• Number 337  |
• May 16, 2011

Doomed bridge sparked Salari’s engineering career

As a student at the State University of New York, Stonybrook, Kambiz Salari saw a video of the 1940 collapse of the Tacoma Narrows Bridge and decided he wanted to learn more about why the structure failed.

“The explanation was simple, yet it was intriguing, and that pushed me to want to understand more about mechanical engineering,” said Salari, who works in the Center for Applied Scientific Computing (CASC) at DOE’s Lawrence Livermore National Laboratory.

Later, Salari learned that 42-mile-per-hour winds rolled over the four-month-old Tacoma, Wash., bridge to cause a small change in pressure. In turn, the imposed force caused the bridge to oscillate violently and then fail.

From that pique of interest, Salari, 53, became a fluid dynamics researcher by training and a mechanical engineer by degrees.

Salari’s work and the efforts of the LLNL team he has led for the past six years paid off last spring when it was announced at NASA Ames that they had developed technologies that could increase fuel efficiency by 12 percent for semi-trucks. The work, done with engineers from Navistar of Fort Wayne, Ind. could save the nation more than $10 billion annually in fuel costs.

Salari, whose expertise is in computer modeling and simulations, was hired in the summer of 2001, within weeks of the hiring of Jason Ortega, a UC Berkeley Ph.D. graduate in aeronautical engineering, whose forte has been experimental modeling.

Both Salari and Ortega were brought to the Laboratory by mechanical engineer Rose McCallen to work on the Department of Energy (DOE) Heavy Vehicle Aerodynamic Drag Consortium. The consortium’s R&D led to the improved semi-truck fuel economy advances.

McCallen, now part of LLNL’s Weapons and Complex Integration Principal Directorate, helped found and led the DOE heavy vehicle consortium from 1997 to 2007, when she moved on to other research. Salari started directing LLNL’s work in the consortium in about 2005.

“Kambiz has many skills,” McCallen said. “He performs code development, as well as simulations and computational modeling. He is very careful and very meticulous, and he has a strong understanding of the experimental work. He can look at the entirety of this complex problem.”

In McCallen’s view, the teaming of Salari and Ortega turned out to be an almost perfect fit.

“By teaming Kambiz and Jason, Kambiz developed more skills in experiments and Jason gained more expertise in simulations. They now have extensive experience in both areas and they complement each other.”

For about a full month last year, Salari, Ortega, other members of their team, Navistar and the U.S. Air Force conducted truck fuel economy tests in the National Full-Scale Aerodynamics Complex wind tunnel, the largest in the world, at NASA Ames.

“What’s wonderful is that both Kambiz and Jason were able to work directly with the wind tunnel experts in defining the experiments based on the simulations they do at the Lab,” McCallen explained. “A lot of our success in this project is because of our computer resources, as well as the modeling and experimental expertise we have at the Laboratory.”

In 2000, while working at Sandia in Albuquerque, the challenging fluid dynamics problem of improved fuel economy for trucks aroused Salari’s scientific curiosity. His work on the project continued when he joined LLNL the next year.

“It took a little adjustment into the Lab culture from Sandia, but having someone like Rose help me made a difference,” Salari recalls. “I knew I had landed in the right place.”

Salari adds that he has been “very happy” working in the Center for Applied Scientific Computing, noting, “They not only care about you, they are supportive of you as a researcher.”

In his studies, Salari says he and other scientists find that understanding the flow of so-called “bluff bodies,” such as a truck or a loaf of bread, is more complex and challenging than a streamlined body, such as an airplane.

“The challenge is: you’re given a box and the interior can’t be touched. We have to preserve the shape, so what we had to do was to trick the flow of air to think this shape is more streamlined than it is,” Salari explains.

To date, the LLNL aerodynamic drag program has garnered five patents for new devices or technologies and has filed 15 additional patents that are currently pending.

Modest and soft-spoken, Salari is quick to give credit for the project’s success to other LLNL team members and McCallen, who led the multi-laboratory, multi-university effort for 10 years.

“Typically, there is a lot of work that gets done. I am only one person and really the credit has to go to our entire team. A large measure of the work was done by Rose. What I did was take her good work and try to extend it. She set a great foundation. I think Rose was very successful in bringing the truck industry together…and that was not an easy task,” Salari added.

Beyond Salari, Ortega and McCallen, other LLNL members of the team have been postdoc and Stanford Ph.D. graduate Paul Castellucci (2004-08), postdoc and UC Diego graduate Craig Eastwood (2006-08), postdoc John Paschkewitz (2007), current employee Tim Dunn and former LLNL employee Dora Yen Nakafuji (2000-02).

Salari’s road to the Laboratory started in 1977 when his family — his parents, three sisters and a brother — emigrated to the United States from Iran, about two years before the Shah fell from power. He was born in Kermanshah, now a city of about 800,000 people in western Iran, 75 miles from its border with Iraq and some 325 miles from Tehran.

“Since I left Iran, I’ve never gone back. There hasn’t been a reason because my whole family came to the U.S.,” he explained.

The Salari family settled in upper New York state, and Salari undertook four or five months of intense training in English. He attended the University of Buffalo for one semester, but after not seeing any of the campus until the snow melted, he decided it was “too darn cold” there. He transferred to the State University of New York, Stonybrook, where he received his bachelor’s degree in mechanical engineering.

Over time, much of his family moved to New Mexico, and Salari decided to study at the University of New Mexico in Albuquerque, receiving his master’s and Ph.D. degrees in mechanical engineering.

Salari later went to work for eight years for a small Albuquerque company, Ecodynamics Research Associates Inc., started by his Ph.D. adviser Patrick Roache. From there, it was on to Sandia and then Livermore.

During his time at LLNL, Salari has worked on the detection of explosives and underground coal gasification as well as the Heavy Vehicle Aerodynamic Drag Consortium and other projects.

“A project like this one (the consortium) usually doesn’t have this type of longevity,” Salari says. “But we received help because of fuel prices going up substantially in 2008; we were looking at $5 per gallon diesel fuel. That makes people think: Is there something we can do to improve fuel economy?”

Along the way, the 14-year consortium has been comprised of a number of partners besides LLNL and Navistar, including Sandia National Laboratories - New Mexico, Argonne National Laboratory, NASA Ames, the California Institute of Technology, the University of Southern California, the University of Tennessee at Chattanooga, the National Research Council of Canada, the tire manufacturer Michelin and a number of truck manufacturers.

The consortium staged two international conferences — in 2002 at Monterey and in 2007 at Lake Tahoe — on “Heavy Vehicle Aerodynamic Drag: Trucks, Buses and Trains.”

Looking back, Salari notes that with the project’s funding and the high cost of testing at top-tier wind tunnel facilities, such as the National Full-Scale Aerodynamics Complex, NFSAC, it didn’t seem possible they would be able to use such facilities.

“We had always hoped we would achieve something like this (the fuel economy results), but to have success at this level and be able to use this wind tunnel had seemed out of reach.”

But that is no longer the case for the consortium or Salari.

 – by Steve Wampler

Submitted by DOE's Lawrence Livermore National Laboratory