Latest revision as of 12:06, 12 February 2017

Turbulent Flow Past Two-Body Configurations

Flows Around Bodies

Underlying Flow Regime 2-12

Description

Introduction

The UFR in question may be considered of relevance to the Application Challenge AC 1–08 (L1T2 3 element airfoil) and, to some extent, to AC 4–01 (wind environment around an airport terminal building). It is characterized by several common features which include separation of turbulent boundary layer from the upstream body, free shear layer roll-up, and interaction of unsteady wake forming in the gap between bodies with the downstream one. Capturing these features is known to be a serious physical (in terms of a proper turbulence representation) and computational (in terms of numerics and computer resources) challenge, which motivates systematic thorough studies aimed at an objective evaluation of capability of existing modelling approaches/numerical techniques to predict them with acceptable accuracy.

A purpose of this document is to provide a summary of findings of such a study undertaken in the framework of the EU Research Collaborative Project "Advanced Turbulence simulation for Aerodynamic Application Challenges" (ATAAC) [1] for the Tandem Cylinders (TC) configuration as a representative example of the considered UFR. The document is organized as follows.

First, a brief overview of the studies of the TC flow is given with a focus on those carried out in the course of the ATAAC project, whose outcome presents a core of the subsequent analysis. Then, an outline of the NASA experiments [2]-[4] is given, after which a summary of CFD methods used, results of the simulations performed, and their comparison with the experimental data are presented. Finally, in the BPA section, some practical advice on computing the considered UFR is given based on findings of the performed analysis. Note that these findings should be considered not as standalone ones but together with those of the Workshop on Benchmark problems for Airframe Noise Computations (BANC-I) [5] and forthcoming BANC-II Workshops.

Review of UFR studies and choice of test case

Among numerous studies devoted to validation of different CFD tools and turbulence modelling and simulation strategies for the considered UFR, two international comparison exercises can be distinguished as the most systematic and well designed CFD campaigns: The EU ATAAC project and the BANC-I Workshop. Both campaigns covered a wide range of turbulent flows but, as far as the UFR in question is concerned, both relied upon experimental data on the high Reynolds number TC configuration accumulated in the course of experimental studies of NASA Langley Research Center [2–4].

Despite the relatively simple geometry, the flow past TC includes major challenging features of the considered UFR (see Introduction), thus presenting a relevant test case for assessment of CFD capability of providing a reliable prediction of these features. Moreover, the experimental study was specially designed for CFD validation. As a result, the database ideally matches this purpose and provides a solid background for assessment of different CFD approaches. The last but not least argument in favour of choosing the TC configuration is that the experimental data are well documented in publications and, what is also very important, are available in digital form on the workshop website.

The experimental studies were carried out for two configurations with different distances between the centres of the upstream and downstream cylinders of the tandem, L=1.435D and 3.7D (D is the cylinder diameter), and the latter was chosen as a mandatory one within both ATAAC project and BANC Workshop. Note that according to, e.g., [6, 7], the flow pattern observed in the TC configuration strongly depends on the separation distance between the cylinders. Particularly, at L/D < 1.1, the two cylinders behave as a single bluff body (vortex shedding occurs on the rear cylinder only). Then, at 1.1 < L/D < 2.5, the shear layers separating from the front cylinder attach to the rear one, but vortex shedding still takes place only on the rear cylinder. At 2.5 < L/D < 3.2, intermittent shedding is observed also in the gap between the cylinders and at 3.2 <L/D < 3.8 (i.e, within the range the chosen value of L/D belongs to) vortex shedding occurs on both cylinders, the shedding on the front cylinder being tangibly affected by the presence of the rear one. Finally, at L/D > 3.8, vortex shedding occurs on both cylinders with the same characteristics as a single cylinder^[1]

As will become apparent, the tandem cylinders case exhibits strong sensitivity to even subtle details of the computational setup (strong model-, code- and numerics-dependency). As such, this UFR is particularly valuable for the purposes of verification.

↑ Detailed quantitative (obtained with the use of PIV) data on the time-averaged and instantaneous flow structure for several of these flow regimes at Re = 10⁴ are available in [8].

Contributed by: A. Garbaruk, M. Shur and M. Strelets — New Technologies and Services LLC (NTS) and St.-Petersburg State Polytechnic University