Industrial Wireless Implementation Guide
So you've dabbled in wireless or thought it might save time, effort, or dollars. This primer helps with technology selection; subsequent articles in this supplement and online at www.controleng.com reveal trends and implementations. Industrial wireless implementation isn't quite as easy as one might hear, with dozens of technologies from which to choose, application requirements, and standards and interoperability considerations.
I've heard some describe industrial wireless technologies as if they were describing freedom itself. Let your thoughts fly! What if you were no longer entrenched with traditional requirements? If you weren't tied down, what would you do?
Just as flying isn't as easy as flapping your arms, industrial wireless implementation isn't quite as easy as one might hear, with dozens of radio frequency (RF) technologies from which to choose, various strengths and weaknesses depending on application needs (including speed, power, and security requirements), in addition to standards and interoperability considerations. Is a site assessment needed? Be aware that sites vary widely by application and even by time of day within an application, depending on what's happening in and around where the signals fly.
Despite these challenges, there's a lot to be said for not having to run wire; some suggest that because of continuing advancements in signal integrity, wired industrial communications will be the exception rather than the rule within 10 years.
The top five industries for wireless communications are oil and gas, primary metals, electric power, food and beverage, and water and wastewater, while the top five applications are tank level monitoring, overhead crane controls, temperature monitoring, flowmeter monitoring, and conveyor monitoring.
That's according to Ian McPherson, president of Wireless Industrial Networking Alliance (WINA, part of the Automation Federation), and vice president of network architecture for Apprion Inc. McPherson adds “Industrial wireless will be everywhere in your plants, pervasive like HMI-SCADA.” (That's human-machine interface and supervisory control and data acquisition software.)
Wireless is a game-changing technology, one that will allow application of more automation to processes because it changes the technological and financial balance, says McPherson. Here's how it will go. The first applications are difficult to reach and costly to implement solutions where there's a 90% cost advantage versus wired.
Next will come traditional safety and security applications, stand-alone or stranded loop controllers or PLCs in remote locations, followed by nontraditional applications in safety and security like mobile asset tracking and human-down systems. Finally wireless will go into noncritical and critical control applications, with continued increases in cost advantages, security, and reliability.
“Wireless is about tradeoffs because power consumption comes at the cost of latency and throughput,” McPherson says. “You have to look at the cost versus application; there is no free lunch.” To help understand this, see the two-part graphic matching RF technologies to applications.
If you were hoping that standards would create interoperability, McPherson suggests looking at the fieldbus wars to see the likely future of wireless. “OEMs are developing their wireless systems with a focus on performance and specialization, which comes at the expense of interoperability.” After all, allowing interoperability so a competitor can easily replace a company's installed base isn't among priorities for major automation vendors.
Even so, wireless benefits far outweigh drawbacks. Market pull will win, McPherson says, citing Wayne Manges, Oak Ridge National Laboratory (and ISA100 co-chair), who contends that within 10 years even critical control communications will be wireless.
McPherson's advice for moving ahead includes continued learning about industrial wireless, developing a wireless roadmap to solve problems, and calculating value created by projects, investment required, and return on investment.
See through the wireless mesh
“Choose the right technology for your need, or you're going to be very unhappy,” warns Joel Young, chief technology officer, senior vice president of research and development, Digi International. He says be wary of sweet-talking pitches from wireless companies saying a specific technology can meet all needs. It doesn't work that way. Network topologies describe the interactions and interconnections among network participants. Wireless topologies aren't always what they seem to be. Wires follow a path, so are relatively simple, Young explains, whereas wireless networks share the air space medium so they have to co-exist.
Topology types include ring (which pass the ring around), bus (peer-to-peer makes rules for who speaks), and star (routing through central point).
Mesh interconnections, partial or full, allow redundant paths via other nodes. Beyond such general explanations, Young says, it gets complex quickly. Some or all nodes may be routers. Some or all nodes may be end points. A star network is a mesh when only one node is the router. And then there's point-to-multipoint, peer-to-peer, and cluster-tree mesh.
Wireless challenges, according to Young, include:
Medium access. Listening is more important than talking. If everyone talks at once, listening is difficult.
Route discovery. Plan the trip in advance or take one step at a time. Adapt to changes, such as new and disappearing nodes.
Changing signal conditions, such as sleeping and waking for power conservation.
Points of comparison for industrial wireless communications include:
Security (as much about the perception of threats as an actual threat). Security includes encryption, authentication (are who you say you are?), and authorization.
Reliability and robustness. Will the message get to where needed on time? Components are frequency agility, message loss potential, adaptability, and single points of failure.
Power management. The number one question is how long will batteries last. The answer depends on network architecture: end nodes, router nodes, coordinator, wake-up methods (cyclic or on demand), listening time versus talking time (listening is more efficient).
Scalability. How big can the network get before failure? There are physical versus practical limits (which are usually smaller). Size also depends on the application's data needs: Dribble data, bursty data, or streaming data.
Data movement variables, such as data rate, latency, packet size, fragmentation, range, and determinism.
Cost, measured by individual unit cost and cost to maintain. Maintenance often is difficult to quantify; deployment cost often is forgotten. A simpler approach is cost per unit, relative to device value and data value.
Enablers and obstacles
Since 1900, there's been more than a 1 trillion times improvement in wireless spectral efficiency, says Steve Toteda, vice president of marketing, Dust Networks Inc. When you can cut the power and communication cords, wireless becomes a key enabler, he adds. As such, wireless will facilitate process optimization, lower costs, increase energy and materials efficiency, improve safety and compliance, and lower system and infrastructure costs.
When exploring the potential of wireless, it's also key to be aware of RF signal obstacles, which could hinder the application. Such obstacles include fixed and moving physical objects, motors, switching transformers, microwaves, and people, Toteda says.
With industrial wireless trends moving as quickly as they are, use the resources in this supplement to get needed information about wireless technologies and applications and don't stand in the way of wireless benefits.
McPherson, Toteda, and Young were presenters at the 2008 Sensors Expo in Rosemont, IL.
More than 30 resources follow to various online articles, new products, podcasts and webcasts on industrial wireless technologies and applications.
Thinking of applying industrial wireless technologies? Application advice from others can teach a thing or two, so your implementation will go more smoothly. Six industrial or rugged wireless applications follow involving Emerson Process Management, Fisher BioServices, Dust Networks, General Electric, Hexion Specialty Chemicals, Numerex, Moxa, Pepperl+Fuchs, Phoenix Contact, and WirelessHART. Each has a summary box of application attributes or benefits.
Wireless Communications for Industry
And... in this issue: Demystifying ZigBee and its Application
Other wireless articles and resources
- 2007 Wireless Communications for Industry (under supplements, lower right, five more articles, covering: How to o ptimize wireless infrastructure for application needs, Machine builders: Driven by consumer standards, 3 Approaches to Process Plant Wireless, WiFi on the plant floor, and Wireless Keeps Machines Coordinated.)