USD

Wireless Accelerometer and Temp Sensors Simplify Deployment of IIoT Machine Monitoring Capabilities

By Art Pini

Contributed By Digi-Key's North American Editors

Machinery monitoring is a well-known technique for maintaining factory machinery and it is a major component of the Industrial Internet of Things (IIoT), or Industry 4.0 initiative. This initiative is driving higher levels of automation including increased levels of data exchange within manufacturing operations and distributed signal processing. One element of the IIoT is the expansion of the ability to measure and data log a host of operating parameters, including the vibration levels and temperature of rotating machinery. This provides understanding into the machine’s current condition and provides insight into pending failure mechanisms, allowing for scheduled maintenance instead of catastrophic failure.

The “catch” with IIoT is the need to mount and wire accelerometer, temperature, and other sensors on multiple machines across the factory floor or external facilities, such as oil rigs or gas pipelines and storage facilities.

The solution to the wiring issues is to use smart wireless sensors that collect and combine both vibration and temperature data that is linked to the control room or to the cloud via a low-power, wide area network with excellent range. Add built-in compute capability for edge-based processing to help interpret the volumes of data and transmit only essential data, and designers can reap the full benefits of the IIoT.

This article discusses machine maintenance fundamentals before introducing wireless accelerometer/temperature sensors from TE Connectivity Measurement Specialties. It then shows how these devices should be selected and applied.

Why machine maintenance is critical

Machinery on a factory floor needs to be kept running in order to ensure that there is no disruption and costly or catastrophic downtime. This requires servicing and repair of critical machines either reactively or proactively. Modern manufacturers, being especially aware of Industry 4.0, tend to be proactive, including machines on the critical path in predictive maintenance programs. This entails monitoring, data logging, and analyzing key machine parameters like vibration levels and temperature, which are key indicators to a machine’s current operational status. This requires that all machine related data be sent to a control room, cloud, or other central location for monitoring and analysis. Historically, this was accomplished by running cables between the monitored machines and the control room. This approach was costly and required a good deal of maintenance. The development of the IIoT has eliminated the need for hardwiring sensors from machines to the control room and replaced it with networked wireless connections.

Consider an example of a conventional machine monitoring application—a typical machine instrumented with an accelerometer. All the vibration data from the transducer is transmitted to the control room and analyzed for any immediately obvious problems and may be archived for reference to analyze long term changes that indicate a need for maintenance. Consider the vibration signature of a three-bladed cooling fan acquired from an accelerometer mounted on the fan frame (Figure 1).

Graph of vibration signature of a three-bladed cooling fan (click to enlarge)Figure 1: The vibration signature of a three-bladed cooling fan running at 1,668 revolutions per minute (right) and its Fast Fourier Transform spectrum (left). The spectral peaks contain all the necessary information to characterize the fan’s operation. (Image source: Digi-Key Electronics)

The accelerometer signal appears in the right-hand grid. This is a time history showing acceleration in units of gs versus time and contains 100,000 samples. The accelerometer output is an electric signal with a scale factor or sensitivity of 100 millivolts per g (mV/g). This voltage signal is rescaled by the measuring instrument to read in gs.

The acceleration time history appears random, but by performing a fast Fourier transform (FFT) and viewing the acceleration signal as a function of frequency (spectrum), as seen in the left-hand grid, the interpretation becomes much clearer. The spectrum plots the linear amplitude of the signal in gs versus the frequency in hertz (Hz). Seven peaks are marked on the spectrum. These peaks are related to the characteristics of the fan—namely the rotational speed and the power line frequency.

The peak at 27.8 Hz (second from the left) is the rotational speed of the fan motor—27.8 Hz corresponds to a rotational speed of 1,668 revolutions per minute. The harmonics of rotational speed at 55.6, 83.6, and 194.7 Hz are also marked, and the relative levels of these signals are indicative of issues such as mechanical looseness. The third harmonic at 83.6 Hz has a higher amplitude because it is also the blade passing frequency. The fan blades pass support structures three times for each rotation of the motor causing vibration. This adds to the third harmonic of rotation, making it higher than the other harmonics. The large peak at 120 Hz is due to the rotating magnetic field of the induction motor. It has sidebands at 92 and 148 Hz from the mechanical rotation.

It is quite clear that the FFT greatly reduces the amount of the data that must be transmitted. The vibration signal’s 100,000 samples can be broken down into seven key peaks that need to be transmitted for this machine. If this processing takes place in the transducer, then only the information about the spectral peaks need be transmitted, reducing the load on the communications channel.

Accelerometers

An accelerometer is a vibration sensor that produces a voltage output proportional to mechanical acceleration. Piezoelectric accelerometers use a known mass to compress a piezoelectric element such as a ceramic or quartz element to produce a voltage proportional to the acceleration of the sensor. Examples of a wireless piezoelectric accelerometer are the TE Connectivity Measurement Specialties models 8911-A and 8911-E. These single, battery-powered devices combine two sensors, data collector, digital signal processor, and radio into one compact device that measures both vibration and temperature (Figure 2).

Diagram of TE Connectivity Measurement Specialties 8911 wireless accelerometerFigure 2: The TE Connectivity Measurement Specialties 8911 wireless accelerometer contains an accelerometer, temperature sensor, microprocessor, and radio in a compact, battery-powered device. (Image source: TE Connectivity Measurement Specialties)

The accelerometer has a maximum acceleration range of ±50 g, a sensitivity of 100 mV/g, and a ±1 decibel (dB) bandwidth of greater than 10 kilohertz (kHz). This is all contained in an environmentally sealed stainless steel and polymer housing that has an operating temperature range of -20° to 60°C. The accelerometer is powered from a single, replaceable 3 volt CR123 battery.

The microprocessor is responsible for operational control and signal processing of the vibration data. Temperature data originates from the embedded temperature sensor in the microprocessor. The microprocessor performs the FFT analysis on the acquired vibration data. The FFT is evaluated showing the center frequency, peak amplitude, and percentage of the total spectral content for the eight most significant acceleration peaks in the vibration data. As described previously, the peak frequencies and magnitudes are the key parameters needed for machine diagnostics. Reducing the amount of data transmitted reduces the communication channel bandwidth, increases the range, and reduces the power consumption of the 8911 accelerometer. The typical battery life of the accelerometer is five years. This long battery life cuts down on required maintenance of the accelerometer, a very desirable condition.

The communications channel

The accelerometer uses the LoRaWAN Class A communication protocol utilizing unlicensed radio frequencies of 868 megahertz (MHz) (8911-E) in Europe and 915 MHz (8911-A) in the U.S. The LoRaWAN Class A protocol offers a simple, reliable, and secure communications channel that allows a means of expanding machinery diagnostics into factory areas where it is prohibitive to install wired systems.

LoRaWAN is an open standard managed by the LoRa Alliance. It uses proprietary spread spectrum technology from Semtech Corporation. The standard uses a frequency modulated “chirp,” which is easily generated to produce a spread spectrum channel with high noise immunity, capable of a 5 to 15 kilometer (km) reliable communications range. Data rates of up to 50 kilobits/s are possible depending on the range.

The 8911 wireless accelerometer is capable of two-way communication. In addition to transmitting vibration and temperature measurements, the transducer can receive remote control signals that set the accelerometer’s sampling period anywhere from once per minute to once per day. In operation, the 8911 accelerometer performs a self-diagnostic routine on power up. It then attempts to join the LoRaWAN network using over-the-air activation (OTAA). It will repeat this operation on a preplanned schedule governed by an internal “Join” timer. Once it is successful in joining the network it enters its sampling mode and begins processing vibration and temperature data.

The programmed workflow is to acquire the vibration signal, perform the FFT on the acquired signal, detect and extract the significant vibration peaks, and finally transmit the data to the network.

The data protocol that is used is fixed (Figure 3).

Table of LoRaWAN data protocol showing the order of the data transmitted to the networkFigure 3: The LoRaWAN data protocol showing the order of the data transmitted to the network. (Image source: TE connectivity Measurement Specialties)

The battery state is the first data transmitted. It is the battery capacity, in percentage. This is followed by the number of FFT spectral peaks, currently set to eight. The third data element is the temperature, which is sent in two bytes. The total spectral energy in the measured band is sent next, again as two bytes. The integration size is related to the width of the peak as determined in the peak detection algorithm, again two bytes. The peak data then follows, starting with the first peak: two bytes for frequency, two for magnitude, and then a single byte for the ratio of the peak magnitude compared with the total spectrum magnitude. The last three data values are repeated for each of the remaining seven peaks. Again, the small amount of data transmitted accounts for the long battery life and narrow communications bandwidth required.

Using the accelerometer

The accelerometer can be mounted in any orientation; common mounting orientations are vertical or horizontal. The accelerometer can be mounted to a machine using any of three methods (Figure 4). The base of the accelerometer is threaded with a ¼-28 NF thread and can be mounted using any of three available dual studs available from the manufacturer—¼-28:¼-28, ¼-28:M6, or ¼-28:M5. There is also an adhesive mounting stud and a magnetic mounting stud. In all cases the accelerometer has to be mounted solidly as any looseness will corrupt the vibration measurement.

Diagram of three mounting options for the TE Connectivity 8911 accelerometersFigure 4: The three mounting options for the 8911 accelerometers are stud, adhesive, and magnetic. (Image source: TE connectivity Measurement Specialties)

The adhesive mounting requires a mechanically ridged adhesive. The use of pressure sensitive adhesives or foam tapes is not recommended as the flexible mounting results in errors in the acceleration readings. Epoxy or cyanoacrylate adhesives which are mechanically “stiff” are recommended.

The magnetic mounting has a 30 lb pull and is compatible with machine frames that are made with ferrous materials.

Sensor controls and status indicator

The sensor has a single reset push button and two LEDs—one blue and one red—that are used to indicate its status. The LEDs are visible through the polymer cap. These controls and indicators are accessible by unscrewing the polymer cap.

The push button on the sensor will automatically initiate a new capture and data analysis at any point in the transducer’s operating cycle.

The blue LED indicates that the sensor has successfully initiated and joined the LoRaWAN network by lighting for two seconds. It will then flash whenever the transmitted data is successfully transmitted and acknowledged.

The red LED lights for two seconds if the sensor fails to join the network. It will also flash if the transmitted data packet is not acknowledged.

Conclusion

Factory automation engineers and designers ramping up for the IIoT need a fast and efficient means of equipping their equipment for monitoring. As shown, the model 8911 accelerometer/temperature sensor offers a simple, reliable, and secure method of adding machine monitoring in factory areas not easily supported by wired sensors. Its built-in signal processing provides the necessary data for plotting and monitoring machinery performance with minimum loading of the network communications. Based on LoRaWAN, its long communications range, extended battery life, and built-in signal processing make it an excellent candidate for IIoT, or Industry 4.0, applications.

Recommended Reading

  1. LoRaWAN Part 1: How to Get 15 km Wireless and 10-Year Battery Life for IoT

Disclaimer: The opinions, beliefs, and viewpoints expressed by the various authors and/or forum participants on this website do not necessarily reflect the opinions, beliefs, and viewpoints of Digi-Key Electronics or official policies of Digi-Key Electronics.

About this author

Art Pini

Arthur (Art) Pini is a contributing author at Digi-Key Electronics. He has a Bachelor of Electrical Engineering degree from City College of New York and a Master of Electrical Engineering degree from the City University of New York. He has over 50 years experience in electronics and has worked in key engineering and marketing roles at Teledyne LeCroy, Summation, Wavetek, and Nicolet Scientific. He has interests in measurement technology and extensive experience with oscilloscopes, spectrum analyzers, arbitrary waveform generators, digitizers, and power meters.

About this publisher

Digi-Key's North American Editors