Flow and Heat Transfer in Tubes with Obstacles
The objective of this research work is to enhance the understanding of heat transfer and pressure loss in heated tubes equipped with flow obstacles by experimentally investigating the axial and circumferential distributions of convective heat transfer in a heated tube, complemented by pressure loss...
Main Author: | |
---|---|
Other Authors: | |
Language: | en |
Published: |
Université d'Ottawa / University of Ottawa
2017
|
Subjects: | |
Online Access: | http://hdl.handle.net/10393/36514 http://dx.doi.org/10.20381/ruor-20794 |
Summary: | The objective of this research work is to enhance the understanding of heat transfer and pressure loss in heated tubes equipped with flow obstacles by experimentally investigating the axial and circumferential distributions of convective heat transfer in a heated tube, complemented by pressure loss and velocity measurements in an adiabatic pipe flow. The heat transfer experiments employed refrigerant R-134a as the working fluid with a Reynolds number range of 14,000 to 97,000. Three types of flow obstructions were investigated: eccentric cylinders with flat and rounded ends and annular obstacles, each having a flow blockage of either 0.15 or 0.3. The axial distribution of heat transfer coefficient was measured downstream from the downstream end of the obstruction over a distance of 3 to 70 tube diameters. The experimental data indicate that heat transfer augmentation downstream from the flow obstructions depends on the obstructed area, the flow Reynolds number, the distance from the flow blockage and, to a lesser extent, the shape and the circumferential location of the obstruction. Our experiments confirm the previous findings that heat transfer augmentation (compared to the bare tube heat transfer case) decreases with an increase of flow Reynolds number. It was found that heat transfer augmentation typically extends up to 30 diameters downstream of a flow obstacle. An improved prediction method that correlates the obstructed flow area, Re number and the distance from the trailing edge of the obstacle has been derived.
Pressure loss and velocity measurements were also collected for a flow Reynolds number range from 11,000 to 65,000, for flat ended (blunt) and rounded cylinders with a flow blockage ratio of 0.3 and a blunt cylinder with a flow blockage ratio of 0.15. The results showed that blockage ratio and shape of flow obstacle affect the obstacle pressure loss coefficient significantly and they confirm previous research findings that obstacle pressure loss coefficient decreases with an increase of bulk Reynolds number. Measurements of the reattachment length downstream from flow obstacles indicated that the reattachment length for three-dimensional turbulent flow around square-shaped cylinders was significantly shorter than two-dimensional flow over a backward-facing step. An important finding of the current investigation is that, for the flow range investigated, heat transfer augmentation could not be correlated with the local pressure loss coefficient of the obstruction, which differs from the smooth heated channel case where the Reynolds analogy usually applies. Additionally, to assess the capabilities of the widely used k- turbulence model, some CFD simulations were performed. The CFD results were generally in satisfactory agreement with the experimental data; however, near the obstacle, close to the separation and recirculation areas, the agreement with the experimental data was less satisfactory.
The current research can be applied to the design and optimization of spacers and appendages of nuclear fuel elements, as well as serve for the improvement of state of the art computer codes employed in the safety assessment of nuclear reactors. |
---|