Current measurement principles and products

In circuit design, current measurement is widely used, and the main fields are divided into three categories: in measurement, the electric meter is used to measure the current; in protection, the current is often directly related to the power. If the current is too large, it means that there is a short circuit in the system and protection is required, so current measurement is used; in control, such as motor control, battery charging and discharging, etc., current measurement is required.

The methods of measuring current are generally divided into direct and indirect methods. The direct method is generally carried out through resistance. According to Ohm's law, the magnitude of the current is proportional to the voltage. Therefore, the magnitude of the current passing through can be obtained by measuring the voltage difference of a small resistor. Indirect measurement is generally obtained by monitoring the magnetic field generated by the current. Since the current itself generates a magnetic field, the magnitude of the current is proportional to the magnetic field. Therefore, the magnitude of the current passing through can be obtained by measuring the magnitude of the magnetic field. The direct method is used to measure relatively small currents and low voltages. The indirect method does not have any conductive relationship, so it can be used to measure relatively large currents and relatively high voltages.

Indirect current measurement

The most commonly used indirect current measurement is the Hall sensor , which measures the current through the Hall phenomenon and outputs an analog output. Another new technology is to use the VAC sensor, which is a newer current sensor developed by the German Vacuumschmelze company. The difference from the Hall sensor is that TI and VAC have a special matching sensor chip DRV401. The system block diagram is shown in Figure 1. The output current is passed through the DRV401 and the integral filter to generate a certain current source as feedback to achieve magnetic balance. The closed-loop control is used to ensure that the entire ferrite is not affected by saturation, thereby ensuring that the output accuracy can be improved.

Direct current measurement

Direct current measurement methods are divided into two categories: analog output and digital output.

Analog output direct current measurement

Analog output is divided into low voltage (Low Side) and high voltage (High Side), and digital output can be divided into isolated and non-isolated types. Low voltage refers to the current measurement using a low voltage current sensor, and high voltage refers to the current measurement using a high voltage current sensor.

Direct current measurement uses a small resistance current sensor, which has high accuracy and temperature drift characteristics. The absolute value change can be achieved through simple compensation later, but the temperature drift is unpredictable, so compensation is relatively difficult. For current sensors, the temperature drift characteristic is the most important. For example: 1 resistor R=1mΩ, accuracy is %26#177;1%, TCR=%26#177;200ppm/℃, output current I=33A, output power P=1W. When the maximum current is 45A, the output power is 2W, and the temperature will change in this case. Assuming the temperature drift is 75℃, if TCR=20ppm/℃, the output accuracy changes to TCR=(75℃)%(20ppm/℃)%(0.0001%/ppm)=0.15%; if it is a common resistor, the temperature drift characteristic reaches 800ppm/℃, then TCR=(75℃)%(800ppm/℃)%(0.0001%/ppm)=6%. According to the different accuracy requirements of the system, you can choose a current sensor with different temperature drift characteristics.

Since the output after passing through the resistor is a voltage signal, the signal is often relatively small and needs to be amplified by an amplifier . Figure 2 lists the basic characteristics of several amplifiers. Assuming that OP A350 is used, its temperature drift characteristic is %26#177;4.

If the same 75°C temperature drift is added to the bias voltage error, the error margin is calculated to be 800μV, which is 1.8% compared to 45mV.

If the resistor itself has a 20 ppm temperature drift and an error of 0.15%, the amplifier's error is much greater than the error of the current sensor, so it cannot be received. For example, the error of OPA 335 and OPA333 is greatly reduced to 0.02% and 0.03%. Relatively speaking, the main source of error lies in the resistor rather than the amplifier, so the supporting circuit must be selected with an error amplitude smaller than the sensor's error amplitude. If OPA335 is selected, the temperature error of the resistor itself is much greater than the output error of the amplifier itself, which can ensure that the system accuracy can be improved; if a resistor with higher temperature performance is selected, it can ensure that the circuit can accurately amplify the output signal.

In addition, there are other different error sources, such as errors caused by parasitic characteristics of solder joints, PCB traces, connectors, etc., which will also affect the accuracy of the system. Using differential input can improve the overall accuracy of the system and eliminate the influence of parasitic characteristics on the circuit. However, this improvement in accuracy must ensure the matching of the resistors used. If you simply choose discrete devices, it is difficult to ensure the matching of resistors. Therefore, all resistors must be integrated into one chip. For example, INA132 integrates four resistors, which not only ensures the matching between resistors, but also ensures the consistency of temperature drift characteristics. In addition, you can also choose instrument amplifier products to directly amplify the signal, such as INA326.

The above discussion shows the difference between high voltage and low voltage of current sensors, that is, using instrument amplifier to measure current at the low voltage end, and using the inherent voltage division of differential amplifier to measure current at the high voltage end.

Common-mode voltage is used in differential inputs and is defined as the sum of the positive and negative inputs divided by 2. If the current in the differential circuit flows through the resistor to ground, the overall common-mode voltage is positive. Conversely, if the current flows from ground into the differential circuit, the common-mode voltage is negative.

Figure 4 lists some products used by TI for current measurement, showing different common-mode voltage ranges. The table lists the maximum and minimum voltage tolerances for the high and low ends, which can reach up to 75V. If the INA326 is powered by a single power supply, it can only reach 5V.

[page]

In addition, TI has INA168/9 and INA138/9 products, whose gain can be controlled by external resistors. The common mode range of the INA19X series is very wide and convenient to use. Since it has internal buffer drive capability, no external buffer is required, but the gain of the INA19X series is fixed inside the chip, as shown in Figure 5. It can be seen that in the gain part, the INA19X series has 20, 50, and 100 times gain to choose from.

There is also a newer INA270 and INA271 product series, which have a similar basic structure to the INA19X series, the only difference is that the buffer output is off-chip, and the user can filter it in the middle to reduce the noise output.

However, in order to further improve the overall accuracy of the system, the "zero temperature drift" technology was introduced in the new product. Since temperature is important when collecting current, and the resistor itself will heat up and cause temperature drift, the "zero temperature drift" technology was introduced to further improve the overall accuracy. INA209 is the first generation product based on zero temperature drift technology.

Figure 6 shows all of TI’s current acquisition products, including the common-mode voltage range and power supply range. It can be seen that the INA19X, INA27X products, and INA203-206 products all provide a wide common-mode voltage range, and the INA209 is the latest “zero temperature drift” bidirectional current collector.

Digital output direct current measurement

For digital output current collectors, isolation or AD conversion is definitely required.

In addition to traditional AD conversion, TI has also developed a series of AD converters specifically for current acquisition applications . The ADS120X series is a product developed specifically for digital current. There are also different product series for different current sensors , such as the non-contact ADS1204/5/8. These products are not traditional serial or parallel port outputs, but ∑?Δ outputs. They use the ∑?Δ working principle to complete AD conversion. Their output is similar to PWM , but the frequency is variable, and AD conversion is performed in the form of pulse density. ADS1208 is a product developed for Hall sensors and has a programmable current source. The reason is that the Hall sensor needs to be driven by a current source for balanced use, so the programmable current source is integrated into the chip. ADS1205 is a product developed specifically for the need for two-way synchronous acquisition, with two differential inputs, such as for motor control. ADS1204 has four differential inputs, which can not only monitor three-phase current, but also sample different parameters of other environments. In addition, TI also has products specifically for single-phase current sampling, such as ADS1202 and ADS1203, and also provides ISO72XX series of digital isolation specifically for high-speed data streams, which uses capacitive isolation and is not affected by external magnetic fields. TI will also launch a product that integrates the data acquisition front end and isolation, thereby realizing the functions of data acquisition and signal isolation on a single chip, such as AMC1203.

The output of the product series mentioned above is a ∑?Δ data stream, which can be restored to a serial or parallel port output using a dedicated chip AMC1210. It not only integrates a ∑?Δ back-end data filter, but also adds some comparators , so it can realize a lot of digital functions.