Tecplot

by TecPlot Inc

• Order Now
• View Pricing
• Download a Demo
• Request a Quote
• Make Enquiry

Tecplot 360 helps scientists better understand flow separation

Tecplot 360 helps scientists better understand flow separation and shed new light on century-old issues in the identification of flow separation.

Understanding the way flow separates is critical to designing vehicles, airplanes and combustors that will consume less fuel and emit less pollution. To reduce fuel consumption, there must be less drag; to reduce pollution there must be more efficient fuel consumption – and controlling flow separation is crucial to the management of both.

 

An example of how the unsteady flow separation in a lid-driven cavity is visualised. The plot verifies the breakthrough theoretical development on unsteady flow separation. While the flow visualised in this cavity is a benchmark problem and does not represent an actual device, the flow is directly relevant to cavity flows, such as flows in pressure ports, cooling flow over electronic devices, and flows in coating processes.

“Drag reduction and the corresponding reduction in fuel consumption and pollution are of great importance to air vehicles, cars, trucks and water vehicles,” says Dr. Gustaaf Jacobs, assistant professor in the Department of Aerospace Engineering and Engineering Mechanics at San Diego State University. “In combustors, flow separation can be a desirable induced effect that enhances mixing of fuel-air and so reduces fuel consumption and pollution.”

Prof. Jacobs recently collaborated on research that generated new mathematical and computational work for predicting where aerodynamic separation will occur and solved a century-old problem in the field of fluid mechanics.

The research team, lead by Prof. George Haller at MIT, included Prof. Jacobs and Dr. Amit Surana at United Technologies. It yielded exciting new insights that could lead to ways of controlling the separation of fluids—gases and liquids—that may ultimately result in better designs for fuel efficiency. They used Tecplot 360 to better understand the complex phenomena through visualisation of computational results.

The theory extends the mathematical conditions developed by Ludwig Prandtl in 1904 that identify the location of flow separation. Prandtl’s work had a major restriction: it applies only to steady-idealised, 2D flows. Ever since this pioneering work was done, there have been intense efforts to extend Prandtl's criteria to real-life problems, i.e., to unsteady 3D flows. It took more than five years for the team to develop the breakthrough theory, which extends Prandtl's conditions to unsteady 2D and 3D flows.

“This new theory is the first to objectively and without ad hoc assumptions identify the location and angle of the separation of fluid particles from walls in unsteady 2D and 3D flows,” says Jacobs.

Jacobs is quick to point out that the group’s work was built upon significant and crucial advances in the development of identification criteria of flow separation by many others in the field.

“Our work rests on the shoulders of hundreds of scientists who have been working this problem for more than a century,” says Jacobs. “So, while we’re excited about the breakthrough results our group came up with, a lot of the credit for development of separation criteria should go to the entire scientific community.”

High-order solver helps solve flow separation mystery

Jacobs, Haller and Surana used an in-house, high-order Navier-Stokes solver to compute the flow in the lid-driven cavity flow, which he says is ideally suited for a high-fidelity computation of unsteady separated flow.

“The direct numerical simulations have been shown to be as good as true experiments,” says Jacobs. “The theory can only be validated with computations, since it requires velocity derivatives at the wall that are extremely challenging and very expensive to measure experimentally. Moreover, with low-order (low-fidelity) methods, it is difficult to accurately determine high-order velocity derivatives.”

The code was developed and validated by Jacobs for Direct Numerical Simulation (DNS) of turbulent flow. In the multi-domain method, an unstructured mesh divides computational domains of any complexity into non-overlapping hexahedral sub-domains. Within each sub-domain, a high-order polynomial approximates the solution.

Domains are connected only by face data, giving the method a local nature that leads to highly parallel code. The method converges exponentially and is high-order accurate. The high-order solver is characterised by small dispersion and diffusion errors, which translate into high spatial accuracy, as well as accurate time integration. This high-order accuracy is particularly important when simulating unsteady, separated flows.

Tecplot visualisation helps validate new theory

Figure 1 identifies the green separation surface, which acts like a barrier along which fluid particles visualised by white streak-lines are transported when they separate from the side wall of a cubical lid-driven cavity. A lid, or the top wall, that oscillates around a mean velocity drives the flow. Due to the oscillatory motion, the flow characterised by separations near the sharp corners of the cavity is unsteady. The origin of the separation surface on the wall is identified by the new separation theory that determines the separation location with wall data only. The high-fidelity computations and visualisation verify that fluid particles separate from the location predicted by theory.

Jacobs explains, “The analysis of fluid particle separation represents an innovative approach from the typical analysis of the instantaneous skin friction patterns on the wall as visualised by the red lines and the corresponding instantaneous streamline separation. In unsteady flow, the separating streamlines often lead to misleading interpretation of flow separation. For example, in a quasi-periodic flow such as this oscillatory lid-driven cavity, the computations and theory show that the fluid particle separation is quite different from the streamline separation. Whereas the separating streamlines are oscillating, the fluid particles separate, perhaps counterintuitively, from a fixed curve.”

So what did the plot results tell the researchers about their data? “In a nutshell, the plot verified that fluid particles in a quasi-periodic flow separate from a steady separation line on the wall that may be determined by the new theory with wall-based data only,” says Jacobs. “Moreover, it shows that true physical separation of fluid particles is different from the somewhat artificial separation in the instantaneous skin friction field.”

Jacobs has used Tecplot software for ten years and believes its greatest strength is its user friendliness. He also uses the software’s macros and layout for faster data visualisation. “The layout files and macro feature are particularly useful when similar visualisations are required in parametric studies,” says Jacobs. “The multi-frame feature is also unique to Tecplot [software]. Streamlines are also more easily visualised than in most visualisation software.”

Though he’s used other data visualisation tools, Jacobs feels that Tecplot 360 is critical for the numerical research performed in his group. Tecplot 360 is used to create 3D iso-surfaces plots of unsteady flow as well as animations of flows laden with particles that are generated by in-house codes. The resulting plots are used for research papers and presentations. While Tecplot 360’s main role at the university is to help scientists understand the physics of their results through visualisation, Jacobs adds that the software is also instrumental in debugging code.