What Are Accelerometers?

Accelerometers are sensors that measure the acceleration of a body along one or more axes, producing an electrical output proportional to the applied inertial force. They detect both dynamic accelerations caused by motion and static accelerations due to gravity, making them useful for orientation sensing as well as vibration and shock measurement. The discipline spans classical electromechanical design, microelectromechanical systems (MEMS) fabrication, and signal processing, with applications ranging from consumer smartphones to spacecraft attitude control.

The operating principle common to nearly all accelerometers is the proof-mass concept: a small seismic mass is suspended on a compliant structure, and when the housing accelerates, the mass deflects relative to the frame. Transduction mechanisms convert this deflection to voltage using piezoelectric materials, piezoresistors, or differential capacitance between the proof mass and fixed electrodes. Each mechanism offers different trade-offs in bandwidth, sensitivity, noise floor, and operating temperature range.

MEMS Design and Operating Principles

Capacitive MEMS accelerometers dominate consumer and industrial applications because they offer low power consumption, small size, and the ability to measure down to DC (zero frequency), which is required for tilt and orientation sensing. The proof mass is micromachined from a silicon wafer using bulk or surface micromachining processes, and its position relative to fixed sense electrodes is detected by a charge amplifier or switched-capacitor circuit. Piezoelectric accelerometers, by contrast, are inherently AC-coupled and excel at high-frequency vibration measurement in ranges up to 50 kHz, making them the standard choice for rotating machinery diagnostics. IEEE Xplore publications on MEMS accelerometer-based vibration monitoring survey capacitive sensor board designs for structural health monitoring, addressing noise, bandwidth, and wireless transmission requirements in deployed systems.

Signal Conditioning and Calibration

Raw accelerometer output requires conditioning to remove bias offset, correct for scale factor error, and compensate for cross-axis sensitivity, which is the unwanted response to acceleration perpendicular to the sensing axis. Calibration procedures rotate the sensor through known orientations in the gravity field or use a precision shaker table to apply reference accelerations at known frequencies. The Allan variance technique, specified in IEEE standard 952-1997 for inertial sensors, identifies the noise contributions from quantization, angle random walk, bias instability, and rate random walk as a function of averaging time, allowing designers to choose the appropriate averaging interval for a given application. A polymeric piezoelectric MEMS accelerometer with high sensitivity and low noise density published in Microsystems and Nanoengineering demonstrates how material selection and electrode geometry affect the noise floor of MEMS piezoelectric devices.

Fall Detection and Human Motion Sensing

In wearable health monitoring, three-axis MEMS accelerometers detect the characteristic acceleration patterns that accompany a human fall: a brief free-fall phase with near-zero acceleration followed by a high-magnitude impact. Threshold-based and machine-learning-based classifiers analyze these patterns to distinguish falls from other activities such as sitting down quickly or stumbling without falling. Research on inertial sensor-based centripetal acceleration as a correlate for lateral margin of stability documents how accelerometer-derived metrics can quantify balance during walking and turning, relevant to fall risk assessment in clinical populations. Sampling rates between 50 and 200 Hz are sufficient for human motion capture, while impact detection may require higher rates to resolve the peak g value accurately.

Applications

Accelerometers have applications in a range of fields, including:

• Fall detection and activity monitoring in wearable health and fitness devices
• Inertial navigation in aircraft, autonomous vehicles, and spacecraft
• Structural health monitoring of bridges, buildings, and industrial machinery
• Automotive crash sensing for airbag deployment and electronic stability control
• Seismology and geophysical exploration using arrays of ground-motion sensors