When selecting a MEMS accelerometer for condition-based monitoring (CbM), what are some often overlooked but crucial parameters to consider? To cut to the chase, these parameters include g-range, bandwidth, and resonant frequency. If these parameters are too low or insufficient, they may have undesirable effects on the measurement results.

MEMS accelerometers play a critical role in detecting fault conditions and preventing accidents that could result in unexpected downtime or other financial losses.

Engineers tasked with selecting and installing sensors for condition-based monitoring (CbM) must carefully consider several important parameters when selecting these sensors, but these are often overlooked.

This article explores some important considerations to keep in mind when choosing a MEMS accelerometer.

■ Condition-based monitoring (CbM)

CbM is the process of using sensors to monitor mechanical systems for defects or damage. CbM is used to monitor defects in ball bearings, gears, pumps, and many other applications.

For optimal monitoring, it's common to use a combination of sensors. These sensors can detect any abnormalities early and take action to prevent damage or failure.

The introduction of CbM enables predictive maintenance (PdM). PdM can predict potential system failures based on data collected by sensors. This helps reduce downtime and increase operational efficiency.

CbM uses a variety of sophisticated sensors, including accelerometers, temperature sensors, magnetometers, and MEMS microphones. This article focuses on MEMS accelerometers.

■ MEMS accelerometer

MEMS accelerometers convert mechanical vibrations into electrical voltages or digital values. These MEMS sensors consist of movable and fixed silicon elements that are linked to form a capacitor (Figure 1a). When mechanical movement occurs, the movable element moves toward the fixed element.

This structure can be mathematically described as a mass-spring system, from which acceleration can be calculated from measured forces.

Analog MEMS sensors (Figure 1b) convert this into a voltage. Digital sensors additionally use an analog-to-digital converter to output a digital value (Figure 1c).

Analog Devices offers a wide range of MEMS accelerometer products, including sensors with low noise floors, high bandwidths, and multiple axes.


■ Key parameters to consider when selecting a MEMS accelerometer for CbM

○ g range

The g-range of a MEMS sensor should be selected to cover all accelerations occurring in the system. If the sensor's g-range is too low, the signal may be clipped. This can result in asymmetric signals/offsets in the measurement results, leading to incorrect acceleration calculations. The acceleration due to gravity (1g) is often ignored.

○ Bandwidth

The frequency at which acceleration occurs in a system must be considered in conjunction with bandwidth. In CbM applications, early detection of defects in components such as ball bearings or pumps is crucial. Initially, defects typically manifest at high frequencies. Therefore, if the bandwidth of a selected sensor is too low, defects may not be detected. In these applications, acceleration is a square function of frequency. For example, a displacement of 250 nm and 1 kHz results in an actual acceleration of 1 g. If this displacement occurs at 10 kHz, the resulting acceleration is 100 g, or 100 times higher. This means that selecting a sensor with sufficiently high bandwidth and an appropriate g-range is crucial for early detection of defects in a system. For critical applications, ADI offers sensors with bandwidths up to 24 kHz and a g-range of 500 g.

○ Sensor resonance frequency

Another factor to consider with regard to bandwidth is the sensor's natural resonance. Acceleration at the sensor's resonant frequency causes amplification, which in the worst case can distort the available signal and lead to erroneous measurements. The solution to this problem is to incorporate mechanical damping/filtering into the system. In addition to bandwidth, a low noise floor is also important for early detection of defects or anomalies. A good MEMS accelerometer has a noise floor of less than 100 g/√Hz.

A general overview of MEMS accelerometer specifications can be found in ' Accelerometer Specifications - A Brief Definition '.

■ Conclusion

MEMS accelerometers have emerged as a promising alternative to piezoelectric sensors. MEMS accelerometers are now available, featuring high bandwidths of up to 24 kHz and low noise floors.

MEMS accelerometers are ideal solutions for predicting and detecting the onset of faults, reducing system failure rates and associated costs. ADI offers a variety of sensor products for CbM applications, including the ADXL100x and ADXL356/ADXL357. See more MEMS accelerometer products on ADI's Accelerometer Product Page.

※ About the author

Benjamin Reiss, FAE, Analog Devices, Inc.

Benjamin Reiss has been with Analog Devices in Munich, Germany, since April 2017. He received his Master's degree in Nanotechnology from Friedrich-Alexander-Universität Erlangen in 2016. After completing his internship at ADI, he joined the Munich office as a Field Application Engineer (FAE) responsible for various market segments.