Partners with Energy

PSC collaborations with the Department of Energy

Improved Actions for Staggered Quarks

Magnetism in the Solar Dynamo:
University of Chicago

Rocket Science:
University of Illinois Urbana-Champaign

Shock Waves in Gas:
California Institute of Technology

Turbines and Turbulence:
Stanford University

A New Picture of How Metals Deform

The Edge of Reality

Turbines and Turbulence

Unsteady flow simulation of the Aachen experimental compressor as simulated by Jixian Yao of Stanford on the CRAY T3E. Download a larger version of this image (247KB).

Transition from smooth to turbulent flow along the surface of a turbine blade. The rear wall of the 3D volume represents the blade surface. Turbulence can be seen increasing from left to right. These simulations seek to isolate the process that triggers turbulence. Download a larger version of this image (205KB).

Gas turbine engines are the reliable workhorse of modern civilization. They produce a large portion of our electrical energy and they keep global business on schedule by lifting giant airliners into the air. The basic idea is as old as water wheels and windmills - rotating blades translate raw energy into circular motion. In the modern version, the raw energy is fossil fuel, and the design involves complex fluid dynamics. Stationary blades (stators) channel high-speed flows into rotating blades (rotors). Variations in the shape or number of blades along with other variables govern how well the turbine works.

The Center for Integrated Turbulence Simulations at Stanford University directs its research toward realistic simulations for the design of the compressor, combustor and turbine components of aircraft gas turbine engines. Ultimately, the goal is to integrate these component simulation technologies and to accurately simulate the complex situations governed by interactions among components. An array of phenomena present themselves as challenges - rotating stall in the compressor, blade vibrations, combustor instabilities due to coupling between heat release and acoustic modes, and heat transfer from the combustion gas to the first blade rows of the turbine.

In one large-scale study at PSC, CITS researcher Paul Durbin and colleagues attacked the problem of "wake-induced transition" from smooth to turbulent flow along the surface of a rotor blade. As flow goes from one stage of blades to the next, the wake from upstream blades strikes downstream rotors, disrupting smooth flow along the rotor surface and creating flows that are inherently complex and unstable. In a series of direct numerical simulations, solved using 52 million grid points, the researchers looked at this problem with an unprecedented level of detail.

In another series of CITS studies at PSC, Jixian Yao carried out calculations on the unsteady flow in a multi-blade-row compressor. These simulations aim at validating a particular "flow solver," and his computations simulate the geometry of an experimental compressor, the Aachen Compressor, composed of three blade rows.

"PSC has been a tremendous resource for our program," says CITS director Bill Reynolds. "For the kind of highly structured code we write, the CRAY T3E is an excellent machine that scales well as you go up in processor count. The consulting people were extremely helpful. PSC is a great asset to the national research effort and has enabled us to make substantial progress."

Other turbine research at PSC is described in The Supercomputing Science Consortium.

NEXT: A New Picture of How Metals Deform