Take the right steps to achieve precise measurements

A variety of devices used to provide control or safety functions often require the use of current measurement devices. The most common method of measuring current through a circuit is to measure the voltage drop across a low-value resistor (Figure 1). The current flowing through the load also flows through the sense resistor (also called the shunt resistor) RSENSE _and develops a voltage drop VM across the resistor.

Figure 1 shows the measurement principles of the two current measurement methods. The traditional current measurement method is to use an ammeter, which was born in the early days of electronics and is still in use today. The ammeter usually indicates full scale when the current flowing through it is 100 μA or 1 μA, so when measuring larger currents, it is necessary to connect it in parallel with a shunt resistor (which divides most of the current). This method can adjust the range of the ammeter and use it to measure currents of any amplitude.

A more modern approach to current measurement uses an operational amplifier to amplify the voltage drop across a sense resistor, with the amplifier output typically connected to an analog-to-digital converter (ADC). When designing this measurement setup, you first need to choose the value of the sense resistor and set the gain of the amplifier.

For example, assume that the current flowing through the load varies between 0 and 10A, and the maximum input signal to the ADC is 5V. Since the current is relatively large, a very small shunt resistor is required. To simplify the calculation, a resistance of 0.025 Ω is selected here. When the current reaches the maximum value of 10A, the voltage VM across the shunt resistor is 10 x 0.025 = 250mV. The physical size of the shunt resistor is determined by the power dissipation (Pd ₌ I2R ⁾ . Based on the above values, Pd ₌ 102 ^x 0.025 = 2.5W. To reduce the heat generated in the shunt resistor, a resistor with a power rating of 5W or more should be used.

Figure 1: The most common way to measure current through a circuit is to measure the voltage drop across a low value resistor.

To meet the ADC input requirements, the amplifier needs to provide a gain of:

If the signal from the shunt resistor is added to the non-inverting input of the amplifier, the gain Av is equal to 1 plus the ratio of the feedback resistor to the input resistor. If the resistance value of R1 is chosen to be 5kΩ and R2 is 95kΩ, the gain is 20.

Figure 2: A Kelvin connection is a 4-wire connection with two excitation leads and two sense leads.

Consider the thermal characteristics of the shunt resistor

The examples given above raise several issues that require further consideration. The shunt resistor will heat up due to power dissipation or ambient temperature changes and change the resistance of the shunt resistor, thus affecting the accuracy of the measurement. The metal used in the resistor determines the temperature coefficient of the resistor (see Table 1).

Table 1: The metal used in a resistor determines the temperature coefficient of resistance.

Kelvin sensing resistor

The second area of ​​concern is the connection between the shunt resistor and the amplifier input. In the example given earlier, a common 5W resistor uses 20-gauge copper leads. If the leads on both ends of the resistor are trimmed to 0.5 inches long and soldered in place, the leads will introduce a 4% relative error (relative to the nominal value of the resistor of 0.025Ω) because the resistance of 20-gauge copper wire is approximately 0.001 Ω/inch.

To avoid introducing such errors, two further measures are needed: use a Kelvin connection to the shunt resistor and add a differential input amplifier. The Kelvin connection is a 4-wire connection with two excitation leads and two sense leads (Figure 2). Two of the wires (W1 and W2) are the excitation leads, which connect the shunt resistor to the high-current end of the circuit. The other two wires (W3 and W4) are the sense leads, which connect the voltage across the shunt resistor to the amplifier. This method separates the wires carrying the load current from the measurement circuit and removes the voltage drop in W1 and W2 from the voltage measurement across the shunt resistor. To further effectively utilize the advantages of the Kelvin connection, a differential amplifier is also added (see Figure 3).

Figure 3: To further exploit the advantages of the Kelvin connection, a differential amplifier is added.

Low-end sensing and high-end sensing

The measurement method of placing the sense resistor between the load and ground is called low-side sensing. Low-side sensing can be used when the input voltage is positive and close to ground. However, there are some problems when using an op amp and these conditions are not met. Low-side current measurements may cause problems that exceed the negative common-mode voltage limit of the op amp.

Due to the design of the op amp, the op amp can only work properly when the input common-mode voltage is within a certain range between the positive and negative power supply voltages. The common-mode input voltage specification of a rail-to-rail input op amp operating on a single 5V power supply may be 0 to +5V.

Figure 4: Distributed load system.

Because common-mode issues can arise in low-side measurement examples, consider the distributed load system shown in Figure 4. The control unit in the dashed box uses MOSFETs to control the power delivered to the two loads. The parasitic resistances R _P1 and R _P2 between the power supply, load, and control unit represent the sum of the ground lead resistance and the connector resistance. In a real circuit, their values ​​may be hundreds of milliohms. For this example, assume that R _P1 and R _P2 are 0.2 Ω. Differential amplifiers 1 and 2 are similar to the differential amplifiers in Figure 3.

As in Figure 1, the current flowing through each load is 10A, and the shunt resistors RS1 _and RS2 _are both 0.025 Ω. The current flowing through the control unit is 1A. When load 2 is operating, the voltage at point V _G2 is 2V higher than the voltage at point V _G1 (0.2 Ω x 10 A = 2 V), and V _G3 is 0.2V higher than V _G2 . Therefore, the input voltage of differential amplifier 1 is -2.2V relative to the common connection point at the control unit, which is higher than the negative common-mode voltage limit of most amplifiers. Unless the amplifier used in the differential amplifier is specially designed for negative common-mode voltage, it will not work properly in this circuit. Another disadvantage of low-side sensing for this example is that wires are required to connect the shunt resistors to the differential amplifier inputs.

High-side current sensing can solve all of these current measurement problems (Figure 5). The shunt resistor is placed at the supply side of the power supply instead of at the ground side. This configuration allows the shunt resistor to be installed in the control unit containing the differential amplifier and MOSFET. The wiring connection to the control module is also simplified.

Figure 5: High-side current sensing can solve all of these current measurement problems.

In this example, the differential amplifier must be able to handle an input common-mode voltage equal to the supply voltage. One way to solve this problem is to run the differential amplifier at the supply voltage and select an amplifier with an input common-mode voltage specification greater than the positive supply voltage. Another approach is to use an amplifier specifically designed for high input common-mode voltages. The advantage of this approach is that the amplifier can be operated at the same supply voltage as the rest of the control circuitry when the input is connected to a higher positive voltage.

In general, when measuring current in a circuit, the amplifier's input common-mode voltage specification is required to exceed the amplifier's positive and negative power supply voltages, while also ensuring that the input bias voltage is low and the gain accuracy is high.