Robust Digital Isolation Adds Safety to High-Voltage Applications

Wherever powered electrical circuits have the potential to interact with other circuits, hardware and infrastructure, or human users, there’s a potential for damaging overvoltage conditions. Physically or electronically isolating the current from potential points of interaction, commonly known as galvanic isolation, is essential for safety and the continued functioning of the circuit. As an added benefit, isolation often reduces unwanted noise in the output signal.

Isolation requirements are prevalent in robotics, high-voltage power grid equipment, factory floor equipment, automotive applications, and consumer products. Application specifics like variable input voltages, the use of battery power, or the need for a compact footprint are further requirements to consider when designing an isolation system.

To choose the right isolation components, designers need to understand the pros and cons as well as the makeup of various isolator architectures. With this understanding, they can incorporate the most effective, reliable, and space-efficient isolators into their electronic designs.

Identifying isolators

Galvanic isolation can be achieved in several ways, but they all share a basic principle: A higher voltage input on the primary side is separated from the lower voltage, low-current secondary side by some physical barrier. The details of the barrier, as well as the method of transmitting power, signals, or both across it, depend on the type of isolator.

Optocouplers use LEDs to convert the signal on the primary side from electrical impulses into photons. On the secondary side, a photosensitive component like a phototransistor, photodiode, or photo-field-effect-transistor (FET) receives the photons and converts them back into an electrical signal. Along with physical isolation of the primary and secondary circuits, optocouplers automatically remove unwanted noise from the output signal and prevent ground loops.

In magnetic couplers, voltage across the primary-side winding of a transformer generates a magnetic field. This magnetic field induces a voltage across a winding on the secondary side, transmitting an electrical signal while maintaining galvanic isolation. Transformers can have two separate windings on a single iron core or can be two inductors, each with one winding around its own iron core, separated by a dielectric material. Designers choose magnetic coupling for its high-voltage capabilities, relatively quick response times, and its ability to filter out signal noise. However, the size of the isolator, the possibility of heat generation, and the production of electromagnetic interference should also be considered.

Capacitive couplers employ capacitors, which are components with two electrodes separated by a dielectric material. Charge builds up on the primary-side electrode due to the input voltage. This creates an electric field that induces a voltage in the secondary-side electrode. Capacitive couplers are known for their small size, low power usage, and their rapid response to changes in input, making them convenient and efficient to deploy in transmitting electrical signals across an isolation barrier. Designers must take steps to protect capacitive couplers from an input voltage that exceeds their capabilities, environmental humidity, and dielectric breakdown.

Deploying digital isolators

Any of the isolator types discussed above can be incorporated into digital isolator systems on integrated circuits (ICs). These topologies can be further integrated with power modules or signal transmission components to form complete digital isolation systems on single chips. Some common digital-isolator system topologies include flyback, half-bridge, and push-pull.

A flyback power supply is a form of magnetic isolation that creates a transformer by combining a split inductor with a buck-boost converter that can increase or reduce the voltage of a direct current (DC) input to match the desired output. Feedback to the buck-boost converter is supplied by a tertiary inductor winding or an optocoupler. Flyback power supplies are recommended for low-power applications, but designers must be aware of the potential for unwanted EMI.

Half-bridge (H-bridge) designs include an H-bridge square wave generator, a resonant circuit containing two inductors and a capacitor (LLC), and two rectifiers that deliver the desired DC output voltage. The rectifiers allow for higher output power than some designs, and H-bridge isolation designs are recommended for medium-power applications.

Push-pull isolated power supplies use two transformers for magnetic coupling. Two switches alternate which transformer receives the input voltage. Two full-bridge rectifier diodes on the secondary side anticipate changes to the voltage and regulate it into a symmetrical output.

For greater control, designers may choose to add a transformer driver into a push-pull setup. The driver integrates an oscillator, a frequency divider, and a logic controller to coordinate the opening and closing of the switches in a break-before-make (BBM) pattern. This pattern produces a relatively constant output signal while protecting internal and downstream components from being damaged by having both switches connected at once.

Systems with transformer drivers may also control the output with low-dropout linear voltage regulators (LDOs) that replace or augment the function of the rectifier diodes. Dropout voltage is the minimum difference between the input and output voltage below which the circuit cannot adequately regulate the output. In LDOs, this difference is extremely small, ensuring reliable operation over a wide range of input voltages.

Leaning into LDOs

An LDO contains a FET, a differential amplifier, and a bandgap voltage reference. The differential amplifier compares the output voltage to the reference voltage, and, if the difference between them is too high, the amplifier signal triggers the FET to adjust the circuit resistance to keep the output voltage steady.

In addition to the dropout voltage, several other specifications should be considered when selecting an LDO for a digital isolation application, including load and line regulation, power supply rejection ratio (PSRR), output noise, and quiescent current (I_Q). Load regulation is an LDO’s ability to handle variations in input current while maintaining a stable output voltage, while line regulation concerns variations in the input voltage. Many specifications also quote PSRR, which measures the regulator’s ability to manage ripple in a rectified alternating current (AC) input.

Designers also want to ensure that output noise is kept to a minimum. A low I_Q, the current needed to operate the regulator’s internal circuitry, simplifies the system and preserves battery life in mobile applications.

One example of an LDO designed specifically for battery-connected systems is 3PEAK's TPL8031Q-S (Figure 1). These regulators generate fixed-voltage outputs of 3.3 V or 5 V with ±2.5% accuracy. They have maximum dropout voltages of 720 mV for the 5 V output version and 900 mV for the 3.3 V output version.

Figure 1: Low-dropout linear voltage regulators (LDOs) provide reliable output voltages for digitally isolated systems like automotive electronic control units. (Image source: 3PEAK)

TPL8031Q-S regulators tolerate input voltages between 3 V and 42 V with transients as high as 45 V, and can output up to 300 mA of current. At the same time, they consume little power, with a typical I_Q of 3 µA. Internal current limits protect the regulators from fault conditions, such as shorting to ground, by stopping voltage regulation. In addition, over-temperature protection shuts down the regulator if its internal temperature reaches a thermal shutdown (TSD) threshold, and allows it to resume operation once it has sufficiently cooled.

Reliability, along with low power consumption and high-voltage capabilities, makes the TPL8031Q-S voltage regulators good LDO candidates for many space-limited automotive applications that rely on battery power. These include electronic control units (ECUs), domain and body control modules, microcontrollers and transceivers, interior and exterior lights, infotainment systems, instrument clusters, and other subsystems powered by or connected to the vehicle battery.

Conclusion

Automotive applications exemplify systems that need robust digital isolation to protect delicate electronics from overvoltage and to ensure human operators, passengers, and others who come into contact with the systems are safe from dangerous voltages. There are many permutations of power and signal isolation that can accomplish this, and LDOs are a critical component of carefully designed digital isolation systems.