Fieldbus in focus: Understanding network similarities, differences
There is no one-size-fits-all fieldbus. Individual buses are optimized for different characteristics. Building automation has different requirements from networks designed to support manufacturing processes. Minimizing cost is a priority for building automation buses such as LONworks and BACnet, whereas networks for the manufacturing environment focus on robustness. Regardless of whether a manufacturer is a factory automation user or a process automation user, networks and buses must be able to withstand potentially harsh environments and must prioritize network availability and data integrity to support the manufacturing operations they control.
A fieldbus is a digital communications network typically used for distributed instrumentation and control functions. The Profibus Trade Organization (PTO) defines fieldbus as “a digital, serial, two-way, multi-drop communication link among controllers and their remote I/Os, sensors, actuators and inter-networking components. Compared to standard local area networks (LANs), fieldbuses are specialized for the rugged industrial environment, determinism and bus powering.”
A fieldbus must accommodate interconnectivity, interoperability and interchangeability. Interconnectivity means that devices from different manufacturers can be safely connected to the same fieldbus (electrically compatible). Interoperability means the ability to connect successfully, and operate elements from different suppliers. Interchangeability means that devices from one manufacturer can be replaced with functionally equivalent devices from another manufacturer.
The alternative to using a fieldbus is point-to-point wiring %%MDASSML%% individual cables between each field device and its controller. One of the primary reasons to use fieldbuses is to eliminate extreme amounts of cabling that would be necessary if each field instrument was wired to its controller individually. Other benefits of using a fieldbus include the capability of device configuration or reconfiguration; device diagnostics; device troubleshooting; reading the values of additional measurements provided by the device; quicker commissioning and device health and status.
The on/off status of factory automation I/O is a “no-brainer” for digital. By its very nature, the task of detecting whether a limit switch is open or closed is a binary operation. However, process automation is a different world. Continuous control requires continuous measurement and continuous actuation of control elements. Control variable measurements are inherently analog. Obviously, these analog signals must be converted to digital if they are to be transmitted on a digital fieldbus.
Primarily, buses are digital to minimize error. Analog signals are subject to distortion, interference, losses or other factors that affect their accuracy. On the other hand, digital signals are not subject to the same distortions or electrical interferences that affect the integrity of the signal. There are factors that affect the integrity of digital signals. However, most fieldbuses include provisions to retransmit data if information is unintelligible. Error-checking techniques enable fieldbus systems to detect if the digital message has been corrupted, discard corrupted messages and ask for retransmission of that message.
Types of fieldbuses
Types of fieldbuses vary depending on the type of manufacturing it supports. An automobile manufacturer and a chemical plant have different constraints and require different control systems. Consequently, the types of data exchanged during control functions are different, which require different types of fieldbus networks.
The term “factory automation” is used to describe discrete logic and automation used in assembly-line type manufacturing operations such as automotive, machining, packaging and bottling. Factory automation is predominantly controlled by discrete logic. Discrete logic is inherently binary in nature, as it represents one of two possible logic states: on or off, high or low, 1 or 0.
Networks designed to handle simple, discrete I/O logic states typically accommodate small data packets, and are not suitable for larger messages such as configuration downloads inherent in process automation. Device buses %%MDASSML%% or byte-level buses %%MDASSML%% are oriented to higher speed advanced protocols and include DeviceNet, ControlNet and Profibus DP. Smaller data packet size accommodates fast moving machinery common to factory automation.
DeviceNet is a digital, multi-drop network that connects and serves as a communication network between industrial controllers and I/O devices. Each device and/or controller is a node on the network. DeviceNet supports multiple communication hierarchies and message prioritization. DeviceNet systems can be configured to operate in a master-slave or a distributed control architecture using peer-to-peer communication.
ControlNet is a real-time, control-layer network that provides high-speed transport of both time-critical I/O data and messaging data, upload/download of programming and configuration data and peer-to-peer messaging on a single physical media link. ControlNet permits all nodes on the network to simultaneously access the same data from a single source.
Profibus DP is based on RS-485 multi-drop serial communications operating at up to 12 Mbaud. Actually, Profibus DP and Profibus PA use the same protocol. Both have extensive parameterization and diagnostic features, 244-byte data packets, fail-to-known-state behaviors, are deterministic and easily configured with one tool for both discrete and process automation.
“Process automation” describes the control processes required in the continuous process industries, such as refineries, pulp and paper, power generation and chemical plants. The networks or fieldbuses used predominantly in process industries include HART, FOUNDATION fieldbus and Profibus PA. These protocols were specifically designed for bus-powered field instrumentation having predefined parameters and commands for asset management information. This asset management information can include diagnostics, identification, materials of construction and calibration and commissioning functionality.
HART stands for Highway Addressable Remote Transducer. It uses a combination of digital communication superimposed on a conventional 4-20 mA analog signal. HART communication is a bi-directional master-slave industrial field communication protocol used to communicate between intelligent field instruments and host systems. The HART protocol communicates at 1,200 bps without interrupting the 4-20 mA signal and allows a host application (master) to get two or more digital updates per second from a field device.
“There is a big difference between a 4-20 mA analog device and a microprocessor-based HART device that uses the 4-20 mA as a power and transmission medium,” said Ed T. Ladd, Jr., director of technology programs at HART Communication Foundation. “HART-enabled devices have traditionally provided a unidirectional 4-20 mA signal representing the primary variable along with a simultaneous/bidirectional digital signal. The 4-20 mA provides full backward compatibility to analog-only systems, while providing the multivariable, device status and diagnostic information on the digital side.
“You can use the HART application layer, which is the part the user sees, over Profibus, Profinet, Ethernet and RS-485 Modbus,” Ladd continued. “When you send HART Communication over one of these physical layers, it is all digital. Profibus has its own application layer for DP and PA.”
The FOUNDATION fieldbus architecture provides a communications protocol for control and instrumentation systems in which each device has its own “intelligence” and communicates via a digital, serial, two-way communications system. Foundation H1 is intended primarily for process control, field-level interface and device integration. Running at 31.25 kbit/s, the technology interconnects devices such as transmitters and actuators on a field network. FOUNDATION fieldbus H1 is designed to operate on existing twisted pair instrument cabling with power and signal on the same wire.
Most current distributed control systems (DCSs) offer FOUNDATION fieldbus digital communications as one of the ways to connect digital field devices into the system. “For a FOUNDATION fieldbus project to be a success it has to be planned, engineered and designed correctly,” said Jim Cahill, marketing manager at Emerson Process Management. “If you follow old DCS practices for design, loop check and commissioning, all you will get is a more expensive DCS without new fieldbus benefits. However, if done correctly FOUNDATION fieldbus will deliver on the great promise of digital control and architecture. These practices have been worked out and fine tuned by leading automation vendors over thousands of projects the past 10 years.”
Profibus PA and FOUNDATION fieldbus H1 use identical wiring and follow the IEC 61158-2 standard. However, there are major differences between the two, depending on system architecture. Only a master device can initiate communication and only one device at a time accesses the bus. The slave field instruments such as transducers, transmitters or valve positioners respond only. Therefore the only two device types recognized are master and slave. Both central controllers and the host configuration tool can be recognized as masters.
Masters pass a token among themselves in a logical ring according to the addressing sequence. When a master receives the token, it can send its message. The master holding the token can poll slave devices. Because of the token passing arrangement, only one device is granted permission to access the bus at any given time, which ensures that there are no collisions.
System architecture, levels
There is a marked difference between the levels of networking in an industrial networking environment %%MDASSML%% this is necessarily by design. The architecture of industrial networks is such that the requirements of field devices such as pressure and temperature transducers, flowmeters and control valves are quite different from the requirements of the host.
The host level network ties the process automation subsystems together. It also communicates with the enterprise computer systems, historians, maintenance systems and manufacturing execution systems. For host level communication, the dominant contender is becoming Ethernet. The use of Ethernet for office LANs and consumer/residential computing has driven the cost down to a point that makes it practical for the plant floor %%MDASSML%% even with the emergence of industrially-adapted Ethernet hardware.
It should be strongly emphasized that Ethernet is not a fieldbus. It is not even a complete protocol. Ethernet specifies different cable options and how devices on the network access the bus. It does not specify data formats or semantics.
Even when used with TCP/IP, the protocol is incomplete. Even with TCP/IP, most of the Ethernet networks for control systems on the market today are proprietary. This is because other devices cannot access and interpret the information.
Introduced in 2001, EtherNet/IP is a member of a family of networks that implements Common Industrial Protocol (CIP) at its upper layers. CIP encompasses a suite of messages and services for a variety of discrete manufacturing automation applications, including control, safety, synchronization, motion, configuration and information.
Profinet builds on the concepts of Profibus; uses the same organization and structure as Profibus; and shares application profiles such as Profisafe and Profidrive with Profibus. However, Profinet’s protocol is different from Profibus. The types of tasks and operations capable over Profinet include real-time I/O, peer-to-peer integration, motion control, vertical (enterprise) integration, safety, security and integration of existing buses.
Profinet achieves industry requirements for speed and determinism using standard Ethernet by using TCP/IP where appropriate, skipping it when necessary, reserving bandwidth for higher performance and scheduling traffic to ensure motion control needs. Profinet does not use standard TCP/IP for real-time operations or communications. However, it does use TCP/IP where it makes sense, such as in diagnostics, non-time-critical data and communicating with the host or other higher-level IT systems. Profinet real time coexists with TCP/IP without restrictions.