A. AGMI Measurements
Amorphous ferromagnetic (Fe0.06Co0.94) 72.5Si12.5B15 wires were used in this study. The wire has a nearly zero magnetostriction and it is supplied by Unitika Company. The wire with a diameter of 125 µm was cut for measurements of 2, 4.5 and 7 cm in length. Electrical connections were made using silver conductive paint. Agilent 4294A impedance analyser was used for MI measurements at 100 kHz. Earth's magnetic field was cancelled out using Helmholtz coils. The applied magnetic field was swept from –Hmax to +Hmax and then to –Hmax in all measurements. The GMI ratio as a function of the DC applied magnetic field H was defined as ΔZ/Z (%) = 100×[Z(H) − Z(Hmax)]/Z(Hmax), where the maximum applied field Hmax was taken to be + 3000 A/m or -3000 A/m.
Figure 1 shows the MI curve of wires 2 and 7 cm long at 100 kHz driving current frequency. The MI curves have a single peak shape and the results are in agreement with those previously found. 2 micro magnets were placed 1cm from the end of the wire to form AGMI (Fig.2). NdFeB magnets supplied from China have a maximum energy product of 406 kJ/m3. Magnets have a length of 1mm and a diameter of 300 µm. Fig. 3 presents the AGMI curves of wires with different lengths with two micro magnets, which are placed at 1 cm away from the ends of the wire as shown in Fig.2. Amorphous wires were kept at a distance of 1.5 cm from each other in all measurements to minimize the interaction between amorphous wires and magnets. Of course, there may still be some wire interactions, but since the purpose of the study is not in this direction, these interactions have not been studied in detail. It was found that changing of direction of magnets leads to change in the asymmetry direction in the AGMI curves, but no difference in the behaviour of the curve was observed. Basically, we have used this property of AGMI to design a magnetic field sensor.
The observed AGMI effect can be ascribed to a non-uniform bias field created by micro magnets. The bias field changes the static magnetization distribution and the effective permeability in the wire. In the presence of the bias field, the field dependence of the permeability becomes asymmetric, and the AGMI is detected, details are given in [42-43].
B. Sensor Circuit Design and Results
AGMI magnetic field sensor was developed using amorphous ferromagnetic wire and measurement circuit. Firstly, the sinusoidal signal with a frequency of 100 kHz and 1.2 V amplitude was applied to the ends of the 7 cm wires, which were attached to two micro magnets as shown in figure 2, and then measurements were made for the wires with the opposite direction of the magnets under the same conditions. The ac signal obtained between the two ends of the wire was applied to the peak detector and the dc output signal was obtained from there. As can be seen in Figure 4, AGMI curves were obtained as a function of the magnetic field. In general, the shape of the curves was found to be compatible with the AGMI curves given in figure 3. After one of these two curves was inverted, the addition of these two curves was done by computer and as seen in Figure 5, a linear curve was obtained as a function of the magnetic field.
A simple circuit was designed to perform the above-mentioned process. The circuit consists of two signal generators, two peak detectors and one differential amplifier. The circuit diagram is given in figure 6. One of the amorphous wires with 7 cm length in the circuit is attached to magnets as in figure 2a and in the other, as in figure 2b. Fig. 7 shows the effect of driving frequency and its amplitude on the output of the circuit (Vout). As can be seen from the curve (a) in Fig. 7, Vout does not show a linear change exactly as a function of the magnetic field when a signal with values of 100 kHz and 1 V is applied from the signal generators 1 and 2. In addition, when the amplitude of the signal applied from the signal generator 2 was increased to 1.2 V, a shift in Vout was observed (curve b in Fig. 7). It has also been found that the Vout signal is quite noisy for some values such as 130 kHz and 1 V the applied from the signal generator 1, 130 kHz and 1.2 V the applied from the signal generator 2 (curve c in Fig. 7).
The output of the circuit was measured for different frequency and amplitude values of the applied sinusoidal signal and the best results were found to be around 100 kHz and 1 V. Figure 8 shows the circuit output signal obtained by using wires with lengths of 2, 4.5 and 7 cm. Samples with a wire length of 2 cm show a linear output between the magnetic field values of approximately ± 250 A / m. The output signal varies between ± 0.31 V. In this measurement of 2 cm wire, 110 kHz and 1.2 V sinusoidal signal was applied to wire 1, and 130 kHz and 1.3 V sinusoidal signal was applied to wire 2. The output signal ranged from ± 0.7 V on wires of 4.5 cm length and showed linearity in the magnetic field region of ± 100 A / m. 115 kHz and 1.3 V sine signal was applied to wire 1, and 107 kHz and 1.4 V sinusoidal signal was applied to wire 2, in wire with 4.5 cm length. In the 7 cm long wires, the output signal showed linearity at lower magnetic field values, in this case, the signal with the values of 110 kHz and 1.18 V was applied to the wire 1 and the signal with the values of 90 kHz and 1.25 V to the 2 wire.