Friday, May 17, 2013

Radome Induced Vortex Shedding

Infinite Technologies, Inc. RCS uses an analysis-driven approach to design that focuses on gaining an in-depth understanding of the physics behind our products and the products of our customers.   An example of this is the simulation of wind loads on large structures, such as radomes.
 
ITI-RCS radomes are designed to withstand wind loads in excess of 100 mph (typically 155 mph).  At such speeds, pressure loads can vary widely from one side of the dome to the other, becoming a significant factor in its structural requirements.  Thus, accurate prediction of the wind load is very important to the success of a radome design.

Simulating wind flow around a radome is not as simple as it might sound.  Due to the high Reynolds numbers associated with these bluff bodies (between 105 and 107), the nature of the flow can be chaotic in the wake.  High fidelity models must be three-dimensional and unsteady since the flow separates behind the domes and forms vortices that shed in an undulating fashion.  As a result, pressure on the leeward side of the radome oscillates in both magnitude and position.

To illustrate this, we have simulated the conditions of one of the wind tunnel tests conducted in Reference 1.  In this test, D'Amato and Fanning subjected a sphere-cone radome to a uniform horizontal 100 mph wind.  The resulting Reynolds number was 2*106.  We modeled the test using a Detached Eddy Simulation (DES) run with a seven million cell unstructured mesh (see Figure 1).  Time steps were on the order of 10-3 seconds.

Figure 1: Unstructured mesh tailored for vortex shedding.


Video 1 shows the chaotic nature of the flow in the radome’s wake, over roughly 0.5 seconds.  The vorticity of the flow is shown along seven planes slicing parallel to the direction of the wind.  The undulating vortices roiling downstream can easily be seen. 



The true structure of the vortices is easier to discern in Video 2.  In this case the 150 Hz iso-surface of the vorticity is tracked through time (the surface is colored by the velocity magnitude).  Vortices of several sizes and shapes can be seen in the video.  Some are small and short lived.  However, the majority are large and persistent.  These horn-shaped vortices form just aft of the dome and are shed semi-periodically.  They are tight swirls near the centerline of the radome but then arch up and outwards until connecting with the ground at the edge of the wake zone and they grow in size as they travel.


  
Figure 2 shows the pressure distribution on the leeward side of the radome at two different points in time.  The unsteady nature of the vortex shedding just discussed causes high and low pressure zones to alternate side to side with time.  This variation causes lateral forces that are as high as 23% of the lift and 28% of the drag.  Obviously, this variation complicates any subsequent structural analysis.


In addition to unsteady flow around bluff bodies, ITI-RCS has the ability to simulate many other fluid flow conditions.  These include porous media, pipe flows, aerodynamics, micro components, and fluid-thermal-structural interactions, to name a few.  For more information on our modeling capabilities and how we can help meet your project needs please visit our website.

Reference 1:  D’Amato, R., Fanning, W. R., “Pressure Distributions on Sphere-Cone Radomes in Uniform and Gradient Flows,” Techincal Note, ESD-TR-68-84, Lincoln Library, MIT, 1968.