With the development of modern industry, real-time monitoring of operating parameters has become increasingly important. Parameter monitoring is divided into two main categories: electrical and non-electrical parameters. For the measurement of non-electrical parameters, the success of the measurement depends on the quality of the sensor and the extraction of the induced signal. Among various non-electrical sensors, capacitive sensors are arguably the most widely used. In industrial settings, they are widely applied as sensing elements for physical quantities such as flow rate, pressure, displacement, liquid level, velocity, and acceleration. In the coal industry, capacitive sensors are extensively used in production mining, safety monitoring, and coal preparation automation. Accurate and timely acquisition of capacitive sensor signals is of great significance for monitoring and control. I. Characteristics of Capacitive Sensors A capacitive sensor consists of two plates, with a simple structure. It can be assembled into various forms such as flat plates, curved surfaces, and cylinders. The plates are generally made of metal, capable of withstanding large temperature changes and harsh environmental conditions such as radiation. Due to geometric limitations, the capacitance of capacitive sensors is very small, typically only a few pF to tens of pF. Because capacitor C is very small, its capacitive reactance is very large, making it a high-impedance element. Due to the small capacitance, the required energy and movable mass are also small, resulting in a high natural frequency and good dynamic characteristics. The sensor's apparent power P = U²⁰ωC, with both C and P being very small, makes it susceptible to external interference, thus making signal extraction difficult. Furthermore, the small capacitance means that distributed capacitance and parasitic capacitance affect both sensitivity and measurement accuracy. Traditional measurement methods use analog circuits, primarily bridge circuits (ordinary AC bridge, transformer bridge, double-T diode bridge); pulse width modulation circuits; frequency modulation circuits, etc. Analog measurement methods have many circuit components and are easily affected by zero drift and temperature drift, especially for measuring small capacitors, making it difficult to guarantee measurement accuracy. II. Digital Measurement Principles Digital measurement first converts the sensor's capacitance into a frequency signal, commonly using LC oscillation and RC oscillation. Taking a 555 multivibrator as an example, if the measured capacitance is Cx, its oscillation frequency is f = 1.443/[(R1 + 2R2)Cx]. The oscillator circuit diagram is shown in Figure 1. The circuit structure is simple, less affected by external factors such as power supply, and the oscillation frequency is stable. Figure 1 C/f conversion circuit diagram. As can be seen from the working principle of the capacitance sensor, whether it is a change in the distance d between its plates, a change in the relative area S of the plates, or a change in the dielectric constant ε, it all manifests as a change in capacitance. Since f is inversely proportional to C, to measure Cx or ΔCx, f cannot be directly counted, and calculating ΔCx using Δf is even more cumbersome. However, the oscillation period T = 1/f = KCx is directly proportional to Cx. Therefore, if a precisely measurable parameter A is defined, and certain measures are taken to make A = (1/K)T = Cx, then measuring A is equivalent to obtaining Cx, and calculating ΔA is equivalent to calculating ΔCx. Most popular microcontrollers have external pulse triggering (INT0, INT1) and timer (T0, T1) functions. By using a pulse trigger with Cx participating in the oscillation to start and stop the timer, the value A corresponding to Cx can be obtained under software control. For example: To measure a capacitor Cx of approximately 1000pF, a 555 timer is used to create an oscillation circuit. During hardware adjustment, a standard 1000pF capacitor is used to replace Cx, and R1 is adjusted to make the output pulse frequency 2kHz. The microcontroller is initialized by defining INT0 as the external pulse input, triggered on the rising edge, and allowing INT0 interrupts; T0 is a 16-bit timer triggered by T0r. The system clock uses a 12MHz crystal oscillator, so the counter T0 increments by 1 every 1μs. The 16-bit timer counts for 65536μs. The design requires that when the capacitor is 1000pF, the parameter A is also 1000, meaning A varies with Cx, with a resolution of 1pF. Input the oscillation pulse to the INT0 terminal. In the first interrupt of INT0, start T0 for a total of 16 pulse cycles. Stop T0 timing at the 17th INT0 interrupt and read the values ​​of TH0 and TL0. When the pulse oscillation frequency is 2kHz, the period is 500μs, and 16 cycles is 8000μs, which is also the timing value of T0. Divide the T0 result by 8, i.e., shift TH0 and TL0 right by 3 bits, to obtain the value of A, which corresponds to the value of Cx. The standard frequency of the circuit can be adjusted using a frequency meter or by running a measurement program to read the value. When A=1000, it is ready. Use a stable monolithic capacitor for the 1000pF standard capacitor, and a multi-turn precision potentiometer for R1. After adjustment, replace C with Cx to perform the measurement. The circuit is easy to adjust, has stable performance, and the detection accuracy is ±1pF for 1000pF. III. Measurement of Small Changes in Capacitance In practical applications, it is often necessary to detect the change in capacitance of a capacitance sensor, ΔC = Ct1 - Ct0. Due to differences in sensor design and installation, the basic capacitance (the sensor's open-circuit capacitance, connecting wire capacitance, and other distributed capacitances) is relatively large, while ΔC is very small. If the basic capacitance is stable, the above methods can accurately measure ΔC. However, due to changes in the environment (medium temperature and humidity, static electricity, etc.), the basic capacitance (mainly connecting wire capacitance and other distributed capacitances) changes significantly, and ΔC is drowned out by noise, making it difficult to measure ΔC using general methods. The following introduces a method for measuring ΔC using a comparison capacitor. The principle circuit is shown in Figure 2. When the sensor is connected to the transmitter (555 oscillator), a double-core shielded cable is used. Core wire a is connected to the positive plate of the sensor capacitor as a signal lead; core wire b is connected as close to the sensor as possible, and its own wire capacitance constitutes the comparison capacitor; the shielding wire is connected to the negative plate of the sensor capacitor (usually the ground electrode). Core wires a and b are connected to the oscillator through an analog multiplexer. During operation, the multiplexer controls the connection of core wire a or core wire b respectively, measuring Ca and Cb at a certain moment, where Ca = Cx + Ca′ and Cb = Cb′ (Cx is the sensor's sensing capacitance, and Ca′ and Cb′ are the corresponding wire capacitance and distributed capacitance of core wires a and b, respectively). Since core wires a and b are completely in the same environment, Ca′ = Cb′. Calculating Ca - Cb = Cx yields Cx at different moments, and thus ΔC can be calculated. Figure 2 illustrates the principle of the capacitance comparison method. In an application of level detection using a capacitance sensor, the capacitance change due to the presence or absence of material is approximately 30 pF. The sensor's basic capacitance is 1000 pF, and the capacitance change caused by environmental influences ranges from 0 to 200 pF. By using the capacitance comparison method to detect ΔCx, the useful signal is accurately picked up. IV. Detection Software Block Diagram The software block diagram for acquiring capacitance Cx is shown in Figure 3, written in MCS51 assembly language. Using a microcontroller system not only allows for precise measurement of Cx and ΔCx, but also enables intelligent operation of systems using this sensor. The acquisition software can be called as a subroutine within the entire system. Figure 3 shows the software block diagram of the C-value coefficient . V. Conclusion Digital measurement of the capacitance sensor's capacity allows for long-distance transmission after the signal is converted into a digital signal locally at the sensor. The conversion circuit is simple and stable. The comparative capacitance method for detecting ΔCx overcomes the influence of environmental changes on conductor capacitance and distributed capacitance, ensuring a reliable detection signal and significantly enhancing the system's anti-interference capability. Both methods have achieved satisfactory results in specific applications of capacitive coal pulverizer level sensors.