Interested in linking to "Armoring Up for Safety"?
You may use the Headline, Deck, Byline and URL of this article on your Web site. To link to this article, select and copy the HTML code below and paste it on your own Web site.
This article was printed in CONTROL's November 2009 edition.
By Chris Williams
Eastman Chemical Co. (www.eastman.com) in Columbia, S.C., uses Foundation fieldbus (FF, www.fieldbus.org) technology and DeltaV (www.emersonprocess.com) hardware throughout one of the plants at our facility. This system functioned so well that Eastman decided to install fieldbus in a second plant. My challenge was to figure out how to install a similar Foundation fieldbus system, but in very different environment.
Our second plant has many Class I, Division 1 and Class 1, Division 2 areas with processes that are not only hazardous, but extremely corrosive. Traditional installations required use of rigid metal conduit, explosion-proof fittings and poured seals, or the use of intrinsically safe circuits. Typically, intrinsically safe FF segments have some signal loss, and only permit a maximum of four devices with short trunks/segments. Calculations are required to prove all components will work together to form an intrinsically safe circuit. We wanted to install a system that would function reliably in a harsh and hazardous environment and maximize segment capacity and that would enable our personnel be able to work on the instruments live without having to gas-check the area first.
In general, Fieldbus Intrinsically Safe Concept (FISCO) provides a way around the limitations of normal intrinsic safety. It makes more voltage and current available in hazardous locations—12.8 V and 115 mA. Trunk and spur lengths are 1,000 m and 30 m respectively.
However, one concern about bringing fieldbus technology with restricted voltage into our plant was its corrosive environment. Connectors get dirty and boxes start to fatigue, degrading communications over time and leading to network instabilities. These conditions also create additional resistance that results in voltage drops. Random and intermittent communication problems are hard to pinpoint and eliminate. I wanted to be able to find those problems before they caused a shutdown. Also, I wasn't sure if 12.8 V would be enough to compensate for corrosion over time, given that the fieldbus devices require at least 9 V to operate.
Trunks and Barriers
While seeking the best solution for our problem, I saw a presentation about Pepperl+Fuchs' (www.am.pepperl-fuchs.com) Advanced Diagnostic Module (ADM). It reported that ADM would save time during start-up and commissioning, and also could detect degrading conditions or faults occurring on fieldbus segments, meaning that any change in the installation, such as those caused by corrosion, would be detected before the change became critical to performance.
ADM is one of Pepperl+Fuchs' FieldConnex products for fieldbus communications. While the fieldbus transmits the data, FieldConnex provides the infrastructure, including power supplies, installation equipment and accessories. Its high-power trunk (HPT) concept and FieldBarriers provide a new approach to creating intrinsically safe fieldbus segments, and the whole installation is continuously monitored by ADM integrated in the FieldConnex power hub (Figure 1).
For example, rather than limiting energy on the fieldbus trunk cable to intrinsically safe or nonincendive levels, HPT limits energy on the spur connections to the instruments. Using FISCO with HPT, energy on the trunk is increased to 500 mA, rather than 115 mA. This increases energy available for field instruments and facilitates consistent installation design regardless of area classification. By limiting the energy in the field with a FieldBarrier rather than in the control room, power is more efficiently distributed to the instruments where it's required.
FieldBarriers are the core of HPT. They're mounted near the field devices to provide short-circuit protection and energy limitation for explosion protection. They guarantee that the segment remains in operation even during a fault condition on a spur. HPT considersg each spur connection separately. FieldBarriers can be daisy-chained on the trunk and enable the field devices to be serviced without the need for a hot work permit.
When we brought our original plant network online, we did it using the standard method—pulling the trunk cables apart, performing resistance and capacitance checks. When we had a problem, technicians checked the system in the field. The instrumentation for Foundation fieldbus eliminated the need for technicians to be out in the field doing calibration checks, putting in the scaling factors inside the transmitter, and validating the 4-20 coming back. Start-up was extremely fast. In just one day, we commissioned and brought over 40 instruments on line.
For our second plant installation, we were told that we didn't need to go through that same process. Supposedly, with advanced diagnostics, we could shoot our networks and scan them before we hung the instrumentation. After hanging the instruments, we could scan the network again and compare the noise levels, power draws and the health of the network.
I wasn't convinced. I do know that installation practices have a significant effect on communication quality, so we were very careful to install our system perfectly. We installed the first four segments and did all the initial checks, and we checked each instrument with a handheld configurator. Then we used ADM to validate all that information, and it matched perfectly.
Monitoring and Predictive Maintenance
During checkout, ADM even detected a problem the handheld configurator missed. One of our bricks was over-terminated, but not enough to cause a failure. We still had a running system. We weren't getting any communication errors, yet the network was deemed unhealthy by ADM. We had failed to turn off the built-in terminating resistor on one of the bricks. My confidence in advanced diagnostics was starting to grow.
The next system we brought on-line had eight segments. This time, we made sure that the terminators were right and the trunk cables were terminated correctly. After we determined that the segments were good, I used ADM to commission all eight segments. It took only 20 minutes, and identified two problems that my field technicians missed. These errors were corrected in 10 minutes, and the segments were back online and healthy.
After that, we hung 40 instruments on the segments, and used only the diagnostic system to validate. Not one had a problem. Even our longest run, 1,500 feet, came on line with no problem as quickly and as easily as the run that was 5 feet from the cabinet.
Currently we've installed FF using HPT in six different areas. We've stopped using the handheld, and don't haul the oscilloscope out to the field. Once the system is installed, I do the sweep remotely, and we're ready to go.
The physical layer diagnostic module captures snapshots of all values and can store that data for up to two years for long-term trending. Because it has the ability to create a visual representation of the digital fieldbus communications coming from the control room, it truly provides an “extra set of eyes” on communications. In fact, we were having trouble with a valve, and the manufacturer tried to tell me that it was because I had an unhealthy network. I showed him some of the screen shots from the software, and he's never mentioned the health of my network again.
The amount of time that we've saved with advanced diagnostics has been dramatic. It actually takes longer to walk down the system to set the jumper switches and the boxes than it does to commission the system. The physical walk down and commissioning for the first, four-segment system took a day and a half. On the following eight-segment, the physical walk down took about 60 minutes, and the commissioning took just 20 minutes. Our confidence in the network also increased dramatically. When the ADM certifies that the system was good, we know it is good.
In addition, advanced diagnostics has enabled our process to upgrade from incident-based maintenance to predictive maintenance. When you have a plant running 24/7, and an instrument failure of any kind will bring your operation to a standstill, you need a communication system that you can rely on. The ADM-FISCO combination has delivered on every requirement.
Chris Williams is an I&C engineer at Eastman Chemical Co.