How Multi-Sensor Asset Monitoring Can Improve Performance in Industry 4.0 Factories and Logistics and in Datacenters

Monitoring machines for parameters like vibration and temperature can provide real-time data on machine performance and health and give manufacturers the data needed to schedule proactive maintenance, reduce downtime, and improve productivity.

Humidity and temperature monitoring in logistics facilities or during transport can improve operational performance and preserve goods like vaccines or fresh produce. Environmental monitoring systems with wired and wireless connectivity are available to suit various applications, including enterprise and Cloud data centers.

Monitoring vibration can be beneficial for identifying potential machine problems before they occur. International Organization for Standardization (ISO) 10816 can be an important reference. It provides guidance for evaluating vibration severity in motors used for pumps, fans, compressors, gearboxes, blowers, dryers, presses, and similar machines that operate in the 10 to 1000 Hz frequency range.

This article presents some key considerations for selecting between wired and wireless connectivity for monitor systems and how using wired and wireless networks is not an either/or choice. It then examines the four classes of vibration severity as defined in ISO 10816. It concludes by discussing various options for implementing both wired and wireless condition monitoring systems, including using multiple sensors for monitoring vibration, temperature, humidity, and representative applications.

Banner Engineering offers a choice of equipment health monitoring (EHM) gateways that provide easy access to the EHM network data. Industrial EHM designers can choose between the company’s SNAP ID wired gateway solutions with a local display for sensor readings or an optional Cloud dashboard and the CLOUD ID wireless gateways designed to connect with a Cloud dashboard (Figure 1) directly. Common features of these two choices include:

• A range of sensors to select from to optimize EHM operation
• Rapid deployment supported by automatic recognition of connected sensors without additional programming
• Sensor data readily available for adjusting equipment or for scheduling needed maintenance to minimize downtime and maximize productivity
• Cloud connectivity support by both systems
• Preconfigured dashboards available and customizable for optimal data visualization

Figure 1: Banner’s SNAP ID wired (left) and CLOUD ID wireless (right) EHM gateways have several common features. (Image source: DigiKey)

Wired or wireless EHM gateway?

While they have some common features, there are essential differences between the wired and wireless EHM gateways. The AMG-SNAP-ID asset monitoring gateway (AMG) supports commissioning, monitoring, and alarms for up to 20 sensors and converters. It supports Modbus and Banner’s SNAP SIGNAL connectivity and scans for individual sensors or converters, auto-detecting model information. Users can change and assign Modbus server ID numbers to build and commission custom EHM solutions. Connected devices can be grouped, and alarms can be assigned thresholds individually. The alarm status is visible on the touchscreen and by the color of the light on the top of the enclosure.

When reaching directly up to the Cloud is required, EHM system designers can turn to the DXM1200-X2 IIoT gateway to connect up to 200 devices from both Banner and third parties to deliver performance and machine health data. It can automatically detect and connect with sensor nodes and deliver data to the Banner Cloud software. Developers can choose between simple or complex programming tools. The IIoT gateway can process information on the edge and then send it via both Ethernet and cellular networks to be monitored anywhere in the world with an intuitive Cloud dashboard (Figure 2).

Figure 2: The wireless (left) and wired (right) IIoT sensor network gateways include several common features. (Image source: Banner Engineering)

Wired and wireless EHM architectures

Wired and wireless EHM architectures are not mutually exclusive. Wired systems can have wireless elements, and wireless architectures often include wired connectivity.

For example, a basic wired EHM architecture can include several junction boxes connected to multiple sensors like the 4-port R50-4M125-M125Q-P and the 8-port R95-8M125-M125Q-P. Banner’s Sure Cross R70SR serial data radios, like the 900-MHz R70SR9MQ and the 2.4-GHz R70SR2MQ, can extend network range without additional cabling. Features of these radios include (Figure 3):

• RS-485 serial interface
• Support for star and tree network topologies
• Support for self-healing, auto-routing radio frequency network with multiple hops to further extend network range
• Frequency hopping spread spectrum (FHSS) technology for reliable data transmissions

Figure 3: Basic wired asset monitoring topology (left) with example of a wirelessly connected remote sensor cluster (right). (Image source: DigiKey)

In a large facility, numerous systems can be spread out over a wide area, including:

• Air compressors
• Pumping systems
• Conveyor systems
• Numerous electric motors and machines
• Gearboxes
• Air filtration systems
• Level measurement and monitoring in storage tanks

In these cases, EHM system performance can be improved by combining wired and wireless technologies. The DXM1200-X2 wireless IIoT gateway mentioned above includes Modbus wired connectivity. If Ethernet is needed, designers can turn to the DXMR90-X1. The  DXMR90-4K can implement IO-Link master/controller functions. In addition to the choice of Modbus, Ethernet, and IO-Link, designers can use the R709 serial data radios to provide wireless connectivity to physically dispersed assets (Figure 4).

Figure 4: IIoT wireless gateways (lower left) are available with Modbus, Ethernet, and IO-Link connectivity. (Image source: Banner Engineering)

ISO 10816 vibration severity

ISO 10816 is an important standard for EHM systems. It quantifies vibration severity for machines like electric motors, pumps, and generators. The standard uses the root mean square (rms) value of acceleration, displacement, or vibration velocity. ISO 10816 also includes considerations for peak-to-peak values. The vibration severity has the highest rms value when measuring two or more parameters. The standard classifies vibration severity into four levels:

• Good usually indicates newly commissioned machinery.
• Satisfactory vibrations indicate the unrestricted operation region.
• Unsatisfactory vibrations indicate a need for restricted operation and scheduling preventative maintenance.
• Unacceptable vibrations indicate that machine damage is likely.

Figure 5: IEC 10816 includes four categories of vibration severity. (Image source: Banner Engineering)

Vibration and machine learning

Even “identical” machines are not exact replicas. That’s where machine learning (ML) comes in. Banner Engineering offers VIBE-IQ, a vibration monitoring software package that uses machine learning (ML) to establish a unique baseline operating value for each machine’s vibrations. The ML software then automatically sets warning and alarm thresholds. It can automate complex EHM calculations and analysis. Some features of VIBE-IQ include:

• Continual monitoring of rms velocity from 10 to 1,000 Hz, rms high-frequency acceleration from 1,000 to 4,000 Hz, and temperature
• Only monitors motors that are running
• Uses data for trend analysis as well as real-time monitoring to identify conditions like:
  • Misaligned or imbalanced systems
  • Worn or loose components
  • Excessive bearing wear
  • Improperly mounted or driven motors
  • Over temperature conditions
• Proactively sends alerts to the host controller or the Cloud

Vibration and temperature

Vibration is not the only clue that a machine might need preventative maintenance. A rising temperature trend can also alert the EHM system of potential problems, especially if the rising temperature is correlated with increasing vibrations.

Combining the two parameters gives a more complete picture of the equipment’s condition. They can alert operators to different sets of conditions and provide multiple benefits:

• Vibration can identify mechanical issues like misalignments, imbalances, bearing wear, etc.
• Temperature increases can identify electrical problems like overheating windings or lubrication problems.
• When detecting anomalous operation, correlating out-of-band vibration and temperature can help identify possible causes. For example, vibration patterns can help identify the root cause.
• Planning preventive maintenance can be aided by monitoring both temperature and vibration. A gradual temperature rise is not necessarily as much of a problem as increasing vibrations that can demand more immediate correction.
• Learn how to improve longer-term asset selection and utilization using sensor data to identify potential operating limitations before they become problems.

When temperature and vibration need to be monitored, EHM system designers can turn to the QM30VT2 sensor in an aluminum housing or the QM30VT2-SS-QP in a stainless steel housing, both from Banner Engineering. Both sensors can connect to a Modbus radio or any Modbus network as a slave device via RS-485. Their small form factor enables them to fit into tight locations (Figure 6). Other features include:

• High-accuracy temperature and vibration measurements
• Temperature measuring range of -40°C to +105°C, with a resolution of 1°C and an accuracy of ±3°C
• Detects dual-axis vibration up to 4 kHz bandwidth with an accuracy of ±10% at 25°C and a default sampling frequency of 20 kHz
• Outputs for rms velocity, rms high-frequency acceleration, peak velocity, and other parameters pre-processed from the vibrational waveforms

Figure 6: Two-axis vibration plus temperature sensors can be mounted directly on motor housing (right). (Image source: Banner Engineering)

Vibration spectral banding is an advanced capability. It allows users to split the wide band fast Fourier transformation (FFT) to get rms velocity or acceleration data for narrower frequency bands in addition to the 10 to 1,000 Hz and 1,000 to 4,000 Hz scalar data. Depending on users' needs, the band frequencies can be input manually or automatically generated based on a dynamic or static speed input. Spectral band analysis can aid in diagnosing problems with rotating machines more specifically.

Temperature and humidity

Monitoring temperature and humidity can be important in data centers, warehouses, cleanrooms, refrigerators, or chillers. A temperature and humidity sensor like the DX80N9Q45THA can help to:

• Preserve goods like fresh produce or vaccines where knowing the temperature and humidity are essential to long-term viability and preventing spoilage
• Protect equipment like servers and storage devices in a data center where excessive temperature or humidity can interfere with normal operation or lead to breakdowns
• Enhance health and safety of people in warehouses and other facilities where high humidity can make it difficult for workers to keep cool in elevated temperatures, potentially leading to heat exhaustion

The temperature measuring range of -40°C to +85°C with a resolution of 0.1°C and accuracy of ±0.6°C from -40°C to 0°C, ±0.4°C from 0°C to +60°C, and ±1.2°C from +60°C to +85°C. The humidity sensor can measure from 0% to 100% RH with an accuracy of ±2% at +25°C, ±3% from 0°C to +70°C and 10% to 90% RH, and ±7% from 0°C to +70°C and 0% to 10% or 90% to 100% RH.

When the unit is turned on, it operates in fast sample mode and sends data every two seconds. After five minutes, the node enters the default mode and sends data at five-minute intervals. Sample rates of 15 minutes or 64 seconds are user-selectable.

Models with 900 MHz radios transmit at 1 W (30 dBm) or 250 mW (24 dB user-selectable). The 250 mW mode reduces the range but improves battery life in short-range applications. For 2.4 GHz models, transmit power is fixed at about 65 mW (18 dBm). When operating in storage mode, the radio is shut off to conserve battery life.

Conclusion

Effective EHM systems in Industry 4.0 factories monitor vibration and temperature to help ensure high levels of uptime. Humidity and temperature sensors can also improve operational performance of data centers and preserve goods like vaccines or fresh produce in warehouse and logistics operations. These systems can use wired or wireless connectivity to monitor multiple parameters.