The Frankentester comes to the rescue  

This gauge indicates the pressure inside the tank holding the oil. An air connection is used to pressurize the oil inside the tank to levels slightly above that of the surrounding water pressure to replicate conditions on the dredge. (Photo by Jennifer Sowell)

Dec. 27, 2007

By Jennifer Sowell
Portland District

Necessity is the mother of invention.  This statement rings most true in the face of real-world examples. 

When seals that were responsible for keeping nearly 1,000 gallons of oil in the U.S. Army Corps of Engineers Portland District's dredge Essayons' pump motors - and therefore out of the water - proved unreliable, a solution was necessary.  In response, the maintenance staff at the U.S. Government Moorings invented a machine to test the seals prior to installing them on the dredge.  The overside dredge pump motor seal test stand, or simply the Frankentester, was born.

The Essayons is unique in that its dredge pump motors go over the side and into the water as part of the dragarm assembly, rather than being housed within the hull.  This over-the-side design, while much more efficient for dredging, requires special seals on the overside dredge pump motors to keep oil out of the water.  The seals are a small but critical for keeping the dredge at work and preventing damage to the environment.

"Sometimes the seals would last a full dredge season, and sometimes they would last just a few days," said Bill McKay, the port mechanical engineer for the Essayons.

"We needed something we knew would last, because it's just too important," said Mac Robison, chief of the plant maintenance section at the Moorings.

Sudden failures of the seals accounted for nearly 30 percent of the oil spills, or 95 gallons, from the Essayons since September 1999.  Replacing the seals usually took three days and cost $80,000 per day, meaning lost time and increased costs for the dredge, as well as causing unacceptable environmental impacts. 

"The seals require special tooling, extraordinary effort and a special touch to install," McKay said.  "The first formulation and initial conception of the (overside dredge pump) motor seal test stand was in 1994, after a particularly exasperating effort to install the original seal design by the original method."

Due to changes in how the seals were installed, the test stand was set aside for nearly 10 years, since it was developed for the original method.

In 2005, the Corps realized a new seal design was needed, not just a new method of installing.  To assess the new designs, they pulled out the Frankentester to ensure the design would meet engineering requirements for installation simplicity, operating requirements and long-term reliability. 

"The goal is zero leakage and a minimum of a full dredge season of continuous use," McKay said.

The Corps used the Frankentester on potential new seal designs and last year found one that appeared to meet all the requirements.  The seal withstood about 2,600 hours of near-continuous use under varying conditions; they subjected it to maximum and minimum pressure differentials, varying oil temperatures and viscosities, and nearly 1,000 hours of start/stop cycles with no significant leakage.  "We burned up a gear box trying to destroy that seal," said Robison. 

While even the best-designed test platform can't fully replicate real-world conditions, the Frankentester comes close, McKay said.

"It spins at the same rpm (revolutions per minute), operates under water at the same pressure differential across the seal and at the same oil temperature," he said.  "It isn't elegant, but it's fully functional and able to simulate operating conditions that are most likely to effect seal life."

The tester logs data for pressure, temperature and time, which provides information about when and how problems occur. 

Everything seemed to be on track, but when the new seals were installed in March 2007 they failed almost immediately. 

The reason for the contradictory results, a flawless run on the Frankentester followed by failure once installed on the dredge, is still unknown. 

Although the seals failed when they were installed on the dredge, the Corps is continuing to work with the vendor to come up with a reliably operating and simple to install seal by May. 

"The vendor has been really good to work with," McKay said.  "They are continuing to rework their seal design to meet our requirement of zero leakage.  That's a tough goal for a mechanical seal."

"This is an extremely important issue to resolve before the next dredging season," Robison said.  "We need a seal that is robust enough to rely on and not have to constantly worry about."