Voltage vs. Current
A fundamental point here is that measuring voltage is easier than measuring current. A voltage is available simply by forming an electrical connection to the appropriate node, and as long as your measurement circuitry presents a very high impedance (which is easily accomplished with an op-amp), the rest of the circuit will not be significantly affected. Current measurements, on the other hand, are often more intrusive because a small resistance is inserted into the current path. Consequently, if you have a choice of measuring current or voltage, opt for voltage. For example, if you need to measure a current that already flows through a resistor with a precisely known value, simply measure the voltage across this resistor and use Ohm’s law to calculate the current.
A primary objective in any measurement system is minimizing the extent to which the process of measuring affects the quantity being measured. When measuring voltage, this means that the measurement circuit should have a very high input impedance. Think about it this way: if the measurement circuit were not present, the insulating material around the wire or PCB trace would provide an extremely high resistance. So as the input impedance of the measurement circuit increases, it behaves more like a mere open circuit, and thus its effect on the measured voltage becomes less significant.
High input impedance can be achieved with the simple op-amp circuit shown below, called a voltage follower. It also provides low output impedance, which helps to ensure that the reading is not altered by subsequent components in the measurement circuit.
Remember, though, that real-life op-amps introduce small offset and noise voltages, so select a low-noise, low-offset op-amp if you need high-precision measurements.
If you are measuring a voltage that is too high for the input range of the op-amp, you can use a resistive voltage divider in conjunction with the voltage follower:
With these resistors in the circuit, you no longer benefit from the high input impedance of the op-amp. To maintain accurate measurements, the combined resistance of R1 and R2 must be significantly larger than the output impedance of the circuit being measured. This involves a trade-off, though, because higher resistance leads to lower measurement accuracy: more resistance means more thermal noise and more offset voltage created by the op-amp’s input bias current. As an example, let’s assume that R1 = R2 (meaning that the measured voltage is reduced by half) and that the output impedance of the measured circuit is 100 Ω. This plot shows the difference between the true voltage and the measured voltage—in other words, the measurement error—with R1 and R2 varied from 1 kΩ to 1 MΩ.
For this example, a good compromise would be R1 = R2 = 100 kΩ, since increasing the resistance beyond that point has little effect on the measurement error.
Most in-circuit current measurements fall into two general categories: resistive and magnetic. In the resistive approach, a small resistor is inserted into the current path, and a differential amplifier measures the voltage drop across the resistor. Magnetic current measurements involve a sensing device that generates a voltage when current passing through the device is influenced by a magnetic field. Both strategies have advantages and disadvantages.
Resistive measurements are generally more accurate and can be used over a wide range of current amplitudes and frequencies, and the required components are inexpensive and readily available. The primary advantage of magnetic measurements is isolation: there is no electrical connection between the sensor and the current path, and no current-sense resistance is added to the circuit. It follows that magnetic sensors are preferable for very high currents that would create excessive power dissipation if measured with a current-sense resistor.
Putting Ohm’s Law to Work
A basic resistive current-sense configuration involves a precision resistor and a differential amplifier:
The differential amplification could be achieved with an instrumentation amplifier or even a circuit based on standard op-amps, but you should first look for an appropriate part among the current-sense amplifiers—which, as the name implies, are specifically designed for amplifying voltages across current-sense resistors.
The fundamental trade-off here is the value of the resistor. A smaller resistor will dissipate less power and have less effect on the current being measured, but lower resistance also generates a lower voltage that is more subject to errors induced by noise and offset voltages.
Putting the Hall Effect to Work
Magnetic current transducers, generally referred to as Hall effect sensors, use an applied magnetic field to generate a voltage in proportion to the current passing through the device.
The voltages produced via the Hall effect are very small, and thus practical devices incorporate circuitry for conditioning and amplifying the signal. Hall effect sensors are considered nonintrusive because the only resistance introduced into the current path is that of the primary conductor in the device—for example, only 1.1 mΩ in the Allegro ACS709. The relationship between output voltage and sensed current for this particular device is illustrated by the following plot:
Closing the Loop
The circuits and devices discussed thus far merely convert a voltage or current into a buffered, scaled voltage signal—we still need to actually make use of this measured quantity. If your only requirement is to detect a fault condition and initiate a response, you could use a simple comparator circuit that activates when the measured signal goes above or below a predefined threshold. Often, though, you will want to do more with your measurements: store data, analyze trends, detect steep increases or decreases, send measurements to a PC, etc. If this is the case, a microcontroller with an integrated ADC is a powerful and versatile solution that allows you to conveniently digitize the signals, analyze the digitized data, and perform a wide variety of corrective measures.