Fieldbus / Industrial Ethernet

The Internet of Things Paradise May Be Farther Away than It Appears

All these sensors and machines may be able to talk in theory, but look to your home network for the reality of this multi-protocol world

By John Rezabek

Attending a recent conference, I had a chat with Gary Cusick, vice president and marketing manager for MACTek Corp., a manufacturer of appliances for HART and WirelessHART devices and networks. Gary has installed a demo site at his residence, where widely deployed HART devices relay measurements and diagnostics from far-flung pool houses and woodsheds. While Gary’s intent is to demonstrate the ease of deployment and reliability of MACTek devices, his network made me think about the promise of the Internet of Things (IoT), which is presumed to be dawning upon us any minute now.

Most of us technology types have some sort of network installed in our residence, most likely an Ethernet network with both wired devices and a wireless 802.11 b/g/n capability. The routers that typically serve as the hub for these networks offer features for interacting with your Internet service provider (ISP), security and a service called Dynamic Host Configuration Protocol (DHCP), which arranges for the devices in your household (and possibly your neighbor’s) to connect to the network. If you’ve never looked at it, you might find it interesting to browse to your router’s “DHCP Clients List.” When I looked at mine this morning, I found a whole menagerie of clients: an assortment of wired and wireless Windows PCs, a satellite TV box, a DVD player, a Wii, a Roku, an iPod, an iPhone, a few Android devices . . . why, perhaps the Internet of Things is just over the horizon at my house.

See Also: Fieldbus Protocol Foundations Merge

So Santa brought me a new “Phablet” (a giant Android phone) which boasts an app for controlling my PC. This is supposed to work over WiFi and requires that a service is installed and running on its Windows victim. But alas, the PC service asked if I wanted to connect over USB or WiFi, and wouldn’t recognize my PC’s hardwired connection to the LAN. Doggonit, I guess the magical coolness depicted in the commercials is not in the cards for me.

Having a path to useful data does not equal getting any value from it.

Do you suppose such disappointing experiences with a simple home network portend anything about what we’ll face when the new day of the industrial IoT (IIoT) is upon us? In the consumer marketplace, we have disparate and non-interoperable protocols, with their tech giant proponents—Microsoft, Apple and Google—vying to be sole provider of our networks, media, interests, contacts, wallets and so on. But we consumers like freedom to choose. The same is true at work. If I want the best-in-class devices for motor controls, wired field device networks and safety systems, it’s entirely possible I’ll be deploying three different protocols that don’t interoperate. Even before every little widget in the plant connects somehow and starts jabbering on our industrial networks, we’ve struggled with integration, reporting and utilization of useful data from what we already have.

Some pioneering end users have developed custom middleware that attempts to use services such as OPC (formerly known as OLE [object linking and embedding] for process control). If you’re an end user configuring OPC, I’d be interested to hear if you noticed any kinship to OLE (when you plop an active spreadsheet into a Word document, for example), or if you’re using it for anything we’d agree is process control. Most of my experience involves a tedious point-by-point browsing through complex File Manager-like data structures that may or may not lead to the parameter of interest. Sometimes after finding a point or two, comma-delimited or Excel spreadsheets can be used for more expedient configuration, but every use case is different.

There’s seldom an easy path to filter what you need: Do I care about the temperature of a positioner? Not me. An ultrasonic flowmeter? Nope. How about a DP flowmeter case temperature? Why, yes. Are parameter locations for data quality for a WirelessHART device similar to those for a Foundation fieldbus device? Not a chance. How do those servers perform when your customized middleware is hammering them for all this device data? Mine—not so great. Having a path to useful data does not equal getting any value from it.

Unlike the mayhem of the industrial IoT, Gary Cusick’s manageable numbers and homogenous protocol make it easy to decide if the status if the water heater in the pool house is critical or not. Maybe the heightening chaos in our networks will finally be enough to solidify users’ resolve for a single, interoperable field device protocol.