In the last several years, industrial facilities have started to push for greater efficiency in their manufacturing processes. With the technology available today, industry is obtaining more automation, better precision, and increased process data volume. All of this is rapidly bringing manufacturers closer to realizing the benefits that Industry 4.0 has to offer including increased safety, reliability, and productivity – all while reducing the environmental impact that harsh industrial processes are associated with.
Even though the final objective is very desirable, significant obstacles still remain that need to be overcome. Case in point, industrial facilities are traditionally conservative which tends to lead to slow adoption of newer technology. This leads to manufacturers incorporating newer systems with legacy ones, often leading to more intersystem communication complexity. More often than not, existing infrastructure has major difficulties capturing and communication edge network data. The result of this is that we cannot expect an overnight transformation of factories. There must be a transition period. Analog Devices Inc. (ADI) can enable and accelerate this transition period with solutions and system domain expertise.
As organizations and markets transition toward Industry 4.0, the preferred communication medium is Industrial Ethernet. However, solving the problem of determinism over Ethernet continues to be one of the main challenges.
The use of proprietary layer 2 solutions is one solution used by many protocols. A drawback of doing this is that significant interoperability issues when extracting data for use at higher levels of the enterprise network or coordination between unrelated manufacturing nodes is required. To help alleviate this problem, the new IEEE 802.1 TSN (Time Sensitive Networking) standards were developed. These standards are meant to enable the transition from propriety solutions to standards-based ones.
Since Ethernet has always been considered a “best-effort” network, it is necessary to add specific features in mission-critical applications. These features include seamless redundancy, scheduled traffic, time synchronization, ingress policing, and others. With these features, network designers can ensure delivery of certain classes of traffic throughout the entire network. Additionally, the design incorporation of these features is scalable to gigabit-plus line rates, unlike proprietary layer 2 solutions. Allowing this mission-critical real-time traffic on the same network as best-effort and streaming traffic is the goal of the emerging IEEE TSN standards. To help users become familiar with the TSN standards, ADI offers the RAPID-TSNEK-V0001 evaluation kit (Figure 1). This kit includes everything needed for evaluation of the features related to the developing IEEE 802.1 TSN standards. The kit currently supports the following TSN standards:
Since this is an evolving standard, ADI plans on offering free downloads of new features and updates as they become available.
Figure 1: ADI’s RAPID-TSNEK-V0001 evaluation kit for 802.1 TSN standards. (Image source: Digi-Key Electronics)
Many challenges are raised when edge devices are connected to TSN-enabled converged Industry 4.0 networks. While edge device communication technologies like fieldbus and 4 mA to 20 mA current loops work reliably, transporting their data to the cloud is regularly impeded as it travels through the many communication layers from the factory floor to the front office. Often, gateways are required for translation from one protocol to another and multiple servers may store the data as it travels to where the analytics are performed. Another thing to consider is that the total cost of ownership to transport data from a sensor to the cloud is not just the hardware needed for data delivery, but also encompasses the software, processing, and manpower necessary to guarantee the data integrity as it travels through the network.
While bringing Ethernet to a simple edge device such as a temperature transmitter may not seem very efficient, it is not about device simplicity or the small data quantities produced or consumed. It is about extracting the device’s data cost effectively on a converged network and then being able to use that data for actionable results. A distributed control system (DCS), for example, may use the temperature transmitter data to verify that the process it is monitoring is being controlled in real-time. Also, this specific temperature could have adverse implications on the overall process. In this case, having a seamless connection between the temperature transmitter and the cloud enables performance of analytics that considers all process parameters very near to real-time which can ensure the overall process is performing as expected. Additionally, to ensure the production processes are optimized or to increase energy efficiency, adjustments can be made on the fly based on the real-time analytics.
These challenges are viewed by ADI as a motivating factor for their investment in new Ethernet technologies. To this end, ADI offers a low complexity Ethernet technology which is a key enabler for connecting simple industrial devices, such as the temperature transmitters used in the example above, directly to an Ethernet network (Figure 2).
Figure 2: Block diagram showing how ADI’s fido5100/fido5200 can interface with many different protocols. (Image source: Analog Devices)
Low complexity Ethernet brings down the total ownership cost of transmitting data to the cloud by reducing the size, power, and cost when compared to the current standard layer 2 Ethernet implementations.
To facilitate a transition to a converged industrial Ethernet network, innovation at the physical layer is also needed in order to deliver a solution matching some of the incumbent system’s inherent capabilities. In many deployed Ethernet networks, the physical layer standards limit the cable length to 100 meters, requiring multiple twisted pair cables for full implementation. In contrast to this, a majority of the automation network infrastructure installed in factories today is built upon single twisted pair cabling, some of which can extend 1000 meters or more with data rates of 31.25 kbps.
ADI, along with some key industrial partners, is working to address this issue by working with the IEEE in development of a new Ethernet standard. This standard, 10SPE, defines an Ethernet network that operates over a single twisted pair cable up to 1000 meters long with a 10 Mbps data rate. Through this collaborative, standards-based approach, ADI is supporting the effort to lower the barriers to adoption and narrowing the time frame in which the goal of achieving converged, plant-wide networks can be realized.
Additionally, other applications that use deterministic Ethernet (Fido5100 and 5200) at 100 Mbps already are pushing the limits of bandwidth and performance. For instance, robotics applications, with their ever-increasing number of coordinated axes, require control at a greater precision than previously possible. To help satisfy this requirement, the transitioning of the control network to gigabit speeds is essential. This increased communication speed is yet another industrial Ethernet major trend.
Another aspect of the industrial Ethernet that needs to be addressed is that of security. With the anticipated rise in the demand for data and sensing at the edge of industrial networks, come greater security risks. Furthermore, the industrial control low latency and jitter requirements can directly conflict with security requirements. Therefore, designers that use these technologies need to address performance and security concerns in these applications. These security risks are attracting increased attention every day.
With the emergence of the Industrial Internet of Things (IIoT) and Industry 4.0, the industrial space is being redefined by dynamic information flows, widely distributed devices, and connectivity across platforms to provide new capabilities. It is no surprise, however, that as these new capabilities are created, new security threats emerge.
Just establishing the identity of edge devices seems problematic when one thinks about the vast numbers of these devices that require secure connections to the network. Using traditional methods such as physically distributed shared encryption keys and managing certificates-exchanges become impractical and a nightmare logically. Therefore, if a trusted Industry 4.0 enterprise is to become a reality, establishing a keyless identity technique is vital. Also, since edge devices are, by nature, highly constrained devices that require both small hardware and software footprints, low, fixed latency lightweight encryption methods are needed. To address these issues, ADI has heavily invested in resource-constrained device security technologies like lightweight block cryptography and identity authentication. In fact, Linear Technology’s (now a part of ADI) SmartMesh IP™ technology addresses many of challenges discussed in this article.
Figure 3: The SmartMesh IP starter kit from Linear Technology/Analog Devices. (Image source: Digi-Key Electronics)
Responding the needs of the industrial automation market, Analog Devices’ Industrial Automation Group has designed and now offers industrial network edge device solutions in the areas of sensing, monitoring, control, and robust real-time communication systems. Additionally, ADI has also offers solutions in other areas necessary for edge devices such as multiprotocol support, security and authentication, functional and intrinsic safety. With all of this, ADI enables and accelerates the manufacturing transition to a trusted IIoT connected enterprise.