The output voltage is tested by a DPO3014 oscilloscope manufactured by Tektronix (Beaverton, OR, USA) with the input resistance 10 MΩ. In order to prevent power frequency interference, a filter circuit is used in the measurement to filter out the 50 Hz power frequency signal. A data acquisition card is used to collect the amount of charge on the electrodes.
3.1. Nano-Iron Suspension Test
When tilted to the left, the positive charge contained inside the liquid and the negative charge on the left PTFE surface cannot reach electrostatic equilibrium, which causes a right-to-left electron shift between the copper electrodes. Similarly, tilting to the right produces electrons moving in the opposite direction. The tilt surface test setup and charge distribution are as shown in
Figure 2a,b. Both the suspension volume and the nano-iron powder concentration have a significant effect on the sensitivity of the sensor. The sensor is tested at a horizontal glass table top. Respectively 1 mL No. 1, 2, 3 nano-iron suspension is injected into the polymer pipe. First, the motion-balanced sensor is tilted to the left to the maximum angle. The output voltages corresponding to liquid 1, 2 and 3 are as shown in
Figure 2c. Then, the motion-balanced sensor is tilted to the right to the maximum angle. The output voltages are shown in
Figure 2d. Secondly, 2 mL of liquids No. 1, 2 and 3 are injected into the polymer pipe and the output voltages of the left and right tilts of the sensor are shown in
Figure 2e,f, respectively.
It can be seen from
Figure 2 that when the motion-balanced sensor is tilted leftmost, the voltage value is the lowest, which is a negative value; when the motion-balanced sensor is tilted rightmost, the voltage value is the highest, which is a positive value. Therefore, the tilt angle of the object can be judged by the output voltage value of the sensor. The comparison test of liquids 1, 2 and 3 shows that the suspension concentration has a significant effect on the experimental results. The higher the nano-iron powder content is, the higher the output voltages are. Therefore, it is advisable to use a saturated suspension for maximum sensitivity. The nano-iron suspension is saturated when 0.05 g of nano-iron powder is added to 50 mL of ethanol and liquid 1 is used in the subsequent tests.
The voltages in
Figure 2e,f are significantly higher than the voltages corresponding to
Figure 2c,d, indicating that the sensitivity on the condition of 2 mL liquid is greater than 1 mL liquid. This is also consistent with the theoretical analysis described above. Because in the equilibrium position, when the liquid is 2 mL, the suspension just covers the part without the electrode in the middle, no matter which direction the liquid flows, a potential difference will be generated on the electrode. Therefore, the sensitivity of the sensor at this condition is the highest. Conversely, if the liquid volume exceeds 2 mL, the liquid covers both the left and right electrodes when the sensor is tilted in one direction. In order to achieve electrostatic equilibrium, the number of electrons transferred between the left and right electrodes will be relatively reduced and the motion-balanced sensor output voltage will decrease. The output voltage of the three curves will drop by about 35 mV when adding 3 mL of liquid. This means that it is best to inject enough liquid into the sensor tube and not cover the electrode.
At the same time, we can see from
Figure 2 that the output voltage will fluctuate greatly in the first few seconds during the deflection experiment. Due to the fluidity, the liquid will flow in one direction. However, the output voltage tends to be stable as the liquid is at rest.
3.2. Balance Sensing
When the sensor is tilted, the voltage between the two electrodes changes, so it can be used to detect whether the object is in equilibrium. Due to the structure and size limitations of the motion-balanced sensor, the maximum tilt angle to the left and right are both 20 degrees. Therefore, the range of tilt angle of the object that can be measured is [−20, 20] degrees. Here the negative angle sign is tilted to the left and the positive angle sign is tilted to the right. Firstly, test and analyze the dynamic characteristics of the motion-balanced sensor. Press the two ends of the sensor with fingers and rotate the sensor at a constant speed between −20° and 20°. The output voltage of the sensor during fast rotation and slow rotation is shown in
Figure 3a,b respectively.
As can be seen from
Figure 3, the amplitude and frequency of the voltage during rapid rotation are higher than the slow rotation due to the larger flow amplitude of the liquid during rapid rotation, indicating that the motion balance sensor can be used to detect the degree of deflection.
Secondly, the motion-balanced sensor measuring the static tilt angle. 2 mL nano-iron suspension is injected into the polymer pipe. The motion-balanced sensor is fixed to a flat plate and tilted at a specified angle. The motion-balanced sensor is tilted from the equilibrium position to the right by 10 degrees and hold for some seconds, then continue to tilt to the right to 20 degrees. The output voltage curve is shown in
Figure 4a. And the tilting process is shown in
Figure 4b. The output voltage rises from around 1200 mV to around 1450 mV and then rises to 2000 mV, then decreases to 1757 mV. The output voltage reaches a maximum and then falls. The test results show that there is a significant voltage changing process when the sensor is tilted from one angle to another, which is due to the fluidity of the liquid.
Figure 4a shows that there is a direct relationship between the output voltage and the tilted angle. The variation of the voltage during the rotation at different angles is directly related to the rotational speed. The output voltage corresponding to each angle remains unchanged when the sensor maintains a steady state for a period of time. Thus, the output voltage can be used to determine the tilt angle of the object.
The motion-balanced sensor is slowly rotated from −20 degrees to 20 degrees. The output voltage is shown in
Figure 4c and the sensor tilt process is shown in
Figure 4d.
Figure 4c shows that the output voltage of the sensor is the lowest at −20 degrees and the voltage gradually rises during the tilt to the right. Due to the flow of liquid during the rotation, the maximum output voltage is higher than the output voltage stabilized at 20 degrees when the sensor is tilted to 20 degrees.
The corresponding relationship between the output voltage and the tilt angle was analyzed. The voltages are measured every 5 degrees. The test result is shown in
Figure 4e.
The tilt angle range is [−20, 20] degrees for the motion-balanced sensor.
Figure 4e shows a linear relationship between the output voltage and the tilt angle. The angle sensitivity of the sensor is 3.65°/100 mV.
The output voltage in
Figure 4 is higher than that in
Figure 3. This is due to the fluidity of the liquid. When in a stationary state, the liquid can reach the position corresponding to the tilt angle of the sensor and in the continuous motion state, the deflection angle of the liquid is smaller than the deflection angle of the outer wall of the sensor. The above test shows that the sensor is capable of detecting the tilt angle of the object. Therefore, the motion-balanced sensor can be used for balance condition monitoring of buildings, transmission line towers, etc. The sensor is useful for giving an early warning of the danger of an object falling over. It can also be used for rehabilitation of stroke patients and assist doctors in assessing the patient’s condition.
3.3. Motion Sensing
The nano-iron suspension flows within the polymer pipe and creates a potential difference between the two electrodes when the sensor is excited by linear acceleration. Therefore, the motion state of the object can be measured by the proposed sensor. The sensor is fixed to a shell by bearings (
Figure 5) instead of being directly attached to the object to increase the output voltage. The two bearings sandwich the central portion of the sensor from both sides of the shell but do not penetrate the side of the sensor so that it can be shaken around the centerline without being affected. The motion-balanced sensor data is amplified by a charge amplifier and then acquired by an oscilloscope. A single degree of freedom piezoelectric acceleration sensor is attached to the outer wall of the shell as shown in
Figure 5. Acceleration sensor data is collected by the data acquisition card and sent to the computer. The sensitive component of this piezoelectric acceleration sensor is PZT-5, with sensitivity, measurement range and capacitance of 2.73 pC/m/s
2, ±30,000 m/s
2 and 1027 pF, respectively.
The shell is fixed on the tester’s body. The body will produce different vibration amplitudes and vibration frequencies when people are doing different exercises and the amplitude and frequency of the sensors connected to them will be different, so the output characteristics will also be different. We can analyze this output characteristic to understand the motion state of the human body. The output voltage of the motion-balanced sensor and commercial acceleration sensor is collected for analyzing the motion characteristics of the tester. The voltage amplitudes are quite different because the sensitivity of the two sensors is different. The voltage of different sensors is normalized for better comparative analysis. At the same time, the output voltage of the normalized motion-balanced sensor is subtracted by 1 and the two sensor output voltages after processing are placed in the same figure.
In the walking state, the processed voltage is shown in
Figure 6a. It can be seen that the two voltages have a similar waveform law, but the fluid response speed is different from that of PZT-5 so that the phases of the two voltages are different. The motion-balanced sensor’s output voltage waveform is smoother. This is because the fluid itself has good low-frequency characteristics, which can filter out high-frequency noise interference in the environment and is more suitable for low-frequency motions detection, such as human motion etc. Fourier analysis is performed on the voltages of the acceleration sensor and the motion-balanced sensor and the amplitude-frequency characteristics are shown in
Figure 6b. The amplitude-frequency characteristics show that the main energy is concentrated around 1 Hz, which directly reflects the frequency of human walking.
In the running state, results are shown in
Figure 6c.
Figure 6c shows that the two voltages have a similar waveform, but the phases are different. The voltage waveform of the motion-balanced sensor is also smoother. The amplitude-frequency characteristics are shown in
Figure 6d. It can be seen that the main energy is concentrated around 2 Hz, which reflects the frequency of human running.
In addition, statistical analysis is performed on the test voltages of the motion-balanced sensor during running and walking. The sampling duration and number of data points are 10 s and 10,000, respectively. During walking and running situation, the average absolute values of the output voltage are 123.26 mV and 172.04 mV, respectively and the variances are 20,264.47 and 45,214.63, respectively. It is clear that the average absolute values and variance are significantly larger than walking in the running state. Therefore, the motion state of the object can be judged by analyzing the amplitude-frequency characteristics, the absolute average value, the variance and etc. of the motion-balanced sensor output signal.
The above analysis shows that the sensor can monitor balance and motion, and if the sensors are placed in three orthogonal directions, three-dimensional balance and motion can be measured. Moreover, the proposed sensor is inexpensive to manufacture. In addition, polymer pipes, nano-iron suspension and electrodes are easy to industrialize and large-scale applications.
The rate of precipitation of metal particles is primarily related to the ratio of dispersant and metal particle size. The dispersant PVP masses of suspensions ①, ② and ③ in
Figure 7a are 1.144 g, 2.288 g and 3.432 g, respectively, with a ratio of 1:2:3. The precipitation is observed every 4 h. It can be seen that the suspension ① precipitates faster and the suspensions ② and ③ precipitates less after 4 h. Compared to suspension ②, the ③ will also have a little precipitation after 8 h. We continue to observe the precipitation in the next few days: suspensions ② will produce a little precipitation after 24 h, which has the slowest precipitation rate and then the precipitation rate will slow down afterwards. It can be seen that the precipitation of the iron powder suspension in the PVP saturated state is the slowest. After 24 h, it works best in the sensing field after removing sediment.
In order to analyze the effect of the metal particle size on the precipitation rate, suspensions ④ and ⑤, shown in
Figure 7b, were prepared using iron powders having diameters of 400–1200 nm and 1000–1500 nm, respectively. The precipitation is observed every 4 h. It can be seen that both liquid precipitates are rare after 4 h. In addition, it is the same after 8 h. We continue to observe the precipitation in the next few days: both liquids will have a little precipitation after 24 h and the liquid No. 5 will produce a little more precipitation. It can be seen that the smaller the particles are, the slower the precipitation rate of the suspension is. It is better to use small particles when making a suspension.
Since the strong magnetic field has an effect on the characteristics of the nano-iron powder suspension, the sensor should not be used around high voltage lines and signal transmission towers. The suspension will produce a large amount of precipitation after 4 days and we needed to resume the sensor to make measurements with ultrasonic wave shaking. If the sensor is shaking for a long time, we don’t need to do any processing on the sensor.