Studying general fluid mechanics and heat transfer as a graduate student in the 1960s, mechanical engineering professor Richard H. Pletcher had limited exposure to the analytical and even less to computation. But though he could solve certain fluid mechanics equations by running reams of punched cards through the school computer, he found little support for that approach in the precincts of academia.

“My adviser tried to give me some advice,” Pletcher remembers, “which was ‘don’t mess with computational things; the future would be in experiments’—he thought the computer business was a diversion that wouldn’t be very profitable.”

Undeterred, Pletcher signed on with United Aircraft (today United Technologies), where he developed algorithms for solving flow problems. This in turn became the basis for a paper on numerical methods for early CFD applications he would submit for publication shortly after coming to Iowa State in 1967.

One of the lesser-known engineering disciplines, today computational fluid dynamics is critical across several areas, from the modeling of airflows over jets to the design of multimillion-dollar reactors for pharmaceuticals. The ability to predict the movement of liquids, gases, and solids, whether singly or in multiphase systems, has allowed innovation in numerous industries to pace and parallel the ever-accelerating power of computation.

Yet given the relative scarcity of computing resources in the 1960s—even in academic settings—the injunction of Pletcher’s adviser was perhaps not as outlandish then as it may sound to modern ears.

Crossing disciplinary boundaries
Nonetheless, although initially hired to teach courses in fluid mechanics, Pletcher continued to dabble in computation, hauling stacks of punched cards to Iowa State’s mainframe. Using the computer, he would solve the boundary-layer equations, a reduced set of the Navier-Stokes equations modeling viscous flow over solid surfaces.

“We’d estimate the friction exerted on a wing or any solid body, and I’d also include heat transfer,” Pletcher recalls. “I’d solve the energy equation that would determine if the surface was hotter or cooler than the air, and we’d estimate the transfer of energy.”

Before long those methods paid off in Pletcher’s classroom, so Pletcher paid a visit to the Department of Aerospace Engineering and Engineering Mechanics. There he met Dale Anderson and John Tannehill, who themselves had begun to investigate the use of numerical methods to model airflows over surfaces.

The three decided to develop electives to be offered across their respective departments, determining that the courses should be team taught—the field was so undeveloped that no one had sufficient knowledge to cover the subject on his own. Their complementary interests would result in one of the seminal textbooks in the emerging discipline. First published in 1984, by 1997 Computational Fluid Mechanics and Heat Transfer had gone into a second edition as well as translation into Russian.

Another factor driving the early development of CFD at Iowa State was the relationship Pletcher and his colleagues enjoyed with U.S. government labs, especially NASA Ames in California. In fact, Pletcher notes, a number of the branch and division chiefs at NASA Ames were Iowa State graduates, a connection that enabled Iowa State faculty and students to work remotely off NASA mainframes. By 1980, that connection resulted in major grants from NASA to Iowa State and a handful of other U.S. academic institutions to develop advanced CFD programs.

“That was a huge step that we got recognized as one of the top five programs,” Pletcher says. “Iowa State would stand up with some very prestigious schools in the field.”

Next Page>>