Technology Enablers for Faster, Safer, High-Efficiency EV chargers
This article looks at semiconductor technologies that are driving EV chargers, including high-voltage semiconductor switches, power converters, and multi-level complex power stages.
As the number of electric vehicles (EVs) increases, there is a growing need to create more energy-efficient charging infrastructure systems that can charge vehicles faster than ever before. New EVs have higher ranges and larger battery capacities than their predecessors, necessitating the development of fast DC charging solutions to support quick-charge requirements. It takes approximately 30 minutes for a 150- or 200-kW charging station to charge an EV up to 80%, which is enough power to travel approximately 250 km. A fast DC charger station, according to Combined Charging System and Charge de Move standards, can deliver power up to 400 kW.
Today, we will look at semiconductor technologies that are driving faster, safer, and more efficient chargers:
- High-voltage semiconductor switches (insulated gate bipolar transistor [IGBT] and silicon carbide [SiC]) are driving up bus voltages (800 V or 1,000 V) in the system. With increased system voltages comes the requirement for isolation technologies to ensure overall safety and reliability.
- As power converters become capable of faster switching frequencies (hundreds of kilohertz to a few megahertz), operating at these high frequencies reduces the size of magnetic components and other passive devices used in the circuit, which then reduces system cost and improves overall power density. Thus, there is a need for high-bandwidth current and voltage sensing to accurately control and protect the digital power stages.
- Higher efficiency requires the use of multilevel complex power stages, which in turn necessitates the need for high-voltage isolated gate drivers to efficiently switch these power stages and reduce overall switching losses, while also including reinforced isolation and short-circuit protection features.
Let’s take a deeper look at these technology enablers.
Safety compliance is critical in EV chargers because they directly interface with the utility grid. Isolation is necessary in order to ensure operator safety, to protect processors from damage in high-voltage power-converter systems, and to prevent ground loops and potential differences between different communication subsystems. Power controllers with a secondary-side control architecture require isolation not only in the power stage (through an isolation transformer) but also in the controller drive circuit and associated signal-conditioning circuitry.
Noise interference caused by the switching action of the power converters can negatively affect system performance. For example, when transients from power-converter switching occur, a high slew rate can cause transient voltages on the signal path and create a common-mode voltage transient that requires an isolator with high common-mode transient immunity (CMTI) to maintain signal integrity.
Increasing DC link voltages in EV charging stations also shows the importance of reinforced isolation for operator safety and reliability. Depending on the operating voltage, there are three basic classes of isolation: functional, basic and reinforced. Functional isolation, also called operational insulation, does not protect or isolate against electrical shock, but is required for a product to function. Basic isolation is a single level of insulation that provides basic protection against shock. Reinforced isolation is a single insulation system that provides electric shock protection equivalent to double insulation.
Various isolation technologies are possible for semiconductors:
- Optical isolation uses LED light transfer across a transparent nonconductive insulation barrier. Its key advantages are high electrical isolation values and low cost. However, optical isolation also has long propagation times, low noise immunity, higher quiescent current, and fast degradation of insulation with temperature and age. These limitations restrict optical isolation technology to cost-sensitive, lower-speed power converters.
- Magnetic isolation uses inductive coupling transfer using a transformer coil design and provides high isolation at high frequencies. It provides better propagation times compared to optical technology but has high electromagnetic noise issues, low noise immunity, and insulation degradation with temperature and moisture.
- Capacitive isolation uses a changing electric field to transmit energy through capacitance. The advantage of this technology is its ability to operate at high speeds and its relatively small package. It has the highest reliability with the best insulation stability over temperature, along with high CMTI and low radiated emissions.
Figure 1 shows capacitive isolation, which Texas Instruments uses in its isolated gate drivers, amplifiers, and digital isolators.
Figure 1. Capacitive isolation
High-Bandwidth Current and Voltage Sensing
An EV charger application uses current and voltage sensing for three main functions: monitoring, protection, and control. In an EV charger, the conversion of energy from the grid usually occurs in two stages. The power factor correction stage converts the grid voltage to a stable DC link voltage. Then a DC/DC stage converts the DC voltage to a voltage suitable for the EV battery pack.
Figure 2 is a block diagram of an EV charging station, with the current-sense locations marked as A and the voltage-sense locations marked as V.
Figure 2. Block diagram of an EV charging station
The increasing use of SiC and gallium nitride (GaN) switches in the power stage has enabled increased frequencies of operation (hundreds of kilohertz to a few megahertz) while offering improved efficiency and higher power density. These power stages require accurate sensing of fast switching currents for reliable operation of the control loop to ensure the stable operation of the converter. Fast response time, linear operation over-temperature, and accurate current and voltage sensing are essential for all high-power systems with high-voltage stages.
Semiconductor technologies that aid with current sensing can broadly be classified into direct and indirect sensing methods. The direct methods include shunt resistor-based sensing by employing either an isolated amplifier or an isolated sigma-delta modulator. The voltage drop across a shunt resistor, which is typically 50 mV or 250 mV (to keep current-resistance losses to a minimum), forms the input to this stage.
For an isolated amplifier, a scaled amplification of a low-voltage signal is sent out to an external controller to make precise measurements of current on high-voltage rails while maintaining electrical isolation.
An isolated sigma-delta modulator modulates the voltage drop across the shunt directly into a digital bitstream that when directly interfaced with a microcontroller’s sigma-delta peripheral enables a much higher bandwidth. A higher signal bandwidth ensures rapid, precise current measurements and an accurate representation of the switching signal for controlling the converter’s power stage.
Shunt-based sensing is preferred because this method can achieve better DC accuracy over temperature compared to Hall-effect-based solutions with basic one-time calibration. Shunt-based solutions are much more accurate, particularly at low currents, because of their limited sensitivity to external magnetic fields. Shunt-based solutions are linear over the entire voltage range, especially at zero crossing and near the magnetic core saturation region. This solution also offers reinforced isolation up to 5 kV and a reduced form factor compared to Hall-effect sensors.
Indirect methods involve sensing the magnetic field around the current-carrying conductor. For example, Hall-effect sensors provide an indirect measurement of the magnetic field generated around a conductor by sensing the current flowing through it. Open-loop Hall-effect sensors are available with a bandwidth of up to 1 MHz. Closed-loop sensors have a bandwidth of 350 kHz and provide better performance compared to open-loop Hall-effect sensors, but also cost more.
Given their superior bandwidth and response time, open- and closed-loop Hall-effect sensors provide better protection for SiC switches over shunt solutions during short-circuit conditions, especially when switched at high frequencies. The short-circuit withstand the time of SiC switches is typically 1-3 µs and will need fast detection in order to prevent short circuits. The voltage drop across the inline shunt results in thermal dissipation and power losses when compared to Hall-effect-based solutions, especially when the measured currents increase.
Isolated Gate Drivers
High-speed gate drivers are critical to building a power module that has high efficiency, high power density, and is reliable and robust. The gate drivers interface between the pulse-width modulator on a controller and the high-power switch. High-power SiC-/IGBT-based power modules require gate drivers with the ability to source and sink peak currents at extremely high speeds, minimizing turnon and turnoff transition times and thereby minimizing switching losses. Gate drivers must:
- Be flexible to use the same driver with wide operating voltages and different types of power switches.
- Be robust to operate in noisy environments and extreme temperature conditions.
- Have minimum turnon propagation delays to enable quicker switching of a field-effect transistor (FET), minimizing the conduction time of the body diode and thus improving efficiency.
- Have good delay matching to ensure the driving of paralleled metal-oxide semiconductor field-effect transistors (MOSFETs) with a minimal turn on delay difference.
For high-voltage applications, reinforced isolated gate drivers provide increased system resilience against surges (CMTI), leakage currents caused by potential differences, and other anomalous events that threaten to damage the system.
Depending on the placement of the controller, isolation is likely required between the controller and the driver. A traditional method for isolation is to use a separate transformer with a non-isolated gate driver. An integrated gate driver has a propagation delay similar to or better than a discrete transformer solution while taking up as much as 50% less area. Furthermore, an integrated gate driver can be tailored to deliver CMTI greater than 100 V/ns, a number significantly higher than that achievable by the discrete solution. CMTI is a key parameter that determines a gate driver’s robustness.
Protection features in gate drivers are required for the reliable operation of the converter. Because of the benefits of improved power density and efficiency, SiC and GaN have become a potential substitute for silicon IGBTs for various applications. A SiC MOSFET has more stringent short-circuit protection requirements; the short-circuit withstand time is 1 to 3 µs compared to an IGBT, which is around 10 µs. A DESAT pin integrated to the gate driver is critical to provide fast response in detecting short circuits. Integrated undervoltage lockout and an active Miller clamp are also vital in preventing false turnon in FETs used in half-bridge applications.
The need for portable DC fast chargers with natural convection cooling (that can be easily picked up and stored in the back of an EV trunk) is pushing the limits of designing EV chargers with state-of-the-art power density and efficiency. GaN-based switches with integrated gate drivers offer on-resistance, fast switching and low output capacitance, aiding the design of EV chargers with up to one-third improvement in power density. Resonant architectures commonly used in EV chargers also stand to benefit from zero-voltage and zero-current switching that mitigate switching losses and improve overall system efficiency.
High-power density, reliability, and robustness are becoming increasingly important in power converters used in EV charging stations. With increasing power and voltage levels, it is important to protect humans as well as equipment against hazardous operational conditions.
Manufacturers that are targeting high-power-density and efficient chargers will embrace IGBT-, SiC- and GaN-based power converters with switching frequencies moving from hundreds of kilohertz to a few megahertz. High-frequency current and voltage sensors will be crucial to development on these platforms.
Smart gate-driver technology will enable the necessary high voltage levels, fast switching speeds, and need for fast protection. Given the leaps that semiconductor technology has taken in the last decade, it may soon be possible to charge an EV to its complete range during a quick coffee break.
This article was co-authored by Harish Ramakrishnan, a systems engineer at Texas Instruments.
Industry Articles are a form of content that allows industry partners to share useful news, messages, and technology with All About Circuits readers in a way editorial content is not well suited to. All Industry Articles are subject to strict editorial guidelines with the intention of offering readers useful news, technical expertise, or stories. The viewpoints and opinions expressed in Industry Articles are those of the partner and not necessarily those of All About Circuits or its writers.