Keeping an Eye on the Tank Farm
Figure 3. At Lubrizol's Deer Park, Texas, facility,
temperatures and pressures at the tank farm are
monitored wirelessly. Courtesy of Lubrizol
MOL will be monitoring corrosion, the temperature of fire heaters, the efficiency of heat exchangers, temperature and differential pressure transmitters and the open/closed position of valves.
The MOL refinery was the 2010 HART plant of the year, so it already has a strong infrastructure of wired HART devices, but there were some problems. Bereznai explains, "We had positioners we couldn't use because the gates between the positioners and the DCS were not HART-ready. Our explosion-proof barriers weren't HART-ready either."
So the company went to Emerson for THUM modules (The THUM can be retrofit on any existing two or four-wire HART device, enabling wireless transmission of measurement and diagnostics). The THUM adapter "allows us to get around this problem," says Bereznai.
The THUM adapter also allows MOL to connect many of its field devices to its Emerson AMS asset management system, and in another adaptation, meet government emissions reporting requirements. "We created an app where we can use pyrometers and THUM adaptors together to measure emission rates," says Bereznai.
And don't forget safety and security functions. Wireless monitoring of eyewash stations and safety showers, as well as personnel location, is an easier sell, perhaps, than applications such as monitoring and maintenance. Such applications are available from many vendors, including Honeywell, Emerson, Apprion and BS&B Industrial Wireless Solutions, among others.
GlitchesAs effective as these solutions are, getting them running is not all roses and lollipops. As with any new technology, there are glitches in the system and a learning curve.
Honeywell Sensing's Citrano observes, "Early on people had bad experiences with wireless because it wasn't really fit for the purpose. When you got into discussions with people [we found] reliability and security were issues. Battery life was a problem too, but as people get better at setting up their networks—they would have too much messaging going over one node—as they get more experience, they can avoid these problems."Lundbeck's Ribon reports, "At the beginning we had some problems with communication between some repeaters/adapters; we discovered some radio noise due to a huge radio antenna place about 300 m from our facility. We solved the problem by moving some repeaters in areas where there was less interference and by placing some shields near the repeaters."
Speaking about MOL's integration of field instrumentation with the refinery's AMS system, Bereznai says, "When we started, the communication was slow. We wanted to create a fingerprint curve, which is very communication intensive. Working with Emerson, we developed a fast-loop function to help with this, but it took some time to find the solution. Emerson in the U.S. helped us find the solution."
The reality is that at present, at least, most wireless users find the systems too slow to use for industrial control. "You can have problems with interference, and you can lose some data," says Ribon. "This is acceptable only for monitoring non-critical parameters."
He adds, "In any case, before you start an installation, you need to do an engineering phase to verify the feasibility, checking for interference."
On the other hand, wireless systems can be surprisingly reliable. "In the last two years, we've had bad storms with very heavy lightening and have had no problems [with the wireless]," says Bereznai. "We also did welding just a few centimeters from the transmitters, and that didn't affect the signals."
Bereznai adds that MOL hasn't replaced the batteries in his system in the two years it has been operating.
The Elephant in the Room
If there is one thing still making many potential wireless adopters uneasy, it is the on-going debate about on which protocol to standardize. According to the Global Automation 2012 survey, 40% of the respondents are waiting for a single international standard, up from 20% in 2010.
There are multiple standards and protocols out there, but the two big dogs in the field are WirelessHART and ISA-100.11a. (See Figure 4.) An entire forest and several billion electrons have been killed off in the debate over which of these two standards is better and which one will "win" in the end. In reality, we shall probably see both of them around for a good long time to come.
The politics of the matter aside, in many respects, they are equally valid and useful. There are differences between them, but they are not insurmountable. The choice the end user makes will depend on what his or her goal is.
Process Automation Hall of Fame member Tom Phinney, who has been instrumental in developing both WirelessHART and ISA 100.11a, says, "WirelessHART is more restrictive in terms of its development potential. ISA100.11a is built around the Internet and IPV 6, [the latest version of the Internet protocol]. In the longer term, ISA 100.11a has more potential. WirelessHART does not use standard Internet technologies, but those will be available to leverage with ISA 100. ISA 100 has great longevity built into it. The truth is, it will be a next-generation protocol."