Author:
Ally Winning, European Editor, PSD
Date
02/01/2022
PDF
There is an ongoing trend in electric vehicle (EV) design to move to a higher voltage operation. Most of the initial tranche of EVs had a battery voltage of 400V for cars and 600V for heavier vehicles, but manufacturers are now making the transition to 800V and in some cases even higher voltages. Even the manufacturers who are still using 400V supplies are stretching the technology’s capabilities. Where 400V in practice used to mean a range from 320V to 400V, it is now more like 380V to 450V. In the same vein, some manufacturer are trying to stretch the capabilities of the newer 800V designs to 900V and over. For example, Lucid Motors uses a 900V supply in its new EV range to offer a potential range of 516 miles. There is also a discussion currently ongoing about where the voltage range for the next generation of EVs will be targetted – whether to use 1,000V or step up to 1,200V. Higher voltages offer benefits such as lower currents when charging the battery, reducing overheating and allowing better power retention for longer ranges. They also use less copper, allowing a smaller motor size and reducing the overall weight of the vehicle.
A solution that would fit today’s 800V+ designs and be a foundation for designs at 1,000V and beyond would still be required to start at 30V and work right up to the maximum voltage. Power Integrations has now launched two automotive-qualified, high-voltage switcher ICs that it claims will provide a compelling solution for manufacturers of cars and heavy vehicles wishing to push the 600V and 800V levels to their limits and start on next-generation 1000V+ designs. The two new InnoSwitch3-AQ family parts are differentiated by the internal SiC MOSFET ratings, one has a 50W output and the other features a 70W output.
The company has only operated in the automotive field since the launch of a 900V rated part last year. However, Peter Vaughan, director of automotive business development at Power Integrations claims that is not such a drawback, as the company has a history of developing products which operate over the industrial temperature range for consumer OEMs that can be just as demanding as automotive ones. The company’s low 1 ppm failure rate in its products has also helped with customers in the automotive market.
The biggest selling point for the initial 900V part was that it decreased the number of components needed to build a complete power supply, even though the company added an external MOSFET and some biasing components to its design. The new 1,700V ICs are even simpler with the stackFET cascode configuration no longer required, simplifying the design further. The 1,700V SiC FET is integrated into the devices and driven directly. PI estimates that this can cut the number of components required for a full power supply design by around half.
Vaugh explains, “Reducing the number of components cuts costs, as well as the number of individual components that have to go through the automotive qualifying process. As the component count is cut, reliability naturally increases as there are fewer components to fail and fewer soldered joints. OEMs are putting tremendous pressure on the tier one vendors to make traction inverters smaller, increasing the power density and making the end design cheaper and lighter, with less material and a smaller enclosure. To achieve that means shrinking the PCB. So halving the number of components for the design is valued over almost anything else for the engineers doing the traction inverter design”.
Another way that the new devices help reduce PCB size is due to PI’s reinforced 5000 VRMS FluxLink isolation allowing the IC to sit across the isolation barrier between the primary and secondary circuits in an area of the PCB that would normally be bare. According to Vaughn, customers have even placed the device directly under the transformer.
The new 1700V ICs offer all the features of PI’s InnoSwitch technology. They incorporate the MOSFET itself, the low voltage primary controller, and the secondary side controller. The secondary side controller is the main controller for the system, telling the primary switch when to turn on and off - the opposite of most designs where the primary controller determines when to switch. InnoSwitch devices can have control on the secondary side of the circuit because of the FluxLink isolation between the primary and secondary sides. The benefit of secondary side control is that it deterministically knows when to turn the synchronous rectification FET on and off, negating the challenge that normal synchronous rectification solutions face to cover fault conditions when turned on at the wrong time. The inclusion of synchronous rectification and a quasi-resonant (QR) / CCM flyback controller achieves greater than 90% efficiency. These new parts consume less than 15mW at no-load, which is ideal for reducing self-discharge in battery management systems. The devices also feature fast transient response which cut the amount of output capacitance needed, allowing smaller form factor capacitors to be used. FluxLink regulation means that the circuit can go from no load to full load and slew the power delivery from zero to maximum in only two switching cycles.
The two new devices are AEC-Q100 qualified and come in an InSOP-24D package. The InnoSwitch3-AQ 1700-volt parts are also suitable for industrial markets, where the integrated solution can replace discrete controller-plus-MOSFET designs. A reference design, DER-913Q, and hardware kit RDK-919Q, are available for the evaluation of the 1700-volt IC.
