Implementing Industrial Networks

Year of the Network

Networks and communications

Fieldbus

Device-level networks

Sensor-level networks

Sidebars: How to Select an Industrial Network Fieldbus helps make oil field profitable How to implement a digital industrial network Understanding a Foundation fieldbus Configuration Tough stuff: Transmission networks for industry require hardened cables, connectors

Successful implementation is key to any project, no less for installing an industrial network—a field-, sensor- or device-level bus.

The irony of a network implementation gone bad wouldn’t be lost on participants: communication networks are supposed to increase and improve information flow and maximize available assets, human, and capital.

At implementation, the network user has already gathered information in several key areas.

Choosing to use an industrial network may be the easiest decision. Benefits abound for automation or process control using field-, sensor-, and/or device-level buses (see below).

Selecting an industrial network among more than 50 choices is more difficult—few devices, at this time, interoperate, or work on more than one network. Application complexities along with industry and regional preferences make use of a single network within a plant, much less the world, unlikely for the foreseeable future. Networks with more installed base generally have more available devices at lower cost but selection criteria are many. (See “How to select a Network.”)

Implementing an industrial network—selecting a topology and devices, training, installation, device tagging, getting software up and running, and maintenance—can challenge the most ardent network proponent. Some networks are easier to work with than others; for most users, the more “transparent” the network is, and the more automated the process of configuring or reconfiguring is, the better.

Industrial network benefits

Benefits of industrial networks have been well documented. According to a recent Venture Development Corp. (Natick, Mass.) study, “The U. S. Market for Industrial Automation Products Incorporating Device/Sensor Buses…,” the identified strengths are:

Less wiring and installation labor;

Fast throughput;

Flexibility in ease of expansion;

Good diagnostics;

Simple to use;

Large data capability;

Easy to install and put into operation;

High reliability and robustness;

Strong vendor support; and

Openness.

Weaknesses cited covered some of the same topics, showing all networks and users’ perceptions, aren’t equal. The report provides information on As-I, DeviceNet, Interbus, LonWorks, Profibus DP, SDS, and Seriplex.

Implementations underway

At the field level, as with device and sensor networks, users are implementing communications.

Ferrous Steinka, Relcom (Forest Grove, Ore.) engineer, Fieldbus products, sees “substantial” interest in FOUNDATION fieldbus 31.25 kbit/sec H1 network protocol (based on the ISA SP50/IEC 1158 standarization efforts).

“Sales of our Fieldbus connection blocks have been doubling each month for the last four months,” Mr. Steinka noted in June. Relcom’s “involvement with the Fieldbus Foundation is in testing of Fieldbus wiring, conformance testing the ‘physical layer’ of Fieldbus devices, helping wire trade shows for Fieldbus demonstrations, etc.,” rather than the network protocol or devices.

Relcom’s well-received “Fieldbus Wiring and Installation Guide,” provides practical advice on wiring (diagram), according to the application, showing the differences between a home run and chickenfoot.

Mr. Steinka says implementations have shown the wiring FF specs are conservative. “Potential users of [FOUNDATION ] fieldbus would be interested to know that the signalling on Fieldbus wiring is extremely robust. The distances that can be spanned and the length of spurs off the main cable are several times those specified in the Fieldbus standard.” (see also ARCO implementation note.)

As users find greater needs for connecting and maintaining the people, assets, and information that comprise their businesses, industrial networks will continue to gain market acceptance.

Sampling of Industrial Networks

Network
Introduced
Topology
Data transfer size
Speed
Original technology developer*
Free information

* Most industrial networks are now governed/guided through groups or consortiums.Implementation of industrial communications technologies vary widely by transport mechanisms, physical characteristics, performance, applicable standards, “openness,” and other parameters. This information draws from a March 1998, 18-column comparison from Synergetic Micro Systems, Downers, Grove, Ill.Source: Synergetic Micro Systems

As-I
Fall ’93
Bus, ring, tree star, or all
31 slaves w/4 in, 4 out
167 kbps
Consortium
252

CANopen
’92
Trunkline or dropline
8-byte, variable message
1 Mbps, 500 kbps, 250 kbps, 125 kbps
Philips/CiA
253

ControlNet
’97
Trunkline, tree, star
510 bytes
5 Mbps
Allen-Bradley
254

DeviceNet
3/94
Trunkline or dropline with branching
8-byte variable messaging
500 kbps, 250 kbps, 125 kbps
Allen-Bradley
255

Fieldbus Foundation
’95
Multidrop with bus-powered devices
16.6 M objects/device
31.25 kbps, 1 Mbps, 2.5 Mbps
Consortium
256

IEC/ISA SP50 Fieldbus
’92-96
Star or bus
64 octets high and 256 low priority
31.25 kbps IS+1, 2.6, 5 Mbps
Committee/consortium
257

Interbus
’84
Segmented with “T” drops
512 bytes h.s., unlimited block
500 kBits/s, full duplex
Consortium
258

LonWorks
3/91
Bus, ring, loop, star
228 bytes
1.25Mbps full duplex
Echelon Corp.
259

Profibus
DP-’94; PA-’95
Line, star, ring
244 bytes
DP up to 12 Mbps; PA 31.25 kbps
Consortium Siemens
260

SDS
1/94
Trunkline or dropline
8-byte variable messaging
1Mbps, 500 kbps, 250 kbps, 125 kbps
Honeywell
261

How to Select an Industrial Network

This article in the Control Engineering Year of the Network series, jumps in after the process of selecting a network. Choosing among 50 or more industrial networks can be a challenge. Industry and application-specific loyalties emerge before, during, and after every industrial networks network discussion.

For more information to help in selecting an industrial network, go to

Search on industrial networks using the “Search Past Issues” function on the left bar.

Use the “Discussions” function on the left bar.

Use the “New products” function on the left bar.

Search the deeper database of Manufacturing Marketplace information by using the “Search” function on the top bar.

Search advertised products at

Use the circle numbers in this article.

Go to

To use a system integrator to implement an industrial network, see Control Engineering’s June 1998 two-part cover story on finding and working with system integrators.

Source: Control Engineering

Fieldbus helps make oil field profitable

Technology often enhances profitability, but ARCO (Atlantic Richfield Co., Los Angeles, Calif.) is using digital industrial network technology to make profits flow from frozen assets.

ARCO Alaska Inc.’s West Sak Phase 1 North Slope project—near Harrison Bay (see map) on the state’s Arctic coast—occasionally sees a 70 8F summer day, but has brutal winters with daylong darkness.

Vehicles, to operate 100 miles north of the Arctic Circle, run all winter; December temperatures, prior to project startup, were as low as–45 8F; windchills fell into the –70s.

A warm–20 °F

“On warm days, it got up to–20 °F,” says Duane Toavs, staff engineer, technical coordinator, ARCO Alaska Kuparuk site. “We needed reliability and a design suitable for not-normally manned operations that would allow us to identify and deal with problems. And we needed to lower development costs about 30%, compared toprevious applications,” says Mr. Toavs, an 18-year ARCO veteran.

West Sak is a large, relatively shallow viscous oil accumulation overlying much of the large ARCO-operated Kuparuk field, allowing use of nearby existing lines and processing centers. Even though oil in place is estimated at more than 15 billion barrels, development cost per well had to be reduced to make a go of it, he explains.

ARCO evaluated five field network technologies; Fieldbus Foundation’s (FF, Austin, Tex.) 31.25 mbit/sec H1 was one among the choices that could work without active components in this field application.

Mr. Toavs also liked FF for its remote monitoring, diagnostics, and hardened field-based components. Digital technology reduces wiring and distributes control to the wellhead, using just one cable for every four wells. Lab testing verified that 9 m spur length would work reliably, far beyond conservative recommendations in the specification.

Results include 70% less control room space; 84% fewer terminations; 98% less cabling; and 90% reduction in system configurations. In addition to cutting costs, “These kinds of reductions allowed project startup within eight months, down from the usual 18 months,” Mr. Toavs says, despite the harsh conditions.

“The wind blows so hard, that within a day, snow completely packed the inside of a closed NEMA 4 junction box that didn’t have screws tightened down all the way. Using another system, we would have had PLC-like components in there; with FF, a Relcom fieldbus spur block and Phoenix Contact blocks operated without failure.”

Other FF-compatible equipment used at the project includes Emerson Electric El-O-Matic ELQ electric valve actuators, Rosemount Model 3051 pressure transmitters, Fisher-Rosemount Systems DeltaV process automation system, Micro Motion Coriolis flowmeters, and Fisher Controls FieldVue valve controllers.

“This departure from traditional technologies helped make this a profitable project. Learning that less equipment is best is a real departure from the way we’ve always done things—it’s not always easy to change 20 years of habit,” Mr. Toavs admits.

Project growth

The application will grow. Phase 1 started at 200 barrels of oil per day (b/d) on Dec. 26, 1997, with the first oil from a single producer; 50 wells (both production and injection) are expected by early 1999, yielding 7,000 b/d gross. Phase 2 would add 500 more wells, producing up to 70,000 b/d by 2007. West Sak will help offset expected declines in North Slope production, prompting ARCO’s slogan: “No decline after ’99.”

For West Sak, lower development and operating costs were critical to making the project profitable.

ARCO Alaska is the second largest oil producer in the state. It operates the eastern half of the Prudhoe Bay field on Alaska’s North Slope, the nation’s largest oil field, and the state’s second and third largest oil fields—Kuparuk River and Point McIntyre Area. Ultimate oil recovery from these fields is expected to exceed 16 billion barrels.

How to implement a digital industrial network

Key considerations when implementing an industrial network include network “behavior,” configuration, and support. Reprentatives associated with two networks provide insight—Mark Knebush, Interbus Group Director, Phoenix Contact (Harrisburg, Pa.), and Bill Moss, executive director, Open DeviceNet Vendor Association (ODVA, Coral Springs, Fla.).

Q How will the network behave?

ODVA: Interoperability is a benefit touted by most networking protocols. The word implies that the network-compatible products will work together seamlessly. Even so, end-users often spend valuable time interfacing products designed for the same network. End-users want assurance that products will interoperate, but not all networks have testing to assure products conform to specification. To indicate DeviceNet conformance, ODVA recently unveiled a Certification Mark, requiring passing more than 10,000 physical layer, software, and interoperability tests conducted at independent, certified ODVA labs.

Interbus: For a network to behave as advertised a user needs to know how much of the protocol is embodied in a single chip or chipset. Interbus has a simple standard implementation for slave devices and for master devices. You should also know how robust the network is to common installation problems such as cable shorts, cable opens and EMI (electromagnetic interference) noise. The nature of the Interbus topology allows identification of the exact node with a problem. Interbus touts speed and diagnostic capability. Phoenix Contact stands behind the performance of the network.

Q How easy is the network to implement?

Interbus: The network will autoconfigure upon connection of all devices. There are no dip switch settings or special configuration software required. The most common cabling scheme used is DV9.

ODVA: Network configuration software tools (available from several suppliers) help users implement a network in a timely fashion.

Configuring a product for some networks is as easy as assigning an address for each product, plugging the product into tap ports and following the software’s point and click instructions for specific configurations. A recent panel of experts, each with experience in three or more bus networks, rated DeviceNet as the easiest network to configure from a user’s standpoint. The ODVA web site publishes several tutorials on how to install and implement a DeviceNet network.

Q With an open network, who provides support?

ODVA: ODVA has assembled 12 DeviceNet experts to answer technical questions—typically in less than 48 hours—via e-mail to “Dr. DeviceNet” atodva@powerinternet.com. A newsletter and web site also offer support.

Interbus: The Interbus Club recommends contacting the technology originator for technical support after using Interbus’ inherent diagnostic capability to determine if the problem is in the network or device. Phoenix Contact operates a 24-hour Interbus technical support hotline.

Understanding a Foundation fieldbus Configuration

Foundation fieldbus, defined as part of ISA standard 50.02, H1 low-power signaling, is a process control network for interconnecting sensor, actuators, and control devices. While each network has its own attributes and requirements, many of the terms and concepts are similar.

One of many common configurations

Fieldbus Terms

Attenuation: signal getting smaller as it travels on the cable.

Bit cell: Length of time taken by one bit, 32 microseconds for H1, or 31.25 kbit/sec, Fieldbus.

Chickenfoot or crowsfoot: a common terminal block that connects field devices at a field junction box.

Daisy-chain: Wiring method where a number of devices are attached at spurs along the home-run cable.

End delimiter: Bit sequence used to signal the end of a frame.

Home run: Twisted-pair cable that connects the control room equipment with field devices.

IS barrier: Device used to keep voltages and currents and wires below the levels that can ignite the atmosphere.

Lift-off voltage: Initial voltage for a Fieldbus device to start operating.

Reflection: Unwanted signal resulting from a cable fault or improper termination.

Repeater: A device to boost the signal to another segment if distance exceeds what’s allowed.

Segment: Portion of the network electrically independent from other parts.

Terminator: Device that absorbs the signal at the end of a wire.

Source: Fieldbus Wiring and Installation Guide by Relcom Inc., Forest Grove, Ore.

Tough stuff: Transmission networks for industry require hardened cables, connectors

Networks have become the backbone of building, office, and factory information exchange and control. Silicon and software are driving the bits and bytes closer all the time. Transmission media—specifically the cable and connectors—are becoming significantly different as the applications drive competitors to devise products using engineered materials to handle specific factory-based challenges, such as:

Robotics’ constant high-speed cable flexing;

C-tracks’ constant flexing and abrasion;

Welding’s hot slag;

Machine tools’ cutting fluids;

Processes’ chemically aggressive vapor, condensation, extreme temperatures;

Clean rooms’ low-emission plastics; and

PCB lines’ chemically aggressive washes.

The key to the cable is the outer jacket. It provides the chemical, temperature and abrasion protection to the data carrying inner conductors. When selecting an outer jacket for your cable always prioritize as follows:

Safety: Most industrial cables are rated by their flame, voltage, and temperature rating. Consult your local government regulations to determine the cable requirements.

Reliability: cable reliability is influenced by its exposure to chemicals, extreme temperatures, abrasion, and flexure during operation. The combination of temperature and chemical concentration determines how each jacket material performs. Cable manufacturers can provide charts detailing resistance of materials used for cable jacket material and connector coupling nut. Testing can help ensure performance. Temperature rating of most industrial cables is–40 to 80 8C. This is suitable for most indoor factory and outdoor environments. Higher temperature jackets are being developed for more extreme temperatures. Tough outer jackets including thermoplastic elastomers with the addition of abrasion resistant elasticizers are available for applicationssubject to continuous flexure and contact with machine parts. Outer jackets with elasticizers added are designed to withstand applications of constant flex.

Cost: Cable cost varies with the construction of the outer jacket. Materials considered engineering plastics are typically more expensive. In addition, jackets that have additives to improve flamability ratings, temperature ratings, and cable life are more expensive than standard jacke

The connector has one of the toughest jobs. It must make a positive connection to transmit the data, keep out unfriendly fluids, withstand the environment over the long haul, and hold the cable under mechanical stress. As data rates increase in the future, the connector will also need to shield the conducting pin and sleeves from noise. Today’s industrial networks are operating from 9,600 bps (bits per second) to 12 Mbps (million bits per second) with respective wavelengths of 31.25 km to 25 m. A few centimeters of unshielded transmission media are not significant at network speeds of today.

The material that bonds or attaches the connector to the cable and the coupling nut should be as chemically resistant as the cable jacket to prevent chemical attack. The connector must latch to another as it enters or exits a node, controller, cabinet, or junction. Snap-together connections for industrial applications are now emerging.

Most industrial buses use one of more of the following connectors:

9DB, 9-pin subminiature D shell connector;

Micro (dc) or euro, 2 through 6-pin connector based on M12 threads; (approximate 14 mm outside diameter coupling nut); and

Mini, a 2 through 7-pin connector based on a 7/8 in. dia. barrel.

These three connectors are well defined in industrial standards. An estimate of daily use for some variant of these connectors ranges from 50,000 for mini to 250,000 for the 9DB with the micro dc somewhere in between. This commonality in the industrial market creates a broad selection of materials, competitive prices, and variations in manufacture quality.

A positive electrical connection maintains the low contact resistance required for data transmission. The pin and sleeves are made of copper and should be gold plated. The pin and socket alignment is dependant on the nonconducting carrier to allow either the pin or socket to float. The flexibility of the carrier material allows the pin and/or socket to move and align.

Future materials

As food, beverage, pharmaceutical, and chemical processing use more networks, we will see more variety of cable jacket materials and combination of materials used in connectors. These special materials will allow networks to offer benefits even in the harshest environments where control is most critical. All users can then benefit from the economy, advanced diagnostics and fast start-up that is being realized in many industrial applications today.

For cable installation tips, see “News.”

Bob Svacina and Ann Feitel are part of InterlinkBT LLC, a Plymouth, Minn., supplier of bus components for all major, open industrial networks.

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our WTWH Media editorial team and getting the recognition you and your company deserve. Click here to start this process.