The effect of boundary layer thickness on the wake of a blunt trailing edge body

The flow around blunt trailing edge (BTE) airfoils is an active area of research because of the superior aerodynamic and structural characteristics of the airfoil compared to sharp trailing edge airfoils (Standish and van Dam,... [ view full abstract ]

The flow around blunt trailing edge (BTE) airfoils is an active area of research because of the superior aerodynamic and structural characteristics of the airfoil compared to sharp trailing edge airfoils (Standish and van Dam, 2003). The dominant characteristic of the wake of a BTE body is a vortex street, which causes fluctuating aerodynamic forces (Williamson, 1996). This is associated with a decrease in base pressure and higher drag. It was shown by Bearman (1965) that the base pressure can recover by increasing the length of the vortex formation region. The vortex formation region can be modified by 1) inhibiting wake vortex interactions, 2) breaking vortex interactions via the leveraging of secondary streamwise instabilities that connects the shedding vortices, and 3) modifying the boundary layer (Durgesh et al., 2013). This study experimentally investigates the effect of the thickness of a turbulent boundary layer on the wake of a BTE body, controlled by changing the chord length, c. This contrasts previous studies where changes in the boundary layer thickness is achieved via a change in flow speed, which also changes the Reynolds number based on trailing edge thickness, d. The present study isolates the effect of boundary layer thickness from those associated with Re_d. The motivation for this study is to serve as a stepping stone for future experiments in the control of BTE body wakes.

The experiments were performed inside the UTIAS closed-circuit wind tunnel (Hearst and Lavoie, 2014), at Re_c between 3.0•10^5 and 4.5•10^6. The thickness of the model, d, is 25.4 mm, and the span is 31d. Measurements of the flow were made with hot wire anemometry. Sample profiles of the boundary layer at Re_d = 10,000 are shown in Figure 1.

The extent of the vortex formation region, L_f, was identified as the point where velocity fluctuations in the wake centerline were the highest. The variation of L_f with the non-dimensional displacement thickness, δ*/d, of the boundary layer 0.5d before the trailing edge is shown in Figure 3. L_f is found to increase with δ*/d. This result is consistent with the observations of researchers such as Rowe et al., 2000, and Mariotti and Buresti, 2013. Rowe et al., 2001 attributed this observation to the longer time it takes for vorticity from thicker boundary layers to be carried across the wake when the boundary layers are effectively farther apart. Additional evidence of this is presented in Figure 3, which plots the Strouhal number, St_d = f_sU∞/d, where f_s is the shedding frequency and U_∞ is the freestream velocity, against δ*/d. Figure 3 confirms that the vortex shedding frequency does decrease as the boundary layer thickness increases.

The final paper will also discuss the effect of boundary layer thickness on the spanwise wavelength of the secondary streamwise instability in the wake, which has implications on control techniques based on breaking vortex interactions.

Fig. 1. Boundary layer profiles at 0.5d before the trailing edge for Re_d = 10,000.

Fig. 2. Variation of L_f/d with the boundary layer displacement thickness, δ*/d.

Fig. 3. Variation of St_d with the boundary layer displacement thickness, δ*/d.