Wireless: Simple, Safe, Secure, Successful

Everybody's doing it. You can too. First, try to ignore all the billowing hype, sunshine pumping, prophecies of doom, and other nebulous and/or extreme baloney presently latched onto wireless. Forget about all the cell phones, laptops, Blackberries, and other convenience-based settings where wireless is traditionally used.

By Jim Montague May 1, 2005
  • Real-world wireless solutions

  • Monitoring before control

  • Savings just the first benefit

  • Standards seeking order

Security for wireless

Everybody’s doing it. You can too. First, try to ignore all the billowing hype, sunshine pumping, prophecies of doom, and other nebulous and/or extreme baloney presently latched onto wireless. Forget about all the cell phones, laptops, Blackberries, and other convenience-based settings where wireless is traditionally used.

Now, try to think of wireless devices that are basically a pure substitute for hardwiring and connectors. They do exist, and they’re multiplying quickly. However, as with any control and automation project, the primary trick is to find the most appropriate solution. This is especially true for wireless, which relies on users to conduct a thorough site survey, and identify individual needs and obstacles in their application.

“When you’re dealing with any potentially hazardous product in a process control application, you must know that you have control. Many things can happen, but if a wire breaks, then the operation shuts down,” says Ed Ladd, HART Communication Foundation’s (HCF) technology programs director. “One concern for users considering wireless is that it can theoretically be hijacked or intercepted. A wire can’t be hijacked because of its physical connection.”

To change some of the perceptions, people must see wireless devices and networks in action. “We do demos in labs and real plants, and show how a wireless network can handle temperature and pressure readings. When people see the signal strength information, and see the data coming through, then it starts to become real for them,” says Cliff Lewis, Accutech’s sales and marketing VP. “People had the same trouble conceptualizing electricity about 100 years ago.”

Simple spread-spectrum

Randy Klassen, Omnex Control Systems’ sales and marketing VP, adds that, “At first glance, implementing industrial wireless solutions may look complicated, but in their simplest form, many industrial wireless solutions are nothing more than wireless links that replace copper wires carrying discrete and analog signals. Wireless isn’t always about moving the network through the air, but it’s often about moving bits and bytes that contain monitoring and control information for industrial processes, such as level, pressure, temperature, flow, on/off, and alarms.”

All types of wireless networking are expected to grow in the next 12 months.

Omnex adds that its frequency-hopping, spread-spectrum (FHSS) radios are integrated into the wireless boom controls of Schwing America’s concrete pumping trucks. These booms contain up to four sections, and can reach up to 198 feet. However, they must be precisely controlled. Omnex’s Len Dueckman reports its wireless solution lets operators get far closer to where the concrete is being placed, which greatly increases efficiency, and eliminates the operators’ direct connection to the truck, thereby increasing safety.

FHSS rapidly changes frequencies to overcome interference and multi-path problems that may exist on one frequency, but not affect other frequencies nearby. This method allows Omnex’s radios, which also are distributed through Phoenix Contact, to serve in harsh settings, and reliably deliver process signals that typically require update rates greater than 50 ms. “FHSS operates without any interference evident to the user, and doesn’t require any frequency coordination,” adds Dueckman. “This allows the radios to function as simply and reliably as wired connection, but without the physical connection problems.”

Monitoring made easy

Logically, the most momentum in wireless involves monitoring, though some actual control and actuation via wireless is happening, usually as a supplement to wired systems. This is because monitoring involves less critical functions, so interruptions don’t have the potential to be disastrous.

Mesh network wireless uses an arrangement of transmitter/receiver nodes, each connected to its own sensor, which all are capable of passing along signals they’ve received. Signals keep hopping until they reach their intended destination. B&B Electronics reports that its IEEE 802.15.4 network nodes have a range of 300 feet indoors and 750 feet outdoors.

“Many people already have some WiFi network infrastructure installed, and so putting a wireless node on a machine is no big deal, and costs are far less,” says Mike Zachan, DPAC Technologies’ Airborne product group VP. “Later, this initial level of acceptance can help fuel future acceptance for using wireless in more critical applications.

“In some control applications, a deterministic network is a requirement. The nature of 802.11 is that it is not by definition a deterministic technology. In our IEEE 802.11 space, we’re not as deterministic as some proprietary, point-to-point wireless technologies, but we are more interoperable and inexpensive.”

In fact, DPAC reports that its Airborne wireless LAN (WLAN) node, basically a rugged, embedded, 802.11b module, forms the heart of a low-cost, wireless pulse oximeter that DPAC and the California Institutes of for Science and Innovation (CAL-(IT)2) are developing to help emergency medical response teams monitor and assess the conditions of numerous patients simultaneously, especially if there too many to immediately transport to emergency rooms. These wireless oximeters are attached to patients with a standard finger clip, measure blood oxygen levels, and then transmit the data to a central server via mobile 802.11b access point and the Internet. This allows doctors to remotely monitor the patients, and direct paramedics at the scene. Engineering students and faculty at the University of California-SanDiego used DPAC’s Airborne evaluation kit to develop the wireless oximeter.

Back on the plant-floor, Accutech, a division of Adaptive Instruments, reports that a major brewing facility recently used its wireless modules to improve clean-in-place (CIP) monitoring in its large-scale, multi-batch operation which runs 24/7. The brewer reported that its former indirect CIP monitoring provided incomplete diagnosis of pH, pressure, and flow in the system’s multi-head, rotating spray nozzles.

If these nozzles fail to rotate properly, tanks aren’t cleaned thoroughly, and the next batch will be contaminated, which can’t be detected until later in the process when volume has increased. CIP failure costs include lost product and disposal costs because contaminated product must be treated as hazardous material and removed using multiple tank cars.

Consequently, acoustic monitoring using Accutech’s modules was validated to detect spray nozzle rotation. Installation costs including wiring of acoustic sensors is about $6,000 per tank. Installed cost of Accutech wireless is less than $1,000 per tank. The brewer reports that its CIP monitoring is now 100% effective.

Savings and beyond

Sure, the main promise of wireless is that it can deliver big savings in cable and labor costs. However, advocates add that the advantages of wireless extend far beyond these initial benefits. They say wireless allows users to install networking and monitoring capabilities in places where they simply couldn’t fit before, and perform measurement that would have been prohibitively expensive with hardwiring. Wireless also allows quicker reconfiguration of plant-floor networks as applications change.

For instance, Elpro reports that ExxonMobil Chemicals’ facility in Baton Rouge, LA, recently had a chance to save $100,000 per month if it could find a way to control a steam valve at its power plant, run by Entergy Inc., and located 1.5 miles away. However, the problem facing Exxon’s engineers was that hardwiring would have been very difficult because of the chemical plant’s existing infrastructure, and that cables would have had to cross several property lines, according to Clint Rabalais and Max Hohenberger, both of ExxonMobile Chemicals.

To implement its wireless solution, Exxon’s engineers used a pair of Elpro’s 905U-1 spread-spectrum radios. These modules use a reliable transmission protocol designed for secure communications, even with external interference, according to Elpro. In addition, because the 905Us have transceivers, they can communicate with each other to manage data flow.

The engineers add that Exxon appreciated that more than one system can operate in radio range or on the same multi-drop wire without cross-talk or malfunctions. As a result, this solution allows a system to consist of a simple, two-unit network with input signals at one module appearing as outputs at the other. They add that Elpro’s modules have been installed and running for about one year, and that they’re pleased with the performance, reliability, flexibility, and upgradeability, as well as the savings they gained by enabling control of their power plant’s steam valve.

Standards a poppin’

To help keep up with all the development activity and solutions, and help users find the best solution for their unique application, several standards organizations are drafting recommendations or standards.

Though wireless might spark the same fights previously waged over several fieldbuses protocols, some observers believe Ethernet and other technical innovations may settle most arguments, possibly before they even begin. This is because the Institute of Electrical and Electronic Engineers (IEEE) has been developing wireless-related standards for years, and has drafted most of the significant rules for using wireless, such as its IEEE 802.11 wireless Ethernet local area network (WLAN) standards and 802.15.1 Bluetooth and 802.15.4 ZigBee standards. ZigBee uses direct-sequence, spread-spectrum (DSSS), which may not reduce electrical noise well as FHSS.

IEEE 802.11’s committees and working groups are presently developing 802.11i for increased wireless security, using 256-bit encryption. They’re also working on 802.11n to allow multiple input/multiple output (MIMO) to enable speedier phase-shifted signals.

Also, the Bluetooth Special Interest Group is preparing to release Bluetooth 2, which increases its power from less than 100 mW to 1 W, and thereby its range from 30 feet to 100 meters, making it useful in many more applications.

Though other standards are still being developed, sources say WiFi and ZigBee will be the protocols on which these future standards and recommendations will be based.

“The key issues in wireless are handling many devices, spanning noisy environments, and sending data reliably to a server, and ZigBee handles these really well,” says Venkat Bahl, Ember Corp.’s marketing VP. He adds that Ember is presently adding security levels, hardware, and software support to its ZigBee-based mesh networking solutions. “Wireless also takes Ethernet to places it could never go before. It gets a lot of data back to an aggregation point, where Ethernet can then send it on up to higher levels.”

To help further define procedures for implementing wireless in field-level control and automation, ISA, the instrumentation, automation, and systems society, recently formed its ISA-SP 100 committee in its standards department. This body will create standards, recommended practices, and/or technical reports, and plans to make its first major release in mid-2005.

Formed earlier by several companies in response to a U.S. Department of Defense (DoD) grant and wireless directive, the Wireless Industrial Network Alliance (WINA) educates users on wireless, and is drafting a user’s guide that it plans to release later in 2005, possibly in conjunction with ISA-SP 100. WINA also is expected to recommend that users adopt ZigBee.

Likewise, HCF reports that its newly launched Wireless HART standard project will ensure seamless, interoperable solutions for connecting HART devices in wireless environments. The foundation’s Wireless HART Working Group plans to produce fast-track draft specifications in late 2005 or early 2006. The group will coordinate activities with other industry wireless organizations, such as the ISA-SP 100, to establish continuity and uniformity with other wireless standardization efforts.

“The working group is tasked with defining an alternate to HART’s existing frequency shift key modem (FSK) physical layer using a wireless modem physical layer. The working group is evaluating all of the wireless protocol technologies available in the industry today (Wi-Fi, Bluetooth, ZigBee, etc.). The key for HART is to balance reliability, range, and speed within the normal power consumption of a two-wire HART, device,” adds Ladd.

In addition, developers of the IEEE P1451.5 sensors and transducer standard are seeking ways to integrate their transducer electronic data sheet (TEDS) technology with the evolving wireless methods.

Also, ProSoft Technology reports it’s even started adding OPC tags to its wireless devices, and set up an OPC server for the radios. These tags and server indicate the wireless network’s signal strength, and can generate a notification via existing HMI software if that strength falls below a certain level, according to Kevin Zamzow, of ProSoft’s global wireless technology sales department. “If you’ve got a link indicator and a signal strength indicator, then you know the data pipe is working. We’re just helping to visualize them in HMI software by using OPC.”

Ultimately, wireless networks will likely develop into a two-tier methodology, in which sensors with ZigBee chips and small, long-life batteries will radiate signals into the air, and then a secondary network will relay these signals wherever they’re needed, according to Graham Moss, Elpro’s president. This secondary network will overcome some of ZigBee’s 50-100-ft industrial distance limits. “Elpro is in that secondary data tier, but we see companies, such as Emerson Process Management, Foxboro, and Endress+Hauser, as occupying the first tier that will embed ZigBee in their sensors and devices,” says Moss. This means ZigBee is going to have to prove itself, but my opinion is that it will.”

Security for wireless

Mike Zachan, DPAC Technologies’ Airborne product group VP, says most wireless security problems can be resolved with some combination of:

Encryption. This typically involves enabling the 128- or 194-bit capabilities already in most wireless equipment, and establishing a process for managing the keys that encrypt or decrypt data running over a network. These keys feed into the algorithms that allow data to be encrypted, and using newly available “dynamic change” keys, preventing unauthorized data sniffing.

Data authentication. Just as passwords verify user identities, authentication makes sure that a wireless access point is truly the authorized point that it claims to be. This is accomplished by passing back and forth keys and other pre-programmed information known only by the client device and its host. Methods for doing this include WiFi Protected Access (WPA) and Cisco’s Lightweight Extensible Authentication Protocol (LEAP).