By Richard anslow and Dara O’Sullivan of ADI
Condition based monitoring is critical to achieving industry 4.0
Government initiatives (such as industrie 4.0 in Germany and made in China 2025 in China) are accelerating the development trend of manufacturing towards universal network automation. In addition, the intelligent sensor system is improving the degree of automation and providing more data to monitor and control the production process. In particular, made in China 2025 aims to rapidly develop high-tech industries including electric vehicles, new generation information technology (it) and telecommunications, advanced robots and artificial intelligence. With a more advanced system, we need to adopt more advanced methods to ensure the reliability of the system.
Condition based monitoring of robots and rotating machines (such as turbines, fans, pumps and motors) records real-time data related to the health and performance of the machine for targeted predictive maintenance and optimal control. Targeted predictive maintenance at the early stage of the machine life cycle can reduce the risk of production downtime, so as to improve reliability, significantly save costs and improve plant productivity.
How to implement a status based wired monitoring solution?
To implement condition based monitoring of industrial machines, a series of sensor data can be used, such as electrical measurement, vibration, temperature, oil quality, acoustics and process measurement (such as flow and pressure). However, vibration measurement is the most common at present because it can most reliably indicate mechanical problems, such as imbalance and bearing failure. This paper mainly studies the application of vibration sensing, but this method is also applicable to the data from other sensors.
This transmission means that the sensor data from the sensing node is sent to the master controller or the cloud highly dependent application. In many applications, some local data will be processed at the terminal node, and the summarized data will then be sent to the network gateway wirelessly, or directly to the cloud or analysis server through the cellular link. In these cases, the amount of data transmitted is usually quite low, and because the terminal node is powered by battery, it is usually required to maintain low power consumption. In other applications, raw sensor data transmission is required. For example, data from multiple sensors may need to be adjusted and fused before analysis. In the application of real-time control using data, the original data transmission is also required. In these applications, wired interfaces are more likely to be used as data transmission solutions.
CBM for industrial applications can use ADI’s optimized micro electro mechanical system (MEMS) accelerometer, low-power microcontroller and wired icoupler ® Isolate the interface signal chain to extract and adjust the machine health data from the remote CBM slave, and reliably return it to the master controller for analysis. Over time, machine health data can be used to create software based models to determine changes in machine behavior and actively maintain machine health. In some applications, such as NC machine tools, data can also be used to optimize system performance in real time.
The challenges of implementing wired CBM interfaces include EMC’s robustness when running on long cables, data integrity when transmitting at high baud rate (for real-time transmission of CBM data streams), and communication physical layer / protocol mismatch. ADI’s signal chain and system level expertise provide several possible options for realizing wired CBM interface.
This paper is divided into two parts. The first part introduces the wired interface solution of ADI company, which helps customers shorten the design cycle and test time, and make the industrial CBM solution enter the market faster. The next article focuses on detailed physical layer design considerations, including master controller and wired CBM slave controller.
Design and implementation of wired CBM
The design and deployment of state-based wired monitoring solutions need to consider multiple system performance factors and make trade-offs.
Firstly, when selecting the appropriate MEMS accelerometer, the fault type to be measured must be considered, so as to select the appropriate bandwidth and noise performance MEMS to meet the requirements of the system. Edge node processing requires careful matching of selected processors to ensure maximum system flexibility.
Secondly, the design of wired CBM system needs to carefully select the appropriate wired communication protocol and physical layer to realize high-speed real-time data stream transmission. Implementing wired interfaces requires careful consideration of EMC performance, data transmission cables, connectors, and power transmission on cables.
Select the appropriate MEMS accelerometer
Selecting a suitable MEMS vibration sensor covers several aspects:
Number of axes
The number of shafts monitored is usually a function of the type of fault and the installation arrangement of sensors. If it is obvious that the fault involves a dominant axis and there is a clear transmission path on that axis, the use of single axis sensor is sufficient. Triaxial sensing is useful for faults containing energy in multiple axes or faults with unclear energy transmission path.
The type of fault monitored has an important impact on sensor selection. The noise density and bandwidth of the sensor are important indicators in this regard, because they determine the vibration level and frequency range that can be reliably extracted. For example, for unbalance and misalignment faults of low speed machines, a low noise density sensor may be required, but the bandwidth requirement is quite low, while gear fault detection requires both low noise density and high bandwidth.
In addition to the fault type, it is also important to understand the performance requirements of CBM. Alarm detection of status indicators of basic traffic light types requires complex prediction through different levels of performance. This obviously applies to the analysis and algorithms being deployed, but it will also affect the selection of sensors. The higher the performance level of the sensor in terms of bandwidth, noise density and linearity, the stronger the analysis ability.
Select appropriate signal processing
Design considerations include:
The output of accelerometer is generally analog or serial digital signal, usually SPI. Analog output sensors will require a digital conversion phase and some signal conditioning. This can be a discrete ADC supporting preamplifier conditioning or an embedded ADC in a microcontroller.
Edge node processing requirements
In order to reduce the burden of data link and / or central controller / server, some basic FFT or signal processing algorithms may be required on edge nodes.
Data transmission protocol requirements
The output of ADC or sensor is usually SPI interface. It does not provide any mechanism for data integrity checking, simultaneous interpreting, time stamping and data mixing from different sensors. It is very useful to encapsulate the sensor data in the high-level protocol of the edge node before transmission. This can improve the robustness and flexibility of the sensor interface, but it is required to properly process and encapsulate the data flow on the edge node.
For more information, please refer to the simulation dialogue article “selecting the most suitable MEMS accelerometer for your application”.
Transplant the accelerometer output to the wired communication bus
As mentioned earlier, the output of accelerometer is generally analog or serial digital signal, usually SPI. SPI output can be processed locally (allowing protocol flexibility) and then added to the physical layer interface, or ported directly to the physical layer.
SPI is an unbalanced single ended serial interface for short distance communication. To migrate SPI directly to the physical layer over a longer distance, RS-485 line driver and receiver are required. RS-485 signal transmission is a balanced differential transmission, which can resist interference and is robust when passing through the length of long cable.
There are some challenges when using SPI over a long distance between SPI master and slave. SPI is synchronous in nature and has a clock (SCLK) started by SPI host. SPI data lines – Master output slave input (MoSi) and master input slave output (MISO) – are synchronized with SCLK, which can be achieved in a short distance. SPI also has a valid, low enable chip select (CS) signal that allows separate slave addressing if required.
In order to restore the synchronization between the master and the slave, the clock signal of the slave can be fed back to the master, or the clock phase shift can be used to compensate the cable delay of the master controller. The phase shift of the clock must match the total delay of the system. An-1397 provides implementation details of delay compensation for the main microcontroller.
Wired communication physical layer
Robust and reliable physical layer is needed for long-distance communication. As mentioned earlier, RS-485 signal transmission is balanced differential transmission, which can resist interference itself. The system noise is equally coupled to each conductor in the RS-485 twisted pair cable. The transmission of one signal is opposite to that of the other, and the electromagnetic fields coupled to the RS-485 bus cancel each other. This reduces the electromagnetic interference (EMI) of the system. Some additional key advantages that make RS-485 ideal for CBM systems include:
- Higher data rates, up to 50 Mbps for shorter cable lengths (less than 100 meters)
- When the data rate is low, the cable length can reach 1000m
- Full / half duplex RS-485 and RS-422 multi driver / receiver pairs can use a minimum number of components to convert bidirectional SPI into RS-485 bus signals
- The wide common mode input range allows ground potential difference between master and slave
- EMC performance of wired interfaces
When transmitting in long cables, the communication network may be affected by hazards, such as large common mode noise, ground potential difference and high voltage transient.
Conducted and radiated noise sources can affect communication reliability within 100m cable length. Using icoupler chip level transformer isolation technology of ADI company can improve the anti-interference ability to these noise sources. An-1398 outlines the resistance to common industrial transients that can be achieved using icoupler technology.
In a factory automation environment, system designers usually cannot control the electrical devices that provide communication networks. It is best to assume that there is a ground potential difference. In the motion control system, hundreds of volts of ground potential difference may occur. RS-485 communication nodes need current isolation power supply and data line to operate reliably in these environments. Signal and isopower isolation devices provide maximum continuous operating voltages up to 600 V (base) or 353 V (enhanced). In case of large ground potential difference, the basic insulation supports reliable communication. Enhanced insulation protects operators from electric shock in the plant.
In wired communication networks, exposed connectors and cables may be affected by many severe high voltage transients. The system level IEC 61800-3 standard related to EMC immunity requirements of variable speed electric drive system requires IEC 61000-4-2 ESD protection of ± 4 kV (contact) / ± 8 kV (air) as a minimum. ADI’s new generation RS-485 transceiver provides ESD protection higher than ± 8 kV (contact) / ± 8 kV (air) IEC 61000-4-2.
Phantom power supply on data line
The distribution of power and data lines between the main controller and remote CBM sensor nodes requires innovative solutions to reduce cable costs. The integration of data and power lines on a single twisted pair means that the system cost can be greatly saved, and a smaller printed circuit board (PCB) connector solution can be adopted at the terminal sensor node with limited space.
The power and data are distributed on the twisted pair through the inductance capacitance network. The high-frequency data is coupled with the data line through series capacitor, and the RS-485 transceiver is protected from the influence of DC bus voltage. The power supply on the main controller is connected to the data line through an inductor, and then filtered from the inductor on the sensor node using the CBM at the far end of the cable.
The inductance at both ends of the cable shall be well matched to avoid differential mode noise, and the self resonance frequency shall reach at least 10 MHz to avoid interference with the real-time burst mode of the new generation vibration measurement system of ADI company. Note that power and data coupling solutions must be added to data lines that do not require DC data content, such as MoSi or miso to RS-485 extensions.
Recommended solutions and performance tradeoffs
Based on the proposed design considerations, the following components provide the best path for a robust wired industrial vibration measurement solution.
- Adcmxl3021, wide bandwidth, low noise, triaxial vibration sensor
- Adum5401 / adum5402, four channel, 2.5 kV isolator, with integrated DC / DC converter
- Adm3066e, 50 Mbps half duplex RS-485 transceiver
- Adm4168e, 30 Mbps dual channel RS-422 transceiver
- Ltc2858-1, 20 Mbps full duplex RS-485 transceiver
- Adp7104, 20 V, 500 mA, low noise CMOS LDO Regulator
The adcmxl3021 MEMS accelerometer is common to all three solutions. This accelerometer has ultra-low noise density (25 µ g / √ Hz) and supports excellent resolution. Adcmxl3021 also has a wide bandwidth (from DC to 10 kHz, 5% flatness) and can track key vibration characteristics on many machine platforms. Adcmxl3021 provides customers with a mechanically optimized aluminum package that provides stable coupling with integrated MEMS sensors over a wide frequency range. This ensures that the vibration characteristics obtained from the tested equipment can be reliably extracted and conditioned.
Adcmxl3021 can provide SPI output, which can be directly connected with RS-485 / RS-422 devices, or connected with RS-485 / RS-422 devices through microprocessor and / or icoupler signal and power isolation, as shown in Figure 1. In order to monitor the vibration characteristics of industrial equipment in real time, adcmxl3021 provides a real-time streaming mode with an operating rate of about 12 Mbps SPI.
In order to connect real-time streaming SPI mode to RS-485 bus, components with excellent data rate must be selected.
Adm3066e / adm4168e / ltc2858-1 RS-485 / RS-422 transceivers operate at data rates of 20 Mbps and above.
For option 1 and option 2 shown in Figure 1 (which can be directly connected to RS-485 via SPI), adm3066e and adm4168e provide a reliable interface to realize SPI 3 receiving and 1 transmitting (3 + 1) configuration at the slave vibration sensor node. SPI CS receiving signal is realized by adm3066e, SPI CLK and MoSi, and miso signal is realized by adm4168e. When operating in real-time streaming mode, adcmxl3021 sends an interrupt signal to the main microcontroller to mark when a new data burst can be captured. The interrupt signal (/ busy) can also be transmitted to the host using adm4168e.
The complete solution consists of three signals (MoSi, CS, CLK) sent by the host to adcmxl3021 and two signals (miso, / busy) sent back to the host from adcmxl3021. five × Single ended signal can be converted into differential signal only by adm4168e and adm3066e. The differential signal can be converted by RJ50 connector and plug. Compared with the industrial standard RJ45 Ethernet connector, they occupy almost the same PCB area. Adm3066e and adm4168e transceivers provide greater than ± 8 kV (contact) / ± 8 kV (air) IEC 61000-4-2 ESD protection and provide necessary reliability when directly connected to wired cable interface.
For option 3, the microcontroller can preprocess the adcmxl3021 SPI output or perform protocol conversion between SPI and other serial interfaces (such as UART). UART is an asynchronous protocol commonly used in RS-485 interface. UART consists of transmit and receive signals and transmit enable signals, all of which can be directly connected to full duplex RS-485 transceiver, such as ltc2858-1. In full duplex mode, ltc2858-1 allows simultaneous two-way data transmission, which matches the requirements of SPI two-way data transmission. The microcontroller can handle the conversion from synchronous SPI to asynchronous UART protocol.
Adum5401 / adum5402 are the smallest signal and power isolation devices in the industry. They contain an integrated DC / DC converter that provides up to 500 MW of regulated isolation power at 5.0 V or 3.3 V (5.0 V input power supply).
In Figure 1, option 2 includes adum5401, which obtains 5 V DC from the data bus and then provides 3 V isolated power supply to adcmxl3021. Adum5401 also includes 4 signal isolation channels, which are configured to support 3 + 1 SPI isolation.
Option 3 in Figure 1 contains adum5402, which is similar to adum5401. The key difference is that the adum5402 provides two transmit and two receive digital isolation channels.
As mentioned earlier, adum5401 / adum5402 can improve the EMC immunity of wired CBM interface and protect adcmxl3021 from high voltage interference and ground potential difference on RS-485 cable interface.
Table 1 compares the three solutions using many key metrics, including design flexibility, PCB area, solution cost, complexity, and EMC performance.
Integrating a microcontroller in the CBM sensor node will increase the design flexibility, but will increase the PCB area and increase the software complexity. Since the main CBM node will be equipped with a processor, this means that option 3 in Figure 1 will essentially be a dual microcontroller system, which will start and run slower than a single microcontroller on the main CBM node.
Options 1 and 2 have less design flexibility, but provide a faster deployment path because they support the integration of low complexity and transparent SPI on RS-485 links. Option 1 and option 2 can also use a PCB smaller than Option 3, which requires additional PCB area to lay the microcontroller and related circuits (for example, a clock oscillator and several passive components).
Adding icoupler signal and power isolation to options 2 and 3 takes up the smallest PCB area and improves EMC performance (more than can be achieved with on-chip protection using RS-485 / RS-422 transceivers).
Figure 1. Options for a reliable, highly integrated, wired MEMS accelerometer condition monitoring based solution.
Figure 2. Ltc4332 SPI expansion interface helps save cable cost.
Low data rate solutions
For wired applications running at a lower data rate (less than 2 Mbps), the ltc4332 SPI extender provides an alternative to strengthen the SPI link between master and slave sensor nodes. The ltc4332 can transmit SPI data, including interrupt signals transmitted over two twisted pairs. This solution provides significant cost savings because it can save up to 50% of the bus cable compared to the standard solution.
Introduction to the author
Richard anslow is a system application engineer in the interconnected motion and robotics team of the automation and energy business department of ADI. His area of expertise is condition based monitoring and industrial communication design. He holds a bachelor’s degree in engineering and a master’s degree in engineering from the University of Limerick, Ireland. Contact: Richard [email protected] 。
Dara O’Sullivan is the systems application manager of the interconnected motion and robotics team in the automation and energy business department of ADI. His area of expertise is power conversion, control and monitoring for industrial motion control applications. He holds Bachelor of engineering, master of engineering and doctoral degrees from cork University in Ireland. Since 2001, he has been engaged in industrial and renewable energy applications in research, consulting and industrial fields.