“Inductive displacement sensors are widely used in the detection of small displacements, but in some projects, the measurement accuracy and sensitivity of existing sensors cannot meet the measurement requirements. In order to solve this problem, the front-end signal processing circuit of the sensor is improved, and the upper and lower coils of the sensor are connected in parallel to form an LC circuit, and the performance of the circuit is improved by using the resonance effect of the LC circuit to improve the sensitivity of the signal source. The performance of the circuit after paralleling capacitors of different sizes is simulated, and the generated curve is fitted by the least squares with Matlab, and the capacitor value and the parallel method that make the circuit performance optimal are obtained by comparison.result

“

Inductive displacement sensors are widely used in the detection of small displacements, but in some projects, the measurement accuracy and sensitivity of existing sensors cannot meet the measurement requirements. In order to solve this problem, the front-end signal processing circuit of the sensor is improved, and the upper and lower coils of the sensor are connected in parallel to form an LC circuit, and the performance of the circuit is improved by using the resonance effect of the LC circuit to improve the sensitivity of the signal source. The performance of the circuit after paralleling capacitors of different sizes is simulated, and the generated curve is fitted by the least squares with Matlab, and the capacitor value and the parallel method that make the circuit performance optimal are obtained by comparison. The results show a doubling of sensitivity with a slight loss of linearity.

The essence of the inductive displacement sensor is to convert the change of the sensitive element into the change of the voltage amplitude for measurement. It is widely used in the detection system for detecting small displacement, so the measurement accuracy and sensitivity of the inductive sensor are very high. . The sensitivity of the inductive displacement sensor refers to the ratio of the increment of the output voltage to the increment of the displacement of the side head. Increasing the sensitivity, other things being equal, can improve the minimum resolution and accuracy of the system. There are many ways to improve the sensitivity of the inductive sensor, but at present, it is mainly realized by improving the signal conditioning circuit of the inductive sensor. This paper attempts to change the output signal of the sensor through the resonant circuit, and increase the sensitivity of the sensor from the signal source. This method is equivalent to improving the sensor itself, so that it can also be compatible with other improved technologies such as sensor excitation source, output signal processing, computer software compensation, etc. to jointly improve the performance of the entire system.

1 Model establishment of the improved circuit

1.1 Half-bridge improved circuit

As shown in Figure 1, if there is no C1 and C2, it is an ordinary half-bridge circuit, and the dotted box is the equivalent circuit of the inductive sensor. The displacement of the sensor probe drives the iron core in the solenoid to move up and down, thereby changing the inductance value of the upper and lower coils. The two coils are equivalent to the series connection of pure resistance and pure inductance. In the figure, R1 and L1 form the upper coil, R2 and L2 form the lower coil, and the output is connected to the upper coil. In the actual sensor, the wiring between the coil and the output will not change, but the inductance is changed by the movement of the iron core, so R1 and R2 are fixed.The output voltage
　

In Fig. 1, after the capacitors C1 and C2 are connected in parallel with the upper and lower coils, a resonant circuit I and a circuit II are respectively formed. If the core is at the bottom: Loop II is resonant, and loop I is detuned. When the iron core is at the top: loop I resonates, loop II detunes. Because the impedance of the resonant circuit at resonance will be much greater than the impedance at detuning. It can be qualitatively concluded that when the iron core is at the bottom, the amplitude of Uout will be smaller than that without the capacitor, and when the core is at the top, it will be larger than that without the capacitor, so the sensitivity will increase. However, the change in the middle of the bottom and top, and its linearity need to be determined by subsequent simulation.The output voltage
　

1.2 Full-bridge improved circuit

Ordinary full-bridge circuit Figure 2(a), the upper and lower coils of the sensor are connected to matching resistors R3 and R4 respectively. When L1=L2, the bridge is balanced. When the upward displacement of △X occurs, the iron core moves up, and L1 increases △L , L2 decreases △L, the change of Uout will increase nearly twice that of the half-bridge mode, and the output voltage
　　

As shown in Figure 2(b) and Figure 2(c), the parallel and series capacitors C1 and C2 are used for the upper and lower coils, respectively, to form the resonant circuit I and the circuit II. The changes in the circuit performance of these two methods are observed through subsequent simulations.The output voltage

2 Simulation of the circuit

2.1 Simulation platform and simulation conditions

The simulation platform uses MulTIsim, which is a Windows-based simulation tool launched by National Instruments (NI) Co., Ltd. It is suitable for the design of board-level analog/digital circuit boards. It includes the graphic input of circuit schematic diagram, circuit hardware description language input mode, has a huge component library and a comprehensive instrument library and rich simulation analysis capabilities. It is used to simulate the circuit before and after improvement.

Before the simulation, make some settings for the simulation conditions based on the actual situation of the project:

(1) Excitation power supply: alternating current with a frequency of 7.5 kHz and a peak-to-peak value of 5 V.

(2) Sensor: the total inductance value is 10mH differential inductance sensor, the linear range is 3 ~ 7mH, and the resistance value of the inductance itself is 54Ω.

R1 and R2 are fixed as described above, so R1 and R2 are 27Ω. The corresponding pure inductances L1 and L2 will change with the displacement line, satisfying L1+L2=10 mH (3

2.2 Simulation process and results

For the circuit II of the half-bridge, since the iron core is expected to resonate when the core is at the bottom, and the circuit I resonates at the top, because the variation range of L1 and L2 is 3 ~ 7 mH. Loop II resonates when L2 is 7 mH, and loop I resonates when L1 is 7 mH. Calculate C1=C2=65 nF according to the simulation conditions. To simplify the simulation, take C1=C2, and conduct simulation at intervals of 5 nF from 55 to 100 nF around 65 nF to observe the performance of the circuit. The simulation results are shown in Figure 3.
　

It can be seen from the figure that different capacitance values ​​have a great influence on the performance of the circuit. If the selection is not appropriate, the system performance will be degraded. Only by selecting the capacitor size of the appropriate capacity can the measurement sensitivity be improved while keeping the linearity error as small as possible. Therefore, when the curve is in the L1=3-7 mH segment, the sensitivity is the highest and the linearity is the best, and the least squares calculation is performed. The comparison between it and the ordinary half bridge is shown in Figure 4.
　

It is calculated by Matlab that the common half bridge is in the 3-7 mH segment, the voltage variation range is 1.5-3.5 V, and the sensitivity of the voltage to the inductance is 0.5 V/mH. Linearity is approximately 1. After fitting the straight line by the least square method to Fig. 4(b), the output voltage varies from 0.77 to 4.39 V in the range of 3.8 to 6.3 mH. The linearity can reach 2.39%, and the sensitivity is 1.448 V/mH.

The simulation of the full-bridge circuit is similar to the half-bridge method. It should be noted that the bridge is expected to be balanced when L1=L2=5 mL, so the selection of matching resistors needs to be calculated according to the simulation conditions

For circuit I: R3=R4=|jw×0.005+R1|=237Ω; circuit II: R3=R4=|(jwL+R1)∥(1/jwC1)|=817Ω; circuit III: R3=R4 =|jwL+R1+(1/jwC1)|=98Ω.

For circuits using capacitors, the circuits under different capacitance values ​​are also simulated, and the one with the best performance is selected as shown in Figure 5.

The common full bridge is in the 3.8 ~ 6.3 mH segment, the voltage variation range is -1.2 ~ +1.3 V, and the voltage sensitivity to the inductance is 1 V/mH. The linearity is approximately 1.38. Using Matlab to perform the least fitting straight line for Figure 5(b) and Figure 5(c) as shown in the figure, in the 3.8 ~ 6.3 mH segment, the output voltage in parallel mode varies from -2.66 to +2 .66V, the sensitivity is 2.130V/mH and the linearity can reach 1.68%. The output voltage range of the series connection is about -1.25 to +1.25V, and the sensitivity is about 2.130V/mH and the linearity can reach 1.33%.

3 Analysis and conclusion

As shown in Table 1, for the sensitivity and linearity of each circuit, the sensitivity can be improved under the condition of less loss of linearity. For the half bridge, although the sensitivity is increased by nearly 200%, the linearity is sacrificed more. There is little increase in sensitivity with the series capacitor approach. The best performance is the full-bridge circuit with parallel capacitors, the sensitivity is increased by 113%, and the loss of linearity is small, only 21.7% higher than the original, and in practical applications, it can be achieved by software compensation and pre-calibration. Make up for the lack of linearity.

How to Improve the Measurement Accuracy and Sensitivity of Inductive Sensors

Comprehensive theoretical analysis and simulation results, when the excitation source is determined and the parameters of the inductive sensor are determined, an appropriate capacitance value can be obtained through calculation. When this capacitance is connected in parallel with the two coils of the sensor, the measurement sensitivity will be significantly improved. , while still maintaining good linearity, so as to achieve the purpose of improving and improving the performance and minimum resolution of the inductive sensor.