Supplier eBooks

ROHM - Driving the Future of Automotive Solutions

Issue link: https://resources.mouser.com/i/1437744

Contents of this Issue

Navigation

Page 13 of 36

For many applications, including data centers, automotive, robotics, and telecommunications infrastructure, 48V is commonly used as the highest standard voltage rail in direct current (DC) systems. It presents an acceptable trade-off for both interconnect and system requirements. With higher voltages, the current needed to provide the same power is lower than for a lower voltage power supply. This enables greater power efficiency because of fewer resistive losses throughout the system. Beyond 48V, however, additional protection, insulation, and a higher class of dielectrics and semiconductor processes might be needed for interconnect and control electronics. Using 48V as a DC power rail poses a challenge because modern digital electronics, including complex systems-on-chip (SoC), multicore processors, and field-programmable gate arrays (FPGAs), use voltages as low as 1.2V. To power these systems, voltage converters that can convert a 48V rail to 1.2V are needed. In many applications, these low voltage power rails also require tens of amps of current, which can lead to very low conversion efficiencies when using multiple voltage- reducing stages. Each converter stage has a typical efficiency of around 90 percent, and multiple stages compound a reduction in efficiency. At the same time, lower power regulation is typically more efficient, but in some cases, regulation of as much as 1000W is required, leading to substantial design challenges in selecting parts and designing a voltage regulation solution that overcomes the intrinsic inefficiencies of converter system components. For instance, many switching converter topologies use metal oxide semiconductor field-effect transistors (MOSFETs) as switching devices in the power supplies. MOSFETs have a drain to source ON resistance (RDS(ON), or R ON ) that is an intrinsic conduction loss in the device. A lower R enables much lower conduction losses in a MOSFET device while also leading to less heat generation and easier thermal management. A cooler MOSFET device also translates to lower R ON because R ON is a function of temperature and rises with increasing device temperatures (positive temperature coefficient). R ON consists of several resistances which, when in series, are compounded. Factors range from the diffusion region's resistance, channel region, accumulation region, and most importantly, the drift region. Other factors include the contact resistance between the drain and source metalization, and with the bond wire contact on the die and the package leads. Reducing the conduction losses in each of these, and conduction losses through the PCB design, can significantly improve efficiency. Figure 1: R ON is a product of the cumulative resistances from the drain to a MOSFET device's source. For many applications, including data centers, automotive, robotics, and telecommunications infrastructure, 48V is commonly used as the highest standard voltage rail in direct current (DC) systems. 10 High Voltage and Curent Efficiency Challenges RoHm Semiconductor gmR50 Chip Shunt Resistors LEARN MORE >

Articles in this issue

Links on this page

view archives of Supplier eBooks - ROHM - Driving the Future of Automotive Solutions