makes up a battery, although these elements (Li-ion cells) are never fully discharged or charged. In this unusual context, a solution may be to use a Kalman filter at the cost of significant computing power. More computation-wise economical observers can beings used, in particular Luenberger observers. The results obtained for this type of specific use, give similar final results, with a significant gain on the computing power required.

Modeling of metallized polymer films capacitor's impedance

Proper design of metallized films capacitors requires an understanding of all parasitic parameters sources and their impacts on circuit operation. This paper presents a modeling approach based on the Nyquist diagram to identify different parameters of metallized films capacitors, and build an equivalent electrical circuit. Two types of capacitors have been evaluated and compared to validate the resulting model.

Study of an Electrolytic Capacitor Model as a Function of Temperature

Summary With to their large capacity and their low price, electrolytic capacitors are used in many fields of power electronics, mainly for the filtering and energy storage functions. Their characteristics move strongly according to frequency, and temperature. Temperature variation is generally not taken into account in electrical simulation software. In many applications such as transport, industry, space one or in military applications, the variations of temperature are significant. A model taking into account these variations seems to be essential. In this study we propose to define a simple model taking into account characteristics variations of the electrolytic capacitors according to the temperature and frequency. A method of identification is proposed based on genetic algorithm. A good compromise between simplicity and precision is one of the objectives of this study. This identification was carried out for large electrolytic capacitor of rated value 4,7mF/500V. The comparison between measurements and the model for different temperatures gives good results. The model used in this paper has a precision for temperatures lower than 0°C much better than a standard model. Moreover the limited number of elements used in this model, allows an easy integration in simulation software.