Wireless networking in Process Automation: Building new applications in 2007

01 February 2007

Control system vendors are driving their stakes in the ground for the biggest new technology thrust in 30 years.

Process automation companies are normally very conservative in their approach to new technology, and they have very good reason for being so careful: they work with some of the most riskadverse customers on the planet. Nobody wants to be the first to try something new; they’d much rather let someone else go first and read about their trials later.

Today, however, there is something else going on. Something new is in the wind. One observes a substantial change in the attitudes and conservative approach of companies in the way they present their new concepts of wireless technology. Convinced of the benefits, Emerson and Honeywell are announcing
products months before they are ready to ship, and years before industry standards (and even some aspects of the technology itself) are in place. What is going on?

Riding on top of an already buoyant economy, the spectre of wireless networking in manufacturing plants is
coming within reach, and it is simply too good to put on hold any longer. Countless PowerPoint presentations have already listed the bullet item benefits: massive savings in cabling costs, greater mobility of equipment, more extensive diagnostics, and so on. But vendors are quick to say it’s their customers, and not themselves, that are behind the surge of interest.

‘We’re not pushing this technology,’ says Emerson’s EMEA president Jim Nyquist. ‘The industry is pulling it.’

‘Customers are fuelling the industrial wireless revolution and we plan to be right there with them the entire way,’ says Honeywell Process Solutions’ president Jack Bolick.

Most instrument and control engineers don’t need lectures on how wireless networking will benefit them. What they have to be convinced of is, that wireless actually works and is secure. Emerson says after extensive testing, it has concluded that its wireless network works 99.9 per cent of the time. With 8,760 hours in the year, this means that it might not be working for eight hours. Is this tolerable?

The next frontier
‘Wireless is the next frontier to enabling total plant optimisation and safety,’ proclaims a Honeywell brochure. ‘It’s the greatest thing to come along in process automation since the launch of the microprocessor-based distributed control system in the 1970s,’ says Paul Orzeske, who heads up Honeywell’s EMEA operation. Emerson’s Jim Nyquist agrees that it’s ‘the greatest thing to come along in the past 30 years.’

Although the two companies appear to be head-to-head competitors, the reality is that each has an extensive installed base and will concentrate its wireless evangelism activities with its own customers. The intense competitive spirit will more likely emerge behind closed doors, at standardisation meetings where
each will try to influence the committees that their version of networking should be adapted as the industry standard.

Radio, of course, has been around for a century, but more recent developments have made it a better proposition for industrial communications. Microprocessor-based radios can perform security feats that were awkward and clumsy to do 20 years ago, and required military-grade equipment to implement. Thus frequency hopping, spread spectrum, and cyclic redundancy checking are no longer regarded as exotic techniques and are routinely incorporated in the smallest industrial radios.

The mesh
Perhaps the biggest breakthrough is the concept of ‘mesh’ networking. The idea of the mesh came along with wired networks—Invensys implemented it several years ago—and now is a key inspiration of wireless industrial networking. The mesh is a network of many intelligent ‘nodes,’ each communicating to one of the nodes ‘next’ to it in the network topography. The more nodes, the better the mesh. With an abundance of communication paths, redundancy is assured, because if a node drops out of the network for one reason or another, the other nodes around it are ‘smart’ enough to realise it’s gone and work out for themselves an
alternative communications path.

There are two interesting variations of the wireless mesh. In the Emerson implementation, based on ultra lowpower IEEE 802.15.4 (so-called ZigBee radio), each field device has a radio in it and is a node in the mesh. Thus every radio-enabled, battery-powered pressure and temperature transmitter is a node that can communicate with other nodes. There is a sort of egalitarianism among nodes; no matter what their purpose is out there in the field, they are also battery-powered nodes that must participate in the network. This means they may be obligated to carry some of the network traffic as intermediaries, picking up a message at one node and passing it along to another. Some of the nodes are clustered about gateways
which pick up their signals and transfer them on to the control system.

In the Honeywell version, like Emerson, all communications are wireless. The individual field devices are
‘slaves’ reporting, by low-power radio, to mainspowered IEEE 802.11 WiFi nodes called iNodes that are strategically located throughout the factory. The iNodes do not themselves contain instruments; their, function is to collect instrument signals and pass them on to other iNodes in the mesh, who will eventually pass them on to the control system. The instruments communicate directly with the powered nodes, and not at all with each other. An instrument will communicate with multiple iNodes to assure redundancy.
Honeywell says that since the instruments’ radios are used only to communicate with the iNode, and not to
participate in any mesh networking scheme, their lifetimes are both longer and more predictable.

The weakest link
Like hydrogen-fuelled cars, some batterypowered instruments perform well and show great promise, but their future depends on technology that is yet to be developed. Batteries are a part of the wireless networking story that promoters aren’t eager to give a lot of attention to. With thousands of field instruments
deployed in large process facilities, and many of them in critical situations, control and instrument engineers are understandably wary of the need to replace them, at some time or another. This is something they never have to think about when it comes to fieldbus instruments, but with the advent of wireless, now they have to think about it a great deal of the time.

Both Emerson and Honeywell have a standard set of answers to give to battery questions, and for the most part, they have a common theme: white-knight technology in the future will come along and solve this problem. Batteries with much longer lives, or fuel cells, or something like that will surely be invented and engineers won’t have to worry about power sources any more. Power management of instruments has
greatly improved; just look at what mobile telephones are capable of doing today. And alternatives aren’t too far away: thermal and vibration energy harvesting, for example. The good salesman will pull an odd looking device out of his bag, an alternative energy source, and say, ‘Hey, look at this puppy, we’re experimenting with it in our labs right now.’

The first generation
Dan Sheflin, chief technical officer for Honeywell Automation and Control Solutions, says his research people have ‘looked at every type of wireless application since 2001.’ The first industrial product his people produced, the XYR 5000, shipped in February 2004 and now is resident in 300 sites.

Honeywell, he reminds us, is not new to wireless and has solid credentials for the technology. He estimates the company holds over 300 wireless patents and has produced over 35 million wireless sensors for a number of different markets, ranging from aviation to building control.

The XYR 5000 was the first wireless sensor system aimed specifically at the process control market. In contrast with the ‘mesh’ networking schemes, the architecture of the XYR 5000 is very simple, and this may be its big advantage since it is not complex to install and configure.

With the XYR 5000 Honeywell has enabled a collection of its different sensors—differential, gauge, and
absolute pressure, temperature, as well as more complex devices such as pH meters and conductivity meters, and the recently acquired corrosion transmitters—so that they can communicate with a base station radio receiver. It also includes an analogue input interface for adding wireless capabilities to wired devices. The target application is for sensors located in areas that have no access to power and in positions that may be dangerous or hard to reach. Measuring tank levels in explosion-proof areas is a good example
of where this type of transmitter could be placed.

Each base radio can connect with up to 42 wireless transmitters at distances up to 600 metres. The radio uses frequency hopping spread spectrum (869.4 to 869.64 MHz) to avoid signal interference. The claim is that the batteries will last from three to five years.

The second generation
With the XYR 5000, critics and observers could be forgiven if they thought Honeywell’s approach to industrial
wireless networking was simple and unpretentious. What they didn’t know was that CTO Sheflin and his people were only testing the waters, and they had designs for a family of much more complex and comprehensive wireless systems, ones that embraced not only industrial sensors, but the whole breadth
of factory requirements, ranging through operations, maintenance, and asset management.

And so, rather quietly during 2006, Honeywell opened the door to its wireless roadmap for the factory, and it
was a plan that dwarfed the XYR 5000 scheme. Honeywell’s second-generation devices use the IEEE 802.11 WLAN infrastructure, with a 2.4 GHz frequency hopping, spread spectrum radio; the backbone device is the iNode, which is a mains-powered WiFi node that is strategically located at various points throughout the factory.

The iNode has three aerials on it: one for communicating with nearby instruments, another for communication with IntelaTrac PKS and Mobile Stations, and a third for connecting with other iNodes in the mesh. The powered iNode handles all mesh communications traffic so that the individual instruments don’t
have to be involved in this activity: they communicate their values to all the iNodes within reach. The XYR 5000 connects into the second-generation network by sending its information directly to one or more iNodes. iNodes can be up to 10 km apart; sensors can be up to 600 metres away from an iNode. For one second update intervals, the network can handle up to 1,000 sensors. By reducing the update to once every
thirty seconds, the network will handle up to 30,000 sensors.

To build the iNodes and mesh technology, Honeywell is co-operating with 3e Technologies International of
Rockville, MD, in the U.S.A. The company specialises in secure mesh technology and has worked with the U.S. military to develop secure protocols.

RTLS and IntelaTrac
The primary instrument in IntelaTrac PKS is a mobile handheld computer used for data collection in the field. It is designed for asset management applications and to help automate operator rounds. The Mobile Station is a tablet computer for DCS operators to take with them when they move around the plant. It emulates the operation of the DCS console, displaying graphics, process information, and historical data.

A fourth piece of the Honeywell wireless scheme is the Real-Time Location System (RTLS), which works with its own wired (Ethernet) or wireless network independently of the iNodes, although in the future it may be integrated into the iNode.

RTLS makes use of a combination of three positioning technologies, GPS, UWB (ultra wide band; 5 - 6 GHz) and an ‘active’ RFID (that is, an RFID that transmits under its own power rather than induced power from a nearby source) to locate equipment and personnel. It is said to be good for a range of up to 200 metres line of sight or 50 metres not line-of-sight.

With RTLS and localisation sensors installed around the plant, the exact location of employees can be quickly mapped and tracked to within a metre of their location, whether they are inside or outside of a building. The system will handle up to 5,000 tags so it should be possible to provide all employees and
contractors with an RTLS tag.

RTLS will be a great security and safety tool. Prior to the startup of a certain unit, it allows operators to check to see if unauthorised personnel are in the area. As soon as an unauthorised person crosses a designated perimeter, the system will alarm. The HTML graphics displays who they are and where they’re located. Injured personnel could report their exact locations with a push of a button. Mustering reports can be
automatically generated in the event of an emergency. Honeywell says RTLS will take security ‘to the next level.’

The technology might also be used to locate and track inventory, as long as it is suitable tagged with a locator. This might reduce the time spent looking for assets during construction projects, for example, or tracking other valuable assets are they are located and re-located on the plant site.

Honeywell Process Solutions
www.honeywell.com/ps


Contact Details and Archive...

Related Articles...

Additional Information...

Print this page | E-mail this page