Wake turbulence, or wake vortices, may have played a role in the American Airlines Flight 587 crash that killed 265 people on Nov. 12, according to crash investigators. The tail fin of the Airbus A300 jet sheared off after the pilots struggled against the wake turbulence left by a Boeing 747 that had taken off less than two minutes earlier.

"The wing we designed could make substantial differences in flight safety and airport capacity," said Omer Savas, professor of mechanical engineering at UC Berkeley. Savas and former UC Berkeley graduate students Jason Ortega and Robert Bristol experimented with wing designs that would quickly render wake turbulence harmless after takeoffs and landings.

UC Berkeley filed a provisional patent application on Friday, Nov. 16, for the design using results from Savas' experiments.

Federal regulations require two flights to be spaced far enough apart during takeoff and landing to avoid the potential hazards caused by wake turbulence. While wake turbulence alone would not have likely caused the crash of Flight 587 in New York, "turbulence in combination with a possible structural problem in the tail fin could be devastating," said Savas.

Savas has been testing a design with triangular extensions jutting behind each wing. He has found that with the bat-like design, the wake vortices generated dissipate two to three times faster compared with traditional wing designs.

A wake vortex results from the mismatch in speed, direction and pressure of air moving above and below a plane's wing. These differences govern the lift generated during flight. Planes that are large, heavy and moving slowly create stronger wake vortices.

"On a very clear, dry autumn day, you can actually look up with binoculars at planes in the sky and observe the behavior of these wake vortices," said Savas. "The water vapor from the engine gets trapped at the center of the vortices and marks them as a pair of thin lines in the sky."

Depending upon weather conditions and the plane's speed and size, the wake vortices generated are relatively stable and can stretch a distance of hundreds of wingspans, or three to five miles for a commercial aircraft, said Savas.

For decades, engineers have sought ways to disrupt the stability of wake vortices in efforts to transform the forceful swirls into benign puffs of air. Wing designs have included small pulsing jets mounted at the wingtips, spars and oscillating spoilers. Most of the designs have been ineffective or impractical; some involve moving parts that require greater maintenance.

In tests, Savas' design effectively created instability in the vortices without generating too much additional drag. The design also has the benefit of involving no actively moving parts.

In one of his earlier experiments, reported in April's American Institute of Aeronautics and Astronautics (AIAA) Journal, Savas towed tapered sheet metal wings - one a traditional rectangular design, the other with the triangular flaps - in a 70-meter-long water tank at speeds of 1.6 meters per second. Fluorescent dyes marked the vortex wake generated by the 40-centimeter-wide wings.

Savas found that the traditional wing created two stable, counter-rotating swirls from the outside tips. In comparison, the wing fitted with triangular flaps created four vortices, two from the wing tips and two from the flaps. When the flap and tip swirls - each rotating in opposite directions - ran into each other, they quickly became unstable.

"It's like two tornadoes shredding each other," said Savas. "One is spinning clockwise, the other counter-clockwise, so each one counteracts the other. But, they first have to be close enough to each other and have the right strengths for that to happen. What we're doing with these triangular flaps is creating a second vortex that's close enough to destabilize the wing tip vortex."

Savas and his former graduate students have since conducted several other experiments that have been submitted for publication in which they refine the geometry of the triangular flaps. In one test where the flaps span up to half the length of a wing, the wake turbulence began to dissipate four to eight times faster than the wake vortices created by the traditional wing.

Test results were duplicated and confirmed with computer simulations using flow models developed by Philip Marcus, professor of mechanical engineering at UC Berkeley and an expert on vortex calculations.

"Our model is a significant improvement over current designs," said Savas. "In addition to improving safety, cutting the distance that the wake vortex remains coherent would allow planes to take off and land closer in time together without compromising safety. That leads to more efficient use of runway capacity, a major problem at congested airports around the country."

Savas said the design would entail some increased drag, but it could be easily compensated for with extra engine thrust during takeoff. After takeoff, the triangular extensions could be retracted so that drag would not be a factor.

The dynamics of how triangular flaps affect wake vortices can lead to other applications, Savas noted. For instance, small triangular flaps fitted to the outer tips of helicopter blades could significantly cut down on noise generated when the blades slice through the air.

Savas is currently working on a pilot program with scientists at NASA Ames Research Center to incorporate the triangular-flapped wings in aircraft designs. He noted that commercial jets have not gone through a significant design change since the Boeing 707 began rolling down the runways in the 1950s.