Enrico Dallago

Lossless current sensing in low-voltage high-current DC/DC modular supplies

New data processing ICs require low-voltage high-current supplies together with high reliability and efficiency. The use of a modular power supply would be the ideal solution, but it requires detecting the current in each module, and resistive shunts are usually used. In this paper, a new lossless current-sensing circuit is presented. This lossless current transducer is obtained by applying the technique used to compensate the parasitic inductance in resistive shunts to the filter inductor of the DC/DC power supply. The causes that can influence transducer response are investigated. The current sensing was implemented in a modular DC/DC power supply and experimental results are reported.

Thermal resistance analysis by induced transient (TRAIT) method for power electronic devices thermal characterization. I. Fundamentals and theory

In this paper, a careful theoretical analysis of the thermal dynamics of an electronic device and its package was carried out in order to study the problem of the equivalent thermal circuit implementation. It was found that the device temperature evolution in time is ruled by an infinite and convergent series of time constants. The knowledge of the first n terms of the time-constant spectrum obtained from the temperature transient measurements allows the complete characterization of a suitable and reliable equivalent thermal circuit structured as a Cauer low-pass network with n cells. The total thermal resistance is therefore evaluated as a sum of several contributions due to given parts of the whole system. The techniques allowing the physical identifications of these contributions are also discussed. Furthermore, the influence of plastic coverage on the device thermal behavior is taken into account. The proposed characterization method is also applied to one-dimensional (1-D) multilayered simulated structures in order to study the influence of the number of time constants used for the analysis and effects of local defects or modifications of the material thermal properties.

Thermal resistance analysis by induced transient (TRAIT) method for power electronic devices thermal characterization. II. Practice and experiments

For pt.I see ibid., vol.13, no.6, p.1208-19 (1998). The TRAIT method for thermal characterization of electronic devices, whose theory was exposed in part I for one-dimensional (1-D) structures, was here applied to systems having heat fluxes with three-dimensional (3-D) dependence in order to demonstrate that the spatial resolution of the thermal resistance analysis is still qualitatively maintained in this type of structure too. The analytical procedure was first applied to simulated structures whose temperature transients and steady-state fields were obtained by means of a finite-element thermal simulation program. In these cases, the knowledge of the steady-state temperature distribution allowed identifying the thermal physical domains which correspond to the cells of the calculated equivalent thermal circuit composed by resistances and capacitances. Furthermore, some experiments on real electronic devices with purposely designed assembling structures were exposed and discussed. The samples were power-integrated circuits with plastic packages mounted on various substrates and Schottky diodes in TO-3 packages. The experiments on both simulated and real devices demonstrated that TRAIT analysis, being able to recognize the localization of some induced defects, maintains its spatial resolution character, despite the large distortion of the thermal domains occurring when the defects are close to the heat source.

Analytical Model of a Vibrating Electromagnetic Harvester Considering Nonlinear Effects

In this paper, an analytical model of a vibrating electromagnetic harvester, taking into account nonlinear effects, is presented. Knowing the acceleration applied to the harvester system, the model is able to predict the performance of the harvester in terms of induced voltage on the load. The energy transducer consists of four magnets, two movable and two fixed, arranged in a way such that both fixed magnets repel the moveable one. The model was implemented in Simulink and exploits the results of a finite element method (FEM) solver (Flux2D) to estimate the non-linear electromagnetic repulsion force and the flux linkage by the coil. To evaluate the effect of a load applied to the harvester, the effect of the Lorentzs force, caused by the interaction of the current that flows in the coil and the flux density of the moveable magnet, was taken into account. As a consequence, the error in the estimate of induced voltage, at resonance, is reduced from about 80% to 7%. Finally, at resonance, the maximum power that could be delivered by the harvester and dissipated on a resistive load was estimated to be about 6 mW.

A Fluxgate Magnetic Sensor: From PCB to Micro-Integrated Technology

In this paper, a double-axis micro Fluxgate magnetic sensor is presented. The device represents an evolution of a PCB dual axis sensor previously realized. In the PCB version the experimental results exhibit excellent agreement with the simulations performed using a tool based on the finite element method. Using the same design approach a version of double-axis Fluxgate structure is here proposed to be realized in IC technology. The accurate study of the magnetic field distribution allows a 75% area saving for the IC version with respect to a direct scaling of the PCB version. Imposing an external magnetic field of about 60 muT, the simulated differential output voltage of the micro-integrated Fluxgate sensor achieves a peak value of 1 mV with 5 mA sinusoidal excitation current peak at 100 kHz. The integrated microstructure shows a linearity error of about 1.15% of the full scale, in the range of plusmn50 muT, with a sensitivity of about 0.45 mV/muT

PWM Power Audio Amplifier With Voltage/Current Mixed Feedback for High-Efficiency Speakers

High-efficiency speakers significantly improve electrical to sound power conversion efficiency by reducing the power dissipation over the speaker coil. However, they require some equalization of the input signal. In this paper, a new driving technique based on a pulsewidth-modulation switching amplifier is presented. This exploits a double voltage/current feedback to obtain automatic equalization. Furthermore, the double feedback allows the output resonant filter to be included in the feedback path, compensating for its nonlinearity. Finally, the mixed feedback acts as intrinsic current limiter. Both theoretical analysis and simulations demonstrate that this system is feasible. The results were also confirmed by measurements using a prototype

Development and analysis of a PCB vector 2-D magnetic field sensor system for electronic compasses

A high-sensitivity vector two-dimensional (2-D) magnetic sensor system for low magnetic field measurements has been realized and tested. The system, made in PCB technology, consists of a double-axis Fluxgate magnetic sensor and the readout electronic circuitry, based on second-harmonic detection. The amorphous magnetic materials Vitrovac 6025X (25 /spl mu/m thick) and Vitrovac 6025Z (20 /spl mu/m thick) were used as the ferromagnetic core of the sensor. By applying a sinusoidal excitation current having a 450-mA peak at 10 kHz with Vitrovac 6025Z, the measured magnetic sensitivity was about 1.25 mV//spl mu/T. This value seems to be adequate for the Earths magnetic field detection (/spl plusmn/60 /spl mu/T). The full-scale linearity error was about 1.5%. By using the thicker Vitrovac 6025X and a sinusoidal excitation current having a 600-mA peak at 10 kHz, a maximum sensitivity of approximately 1.68 mV//spl mu/T with a linearity error of about 1.55% full scale in the range of /spl plusmn/60 /spl mu/T were measured. Due to the use of commercially available ferromagnetic materials, the vector 2-D magnetic sensor system presented is characterized by a very simple fabrication process, thus allowing low-cost devices to be designed.

Electronic interface for Piezoelectric Energy Scavenging System

The paper focuses on an electronic interface for systems, called Piezoelectric energy scavenging systems (PESS), which convert the energy of mechanical vibrations into electrical energy using a piezoelectric transducer. The output of the transducer is a strong and irregular function of time hence, to obtain a suitable supply source, an AC-DC conversion is needed. Classical rectifiers (half/full bridge or voltage doubler) with an output storage capacitor do not fit very well, since they work as peak detectors, converting only input voltages which are higher than their output voltage. The paper shows an electronic interface which is able to efficiently harvest the energy associated to the randomic voltage waveform delivered by a piezoelectric transducer. Its working principle is based on an inductive step-up converter; an active driving circuit is used to set the phases of the converter. The energy is stored into a capacitor which is also used to supply the active elements of the step-up converter, realizing a completely autonomous energy scavenging system. For this reason the whole circuitry has been designed with a very low-power consumptions, about 700 nA. A prototype was diffused in 5 V CMOS STMicroelectronics technology and measurements showed its effectiveness.

A Self-Powered Electronic Interface for Electromagnetic Energy Harvester

This paper presents a self-powered, active electronic interface for an energy harvesting system including a vibration-based electromagnetic transducer. The transducer provides a peak voltage of 3.25 V when operated close to its mechanical resonance frequency (about 10.4 Hz) and the power converter has been designed to transfer the harvested energy to a storage capacitor. The circuit is a full-cycle inductive step-up ac/dc converter able to process every voltage pulse coming from the transducer; furthermore, it is supplied by the harvested energy, making the system fully autonomous. The interface has been designed exploiting an accurate model of the transducer in simulations. A printed circuit board version of the interface has been simulated and built to gather experimental results and validate the idea. The system demonstrated to be able to build a voltage across the storage capacitor, which is limited only by the safe operating area of the devices.

Single-Variable Optimization Method for Evaluating Solar Cell and Solar Module Parameters

This paper presents a simple and accurate method to assess the single-diode model parameters of an illuminated and dark solar cell (SC) and its extension to a photovoltaic solar module (PVSM). The proposed method is based on the acquisition of experimental data related to the v -i characteristics of illuminated or dark SC/PVSM at fixed climatic conditions. The method presented in a previous work is generalized, the effects of measurement uncertainty of the results are discussed, and its application to different technologies are shown. A micropower monocrystalline SC, a monocrystalline SC, a polycrystalline PVSM, and an amorphous PVSM have been characterized using direct insulation in an outside environment with a correlation coefficient up to the noise limit.