By Phil Burgert
The extra cost of industrial-grade network components sometimes can be a problem for designers of manufacturing networks. They might be encouraged to learn that a surprising number of vendors will endorse some limited use of less-expensive, commercially available products.
Commercial vs. industrial component decisions boil down to two factors, says Tony Oberkirch, market analyst for N-Tron. Those, he says, are meeting environmental challenges and ensuring maximal uptime.
“Commercial devices typically are located in wiring closet-type areas,” he notes, adding that the temperature is usually controlled, they aren’t located in the proximity of devices that can generate radio frequency fields or electrostatic discharge, and they aren’t subject to shock and vibration. Industrial devices are subject to all those environmental challenges. “Many IT professionals would argue that their networks are as mission-critical as the industrial network, but an Ethernet switch that fails in an IT office environment usually doesn’t result in a truckload of scrap being produced instead of top quality goods,” says Oberkirch.
“Industrial-quality components are known to operate at higher temperature ranges and, as such, can be expected to have a much longer life,” says Lee House, vice president of engineering and chief technology officer at GarrettCom.
Industrial-grade components often need to operate in environments where fans, the traditional cooling method for office-grade products, aren’t a viable alternative due to dust, moisture or corrosive conditions, notes House. In those cases, fanless designs are required, and they rely more heavily on extended temperature ranges and heat-dissipation properties of industrial-grade components and designs, he says.
Christian Vitale, senior product manager for Turck, says a consideration for commercial products is moisture. “Commercial products generally can’t be washed down,” he states. Those products are not going to survive.”
Bill McGovern, national sales manager for Dataforth, recommends that products used in factories carry agency certifications. “The toughest ones are from Europe—CE mark requirements mean that products have to pass susceptibility tests. They’re not affected by radio frequencies, transients, surges in the power line or voltage dips in the power line,” he notes.
Reliability is the biggest advantage of industrial quality components, adds McGovern. With an industrial product you often have a whole library of certifications, not only from agencies but also from lifecycle experience, what the history of the product has been and how reliable is it.”
Customers sometimes call for emergency overnight delivery of products after a commercial grade switch failed and brought down a major production machine, adds Scott Killian, Ethernet switch division manager for Sixnet. “For the cost savings of an unmanaged switch, which might have been $100, they risked an entire line that was their primary revenue-generating process,” he says.
But there are applications where commercial and industrial products can be mixed, says Eddie Lee, senior marketing manager for Moxa Americas. Lee says realistic risk assessment of factors that could hinder production is important when considering combining commercial and industrial quality components in single networks. But he notes that enterprise integrated networks that connect factory floor devices to the enterprise level and share data from the shop floor to the top floor have areas that combine both types.
A system is only as reliable as its least-reliable component, says House. But where the device can be installed in a controlled environment, less expensive office-grade systems could be preferred, assuming a good maintenance or service plan is in place, he says. Even in these conditions, though, for mission-critical functions in industrial applications, high availability and fail-over capability are important, which frequently also argues against the use of typical office-grade hardware.
Mike Hannah, business manager for networks with Rockwell Automation, says his company pushes the use of standard, unmodified Ethernet to permit use of commercial products along with industrial automation products. “When you use our network, you don’t have to buy a network switch from us,” he says.
“There are times when it makes sense to use non-industrial products, especially when putting a whole factory infrastructure together. At the enterprise level, you might use some non-industrial products because they’re located in an office environment, but not down on the machine level.”
Industrial and commercial-quality components can perform similar tasks, he says. “It’s pretty much all about the environment that they are going into,” adds Hannah, and he notes that Rockwell Automation works with Cisco Systems on some reference architectures for industrial networks and co-developed an industrialized high-end Ethernet switch.
Phil Burgert is a freelance writer, specializing in the technical trade media.