MIT aerodynamics breakthrough could improve efficiency, speed

A century-old problem in the physics of aerodynamics, known to scientists as fluid dynamics, has been solved by MIT scientists, both on paper and in the lab. The breakthrough makes it possible to calculate when the otherwise smooth flow of air over a complex surface will cease and separate, hugely affecting aerodynamic drag and lift on, for example, a vehicle. Previous solutions had been largely theoretical, and required low speeds, perfectly two-dimensional surfaces and ideal conditions. Such features are nowhere to be found in the real world, so the MIT breakthrough is a major step forward in understanding - and therefore dealing with - how air flows around objects in three-dimensional, real-world conditions. The man at the head of the project, George Haller, developed the theory along with several other professors and students in a two-dimensional format in 2004. Since then they have been working on expanding that theory to three dimensions, and the latest results are the culmination of the last four years' work. The mathematical calculations had to be tested experimentally to verify that they were more than internally consistent, however, the work of MIT colleague Thomas Peacock has now done that as well. While the science of the breakthrough is beyond the grasp of lay people, the implications are clear: by being able to anticipate when, where, and under what conditions, the flow of air around a car will begin to separate, engineers can design around the problems of drag and lift to produce more efficient vehicles. That would mean either reduced fuel consumption, high speeds with the same amount of power, or potentially both. The scientists behind the new findings are quick to point out that the findings are still young, and it will take time to determine how much it could affect performance or efficiency in cars or other vehicles. "This is the tip of the iceberg, but we've shown that this theory works," Peacock said.

by Gus Chyba #1, Posted: 9/26/2008

Wow, this is a big deal.
Big steps like this will become more commonplace with the advent of faster computers and more money to motivate scientists to find solutions to ever more publicized problems...
Can't wait for a room-temp superconductor, for example...
by Stephan Wilkinson #2, Posted: 9/27/2008

It seems to me that the breakthrough is not only beyond the grasp of lay people but beyond the grasp of the writer.

Fluid dynamics is not "a problem in the physics of aerodynamics," it's a discipline and science in its own right. And discerning when airflow separates certainly doesn't require perfect two-dimensional surfaces and low speeds, I've myself done tuft-testing to determine airflow separation airplanesf and it worked just fine on real-world dimensionality and real-world speeds.

I have no idea what this "writer" is trying to communicate, and I doubt that he does either.
by Nelson #3, Posted: 9/28/2008

It seems to me that the breakthrough is not only beyond the grasp of lay people but beyond the grasp of the writer.

Fluid dynamics is not "a problem in the physics of aerodynamics," it's a discipline and science in its own right. And discerning when airflow separates certainly doesn't require perfect two-dimensional surfaces and low speeds, I've myself done tuft-testing to determine airflow separation airplanesf and it worked just fine on real-world dimensionality and real-world speeds.

I have no idea what this "writer" is trying to communicate, and I doubt that he does either.

While I appreciate your enthusiasm, Stephan, you might want to take another moment to re-parse that first sentence. If you do, you'll find that 'fluid dynamics' isn't another name for the problem, but for the physics of aerodynamics.

Second of all, actually building complex surfaces (even scale models) and then sticking them in wind tunnels with little tufts of fiber attached to them is very, very expensive in relation to the price of automobiles (even very expensive ones). That isn't so much the case in relation to airplanes, which may be why you didn't consider this point.

On the other hand, however, it must be quite obvious how valuable it would be to not HAVE to do tuft testing, but instead use highly accurate, lab-proven mathematical models run on computers. It cuts development time and means fewer expensive prototypes have to be made on the road to a real-world solution.

And even if I had no idea what I was talking about, the fact that the big brains at MIT are so excited about the discovery is reason enough to think it's important.

Before criticizing others, one must always ensure they, themselves, are not equally subject to their own complaints.