Low power application design using MEMS accelerometers

Author: STMicroelectronics Petr Stukjunger

Low-power MEMS accelerometer sensors can be used to increase battery life. Sensors become more and more power efficient, and embedded features can also help reduce overall system power consumption. For example, when the user does not use the When the device is installed, the motion-sensing wake-up function keeps the entire system asleep. However, there are many other possibilities. Use MEMS accelerometers to reduce overall power consumption.

Starting from the MEMS accelerometer sensor itself, the operating mode should be flexible. As shown in Figure 1, we know that the resolution of the sensor and the output data rate, compared to the current consumption of the other side, must be between the two The trade-off - the higher the resolution or data rate, the greater the current consumption, and vice versa. Fortunately, some sensors on the market can operate with very little micro-environment, and also when the power is turned off or in standby mode. Only consume a few nanoamps of electricity.

Figure 1: Sensor parameters affect battery life

For some demanding applications, the sensor's operating mode can be quickly replaced, and the resolution and data transfer rate will only be improved if it is really needed. Some sensors can even automatically switch modes. The customer can configure the active state The resolution and data transfer rate, and customize the conditions for starting it. At this time, the sensor will go to a standstill, but it will continue to measure the data and proceed with extremely low data transmission rate and resolution, etc. Action events) switch back to startup.

Another good design principle is to use low power levels because lower power levels also mean lower current consumption. This is why 1.8V power is preferred for low power applications.

In some designs, the sensor's power cycle can be used. The sensor's power supply is only activated when it is needed to measure, otherwise the sensor will be off. We can provide the power of the sensor through the pins of the microcontroller. Implementation. As shown in Figure 2. When applying this technique, it is necessary to correctly calculate the power budget, because each sensor's startup needs to be configured and waiting until the output is determined and the correct data is provided.

Figure 2: Control of Sensor Power Supply Through Control Microcontroller Pins

Most MEMS accelerometers are digital sensors, which means that they can convert the measured analog signal to digital data. Because there is an integrated analog signal converter, coupled with low sensitivity to signal distortion, bill of materials items can be reduced, However, this is not the only advantage. The embedded interrupt generator MEMS accelerometer can generate a trigger signal when a user-defined parameter condition is met. This is how the motion-sensing wake-up function is used. The microcontroller (MCU) configures the sensor To generate a wake-up trigger and enter the sleep mode with very low power consumption. When an action is detected, the sensor will generate an interrupt signal, the MCU will switch to a suitable operation mode after receiving the signal, and then the processing just occurred. Case.

Digital sensors can also take over tasks related to motion processing performed by microcontrollers. Of course, MCUs can perform the same tasks, but the power efficiency is much lower—the power consumption of the MCU is one milliampere, and the sensor is microamperes. Detection of free fall, single point, double click (user actions like mouse clicks), portrait/landscape direction detection, etc., is achieved through the internal logic of the sensor. The MCU does not need to perform any calculation, it only needs to wait for an interrupt trigger, and only when it occurs React to actions.

Digital sensors often incorporate configurable filters that are used to measure acceleration data. Low-pass, high-pass, and even alias filters can be used for MCU preprocessing. Data, and increase unloading diversion.

The data buffer embedded in the sensor is mostly of the first-in-first-out (FIFO) type because it allows the MCU to reduce the frequency of reading data, thus reducing the current consumption. This allows the microcontroller to perform other tasks and extend the sleep time while Also saves the time required to communicate with the sensor serial port.

Serial communication between sensors and microcontrollers also increases overall power consumption. For ultra-low power applications, serial communications can have a significant impact every time a microampere-ampere is processed. Most MEMS accelerometers are Is through the Serial Peripheral Interface (SPI) and I2C interface for communication. SPI interface is more efficient in terms of power consumption, there are three reasons: First, there is no lead on the communication line will cause additional current consumption; The second is to support higher Data rate; Third, the serial protocol overhead is reduced.

Regardless of which interface is used, we can significantly reduce serial communications by allowing applications to use data ready interrupts without sensor polling, which is to continuously request the availability of new data. After the sensor completes the data measurement and conversion, the data preparation interrupt is generated automatically. The new data set will be read by the MCU. When this interrupt is activated, the MCU will immediately read the output data from the sensor through a single reading action.

As mentioned earlier, the data rate output by the sensor is low, which means that the current power consumption is low. Therefore, the so-called single data conversion mechanism can be a perfect match between the sensor and the application data, as shown in Figure 3. Using this mechanism, either an external trigger signal on the sensor pin or a register written from the MCU using serial instructions. The data thus obtained is stored in the sensor. The sensor can also initiate a data preparation. The interrupt signal informs the MCU that the data conversion has been completed and the data can now be read by the application program. With this function, data rates lower than 1 Hz or any other predefined range can be achieved.

Figure 3: Single data conversion mechanism

This article discusses MEMS accelerometer sensor functions that are important for low-power applications and how they can be used in system design. STMicroelectronics' latest LIS2DW12 ultra-low-power 3-axis MEMS accelerometer utilizes accelerometer sensors as The new application design offers flexibility because it consumes up to 1Ua of current, plus multiple operating modules, an extremely wide range of output data rates, rich embeddedness, high temperature stability, and various enhancements such as digital filtering And FIFO buffer. We believe that many low-power applications can enjoy the advantages of the LIS2DW12. This sensor will provide users with advantages, especially in the following areas: motion sensing and user interface, handheld smart energy-saving features , Electrical related motion monitoring, and impact recognition logging for wireless sensor nodes.

2016 GoodChinaBrand | ICP: 12011751 | China Exports