Date of Award


Publication Type

Doctoral Thesis

Degree Name



Mechanical, Automotive, and Materials Engineering

First Advisor

David Ting



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


Many engineering systems involve proper transfer of heat to operate. As such, augmenting the heat transfer rate can lead to performance improvement of systems such as heat exchangers and solar photovoltaics panels. Among the many existing and studied heat transfer enhancement techniques, a well-designed passive turbulence generator is a simple and potent approach to augmenting convective heat transfer. Two of the most recognized passive convective heat enhancers are wings and winglets. Their potency is attributed to the long-lasting induced longitudinal vortices which are effective in scooping and mixing hot and cold fluids. Somewhat less studied are flexible turbulence generators, which could further the heat transfer enhancement compared to their rigid counterpart. In the current study, the flexible rectangular strips are proposed, marrying the long-lasting vortex streets with the periodic oscillation, to maximize heat convection. This study was conducted in a closed-looped wind tunnel with 76 cm square cross-section. The effects of flexible strips on the turbulent flow characteristics and the resulting convective heat transfer enhancement from a heated flat surface are detailed in four papers which are presented as Chapters 3, 4, 5 and 6. In Chapter 3, the effect of the thickness of the strip is detailed. The 12.7 mm wide and 38.1 mm tall rectangular strip was cut from an aluminum sheet with thickness of 0.1, 0.2 and 0.25 mm. The incoming wind velocity was maintained at around 10 m/s, giving a Reynolds number based on the strip width of 8500. It is observed that the thinnest 0.1 mm strip could induce a larger downwash velocity and a stronger Strouhal fluctuation at 3H (strip height) downstream, leading to a better heat transfer enhancement. The peak of the normalized Nusselt number (Nu/Nu0) at 3H downstream of the 0.1 mm strip was around 1.67, approximately 0.1 larger than that of the 0.25 mm strip. In Chapter 4, the height effect of the strip is disclosed. The strip was 12.7 mm wide and 0.1 mm thick, with a height of 25.4 mm, 38.1 mm and 50.8 mm. The Reynolds number in this chapter was also fixed at around 8500, based on the strip width and the freestream velocity. It was found that the shortest, 25.4 mm strip could induce the closest-to-wall swirling vortices, and the largest near-surface downwash velocity toward the heated surface. Thus, the largest heat transfer augmentation was observed. At 9W (strip width) downstream, the 25.4 mm-strip provided the Nu/Nu0 peak of around 1.76, 0.26 larger than that associated with the tallest, 50.8 mm-strip. In Chapter 5, the effect of the transversal space of a pair of strips is expounded. A pair of 0.1 mm thick, 12.7 mm wide, and 25.4 mm tall aluminum rectangular flexible strips was placed side-by-side with a spacing of 1W (strip width), 2W and 3W. The Reynolds number based on the strip width was around 8500. The results showed that the 1W-spaced strip pair induced the strongest vortex-vortex interaction, the largest downwash velocity, and the most intense turbulence fluctuation. These resulted in the most effective heat convection. At Y=0 (middle of the strip pair) and X=9W, the largest Nu/Nu0 value of around 1.50 was identified when using the 1W-spaced strip pair. This was approximately 0.24 and 0.33 larger than that of the 2W- and 3W-spaced strip pairs. Chapter 6 presents the effect of freestream turbulence on the flat plate heat convection enhancement with a 12.7 mm wide, 25.4 mm tall and 0.1 mm thick flexible strip. A 6 mm thick sharp-edged orificed perforated plate (OPP) with holes of 38.1 mm diameter (D) was placed at 10D, 13D and 16D upstream of the strip to generate the desirable levels of freestream turbulence. The corresponding streamwise freestream turbulence intensity at the strip was around 11%, 9% and 7%. The Reynolds number based on the strip width and freestream velocity was approximately 6000. The freestream turbulence was found to diminish the effect of flexible strip in terms of the relative heat transfer enhancement (Nu/Nu0). This is due to the significant increase of Nu0 with the increasing freestream turbulence. In other words, the flexible strip could always improve the heat transfer, and the relative improvement is greatest for the largely laminar freestream case in the absence of the OPP. Chapter 7 summarizes the effect of all the parameters in previous chapters on the convective heat transfer enhancement. The results show that the freestream turbulence intensity (Tu) had the most significant effect in augmenting the averaged Nu/Nu0, and the local Nu/Nu0 correlated best with the local ke. The maximal averaged Nu/Nu0 over 23W downstream, within ±1 and ±4 strip widths cross-stream was found for Tu=7% case and Tu=11% case, respectively. Conclusions are drawn and recommendations are provided in Chapter 8.

Available for download on Wednesday, March 02, 2022