Search

International audience; In the Giant Magneto-Impedance (GMI) sensors, one of the important features that limit thedetection performance is the equivalent magnetic noise level. In the white noise region, theperformance is mainly limited by the noise from the electronic conditioning not by the intrinsicnoise of the sensing element [1].Digital approaches for the electronic conditioning of these sensors were introduced in recentworks [1]-[2]. These promising approaches and the related software are commonly known asSoftware Defined Radio (SDR). They are based on a direct sampling of the voltage across thesensitive element using an Analog-to-Digital Converter (ADC). The amplitude-demodulation isthen digitally achieved using a Direct-Down-Conversion (DDC) to obtain the final sensor output.By far, this DDC generally includes a local Numerically-Controlled Oscillator (NCO), a set ofdigital filters, multipliers and digital gain as well asrounding circuits(necessary for the real-timeprocessing since only a limited number of bits can be processed by the hardware). All of theseinvolved digital functions are sources of noise [3].In this paper, thekey rules for the real-time implementation of these digital functions arepresented. The idea is to clearly show that the optimization of the noise performance requiresstringent implementation rules. These rules include, but are not limited to, the choice of theparameters of the NCO, the type andthe number of coefficients of the digital filters, the size ofaccumulators of the multipliers and of the rounding circuits. By respecting the proposed rules,the noise of the DDC can be less than the Least Significant Bit (LSB) of the ADC. In this case,the noise of the sensor in the white noise region is mainly dominated by the ADC.To validate the concept, a prototype of an on off-diagonal GMI sensor was developed using a 12-bit ADC having a sampling frequency of 125 MHz. The DDC was implemented using a Field-Programmable-Gate-Array (FPGA). With this realization, performance of 1.8 pT/√Hz isobtained in the white noise region (for sensitive element of 25 mm long). Improving the noiseperformance requires the use of an ADC with high resolution and/or higher sampling frequency.We show that with a technological jump of the ADC technology, the intrinsic noise of the sensitive element could be achieved with the digital concept.

International audience; The digital electronic conditioning for the high sensitivity Giant Magneto-Impedance (GMI)Similar interest in digital electronics wastioning of orthogonal fluxgate sensors in fundamental mode[3]. Byfar, such electronics is mainly composed of a Direct Digital Synthesizer (DDS) which supplies, to the sensitive element. The voltage across theDigital Converter (ADC).Down-Converter (DDC).This DDC mainly includes local digital oscillator (or NCO for Numerically-Controlled-, and digital low-passdecimation filters. These involved functions in such digital conditioning can be realized usingspecific integrated circuit)[1-2] or high-costIn this paper, a new concept for the implementation ofthis digital electronic conditioning for the GMI sensors is presented. This concept is based on theArray (FPGA) reconfigurable circuits. When compared withASICs, the FPGA exhibits advantages such as the high degree of flexibility in theThe architecture of a developeddiagonal GMI sensor using FPGA is illustrated in Fig. 1.

International audience; Fully digital electronics forworks [1-3]. The promising achievedwhite noise region [4]) in openfullmagnetometer.In this paperThe sensitive element was a CoFeSiB wirepick-up coil. The excitation currentvoltage across the pick-up coil was directly sampled by an AnalogThe digital demodulation (or Digital Down Conversion)from a Numerically-Controlleddigital filters). The digital output was converted into theAnalog Converter (DAC). The resulting voltage,field, Bf, using a resistor, Rf, and a 220static gains of the direct chain and of the feedback,functioning of the digital magnetometer, tT/V) so as the theoretical sensitivity,case, the measured sensitivityIt is to note that, on one hand,optimization of the sensitivity andof the noise level of themagnetometer was performed atthis stage. On another hand, thedigital processingintroduces sometime-delay which could impactthe stability of the loop. In thecurrent design, a delay of44μswas estimated andmeasured.This small delay shouldimpose a maximum frequencythe measured fields

National audience

National audience

National audience

International audience

International audience; Giant Magneto-Impedance (GMI) is a large change of the high frequency impedance of somesoft magnetic materials when they are subjected to an external magnetic field. This effect can beused to realizemagnetic sensors based on the impedance measurement of the sensitiveelement.Despite the great advantages of the GMI such as the high sensitivity, the potential ofminiaturization as well as the large bandwidth, there is still only a few concrete or industrialachievements of these sensors.One of the key features of these GMI sensors is their ability to measure both DC and ACmagnetic fields by the same sensitive element. This feature is of great value for a large palette ofapplications like the contactless current measurement(based on the measurement of the magneticfield produced by the measured current [1]).For this application, the mechanical flexibility ofsome GMI elements (like amorphous wires) is also another key andvery useful feature forrealizing toroidal flexible sensors (the sensing element forms a loop around the conductor ofinterest). Such flexible sensors are needed in many industrial environments where rigid toroidalstructures of the current sensor are not convenient.This paperpresents an overview of a full attempt of design andrealization of a quasi-industrialGMI-based currentsensor. The sensor combines DC/AC measurements in a same flexiblesensitive element. The paper addresses several practical problems encountered in a realmeasurement environment. In this context, main attention is paid to the impact of the influencequantities that largely condition the implementation strategies and solutions of the sensor. For theintended industrialapplication,these influence quantitiesfirstly include the operating temperature[2]. Its impact on the intrinsic sensor response is addressed in a quasi-industrial temperaturerange. Secondly, bending stress on the sensitive element is another involved influence parameterin the toroidal and flexible structure. This constraint effect is also characterized [3] - [4]. Finally,the third considered influence parameter is the effect of magnetic disturbances such as theeccentricity of the conductor of interest and the impact of the surrounding magnetic fields (earthfield or fields produced by others conductors in close proximity to the sensor). Differentstrategies of implementation and solutions, taking into account the impact of influenceparameters, are proposed and compared. A sensor prototype is realized and its performances aregiven and discussed .

International audience

The Giant Magneto-Impedance (GMI) is a large change of the impedance of some soft ferromagnetic materials, supplied by an alternating high-frequency excitation current, when they are submitted to an external magnetic field. This thesis presents the design and performance of an original digital architecture for high-sensitivity GMI sensors. The core of the design is a Digital Signal Processor (DSP) which controls two other key elements: a Direct Digital Synthesizer (DDS) and a Software Defined Radio (SDR) or digital receiver. The choice of these digital concepts is justified by the will to reduce the conditioning electronics noise that limits the equivalent magnetic noise level. The latter characterizes the smallest measurable field by the sensor. The developed conditioning system is associated with the off-diagonal magnetic configuration in order to increase the intrinsic sensitivity of the sensitive element. This magnetic configuration consists of the use of an additional a pick-up coil wound around the ferromagnetic material. This association also makes it possible to obtain an asymmetrical characteristic (odd function) of the sensor response near the zero-field point and to consequently allow for sensor implementation and use without any bias magnetic field. Thus, this choice eliminates, or at least minimizes, the problems related to the offset cancelling of the GMI devices. Also, it validates the advantage of this magnetic configuration, especially the choice of the operating point. Modeling of the noise performance of the entire measurement chain, including the digital conditioning, is performed. A comparison between the expected and measured equivalent magnetic noise levels is then carried out. The results yield general optimization laws for a digital GMI sensor. Using these laws, an optimized prototype of a GMI sensor is designed and implemented on FPGA. An equivalent magnetic noise level in a white noise zone region of approximately 1 pT/√ Hz is obtained. Furthermore, this work also makes it possible to validate the interest of digital techniques in the realization of a high-sensitivity measuring devices., La Magneto-Impédance Géante (GMI) consiste en une forte variation de l’impédance d’un matériau ferromagnétique doux parcouru par un courant d’excitation alternatif haute fréquence lorsqu’il est soumis à un champ magnétique extérieur. Ce travail de thèse introduit de nouvelles techniques numériques et les pistes d’optimisation associées pour les capteurs GMI à haute sensibilité. L'originalité réside dans l'intégration d'un synthétiseur de fréquence et d'un récepteur entièrement numérique pilotés par un processeur de traitement de signal. Ce choix instrumental se justifie par le souhait de réduire le bruit de l’électronique de conditionnement qui limite le niveau de bruit équivalent en champ. Ce dernier caractérise le plus petit champ mesurable par le capteur. Le système de conditionnement conçu est associé à la configuration magnétique off-diagonal pour accroître la sensibilité intrinsèque de l’élément sensible. Cette configuration magnétique consiste en l’utilisation d’une bobine de détection autour du matériau ferromagnétique. Cette association permet en outre d’obtenir une caractéristique impaire de la réponse du capteur autour du champ nul, et par conséquent de pouvoir mettre en œuvre et d’utiliser le capteur sans avoir recours à une polarisation magnétique. Ce choix permet ainsi d’éliminer, ou au moins de minimiser les problématiques liées aux offsets des dispositifs GMI, tout en validant l’intérêt de cette configuration magnétique, notamment sur le choix du point de fonctionnement. Une modélisation des performances en bruit de toute la chaîne de mesure, incluant le système de conditionnement numérique, est réalisée. Une comparaison entre les niveaux de bruit équivalent en champ attendus par le modèle et mesurés est effectuée. Les résultats obtenus ont permis de dégager des lois générales d’optimisation des performances pour un capteur GMI numérique. Partant de ces pistes d’optimisation, un prototype de capteur complet et optimisé a été implémenté sur FPGA. Ce capteur affiche un niveau de bruit équivalent en champ de l’ordre de 1 pT/√Hz en zone de bruit blanc. En outre, ce travail permet de valider l’intérêt des techniques numériques dans la réalisation de dispositifs de mesure à haute sensibilité.

In this paper, we propose a general model to evaluate the equivalent magnetic noise for an off-diagonal giant magneto-impedance (GMI) sensor utilizing a software-defined radio (SDR) as a digital receiver. Based on this SDR approach, an analog-to-digital conversion (ADC) of the voltage induced by the sensing element is directly performed. The signal processing is digitally achieved in real time. It includes a digital quadrature demodulation, a decimation, and a filtering. The approach allows to give the key rules for quantifying the white noise level of this digital implementation. The voltage noise spectral density of the ADC and of each SDR stage is given. The white noise level expected by the modeling was in good agreement with the measurements. With a sensing element composed of a 400 turn coil wound around a co-rich GMI wire, equivalent magnetic noise levels of 220 and 4.2 pT/ $\sqrt {Hz}$ were obtained at 1 Hz and in the white noise region, respectively. The potential improvement of these performances, allowed by this digital conditioning, as well as limitations is also discussed.

International audience

International audience

International audience

International audience

National audience

National audience