Experimental study of a new screen for turbulence attenuation at higher Reynold numbers

The purpose of this research effort was to perform a preliminary investigation into the nature of the flow field generated by a cylindrical rod screen model, modified with a helical wrap, in a steady , axial cross flow. Testing was conducted in a buildup in test model complexity starting with single rods, then with single grid test articles (75% porosity) and finally with single screens (56% porosity). Identical models without the helical wrap modification were used to provide the baseline downstream flow fields for direct comparison purposes. First flow visualization using a laser/dye technique was conducted to better understand the complicated flow structure behind the baseline and modified test articles. Initially, the screen’s primary component (ie. single cylindrical rod) was evaluated using various helical wrap patterns in an attempt to optimize the wrap design for turbulence reduction manipulation. Minimized coherent structure in the shedding vortices behind the cylinders was the optimization criteria dictating the final wrap selection. The 1 inch helical wrap (double) was ultimately selected based on the flow visualization results. Next the single wrap modification was evaluated with the single grid configuration at a Reynolds number of 500. The downstream flow fields were then characterized at three axial locations (20,100 and 192 rod diameters) downstream of the baseline and modified single screens. Axial and lateral flow velocities were measured using a single component hot film anemometer at two distinct Reynold numbers (75 and 600). Statistical analysis of the gathered axial velocities, turbulence intensities and velocity dynamics (frequency spectra) was used to formulate a direct preliminary comparison between the test models at these flow conditions. Qualitative results from the flow visualization showed dramatic reductions in the organized flow structure and coherence behind the modified test articles. The quantitative results, however, were less definitive and not as conclusive. Low Reynolds number (75) data were invalidated due to unmanageable temperature gradients in the water tunnel created by severe environmental temperature fluctuations during testing. The high Reynolds number (600) data revealed the same general trends observed during single grid testing, however these trends were less pronounced. The non-uniform nature of the hand modified test articles and the associated flow field impacts were unknown and potentially significant. Non-the-less, repeatable trends supporting the flow visualization were documented. As a result, a basis for future follow-on research has been established by demonstrating potential reductions in organized flow structures and coherence behind screens using this helical wrap modification.