Life on the bleeding edge...

Digital Managing Editor Rick Pedraza and I took a drive in the country this morning to visit B&B Electronics. Regular readers will recall that Gerald Niemi from B&B gave one of the best presentations at the CSIA meeting last month in Memphis. You can still read the blog sections about it...it was about the state of wireless in the industrial environment. I think Niemi knows what he's talking about, so I scheduled a meeting with him and B&B CEO Don Wiencek so we could talk more about wireless, and B&B's involvement in it. Most people in automation know B&B. From a little building in the middle of the cornfields in Ottawa, Ill., they have been one of the most successful niche catalog direct marketers in automation. Certainly not as large as Omega, or as AutomationDirect, they have closely conformed to their mission of becoming automation's serial and data communication experts. One of the things that most people don't realize about direct marketing of automation products is that, unlike typical salesman/customer interactions with reps and distributors, or even company-employed salespeople, direct marketing is incredibly sophisticated in ways not usually seen in the automation industry. As Don Wiencek noted, "We have to very carefully plan everything we do from the very beginning." B&B (as noted in the blog about Niemi's talk, stands for Bill and Bill, the founders) has been making data communications gear and protocol converters since 1981. That makes them a senior citizen in industrial networking. Looking through their catalog is fascinating, because it gives a real clear look at what it takes to do heavy networking in the industrial environment. Some of the products they distribute, but many of them they make themselves in Ottawa. For example, they devote over 10 pages of their catalog to serial converters, isolators, bus converters, and current loop converters. Wiencek said, "We make a whole lot of converters, because they are all needed, every day, to make networking work in industrial automation. Nobody else wants to do this. We make such a wide array of products because we have to." Since I am working on the August cover story, "Tomorrow's Plant," I wanted to hear what Niemi and Wiencek thought. "We think wireless is the next disruptive technology in industrial automation," Niemi said, "So we've decided to move into that market." "It is a real direct extension of our product line," Wiencek noted, "because we are simply adding 'wire replacement' to our communications products." Niemi and Wiencek broke down the market for wireless in automation this way: "First, there will be the early adopters who are looking for a way to monitor a variable where they just can't run wires," Niemi said, "Then there will be the early adopters who want to add to a system, or to monitor machinery health. Finally," he concluded, "there will be the people who see the advantages of wireless communications in everyday applications." The first wave of industrial strength "mesh networking" wireless products is here. They aren't entirely ready for prime time, and they are a little expensive, but they are already doing applications that you can't do any other way. Niemi showed me a set of pictures of a wireless mesh networking application from a "steel company that must not be named." The problem is that they need to measure the temperature of the water jackets on the electric arc furnaces. If the temperature spikes in a specific location on the furnace wall, it is symptomatic of "burn-through" where the wall of the furnace itself melts and hot molten steel is able to penetrate the jacket. This can cause explosive results, as well as instant unplanned downtime. If the furnace does burn through, that downtime can make the difference between a profitable steel mill and a closed steel mill. Wired sensors have never worked properly because of the heat and steel splashover problems. Also, the EMI/RFI problems from an electric arc furnace are, well, intense. B&B partnered with Sensicast to do this application. In fact, B&B manufactures the mesh networking nodes and private labels them for Sensicast. They wrapped the node in heat-resistant insulation, plastic wrap, and put it inside a steel box (!!!), and then located them between the inside and outside walls of the furnace. All of these things would have made a typical wireless system unworkable, but the self-organizing and self-repairing properties of mesh networks made these nodes survive heat from melting steel, wide temperature swings, powerful magnetic fields, water spray, and vibration. And this is not a "dancing bear application" either. You know the story about the dancing bear. It isn't so much that it is dancing gracefully, but that the bear can get up on its hind legs and dance at all. The system, according to "the steel company that must not be named," hereafter referred to as TSCTMNBN, works so well that the system is being enlarged to deliver data directly to the PLCs that control the furnaces. This is an example of the first kind of early adopter. "It remains to get the real feeds and speeds up to snuff to serve the rest of the automation marketplace," Niemi said. "But most of the companies in this market are making the mistake of not doing with what we have, while they wait for the '2-buck node' or the 50-cent RFID chip." "What we have to do," Niemi concluded, "is to make wireless easy enough to deploy that the maintenance mechanics and operators can do it without involving the plant IT people, and without special training."