CONTROLLING POWER

The second challenge in designing a BCM module is creating the power section. This portion of the design is intimately tied to the types of loads the module must drive. One common type of load is a simple LED light. Controlling an LED can be as straightforward as using an MCU output pin to switch on and off the current supplied to the LED. Another approach is to use PWM signals to create a more pleasant lighting effect. Using a PWM signal allows the LED to be switched on at a frequency that, to the human eye, appears to always be on. By increasing or decreasing the duty cycle, designers can increase or decrease the average current to the LED, effectively dimming and brightening the light, much like theatre lighting. By using red, green and blue LEDs, each under PWM control, designers can create a blended light of an arbitrary color. This function further increases the needs for PWM channels in the MCU.

A second type of load for a BCM module is a motor, such as a fan motor in a heating ventilation and air-conditioning (HVAC) system. Motors are also used to adjust car seat positions, and are the driving force for the wiper system. Similar to controlling LEDs, by using a PWM, designers can ramp the speed of a standard dc motor up or down efficiently. In addition, sampling the PWM signal with an analog-to-digital converter (ADC) allows designers to detect potential failures. A variety of motors are found in body electronics applications. These include brushed dc, brushless dc and even three-phase motors. Each motor type requires unique control characteristics, which must be taken into account during the design of the power section.

A third type of load used in body electronics is a heating element, for example, the devices that produce the heat for a heated seat. These are high-power resistors, which put a large demand on the body module to source enough current to effectively heat the seat. Traditionally, simple 12 V relays were required to supply the necessary current for power-hungry applications. Relays are electromechanical devices that are large, heavy and not as reliable as an all-electronic solution — a critical shortcoming in a device for automotive applications. Given these limitations, some of the traditional relay applications have been replaced with power MOSFETs, which are designed to carry the high current required and provide a complete solid-state solution. MOSFETs solve the problems of size, weight and reliability that exist with a relay. A further refinement of this solid-state switch is adding intelligence, also called an intelligent power device or IPD. IPDs typically contain a power MOSFET combined with a control circuit in one package. Just like MOSFETs, IPDs are a smaller, lighter weight and lower-power-consuming replacement for a typical relay. They go one step further by combining the high-current capability and high reliability of a MOSFET with protection and diagnostic features for thermal runaway and short-circuit detection.

Figure 3 illustrates a high-side IPD with built-in short-circuit and overtemperature protection and load-current sensing. Additionally, for reducing EMI within the module, the IPD has a switching control function that limits rapid fluctuations in output current.

This type of IPD is often used to replace relays for body applications such as exterior and interior lighting and heating elements (Figure 4). Due to the small size of the IPD die, it is possible to replace up to four standard relays with a single quad-package IPD. In this example, the four relays used to drive the stoplights and turn signals could be replaced by a single IPD. Additionally, due to the smaller size and height of IPDs compared to relays, ECU designers can reduce the PCB and overall module size. Reducing the module size and number of components, combined with the increased reliability of using an IPD, can also lead to a higher-quality and more cost-effective end product.

As IPDs are relatively new to many designers, it is important to understand their key features when making the decision on which kind to use. Typically, the first step is to determine how much current the IPD has to support and at what voltage. Most suppliers will list their devices according to these features first. Once that decision is made, there are a number of features to consider when deciding on a specific IPD. As mentioned, IPDs are capable of providing diagnostics data to the control section. This can be done over a network protocol, such as a serial peripheral interface (SPI), or by discrete port communication. A SPI connection can be convenient for systems that employ a SPI bus. However, using standard port signals can also be preferred for systems that need to receive feedback from the IPDs faster than a SPI can support. There are IPDs that support either method, so a module designer has to consider the overall system requirements before choosing the communication method that works best.

On-resistance, sometimes referred to as R(ON), is the equivalent resistance across the device when it is operating. A large on-resistance is disadvantageous, as it creates a significant voltage drop across the device, causing larger power dissipation and, therefore, elevated device temperature. To address this issue, devices currently in production provide R(ON) values as low as 8 milliOhms. When choosing an IPD, a designer should always look for the device that meets the system requirements with the lowest R(ON). Another important factor is the connection of the control circuit to the analog power portion. Two common types of IPDs include: monolithic and multi-die. In monolithic IPDs, both the control and the power portions of the device are developed on the same piece of silicon. In multi-die IPDs, it is the opposite: the control portion is a separate die than the power portion. The reason for the two methods is due to the underlying technology. The process technology for high-density logic is not capable of supporting high current, creating the need for a multi-die approach for higher current devices. When the current requirements are not as high, the power and logic can be designed in the same technology, and designers can avoid the cost and complexity of two separate dies and the associated bonding and packaging. While the choice for monolithic vs. multi-die is often fixed by the IPD suppliers, designers should always be aware of the process technology used to ensure that the device will meet their requirements.

Last, packaging is also important. IPDs are being manufactured in smaller and smaller processes, allowing for small packages and support for multichannel packages. Designers almost always look for the smallest packaging, as well as any opportunity to replace multiple IPDs with a multichannel package.

In summary, to design a reliable, cost-effective system requires an understanding of both main portions of the design, the control section and the power section. Each section has its challenges when trying to choose the components that will work best together in the complete system.

ABOUT THE AUTHOR

Adam Prengler is a platform solutions engineer with NEC Electronics America's Automotive Strategic Business Unit. Prengler holds a Bachelor of Science degree in Electrical Engineering from The University of Texas at Dallas.

Click here for the enhanced PDF version of this article