Wireless / I/O Systems

Advances Make I/O Easier, But Still Not as Simple as It Could Be

But no interoperability between WirelessHART and ISA100 is holding back more universal adoption.

By Ian Verhappen

Continuing with our evolution conversation from last month of how control systems continue to change, the same transformation is certainly happening with the Input/Output (I/O) systems used to connect the field sensors to the associated controllers. Wired systems now offer ‘configurable I/O' such as Schneider Electric's "intelligent marshalling" and Honeywell's "Universal I/O" where the type of signal landed on the terminals is configured in software. Emerson is offering CHARMS a solution where you purchase and install the right type of I/O module in their backplane for each type of signal. If you want to be a bit more adventurous you can get ‘similar' functionality with analogous products from other companies with ‘smart terminal blocks' and a backplane that also serve as data concentrators, However, you then need to use a "third-party tool" different from the one you use to configure the rest of your system and perhaps a gateway to convert protocols. The other big change, of course, has been the addition of wireless field sensor networks.

These new technologies are making it easier to access measurements more effortlessly than ever before. We can not only provide measurements via wire, but now with wireless systems, it is possible to consider signals from locations previously inaccessible because location made them prohibitively expensive to connect with a cable or, more importantly, by removing the cable tether, it is now possible to make devices of a size and form factor to fit in smaller spaces. Going to a smaller form factor may require losing some functionality, however if the cost is low enough the "signal" can be determined from the pattern of a number of individual nodes, much like we sense a smell or taste food.

See Also: Fieldbus Protocol Foundations Merge

Until ISA 100 and WirelessHART find a way to interoperate, adoption of industrial wireless technology will not be as great as it would be otherwise.

Bringing this pattern data to a more central device for processing in a single environment is also now feasible. With the recent adoption of ISA100.15 and HART/IP, it is now possible to have the two major process industry wireless networks connect from sensor to controller without having to change protocols through a gateway. Again, this certainly makes system configuration much easier because it effectively means that the wireless systems are no longer the equivalent of ‘third-party networks.'

Though we no longer have to map data across gateways, unfortunately, we still have the two wireless camps of WirelessHART and ISA100. Until these two groups find a way to interoperate, adoption of industrial wireless technology will not be as great as it would be otherwise. Hopefully this month's official formation of FieldComm Group as single entity from which both HART/IP and ISA100.15 were largely developed is a precursor to at least developing a common backhaul technology so we can connect to the wireless field sensor network from a single common access point. I was also encouraged while at a recent IEC meeting in Japan, held in conjunction with the JEMIMA, exhibition to see a demonstration model of a HART to ISA100.11a access point node, so the standards and technology support the potential for a single interface.

Ideally however, a single interface means the protocols must be from the same family such as HART-WirelessHART-HART/IP, ISA100.11a-ISA100.15, or the CIP protocols (CompoNet, DeviceNet, ControlNet, Ethernet/IP), to name a few protocol suites, or as a minimum, they must "transport" the characteristics of the protocol so that the device appears as if it is in its native protocol when it is configured or accessed for maintenance/operations. Again, the desired outcome is to make it easy for the system and, in the end, the person working on the control system to gather the information required with minimum effort.

More signals, more easily is certainly the trend and with more signals, especially digital signals that normally support associated diagnostic information, mean that the challenge will soon become how to identify what signals are required by whom, how to effectively manage all the data, and most importantly use the resulting information to operate our facilities better.