This article focuses on field bus technology, a digital two-way conul1unications protocol. It is designed to take advantage of intelligent instrumentation equipped with microprocessors to communicate between the field and the control room. Because it can communicate large quantities of data from the field devices, industrial networks tying together field bus instruments open the way for complex control and automation tasks. As field bus evolves, some think that computer networking will play an increasingly important role and improve the performance in the plant. Both Profibus and Fieldbus Foundation provide for Ethernet capability. Wireless communication is also likely to play a potential role in field bus systems. There is a potential for wireless in data acquisition. Issues still needed to be worked out: electrical storms can interfere with communications and wires must be strung to power the devices.
Process Industries Present difficult environments for industrial networks. Instrumentation may be required to monitor conditions on a pipeline stretching across the desert in Saudi Arabia or the frigid slopes of Alaska. Physical distances may cover hundreds of yards or miles. Instrumentation must operate in areas where lightning may strike, or in the vicinity of hazardous materials.
Field instruments—flow meters, pressure transmitters, valve positioners, and the like that take readings out in the field—still widely rely on an analog standard, known as 4-20 milliamp, to communicate signals back to the control room. This has worked fine for nearly three decades, but is limited to communicating a single data value.
That is changing with the implementation of fieldbus technology, a digital two-way communications protocol. It is designed to take advantage of intelligent instrumentation equipped with microprocessors to communicate between the field and the control room. Because it can communicate large quantities of data from the field devices, industrial networks tying together fieldbus instruments open the way for complex control and automation tasks. Major players in a range of process industries—oil and gas, food and beverage, pulp and paper, and wastewater treatment, to name just a few—have embraced fieldbus architecture.
One example of what it provides is improved diagnostics. A fieldbus pressure transmitter attached to a line detects an irregular noise, indicating that the line is plugging. It sends a signal to the central controller, identifying the location and supplying detailed diagnostic information to a technician, who can perform the necessary maintenance. Other attractive benefits of fieldbus devices include much lower wiring, engineering, and assembly costs compared with analog systems. What’s more, with all that information, it’s easier to keep tabs on systems in hazardous conditions.
There are two fieldbus standards designed for the process industry. One standard, Foundation Fieldbus, is designed to be compatible with the SP-50 standards project of the Instrumentation, Systems, and Automation Society and has strong acceptance in the United States.
The other major fieldbus standard is Profibus PA. Profibus, which originated from academic and research institutions in Germany in the 1980s, developed a PA protocol for process automation in 1996.
The two standards are mutually exclusive, differing in their architecture and, to some extent, in their applications. While it is true that the standards compete for many of the same installations, proponents of each acknowledge that no single bus technology is best suited for all process applications. Each has its strengths.
Foundation Fieldbus, for example, has more intelligence distributed at the device level. Foundation Field-bus instruments have found strong support among people in the oil refinery and chemical industries, according to Jim Gray, director of marketing for the I/A Series distributed control system at Foxboro Co., a unit of Inven-sys Process Systems in Foxboro, Mass.
Profibus PA is a more centralized control architecture that has found adherents among the water treatment, pharmaceutical, chemical, and food and beverage industries, according to Michael Bryant, executive director of the Profibus Trade Organization in Scottsdale, Ariz.
The two protocols mark their differences in other areas, such as the way the intelligence is distributed over the network. One advantage of Foundation Fieldbus is its function block capabilities embedded in the devices that allow control at the device level in the field, said Richard Timoney, president of the Fieldbus Foundation in Austin, Texas.
According to Tom Wallace, worldwide PlantWeb marketing manager of Emerson Process Management in Eden Prairie, Minn., with Foundation Fieldbus, there is a lot more intelligence in the device itself. “The individual nodes are intelligent,” he said. “They have within them a schedule of what is supposed to happen with the bus and the functionality to perform process control in the field.”
Foundation Fieldbus uses a protocol called publish/subscribe, which will publish information—for example, a temperature reading that has four subscribers—so each device that needs the information will grab it and use it.
“It’s a very efficient protocol in the amount of bandwidth that is used to do the job,” Wallace said. Because the intelligence is distributed in the devices themselves, even if the host goes down, the bus can keep operating, and safe and effective process control can be maintained, he added.
Profibus PA is a more centralized control over distributed fieldbus architecture, according to Bryant of the Profibus Trade Organization. Profibus PA is an extension of Profibus DP, a protocol that was designed for the discrete manufacturing market. In Profibus PA, a smart transmitter may be capable of providing high- or low-level warnings, but would not have the intelligence to take over a system or to shut itself down. “It’s all programmed through the host control,” Bryant said.
One advantage of Profibus PA is that it saves on the power budget, because the system does not require large amounts of memory in each instrument. “In the long run, we provide a good solution that is also a cost-effective solution,” he said.
Bryant said that Profibus PA is positioned as an alternative to Foundation Fieldbus. “Many industries do not require the kind of control that you can get with Foundation Fieldbus,” he said. “There is an increased functionality requirement versus a cost requirement.”
Profibus PA has been used in the oil industry, in operations such as oil blending facilities or offshore drilling platforms. On the other hand, it would not be suited for an oil refinery, which has a need for devices capable of taking over a control function from a master control device, or the ability to shut down an operation in an orderly fashion, he acknowledged.
According to Todd Hubbell, a systems communications engineer with Endress+Hauser USA in Greenwood, Ind., Profibus offers some advantages in the programmable logic controller environment, because it is easier to use. He said the company sold roughly equal numbers of Profibus PA and Foundation Fieldbus instruments.
“Profibus doesn’t have function blocks in the device and you don’t have to train your operators in that,” Hubbell said. “Foundation Fieldbus gives you the ability to do field control. You can look at it both ways.”
Both bus technologies provide for intrinsic safety—a key issue in process plants because process industries often deal with hazardous materials.
"If something goes bump in a process plant, it could affect the integrity of the process equipment or be harmful to the environment or to human life,” said Wallace of Emerson Process Management.
Basically, intrinsic safety means that the power carried on the wires is limited, so that not enough energy is present to cause ignition, even if a fault occurs.
A big concern in the process industry is losing communications with the field instruments if there is a short in the cable, according to Bernd Schuessler, product manager of bus systems at Pepperl+Fuchs Inc. in Twinsburg, Ohio. Equipment such as power repeaters create independently isolated bus segments, so that if one bus segment shorts out, other segments will remain up and running, he said.
Serving Two Markets
Instrumentation manufacturers, for their part, differ in their overall fieldbus strategies, with some companies supporting both standards throughout their product line, and others leaning more heavily toward Foundation Fieldbus or Profibus PA, depending on the markets they serve or their installed base.
Both Foundation Fieldbus and Profibus PA share the same physical layer, using the same electrical representation of the signals on the wire. Beyond that, however, the protocols are much different in their capabilities, which is the reason that Profibus PA and Foundation Fieldbus devices are not interoperable on the same line.
The Foxboro Co., for example, covers both bases for process automation, said Jim Gray. Emerson Process Management, on the other hand, has a broad offering in Foundation Fieldbus, although it will supply some of its devices in Profibus PA, according to Tom Wallace.
Siemens has a broad portfolio of Profibus products, with a large installed base in Europe and a growing market share in North America. “Profibus is the native bus for our systems, and that is the path forward,” said Tanmoy Basu, product manager for fieldbus systems at Siemens Energy and Automation Inc., at Spring House, Pa. However, Basu added that Siemens is a member of both the Profibus Trade Organization and Fieldbus Foundation. Siemens has products in the pipeline that would let customers use the appropriate bus technology in their plants, depending on the automation application.
Mixing and Matching
Interoperability, of course, is a key concern to the end user. Both the Fieldbus Foundation and Profibus Trade Organization say that they test devices to ensure interoperability on the same network. The instrument manufacturers themselves often run their own tests.
Fieldbus Foundation has established registration-testing procedures to ensure that devices work together. The devices undergo a day of several hundred test cases, said Timoney. There is also a series of tests, called the Host Interoperability Support Test, to certify that registered Foundation Fieldbus instruments and host devices communicate properly.
Profibus also tests instruments to be in conformance with the PA profile, said Bryant, who estimated that perhaps 50 instrument manufacturers supply Profibus PA-registered devices.
Device manufacturers often supplement these tests with their own. “It’s in our own interest to make sure our devices can talk to everybody else’s,” said Hartmut Wuttig, vice president of technology at ABB Automation Technology Products in Cleveland.
According to Wallace of Emerson Process Management, different device manufacturers cooperate in interoperability testing. Emerson conducts tests to ensure that Foundation Fieldbus devices operate on its Delta V host system.
The company has agreements with other device manufacturers under which the companies test each other’s devices.
“The sole purpose is determining how well these products work together, and conditions under which there are problems. This way, problems can be fixed before products reach a customer,” said Wallace.
In addition to interoperability testing, Emerson applies stress tests, he said. Stress testing involves putting devices under extreme conditions of the bus in heavily loaded configurations, to verify how well they work.
Hubbell of Endress+Hauser said he is seeing more and more multivendor installations. “Host testing from Emerson certainly does ensure that the integration of the devices works in the host system,” he said.
Divide and Market
As a digital standard, fieldbus offers a fundamental departure from the 4-20 milliamp standard, which has been in existence since 1972 to connect field instruments to the control room. The 4-20 milliamp standard is relatively simple and completely analog, according to Richard Caro, vice president of ARC Advisory Group, a market research and consulting company based in Dedham, Mass. The system’s downside is its limitation to a single datapoint.
That began to change with the realization that instruments were becoming smart, with the advent of digital transmission in the field, explained Caro, who chaired the SP-50 committee of the Instrumentation, Systems, and Automation Society, which developed the standard for fieldbus. “We started off with the idea that there had to be a digital standard,” not something that would be superimposed on top of the 4-20 milliamp standard, he said.
a company that recently made a Profibus PA installation in one of its plants is Reliance Energy Pipeline Services of Dubach, La. The plant is an underground storage facility for natural gas, pumped from 72 wells in the field. The plant used 46 meter runs, each with an orifice plate as the primary element, to track natural gas inventory by well. Project manager Jay Strickler wanted to replace the “ancient and failing" 4-20 milliamp instrumentation.
“Rather than replace the existing system with traditional 4-20 milliamp analog information systems, we wanted to look at what fieldbus possibilities there were,” said Strickler. He investigated Foundation Fieldbus and Profibus PA, and settled on the latter, using all Siemens instrumentation.
An orifice run consists of three instruments: a differential pressure transmitter, a static pressure transmitter, and a temperature transmitter. Data are collected from the transmitters, passing through couplers that communicate directly with a programmable logic controller, and are sent on to the human-machine interface for process control and inventory management. Operators can initiate change in conditions such as flow rate, and temperature and pressure setpoints.
The job, from initial investigation to final commissioning, took about 15 months, Strickler said. He named three benefits of installing a fieldbus industrial network.
One is wiring. “We saved a tremendous amount of money by being able to daisy-chain everything instead of having a dedicated pair of wires to every transmitter, which is what you would have to do with an analog system," he said.
Another benefit is the availability of more information about the transmitter and the process, rather than just one process variable such as temperature. Now, status information about the transmitter, error codes, and other troubleshooting data itself gives field technicians the ability to do remote diagnostics.
The third benefit, he added, is that information is collected at one central computer, providing details of the devices’ working status. “You can do it all in one place, instead of having to go to every transmitter,” he said.
The new fieldbus standard incorporated elements of Open Systems Interconnect, or OSI, architecture for digital communications of the International Organization for Standardization, providing physical, data link, basic protocol, and application layers. Although it was originally intended for all application areas, it quickly became focused on process industries after a user layer that is defined in process control terms was added, said Caro. A factory automation layer was never prepared. The SP-50 standard was completed in 1993.
Yet implementation of the SP-50 standard was opposed by some members of the International Electrotechnical Commission, an international standards organization, who favored an alternative standard, Profibus. An attempt to reach a compromise fieldbus solution was not really successful, Caro said.
A compromise fieldbus standard, called IEC 61158, completed in 1999, was a multipart standard that incorporated eight different bus technologies, including SP-50 and Profibus. “It’s a piece of history, of no value,” Caro said. “No one in their right minds would build something to include all eight parts.”
Foundation Fieldbus, established in 1994, implemented SP-50 as a basic process control standard.
Big Growth Predicted
By several accounts, the process industry is still at an early stage in the implementation of fieldbus devices. Foxboro’s Jim Gray said that fieldbus technology has gone from the experimental stage to the point where there are dozens of suppliers. “We are in the second phase of the life cycle, which is typically where the growth occurs,” he said.
Hartmut Wuttig of ABB estimated that only 5 percent or so of process plants worldwide have installed fieldbus industrial networks in their plants. But, he added, in projects concerning either new plants or significant refurbishment, more than 50 percent of the proposals are based on field-bus systems.
Gray cautions his customers that fieldbus is a new technology. “It’s tempting to jump to the endgame right from the beginning,” he said. “Customers should pick things that are relatively easy to do. Keep it simple and, as you gain more experience with it, you can step it up.”
There is also training associated with engineering and setting up a fieldbus control strategy, he said. Instruments are traditionally specified late in a project. Because instruments are now an integral part of the control strategy, it’s necessary to change the way a project is engineered. Instruments have to be considered earlier, when laying out the control strategy, said Gray.
Customers will also need to reconsider plant practices and standards, and perhaps develop new ones that are appropriate to the fieldbus environment, Wallace said. Some tasks will happen in a different order or may take more or less time, making it necessary to rethink project schedules and staffing. Generally, Foundation Fieldbus schedules are shorter and staffing reduced, compared with traditional analog installations.
The Next Steps
As fieldbus evolves, some think that computer networking will play an increasingly important role and improve the performance in the plant. Both Profibus and Fieldbus Foundation provide for Ethernet capability. “High-speed Ethernet will become more popular as we move forward,” said Jim Gray of Foxboro.
Caro of ARC said, “I look for Ethernet to be the great unifying aspect for higher-level fieldbus.”
A point of disagreement is just how far networking will go. Caro predicts that Ethernet could reach all the way down to field devices within the next few years.
Tom Wallace questions whether Ethernet will ever run down to the cell or sensor. Although Ethernet is an inexpensive solution for commercial information technology, beefing up Ethernet to operate in an industrial environment could erode its cost advantage, he said. “Ethernet is a tremendously valuable thing, but it probably won’t make it out to the sensor, because it’s not designed to go down that far,” he said.
Wireless communication is also likely to play a potential role in fieldbus systems. Richard Caro, for one, expects to see wireless communications in the field for certain applications.
Wallace sees potential for wireless in data acquisition. He added that issues still needed to be worked out: Electrical storms can interfere with communications and wires must be strung to power the devices.