EETimes India - September 16-30, 2008 - (Page 12) In Focus | Green Tech Use non-maskable interrupt to manage dynamic power in MCUs By Ajit Basarur Shantanu Prasad Prabhudesai Nazmul Hoda Multimedia Systems Group Ittiam Systems In typical embedded systems, the main processor derives power from either AC/DC wall adapter or from a battery, whereas the MCU is powered from either of the above sources and—in their absence—from a coin cell to keep it powered on at all times. Coin cells used in embedded systems are of low capacity and usually not rechargeable, but they must be capable of powering up the MCU for the expected portable product life cycle. During normal operation of the system, the MCU is used to control interfaces such as Infrared (IR) remote control, keypads, PS/2 keyboard etc., and it consumes a significant amount of power in this state. When the rest of the system is switched off, only the real-time clock (RTC) is expected to be running. Since the MCU handles the RTC and other user interfaces, it makes sense to switch off all its peripherals except the RTC and to configure it to a lower-power consumption mode. When the system transitions to an ON or OFF state, the MCU should be informed of this change by means of an external component such as a reset controller or a dedicated standby key. Any general-purpose input-output (GPIO) pin can be used for this signalling, but there are specific advantages when the non-maskable interrupt (NMI) pin is used for this purpose. Figure 1: Typical embedded system overview. The MCU can also be powered from an AC/DC wall adapter. als are off except for the RTC. Figure 2: When the RESET signal goes low, it tells the MCU to enter its low-power state, where all the peripher- Figure 3: MCUs generally have the NMI and RESET functionality multiplexed onto a single pin with the RESET function as the primary pin configuration applications. The RESET signal of an on-board system reset controller is used to signal the MCU of change in system power state and it is connected to RST/NMI pin of MCU. When the system reset is asserted (using a push button switch) or when the system is powered off, the RESET signal output from the reset controller Coin cell battery Consider the application circuit in Figure 2, in which MCU is being powered up from a coin cell battery and it is handling remote control, keypad and RTC goes low. This signals the MCU to be put into its low-power state, where all the peripherals of the MCU are off except for the RTC. The MCU now draws minimal power from the coin cell, which is the only available source of power to it. When the system is powered up or when system RESET is deasserted, the MCU has to change back to its normal state where all its peripherals are active and the CPU can work at maximum computational power. Thus, the MCU has to be put in either normal state or low-power state, depending on the NMI pin status. MCUs generally have the NMI and RESET functionality multiplexed onto a single pin with the continued on page 14 12 EE Times-India | September 16-30, 2008 | www.eetindia.com http://www.eetindia.co.in/SEARCH/SUMMARY/technical-articles/MCU.HTM?ClickFromNewsletter_080916 http://www.eetindia.co.in/SEARCH/SUMMARY/technical-articles/GPIO.HTM?ClickFromNewsletter_080916 http://www.eetindia.co.in/SEARCH/SUMMARY/technical-articles/coin cell.HTM?ClickFromNewsletter_080916 http://www.eetindia.co.in/SEARCH/SUMMARY/technical-articles/coin cell.HTM?ClickFromNewsletter_080916 http://www.eetindia.com/STATIC/REDIRECT/Newsletter_080901_EETI02.htm?ClickFromNewsletter_080916
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.