Understanding the Many Benefits of Residual Current Detection

Obscure Circuit Damage

Circuit protection is an area of electronic design dedicated to the protection of circuits from sources of electrical damage. The most common types of electrical damage include electrostatic discharge (i.e. static shock) and over-current, both of which are relatively easy to defend against. 

For example, ESD can be defended against with the use of zener diodes. On the other hand, PTC fuses can be used to protect a circuit from excessive current draw. As these sources of damage are extremely common, the majority of electrical products on the consumer market protect from these sources only. They’re  just sufficient enough to make those products reliable. As technological capabilities improve, some applications may be more critical and require further levels of protection to ensure that they are protected under all possible fault conditions. 

Since all forms of potential circuit damage cannot be predicted, it is next to impossible to design circuit protection circuits for all these events. The good news is that many events can be defended against by recognizing what normal circuit operation looks like and then shutting the circuit down when those operating conditions are not met. Over-current and over-voltage are easily handled with the use of fuses and clamping diodes respectively, but what about current imbalance? Imagine a scenario where a failing circuit (possibly due to a short) does not return the same amount of current that it draws from the power supply. In this situation, the current could be leaking from a power module, conducting through a damaged PCB, and then back into the ground. Worst yet, a loose power connection could be making contact with a metal enclosure that would leave users to a potential shock. This is where a residual current detection (RCD) could be highly beneficial.

Basic RCD Operation

An RCD is a circuit that disconnects the input from the output if the current flowing into the output is not the same as the current flowing back from the output. These devices are critical in electrical installations as they protect users from electrical shock when contact with the live (hot) wire is made. The current flows through the RCD, into the live wire, into the person who is touching the wire, and then into the earth to return to the electrical source. Because the return current is not flowing into the neutral wire, the RCD sees more current flowing through the live wire than the neutral wire and therefore disconnects the RCD output. 

Large RCDs for home installations are based on inductors which have a live/neutral pair wrapped around a coil as well as a smaller coil referred to as a “sense coil”. Under normal conditions, the live/neutral pair have the same currents (but in opposing directions) and therefore cancel out each other’s magnetic field. With no magnetic field, the different current magnitudes result in a magnetic field which induces a voltage in the sense wire. This sense voltage triggers a mechanical breaker which then disconnects the output from the input of the RCD.

Simple RCD—Image courtesy Wikipedia

This type of RCD is often used for detecting current imbalances as much as 30 mA; as such, they are often very large. But the concept of a RCD could be brought onto an electronic product for increasing the reliability and safety to both the circuit and user. If a coil type is too large, then how could current imbalances be detected? Would this method have any inherent issues?

A simple approach to a solid-state RCD

A solid-state RCD is required to measure the current flowing into a circuit and the current flowing back from the circuit. The two currents are then compared to see if they are equal (or within a defined tolerance) and if not, the power entering the circuit can be disconnected. The schematic below shows a very simple implementation of a circuit for single supplies.

The first section of the circuit is at the far-right end of the schematic and consists of two sense resistors, R1 and R2. These are low-ohm, high-precision resistors, which produce a small voltage whose magnitude is proportional to the current flowing through them. The voltages produced across each sense resistor needs to be compared to each other to see if there is an imbalance, though it is worth noting that these voltages need to be handled before they can be compared. The low side resistor (R2) produces an absolute voltage at the VOUT- output, whereas R1 produces a differential voltage between the power supply and the VOUT+. This differential voltage needs to be converted into an absolute voltage so it can be compared with the VOUT- voltage, and this is done with a simple differential amplifier with a gain of 1.

The two voltages—which both represent the current flowing through the sense resistors R1 and R2—are then passed into a secondary differential amplifier whose gain is 10. This is done so that it is easier to read small currents which may produce sense voltages in the millivolts range. The output of this amplifier indicates how much current imbalance there is between the VOUT+ and VOUT- outputs. 

The next stage in this circuit is a simple op-amp comparator, which is used to compare the differential voltage with a predetermined voltage. The predetermined voltage is generated with the use of a potential divider R7 and R8 and this circuit example has a difference voltage of 10 mV (which corresponds to a trip current of 1 mA). If the voltage from the differential amplifier goes beyond the voltage set by the potential divider, then the output of the op-amp U4 goes high, which, in turn, sets the flip-flop A1. The inverting output of the flip-flop is connected to the NMOS M1, which controls if power can go to the external circuit. If the inverting output of the flip-flop goes to 0 (when the trip circuit has detected a fault), then the NMOS is switched off and the external circuit can no longer draw power. The system can be reset by applying a logical pulse to the RESET input of the flip-flop A1.

Practicality of this RCD example

The circuit shown above can protect a circuit from multiple sources of damage and fault scenarios, including short circuits of the power supply that can damage traces and connections, but not the components themselves nor will it prevent metal enclosures becoming live. Indeed, the RCD example above is not without its flaws and these should be understood before using the circuit in a practical environment.

The RCD example uses a low side sense resistor which are incredibly easy to work with (as they do not produce differential voltages) but they can interfere with the ground of the connected circuit. When circuits are designed, the ground connection (which is sometime to referred to VEE, VSS, and 0 V) must be at 0 V, as this is the circuit reference. When a resistor is placed in series with this 0 V connection, a small voltage is produced, which means that all 0 V references are no longer a true 0 V reference. This impact can be reduced by having a sense resistor with an extremely small resistance, though it make it difficult to sense the current.

Another issue with the design is the op-amp choice; the circuit shown here does not specify an op-amp. That said, great care should be given when choosing one. Many op-amps (such as the LM358) have real-world limitations, such as outputs that cannot go to rails and minimum input voltage bias. If, for example, the sense resistors produce a voltage that is smaller than the op-amps input bias voltage, then the op-amp won’t be able to register the voltage drop. 

Conclusion

RCDs are able to protect circuits from bizarre situations which can be hard to predict. While they are mainly aimed at preventing users from being electrocuted by mains voltages, they can be implemented into circuit protection for devices that require the utmost reliability. But if a solid-state type is being designed, then designers should be aware of the issues with low-side resistors and how they can affect circuit performance.