How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

[Introduction]Today’s MEMS accelerometers have rapidly improved performance, with lower power consumption, smaller size, higher integration, wider bandwidth, and noise levels below 100µg/√Hz, etc., and can replace wired systems based on wireless solutions , replacing the bulky single-axis piezoelectric sensor with a small and lightweight three-axis analog device, making it possible to cost-effectively monitor the explosion of machinery and equipment continuously. For maintenance and facility engineers, this means detecting, diagnosing, predicting and ultimately avoiding machine failures through a new paradigm of Condition Monitoring (CbM).

Trends in Condition Monitoring

There are millions of electric motors in continuous operation around the world, consuming approximately 45% of the world’s electricity. Among them, some of the most important electric motors can enjoy the care of wired condition monitoring system. The study showed that 82% of companies surveyed had experienced unplanned maintenance, costing up to $250,000 per hour. For those companies that experienced unplanned outages, based on the average of two outage events, the outages lasted an average of 4 hours and the average cost was up to $2 million.

Another study found that 70% of companies do not know when an asset needs maintenance or upgrade work. Lack of awareness, coupled with the cost of downtime, is driving companies digital, with around 50% planning to invest in digital twins and artificial intelligence (AI). With the large-scale development of the Industry 4.0 movement, business organizations are actively studying the digitization of the industrial landscape to improve productivity and efficiency.

A key aspect of this movement is the trend towards wireless sensor systems. The condition monitoring industry is set to witness significant growth over the next few years, with wireless installations accounting for a significant portion of the growth. It is estimated that by 2030, nearly 5 billion wireless modules will be deployed in the global smart manufacturing industry. We all know that the most critical assets require wired condition monitoring systems, but what about all the other assets currently deployed? For some older facilities, it is not feasible to install a wired solution, which increases the need for a wireless condition monitoring solution.

Installation and maintenance of condition monitoring systems

Wired condition monitoring systems excel in performance, reliability, speed, and security, and are therefore deployed on the most critical assets. Because of these advantages, wired systems are still more likely to be deployed on new-build facilities. When installing a wired condition monitoring system, the factory floor may have to run cables all over the place, especially when certain machines cannot be disturbed. Industrial wired sensor networks typically use 60m cables, and the cost of a single run ranges from $3,000 to $20,000, including materials and labor. Wire harnesses are required in some cases, which add additional complexity and can be time-consuming to install. If cables run through existing infrastructure, they may not be able to be replaced or rerouted if damaged or in need of an upgrade.

While wireless systems may seem more expensive at first, simpler maintenance procedures, coupled with the ability to easily expand, can result in significant cost savings over the lifetime of a condition monitoring system. Fewer maintenance routes and less wiring and associated hardware save costs. The battery can last up to several years, depending on the level of reporting required. If wireless systems based on energy harvesting could be deployed, maintenance would be easier and less expensive. After choosing a wireless system, the next area to look at is which technology is best for your condition monitoring application?

Wireless Sensor Network Comparison

Wireless networks, despite having been deployed for decades, have only recently been widely deployed on the factory floor, thanks to advances in low-power technology and the increased tolerance of wireless networks to harsh RF interference. This section will discuss the advantages of various mesh networks.

grid technology

There are several common technologies that can be used to create low power, low data rate networks such as Bluetooth® Low Energy, Zigbee and 6LoWPAN. These low-data grid or many-to-many network technologies are good choices if you want to develop a dense cluster of wireless sensor nodes with relatively low data volume and short transmission distances (such as those required on factory floors) .

A mesh network can be used for infrastructure nodes and wirelessly connected to each other, as shown in Figure 1. If the communication link between two specific nodes is affected by interference or noise, the nodes can help each other expand the radio signal and even reroute the signal. One of the most important properties of grid technology is the ability to send data from one node to another through the various nodes in the network, enabling the creation of a large interconnected network covering a large area while consuming very little power. For example, in Figure 1, the distance between node 1 and node 3 is long, so they cannot communicate directly. However, in the absence of a direct link between node 1 and node 3, node 1 can transmit data to node 3 through node 2.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 1. Example of a cluster of nodes in a mesh network showing many-to-many communication

Figure 2 shows an example of a factory floor where Node 1 measures the vibration of a motor. This data needs to be transmitted to node 6, but the distance in between is beyond the capability of the transceiver. Direct transmission of data from node 1 to node 6 requires higher transmit power and higher receiver sensitivity. Higher transmit power generally means higher peak current consumption and thus requires a larger battery. With a mesh network, this data can hop along each node between node 1 and node 6 to reach its destination. The power required for each device to transmit over a smaller range is far less than the power required to form a longer direct wireless link across a factory floor.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 2. Grid network implemented on the factory floor, showing data hopping

The main advantages of mesh networks are as follows:

● Autonomous configuration: As Industry 4.0 becomes a reality and enterprises become more digitalized, factory managers must seek better performance. An important aspect of this quest is the ability to add high-density clusters of wireless devices in smaller geographic locations while maintaining highly reliable performance—in some cases nearly as good as wired systems and requiring little manual configuration , because nodes can configure themselves.

● Self-healing: The mesh network is constantly routing data and is therefore constantly disturbed by noise, interference, multipath, fading reflections, etc. from the factory floor. SmartMesh® IP systems (managers and nodes) continuously monitor the noise level of each node and share this data in order to reroute signals away from potentially noisy paths.

● Coverage: The network can be easily resized by simply adding or removing nodes. As shown in Figure 2, the coverage area can be easily expanded without increasing the power consumption of the wireless device.

Table 1 summarizes the various grid technologies and their capabilities.

Table 1. Comparison of different mesh networks

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Other low-power wireless technologies

LoRa or LoRaWAN enables low data rate communication over long distances (up to 6 miles) while consuming very low power. It is based on various frequency bands and enables point-to-point communication. Therefore, these solutions are ideal for low-power, long-range point-to-point communications. NB-IoT or cellular networks are more expensive and complex to implement, consume more power than mesh technologies, and transmit a smaller amount of data. However, it does offer high-quality cellular service and direct access to the cloud. If your wireless solution requires long-range cellular access and higher data rates than Zigbee, then LTE-M might be worth considering.

The Evolution of MEMS to Replace Piezoelectric Vibration Sensors

Until recently, MEMS sensors were not strong enough to compete with IEPE vibration sensors for detecting early vibration failure signatures in critical assets and rotating machinery, as shown in Figure 3. The main limitations of MEMS sensors are noise, bandwidth, and g-range. Low noise is the key to detecting low levels of vibration for earlier fault detection and even prediction. Bandwidth is important because many asset/motor failures, such as cavitation, bearing problems and gear meshing, often occur at frequencies above 5kHz at the earliest, and of course time is critical in detecting failures. The g range is also important, as larger assets can generate shocks or impacts of up to hundreds of g, which can damage MEMS sensors designed for less demanding operations.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 3. MEMS performance evolution for condition monitoring applications

Historically, most MEMS sensors have been designed for multiple applications, so there are usually not multiple application-specific characteristics, but at least three are required for condition monitoring. Automotive crash detection MEMS sensors are an example of an application-specific single part with advanced features. Such sensors are designed to have a high g range but insufficient bandwidth and/or noise performance for condition monitoring and many other applications. Developing MEMS sensors suitable for condition monitoring applications is difficult, which is why there have been few successful suppliers to date.

To highlight these advances in MEMS performance for condition monitoring, ADI compared two single-axis analog-output MEMS vibration sensors released in 2010 and 2017, as shown in Table 2. Both MEMS accelerometers are designed for vibration detection in condition monitoring applications. The bandwidth of both sensors is quite high, but the noise improvement is the most significant, so that MEMS sensors can now compete with piezoelectric IEPE vibration sensors.

Table 2. Comparison of first- and second-generation MEMS sensors for condition monitoring

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

This noise improvement has also been achieved on some high-performance industrial triaxial MEMS sensors, as shown in Table 3. These sensors are not specifically designed for vibration detection, but they are high performance MEMS sensors capable of detecting vibrations below 1 mg rms at full bandwidth. Combined with excellent stability and reliability, these sensors have proven to be very effective in a variety of condition monitoring applications of machinery, whether used as the sole vibration sensor or paired with other wide bandwidth MEMS/IEPE sensors. Ultra-low noise, narrow bandwidth (<5kHz) MEMS sensors can play a key role in detecting vibration in many assets, typically at low rotational speeds and sub-hertz, or where DC response is favorable, such as paper/factory Processing, food/pharmaceutical, wind power, metalworking industries. Table 3 highlights the improvement in multi-axis MEMS sensor performance from 2009 to 2017. It should be noted that to achieve wider bandwidth, lower noise, and higher g range, specifications such as standby current will be higher compared to more general-purpose MEMS sensors.

Table 3. Improvement of MEMS triaxial sensor performance

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

What level of vibration sensor is typically used in a condition monitoring system?

Companies whose unplanned downtime often results in significant lost revenue continue to rely on wired solutions based on 12- to 20-bit resolution sensors that provide the most reliable and accurate performance. Also, the higher cost of wired installations can easily be justified. For less critical assets, performance requirements are less stringent and capex caps may be much lower. Vibration sensor resolutions of 10 to 16 bits are acceptable, which is the range covered by most MEMS-based wireless condition monitoring systems today.

Less critical assets also have demand for high-performance vibration detection, a trend that will continue to grow as industrial companies look to digitize and intensify their efforts to improve performance, production, and efficiency. Historically, cost has been a limiting factor in the use of piezoelectric vibration sensors on less critical assets, but as more designers realize the value and flexibility that MEMS sensors can provide in such situations sex, this situation has now begun to change. Figure 4 shows potential vibration sensor resolutions from 10 bits to 24 bits. Despite the significantly lower resolution of MEMS, the price-performance advantage is attractive enough to justify monitoring of low- to medium-criticality assets.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 4. Sensor Types and Corresponding Resolutions

One of the main advantages of MEMS sensors is low power consumption, usually in the µA range, and even the nA range is possible. This makes them ideal for wireless condition monitoring applications. Some piezoelectric sensors consume power as low as around 200µA, but they lack integration and are expensive compared to MEMS. Some dedicated wireless vibration sensors based on piezoelectric sensors do exist that can provide 24-bit resolution at sampling rates up to 104kHz, but have very limited battery life compared to MEMS solutions. This wireless vibration sensor system typically has 8 hours of continuous battery life. Another key advantage of MEMS is that up to three axes can be integrated into a small package. Triaxial piezoelectric sensors can be more expensive, larger, and require more signal conditioning circuitry, making them less suitable for wireless applications.

Future trends: Desire for new revenue streams

Pumps account for a large proportion of rotating machines currently deployed in factories around the world, and the global market is expected to grow from $38.34 billion to $46.92 billion by 2025. Some of these pumps are critical to ensuring that the process continues to run unimpeded, which requires condition monitoring to avoid unplanned downtime. What does the future hold for such pumps? According to a recent report by Frost & Sullivan, pumps will be analytically capable and intelligent. Growth in pump OEMs will be driven by services based on analytics, artificial intelligence or machine learning (ML) aimed at providing diagnostic information on improving pump performance and reliability. The study found that after 2025, 60% of pump OEM revenue could come from service-related activities, and the pump industry will shift from a product-based model to a service-based model. The main factors driving this shift are the rapid digitization of manufacturing (IIoT), and advancements in condition monitoring hardware and algorithms, artificial intelligence, and machine learning. Traditional heavy industries such as water/wastewater treatment plants, oil refineries, and gas production plants are expected to use these smart pumps as they seek to digitize their operations. New facilities are likely to use wired condition monitoring systems, but what about existing equipment on older facilities? To apply this service-based model to deployed pumps and other rotating machinery, wireless condition monitoring systems can provide a fast, seamless and reliable solution.

EV-CBM-VOYAGER3-1Z Wireless Status Monitoring Module

The Voyager platform (shown in Figure 5) is a robust, low-power wireless mesh network vibration monitoring platform that enables designers to rapidly deploy wireless solutions into machines or test setups. Designers can quickly evaluate ADI’s MEMS sensor technology for vibration monitoring while evaluating SmartMesh IP technology for industrial wireless detection. The overall goal is to accelerate customer asset monitoring and solution development. Nodes include a mechanical housing and mounting hardware with ¼-28 industry standard stud attachments. Voyager solutions can be easily mounted directly onto motors or test circuits.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 5. Voyager Wireless Status Monitoring Module

SmartMesh IP

SmartMesh IP wireless sensor network products are ICs and pre-certified PCB modules with mesh networking software that enable sensors to communicate in harsh Industrial Internet of Things (IIoT) environments. They are built for IP compatibility and are based on the 6LoWPAN and 802.15.4e standards. Consisting of Internet Protocol version 6 (IPv6) and Low Power Wireless Personal Area Network (LoWPAN), 6LoWPAN is an Internet Protocol (IP) based network similar to Wi-Fi. The SmartMesh IP product line supports low power consumption and provides over 99.999% data reliability even in harsh and changing RF environments.

Figure 6 shows a highly scalable, self-forming mesh network of multi-hop wireless nodes combined with a network manager that monitors performance and security and exchanges data with host applications, which can collect and relay data. The grid is automatically formed when the manager and nodes are powered up. Nodes that are outside the manager’s range will forward packets through nodes that are within range. In addition, if the communication link of a node is disturbed by noise, another link/path can be used to redirect the data/packets at a different operating frequency, so that the data can bypass or stay away from the source of interference, the self-healing ability of SmartMesh IP or as The same reliability as wired network (99.999%) comes from this.

The Voyager kit has been tested for SmartMesh IP node hopping. In this test, nodes that are out of range of the network manager can skip nodes that are in range, as shown in Figure 6. A multi-hop network ensures that out-of-range nodes can transmit data to the network manager.

Where is SmartMesh IP best suited?

SmartMesh IP networks are targeted for Industrial Internet of Things (IIoT) applications. In a factory environment, sensors are typically deployed on assets in clusters, as shown in Figure 7. Assets that require regular or even continuous monitoring can be placed in different locations on the factory floor, but in most cases they will not be more than 100m apart. For example, SmartMesh IP has been successfully deployed in data centers with thousands of nodes forming high-density clusters.

In the past, low-power wireless communication devices struggled with interference on the factory floor. Not only is this what SmartMesh IP excels at, but it is specifically designed for deployment in dense clusters that require wire-like reliability and simultaneous monitoring or control.

SmartMesh IP networks communicate using the Time Synchronized Channel Frequency Hopping (TSCH) link layer, a technology pioneered by the Analog Devices SmartMesh IP team and fundamental to wireless mesh networking standards such as WirelessHART (IEC 62591) and IEEE 802.15.4e building blocks. In a TSCH network, all nodes in the network are synchronized within microseconds. Network communications are organized into time slots for low-power packet switching, pair-wise channel hopping, and full path diversity. By using TSCH, SmartMesh IP devices can sleep with ultra-low power consumption between scheduled communications, resulting in a duty cycle typically less than 1%. Network managers utilize TSCH to ensure that nodes know exactly when to communicate, listen, or sleep. This ensures that there are no packet collisions on the network and that each node consumes very low power—typically less than 50µA for routing nodes.

SmartMesh IP networks are among the most secure mesh networks in existence. All traffic in a SmartMesh IP network is protected by end-to-end encryption, message integrity checks, and device authentication. In addition, SmartMesh Network Manager includes applications that support network secure connection, key establishment and key exchange.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 6. SmartMesh connection

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 7. High-density sensors placed near the factory floor

Voyager Signal Chain

Figure 8 shows an overview of the wireless vibration monitoring platform. It also contains a three-axis ADXL356 vibration sensor board and a low-power microcontroller, the ADuCM4050. There is also a stable low power SmartMesh IP LTC5800 board and chip antenna. The kit includes a SmartMesh IP USB adapter that acts as a manager for the wireless network. Embedded firmware and GUI code are available on GitHub.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 8. Voyager hardware and GUI overview

Battery life of the Voyager module is a key design feature, so it selected high-performance, low-power devices to detect, condition, process, and transmit vibration data, as shown in Figures 9 and 10.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 9. High-level block diagram of the ADXL356 signal chain

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 10. High-level block diagram of the ADuCM4050/SmartMesh

Voyager Signal Chain Power Consumption

The active and standby power consumption (worst-case performance taken from the data sheet) of each signal chain device is shown in Figure 11 and Figure 12, respectively. Note that this does not include SmartMesh IP transceivers as their power consumption is more subtle than simple active or standby mode power consumption. The actual power consumption of the signal chain will be lower. In active mode, the ADuCM4050 consumes the most power because it samples vibration data at up to 1.8 MSPS, filters it, and then performs DFT before sending the data over the UART to the SmartMesh IP transceiver.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 11. Signal Chain Power Consumption in Active Mode

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 12. Signal Chain Power Consumption in Standby Mode

Figure 11 and Figure 12 show that the active and standby currents of a MEMS accelerometer are important when the system is transmitting data and in standby mode. Whether you intend to run a periodic monitoring scheme (e.g. every 6 hours) or a continuous monitoring scheme, these metrics are critical to ensuring that battery-powered sensors are operating effectively. In active mode, the ADXL356 consumes approximately 1.4% of the signal chain power consumption. Compared to typical piezoelectric sensors, the ADXL356 consumes much lower power. A typical piezoelectric sensor uses 4mA constant current and a 24V to 30V power supply and consumes close to 100mW. While there are lower-power piezoelectric sensors that can reduce power consumption by 90%, they are still not suitable for long-term use in battery-powered sensor networks.

In standby mode, the ADXL356 consumes 39% of the signal chain current. This may seem high, but to better understand the resulting noise versus power performance trade-off, a comparison and characterization of various MEMS sensors suitable for vibration detection in condition monitoring applications should be performed, as shown in Table 4.

Table 4. Comparison of Active and Standby Power Consumption of a MEMS Accelerometer Supporting Status Monitoring with Voyager Signal Chain Active and Standby Power Consumption

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 13 and Figure 14 show the current consumption and noise of the MEMS sensor in active and standby modes. The ADXL356 and MEMS C4 have the lowest active power consumption, although the latter is no longer recommended for new designs. MEMS B has the highest active power consumption (11.5x higher than the ADXL356), but it should be noted that MEMS B has the lowest noise and wider bandwidth, resulting in higher performance compared to all MEMS C sensors.

Although the ADXL356 and MEMS B have the highest standby current, the noise performance of these sensors is 1.6 to 9 times better than the other devices shown in Figure 14. The inverse relationship between current consumption and noise density is evident and should be considered when choosing a MEMS vibration sensor for battery powered applications.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 13. Comparison of MEMS Sensor Standby Power Consumption and Noise Density

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 14. Comparison of MEMS Sensor Active Power Consumption vs. Noise Density

Another key advantage of the ADXL356 is the ceramic package, which allows it to provide excellent stability and performance over temperature. Considering that most MEMS sensors used in wireless devices are housed in IP6x-rated enclosures, ceramic packaging is critical. In some cases, the shell also potts compounds. The ceramic package can withstand external forces from the potting compound to maintain the sensor’s data sheet performance. For plastic-encapsulated MEMS devices, potting may not be suitable, as package flexing can degrade sensor performance.

MEMS turn-on/power-up time

For MEMS sensors, power-up time refers to the time required from shutdown to standby mode. Turn-on or start-up time refers to the time it takes to go from standby to measurement mode, as shown in Table 5. For the ADXL356, this specification is valid when the output is within 5 mg of the final value.

Table 5. MEMS sensor power-up time

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

These times should be considered when monitoring critical equipment, as critical vibration data may be lost when the system enters measurement mode from standby if it is on for too long. In systems where wireless nodes are periodically powered to conserve power, power consumption when transitioning between different power consumption modes becomes even more important. Considering the turn-on time shown in Table 5, when MEMS C1, MEMS C2, and MEMS C4 measure valid data after more than 1.3s (worst case), the other sensors have completed the measurement and are in standby mode for a while. This saves more power. Figure 15 compares the ADXL356, MEMS B, and MEMS C1 transitioning from standby to measurement mode, measuring acceleration data for 1 s, assuming that the power supply ramps linearly during this transition, and then returns to standby mode after 4.5 s. Although MEMS B has a faster power-up/start-up time, the active current consumption measured at 1s is significantly higher than the ADXL356. Also, in the worst case, MEMS C1 takes 1.3s to enter measurement mode, which means it has to stay on for longer to measure the same data as ADXL356 and MEMS B, thus consuming more power, as shown in Table 6 Show. If MEMS B and ADXL356 measure data at the worst-case speed of MEMS C1, both are in standby mode 55% of the time, while MEMS C1 is in this mode for only a few milliseconds.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 15. Current consumption for ADXL356, MEMS B, and MEMS C1: start-up, then 1s measurement with worst-case start-up time for MEMS C1, repeated twice within 4.5s.

Table 6. Average Current Relative to Figure 15

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 16 shows the current consumption for 5s of activity data measurements per minute, with the device in standby mode the rest of the time. The average current is shown in Table 7.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 16. Current consumption for ADXL356, MEMS B, and MEMS C1: start-up, then measured at 5s worst-case start-up time for MEMS C1, for a total of 60s.

Table 7. Average Current Relative to Figure 16

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Even when measured at a lower frequency (5 seconds every 60 seconds), the average current consumption of the MEMS C1 and ADXL356 is very close, although the active and standby current consumption is different. If the measurement frequency is low, it is more feasible to turn off the MEMS sensor between measurements to reduce current consumption, as shown in Figure 17, where the ADXL356 has the lowest average current consumption.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 17. Current consumption for ADXL356, MEMS B, and MEMS C1: start-up, then measure for 5s, then turn off, for a total of 60s.

Table 8. Average Current Relative to Figure 17

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

SmartMesh IP transceivers such as the LTC5800 have several different power consumption profiles. Figure 18 shows the maximum power consumption for each mode in the data sheet. However, for reasonable operation, the power consumption of a typical SmartMesh chip configuration in a network will be much lower. Various factors will determine actual power consumption, including: reporting interval (1 packet per minute or 1 packet per second), how many hops are required to transfer data, payload size (1 byte to 90 bytes), and path stability (e.g. 80% of indoor environments have dense networks).

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 18. SmartMesh IP current consumption (worst case data sheet specification)

Actual battery life depends on many factors, such as the time the node collects and transmits data versus the time the node sleeps. Payload size, path stability, transmission interval, hop depth, and many other factors can affect the power consumption of a SmartMesh IP node. Based on a few key factors, a very useful and accurate tool—the SmartMesh Power and Performance Estimator—can be used to estimate performance and power consumption, as shown in Figure 19.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 19. SmartMesh Power and Performance Estimation Tool

Voyager module: send a complete dataset

To assess power consumption, it is useful to know how many packets are required to transmit a complete dataset from the wireless node to the SmartMesh IP manager. With a reporting interval of 1s, the data rate sent from the node to the manager is 60 packets per minute. The x-axis, y-axis, and z-axis sampled data each consist of 512 time-domain samples of 16 bits (2 bytes) per sample. The FFT data is also calculated and sent, as shown in Figure 20.

(512 + 512/2) × 3 = 2304 samples, so 2304 × 2 bytes = 4608 bytes. 90 bytes are sent in a SmartMesh packet. 4608 bytes / 90 bytes = 51.2 packets. 52 SmartMesh packets are required to transmit a complete dataset from the wireless node to the SmartMesh IP Manager.

For power consumption estimation, a network with 20 nodes is used as an example, the nodes are arranged in 4 hops with 5 nodes per hop. With the data payload size set to 90 bytes and the reporting rate set to 1 packet per second, the SmartMesh IC (static condition) at hop 1 node consumes 587.9 μA. For worst-case dynamic conditions, it is recommended to increase the power dissipation by 30%, resulting in 587.9μA × 1.3 = 764.3μA. SmartMesh power consumption and performance estimation tools confirmed these results.

Figure 21 shows the worst-case battery life estimates for a Voyager module with 4 hops (2 × Saft LS14500), one where the node activates every 60 minutes, and one that activates every minute for 60 minutes. As expected, with the node transmitting every minute for 60 minutes, the battery life was much shorter. The node at hop 1 wants to receive all the data sent by nodes 2, 3, 4, so it does more work. Hop 1 had a battery life of 19.1 days (0.052 years), while hop 4 had a battery life of 20.1 days (0.054 years). When the node transmits 1 minute per hour, the battery life of hop 1 is 1.38 years and the battery life of hop 4 is 2.12 years.

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 20. Voyager GUI showing time and frequency domain data

How to Choose the Best MEMS Sensor for a Wireless Condition Monitoring System (Part 1)

Figure 21. SmartMesh battery life versus hops required to transmit data

in conclusion

There are three articles in this series. This article concludes by discussing some of the key trends currently driving the rapid development and growth of the condition monitoring market, as well as some design considerations for selecting the right MEMS sensor and wireless transceiver for use in harsh RF environments.

Low-power, high-performance MEMS sensors and high-fidelity, low-power signal chain devices are key to providing the condition monitoring industry with wireless capabilities that rapidly deploy assets and begin to recover $50 billion a year from unplanned downtime Required for dollar losses. The Grid Technology Overview provides an overview of the key differences between competing wireless technologies and highlights which technologies are best suited for harsh RF environments that require simultaneous monitoring and control and wireline-like reliability.

Choosing the most suitable MEMS sensor can be difficult, there are many factors to consider, such as noise, bandwidth, and g-range, but also some less-mentioned data sheet specifications (such as turn-on time) and the required data rate of the wireless system , as this helps determine the most feasible mode of operation and data rate.

Wireless devices used in harsh RF operating environments such as factory floors must provide robust communications with low power consumption. This article presents data sheet worst-case values ​​for SmartMesh devices and power estimates calculated using the SmartMesh power and performance estimation tool to give the reader a rough idea of ​​what is possible. This tool is recommended for further exploration, as sensor networks can be tailored to specific needs, resulting in a better estimate of possible battery life and performance. In the next installment of this series, we will describe how the Voyager platform can detect various machine faults such as imbalance, misalignment and bearing defects early. The next part will discuss the actual power consumption performance of the Voyager module, as well as a variety of different operating modes – both higher data rate mode and ultra-low power mode, so stay tuned.

About Analog Devices

ADI is the world’s leading high-performance analog technology company dedicated to solving the toughest engineering design challenges. With outstanding detection, measurement, power, connection and interpretation technology, build intelligent bridges between the real and digital worlds, thereby helping customers to re-understand the world around them. For details, please visit ADI’s official website www.analog.com/cn.

About the Author

Chris Murphy is an Applications Engineer at the European Central Application Centre in Dublin, Ireland. He joined Analog Devices in 2012 to provide design support for motor control and industrial automation products. He holds a master’s degree in electrical engineering and a bachelor’s degree in computer engineering.

Richard Anslow is a systems applications engineer in the Interconnected Motion and Robotics team in the Automation and Energy business unit at Analog Devices. His areas of expertise are condition-based monitoring and industrial communication design. He holds a Bachelor of Engineering and a Master of Engineering from the University of Limerick, Ireland.

By: Chris Murphy, Applications Engineer, Analog Devices, Richard Anslow, Systems Applications Engineer, Analog Devices

The Links:   CLAA150XP03 CXA-L0524-NJL TFT-Panel