Managing Automation - May 2008 - (Page 39)

wireless networking standard focused specifically on automation applications, not mere information collection. The nascent study group is currently seeking input from industry — users and vendors, alike — to help determine what is needed. Discrete manufacturers say they understand why the original draft .11a standard — likely to be voted on in November — focused on process industries. Under ISA100, there will be myriad application-specific variations on the standard when all is said and done, experts say. And the standards committee had to start somewhere. Not only was the process industry lobby strong, but also hammering out a standard that met process manufacturing needs was easier than a discrete manufacturing standard will be. One of the primary differences is the speed at which process and discrete manufacturers need information transmitted from remote sensors to host systems over a wireless network. The .11a standard for process networks determined that the speed had to be suitable for closed loop control and no faster than once every second, which is fairly slow, particularly for discrete manufacturers, says Dick Caro, president of CMC Associates and co-chair for the ISA100 user working group. Proponents of a discrete standard recognize the technical challenges involved, but they also are well-aware of the chaos that arises when multiple standards emerge — or worse, no standards. In fact, despite the ISA’s work on 100.11a in process manufacturing, dueling standards have developed. Last September, prior to the release of the 100.11a standard, the HART Communication Foundation released a competing standard, WirelessHART. Now the IS and the HART group are trying to reconcile their competing standards. “We thought we were starting far enough ahead of time, but we didn’t,” Caro says of the ISA100 being upstaged by the Wireless HART standard last fall. The problem for manufacturers is that wireless equipment based on different standards won’t talk to each other very well. For large manufacturers, “this is a nightmare,” O’Hearne says. Radios and other devices have to work together using the same frequencies. If they don’t, there’s great potential for interference and security breaches. As the standards work moves along — some say glacially — process manufacturers, at least, aren’t waiting around. “Users are proceeding with whatever they can get their hands on right now,” Caro acknowledges. “Users can justify it by the increased safety, improved process control, and the magic “When you’re buying products from of acquiring data they were never vendors offering five different flavors, able to get before. But they’re you know it costs you more.” adopting the technology in a vacuum.” Eventually, a retrofit kit — Millennial Net’s O’Hearne will be needed to make the sysIn process automation, wireless sensors are tems comply with the standards. used to monitor things that have been “just too ARC Advisory Group analyst Harry Forbes expensive or downright inconvenient to measure” terms wireless sensors a “small business right — for example, a safety relief valve at the top of a now,” with more buying coming from process manstorage tank or the flow of water in safety showers, ufacturers. He estimates that the market is in the Caro says. Sensors might send analog data on “low-tens of millions,” but adds that in about five temperature, pressure, and flow. So the .11a stanyears, 10% to 20% of all process measurements will dard is based on relatively slow data rates. be done wirelessly, stimulating a much bigger However, discrete manufacturing environments market. Discrete manufacturers, however, are waitrequire a greater number and variety of sensors, ing for viable products to come along, he adds. transmitting data at faster rates. And that adds to P&G, which runs both process and discrete the complexity and expense of implementing wiremanufacturing plants, is evaluating wireless equipless networks in discrete environments. ment in a test lab in its corporate engineering ofAn example of how wireless technology might fices, Reizner says. Its wireless deployments, thus be useful in discrete manufacturing is in the control far, have been primarily for human-machine interof a robot arm. A sensor on that arm might send bifaces, data acquisition, and logging. nary data about its operations across the network Eventually, one of the biggest boons from wireat 10-millisecond intervals. And if the hand needs to less technology at P&G may be in easing the inbe changed, there’s less likelihood of error if you stallation of manufacturing equipment that’s built in don’t have to worry about reconnecting sensor one part of the world and then shipped to other wires, O’Hearne explains. parts. “We build local manufacturing facilities and And because sensor density in discrete envitry to standardize those facilities and how they’re ronments is often far greater than in process plants, built. But people in far-away places don’t undera standard would have to incorporate ways to destand wiring methods and requirements in other tect and avoid device interference. parts of the world they’re delivering to. It’s very Photo courtesy: Millennial Net 39 May 2008

Table of Contents Feed for the Digital Edition of Managing Automation - May 2008

Managing Automation - May 2008 Contents Take 1 Award-Winning Shoe-Maker Otabo Alters Course, Shifts Production to China IBM Partners with Universities for Cloud Computing Getting Noise in Production Under Control Incuity Embarks on a Vertical Market Strategy Foundation Intensifies OPC Standard Testing Mailbox Notes PM Roundtable Cover Story: The Business of Going Green Special Report: Night and Day Delivering on Promises Finding the Right Fit for Wireless Driving RFID Product Scan Advertiser Index Next

Managing Automation - May 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.