Arkema’s chemical plant in Calvert City, Ky., has four boilers, which produce steam at slightly different pressures and flows because they’re different sizes and were installed at different times. The first two, both installed in 1952, are brick-set with forced-draft intake and induced-draft fans. Both are rated at 40,000 lb steam/hour (klb/hr). The third boiler, a 1965 economizer, includes only a forced-draft fan and is rated at 75 klb/hr, but it was typically operated at a maximum of 60 klb/hr. An additional 1996 economizer FGR boiler was operated identically to the third.
“All four boilers produce steam at about 165 psi,” explains Robert Horton, business manager, ABB Optimization Services (www.abb.us), “but they usually don’t run all four boilers at maximum load. Boiler 2 is the one used most often, but it was the least efficient, so that’s where we began our focus.”
ABB completed a “Fingerprint” analysis of the boiler operations in mid-March 2008, analyzed the collected data and, based on its findings, forecasted annual energy savings of as much as $100,000 from improved combustion efficiency, smoother load responses and reduced induced-fan load. The Fingerprint is a four-quadrant approach that examines the state of hardware and controls, the stability and operation of the boiler, combustion load tests and dynamic tests for step responses.
Boiler 2 was examined first, and Horton’s team identified several issues for resolution. The induced-draft fan positioner movement was jerky, indicating a potentially faulty pneumatic cylinder/piston assembly. “The positioner drives weren’t operating smoothly,” explains Horton.
The team also noticed a loose hatch door near the oxygen sensor that leaked air into the ductwork before the induced-draft fan. The two boiler oxygen sensors also continued to read about 2% higher than a portable analyzer. “They had two oxygen analyzers that were positioned in the wrong places, and there was leakage,” says Horton. “Plus, the analyzers were out of calibration.”
Air also was being added for an oil flow that wasn’t truly going into the boiler. “We could see oxygen going up and down,” explains Horton. “Air flow and fuel flow measurements were going up and down. They were both exhibiting hysteresis. They were fighting each other and introducing variability.”
A furnace draft test also indicated leakage was being sucked in by the induced-draft fan and thrown out the stack as wasted power. “Based on the load tests, we were able to update the air/fuel ratio,” says Horton. “We realized the boilers didn’t need quite as much air. The oxygen trim is that final adjustment of air/fuel ratio, and the Arkema team hadn’t used it for 10 years.”
Horton’s team recommended a variety of corrections affecting boiler furnace pressure, oxygen settings and transmitters, fan positioner mechanisms, leakage, steam flow measurements, control logic and loop tuning.
As a result, oxygen readings, which had measured in the 6% to 7% range, were brought down to less than 5%. “They’re getting really nice control now,” says Horton. “A reduction in oxygen means you’re not bringing in more air than you need, so you don’t have to heat it up, and you don’t have to suck it in and blow it out, so it means real fuel savings.” The approximate value in savings was $75,000 for Boiler 2 alone, and all without major capital investment, says Horton.
In addition to other efficiency improvements across all four boilers, the third boiler, which was underutilized, would trip inexplicably during storms. ABB’s team identified an exposed forced-draft fan with roof intake as the problem.