ChevronPhillips Relates Early Wireless Lessons

“And when Hurricane Ike hit, the network still ran fine.” Despite early battles with mold and condensation, ChevronPhillips’ Alice McWilliams ultimately prevailed in her first wireless implementation, saving 65% of installation costs to boot.

She and John Scott, senior account manager from Emerson Process Management’s Rosemount division, presented the lessons they learned in a session they called “Wireless Wars: The Engineer Strikes Back” at the 2008 Emerson Global Users Exchange this week in Washington, D.C.

Initially, the impetus for the project involved the need to obtain two readings, but things escalated quickly.

“We had a 40-foot-tall tank that had to be climbed whenever we unloaded a barge,” explained McWilliams. “I wanted to automate the recording of measurements to record temperature and valve opening on the tank. We needed accurate data, but it was at the very top of this tank. Then two additional applications—a bearing temperature for a pump and the barge unloading line pressure—fell into our laps. We’d have to put up a huge scaffold to get cable and conduit to the top of the tank, so a wireless network was the right thing for this project.”

McWilliams and Scott first looked at both wired and wireless solutions. “We began to look at a wired solution for cost reduction,” said Scott. “But we came up with a 65% installation savings by going with wireless. One of the bigger costs for a wired solution would have been the scaffolding.”

McWilliams wanted to run a minimum amount of cable, which meant she needed a wireless gateway installed in the rack room, which had access to 24-V power, but is 600 ft away from the transmitters, and the 900-MHz radio waves weren’t strong enough.

“That’s when we began talking about a self-organizing network to give data reliability,” explained Scott. “We built the network so it could receive the information and give 99% data reliability.”

Then concerns arose over the frequency. “Our plant radio walkie-talkies use the same 900-MHz frequency range as the network,” said McWilliams. “Our radio people were very concerned when we brought in a network that ran at the same frequency as theirs. A radio frequency analyzer showed no interference, but they still were not happy.” To put their fears at ease, Scott explained five key aspects of the network, including protection/encryption, authentication, verification, key management and anti-jamming, which allayed the concerns.

A Heated Battle

But then the battle began to heat up. “I really thought putting the gateway on top of the rack room to receive all of the different signals would be a good idea, but no one really liked to climb up on that roof,” explained McWilliams. And when the first gateway developed mold on a card, on top of its inconvenient location, it was replaced and moved into the rack room. Then the second gateway quit a little more than a month after it was installed. In an attempt to correct the problem, McWilliams cycled power in the hope of resetting it. “What that did was to vaporize all stored data on what had gone wrong with the gateway,” said McWilliams.

When the third gateway was installed, it included a remote antenna, a feature that McWilliams had wanted from the start, but which had only recently become available. “Getting the gateway down from the roof and adding the antenna certainly helped everything,” said McWilliams. The 1420 wireless gateway was installed and operational. It seemed The Force was with them.

And then the transmitters had their say. “The temperature transmitter that started it all was the first one to give me trouble,” said McWilliams. “One day the temperature was up at 70 degrees, and the next day it was down to 35. This is the condenser temperature reading that was the cornerstone of the whole project.”

The batteries were replaced, but less than three months later, the same transmitter “flat-lined” for 16 hours before coming back by itself. At the same time, one of the bearing temperature transmitters on the catalyst pumps failed for a short time in the cold weather and then recovered.

“Then the same pump-bearing temperature transmitter experienced a significant drop in battery voltage, so the batteries were replaced again,” said McWilliams. “We were concerned why we were losing signals and batteries, so we looked at them and removed the covers and found condensation. We put Teflon tape on the conduit plugs because temperature plays a big roll in battery voltage levels.” And once the condensation problem was corrected, battery life improved.

But with all of these lessons came valuable experience. “We developed troubleshooting experience in wireless,” said McWilliams. “Plus, we added the temperature transmitter without Rosemount to prove the system is easy to maintain and grow. We also got a lot of valuable process information at a much lower cost due to the 65% installation savings. And when Hurricane Ike hit, the network still ran fine.”