This online discussion mainly introduces the principle and application of ADI’s advanced impedance and capacitance measurement converter. This paper includes two parts: the first part mainly discusses the impedance converter and the second part mainly discusses the capacitance converter. In these two parts, we first review the main characteristics of resistance and capacitance measurement methods, and then introduce the advanced impedance digital converter and capacitance digital converter launched by ADI for these two applications.

One. Impedance converter

Impedance definition

The circuit components in the real world are very complex. In addition to the resistance characteristics, they also show the capacitance characteristics and inductance characteristics. Therefore, the concept of impedance is introduced. Impedance is a general concept. It not only considers the resistance value of the element at a specific frequency, but also considers the phase relationship at this frequency.

By measuring the impedance at a series of frequency points, the characteristics of the component to be tested can be obtained. This is not only the basis of impedance spectrum method, but also the theoretical basis of many industrial, instrumentation and automotive sensor applications.

Impedance spectrum method impedance spectrum method makes use of the different frequency characteristics of resistors, inductors and capacitors. The ideal resistor has a constant impedance for all frequencies, the impedance of the ideal inductor will increase with the increase of frequency, and the impedance of the ideal capacitor will decrease with the increase of frequency.

By sweeping the frequency of unknown elements, such as investigating the relationship between impedance and frequency of a chemical sensor, we can determine whether it is resistive element, inductive element or capacitive element. The relationship between the real part and imaginary part coefficients of the usually generated response signal and frequency is shown in Fig. 1.

Impedance spectrum method includes two levels of applications, including:

1 qualitatively determine the impedance characteristics of the sensor. Firstly, it is determined that the characteristics of an element or sensor are “normal” under normal working conditions, and then the system detects its impedance characteristics under acceptable constraints. Its typical applications are metal identification and proximity detection.

2. The impedance spectrum method is used to quantitatively measure the actual impedance parameters of the components to be measured. In this case, it is necessary to establish an equivalent circuit model to simulate the components to be tested. The element to be tested is usually an electrochemical or biomedical phenomenon, so it is necessary to adjust the equivalent circuit according to the measured impedance characteristics to make it best match with the measured data. This method can be used to analyze specific substances to be measured.

One of the important applications of impedance spectrum method is impedance analysis.

Typical impedance analysis system

Figure 2 shows a simplified functional block diagram of a typical impedance analysis system. The frequency excitation is generated by DDS. The output frequency of DDS is usually filtered and amplified before being applied to the unknown impedance. The waveforms before and after the unknown impedance are sampled by ADC, and then sent to DSP for further processing. This simple functional block diagram masks several basic problems. First, the ADC must synchronously sample the signal in all frequency ranges, so as to compare the excitation waveform and response waveform in order to obtain phase information. The optimization of this process is the key to improve the overall performance of the system. Second, because a large number of discrete components are used, component error, temperature drift and additional noise will have an adverse impact on the measurement accuracy, especially under the condition of small signal operation. In addition to component selection and PCB size, a large number of discrete components will also bring computational difficulties to error analysis.

Therefore, ADI has developed a new impedance converter structure to solve the above problems.

The impedance converter AD5933 is an impedance converter with 1m sampling rate and 12 bit resolution, referred to as IDC for short. It solves many problems listed above, and integrates excitation circuit and response processing circuit. It uses DDS to generate a predetermined sweep frequency, with a control resolution of 0.1 Hz and a maximum frequency of 100 kHz. The unknown impedance is excited by the above frequency, and then the response waveform is digitized by the on-chip ADC.

One of the key features of the device is that it can do discrete Fourier transform (DFT), and provide the real part and imaginary part values of each sweep point to the user. Using these values, the amplitude and relative phase information of the response signal can be easily calculated. Its working principle is to first apply the frequency to an unknown impedance, which can be resistive, capacitive, inductive or a combination of several. Users need an external feedback resistor to prevent the response signal from exceeding the range of ADC and ensure the linear characteristics of the system. Through resistance selection, the impedance from 100 to 10 m can be measured, and the measurement accuracy can reach 0.5%. The output of ADC will be sent to the on-chip DFT module for digital processing to extract its real part and imaginary part. With the support of the evaluation software, the external impedance can be connected to the evaluation board, so it is easy to generate the relationship curve between impedance and frequency.

Capacitance digital converter

Disadvantages of single electrode capacitance sensor

1) Virtual capacitive coupling from sensor to ground is required.

2) Parasitic capacitance exists in the signal measurement path, so it will lead to unrepeatable and unpredictable measurement results.

3) Additional input protection circuits cannot be added.

Advantages of double electrode capacitance sensor

1) It does not depend on capacitance to ground coupling.

2) It is insensitive to parasitic capacitance, which means that it only transmits the signal charge to the converter. This enables predictable performance and much simpler design.

3) The design engineer can add additional input protection circuits as needed.

Capacitance measurement method

Traditionally, the difficulty of detecting the charge change of capacitive sensor is to realize the signal processing front end of capacitive input with high performance and low cost. Generally speaking, the measurement of capacitance requires an excitation source to be applied to the capacitor electrode. Then, the change of capacitance is converted into the change of voltage, current, frequency or pulse width.

Typical capacitance measurement methods include:

a. In the “direct” method, first charge the capacitor to be measured with a specific current source according to the specified time length, and then measure the voltage at both ends of the capacitor. This method requires low current, high precision current source and high input impedance to measure the voltage.

b. Firstly, an RC oscillator is composed of the capacitor to be measured, and then the time constant, frequency or period are measured. This method is simple, but usually can not achieve high accuracy.

c. Measure the AC impedance of the capacitor to be tested. A sine wave signal source is used to excite the capacitor, and then the current and voltage of the capacitor are measured. A four wire system is used to connect to the capacitor, a proportional measurement method is used, and a synchronous demodulator is used to provide the most accurate results. However, this circuit is very complex and requires a large number of devices.

∑- Δ ADC

a.∑- Δ Architecture

∑- Δ Is a mature technology, which has been used for many years in high-performance ADCs that usually require 16 bit or higher resolution. Figure 3 shows the simplified architecture of industry standard single chip ADC. The CIN and CREF of the capacitor are periodically switched to the voltage input VIN and the reference input VREF, which store the charge in the integrator cint. The comparator detects the output of the integrator and controls the phase of the input switch to form a closed-loop feedback loop, so that it balances the charge flow through the voltage input path and the reference input path. The purpose of all closed-loop feedback loop systems is to achieve balance, or in other words, to achieve zero error. This is sigma- Δ ADC should strive to achieve its goals.

The comparator will output a code stream consisting of ‘0’ and ‘1’, which will vary with the amount of charge used for loop balance. The amount of charge is proportional to voltage and capacitance. Since the value of capacitance is fixed, the density of ‘0’ and ‘1’ represents the ratio of input voltage (VIN) to reference voltage (VREF). Therefore, the constant code stream of all ‘1’ represents full scale, and all ‘0’ represents zero or zero. After the subsequent digital filter processing, we can get the input voltage conversion results.

The inherent characteristics of this architecture are high linearity and high precision, but there will be a compromise between resolution and speed. In order to obtain high accuracy, the digital filter will take a long processing time. The resolution of the converter is limited by system noise. In addition, the output data rate is limited by the clock frequency, which depends on the switching speed, integrator bandwidth and comparator setup time.

b.∑- Δ Capacitance sensor

Standard Σ- Δ The ADC realizes the conversion by switching between the fixed capacitor in the chip and the external input. If the charge is proportional to the voltage and capacitance, in this case, since the capacitance changes, why not use a fixed voltage instead of a fixed capacitance?

Based on this, an improved Σ is proposed- Δ ADC circuit. The fixed input voltage can be regarded as a voltage excitation source, and the variable capacitor moved out of the chip can be regarded as a capacitive sensor. As a result, the output data will represent the ratio of sensor capacitance to CREF change. The charge at the input is the sum of constant capacitance and variable capacitance. Among them, the capacitance to be measured is variable capacitance. Through the capdac in the chip (not shown here), the charge generated by the constant capacitance can be subtracted from the charge feedback loop.

∑- Δ The above innovative idea of ADC circuit allows capacitive sensor and Σ- Δ ADC is directly connected. It has the inherent characteristics of high resolution, high precision and high linearity. When actually using this circuit architecture, there are the following two characteristics:

a. The interface is insensitive to the capacitance between the sensor node and the ground and the leakage current to the ground, which will cause specific restrictions according to the actual circuit.

b. The complete capacitive digital converter can be implemented in a single chip, so it has the characteristics of high integration, easy system implementation, high repeatability and high reliability. Finally, it significantly reduces the system cost.

Application examples of capacitive sensors

A typical application of capacitive sensor is pressure detection. The following takes this as an example to introduce the specific application of capacitive sensor. The circuit diagram of the pressure sensor is mainly composed of two capacitor plates, as shown in Figure 4. When pressure is applied to the sensor, the capacitive plates will be close to each other. This effectively reduces the distance’d ‘between the two plates, thereby increasing the capacitance. A temperature sensor can be used to detect the change of sensor temperature, because its characteristics will change with temperature. An ADC voltage channel of the CDC is used to periodically measure temperature. Pressure sensors are widely used in industry, automobile and medical applications.

ADI has launched the first batch of high-precision single-chip capacitor digital converters. The capacitance sensor allows a constant common mode capacitance input range of 0 ~ 17 PF and a full capacitance input range of 4 PF. The maximum effective resolution (ENOB) of the chip is 21 bit. In terms of capacitance value, this means that the minimum change value of input capacitance that the chip can distinguish is 4 AF (alpha) – about 25 AF of “actual” noise free capacitance. Someone asked what 1 AF is. 1 AF equals 10 “18F (farad). Therefore, 25 AF equals 0.025 FF.

The device can reach the specified technical indexes in the temperature range of 40 ~ + 125 ℃, and the maximum power consumption current is only 850 A. It has an I2C interface and is packaged with 16 pin tssop. There are three devices in the first batch of CDC series products, namely AD7745, ad7746 and AD7747.

Responsible editor: GT