• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.1625-1630
• /
• 2004

An experimental study was carried out to investigate influence of flow conditions on a boundary layer to the near-wake of a flat plate. The flow condition in the vicinity of trailing edge that is influenced by upstream condition history is an essential factor that determines the physical characteristics of a near-wake. Various tripping wires were used to change boundary layer flow condition of upstream at the freestream velocity of 6.0 m/sec. Measurements of the boundary layer and near-wake according to the change of upstream conditions were conducted by using both I-probe(55P14 for boundary layer) and X-probe(55P61 for wake). Normalized velocity profiles of the boundary layer were shown the flow types such as laminar boundary layer, transition, and turbulent boundary layer at 0.95C from the leading edge. The velocity and turbulence intensity profiles of the near-wake for the case of laminar boundary layer at the flat plate surface exhibited a defect and a double peak showing perfect symmetry, respectively.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.5
• /
• pp.722-730
• /
• 2001

The local heat transfer of an axisymmetric submerged air jet impinging on a heated flat plate is investigated experimentally with the variation of mesh-screen solidity. The screen installed in front of the nozzle exit modifies the flow structure and local heat transfer characteristics. The mean velocity and turbulence intensity profiles of streamwise velocity component are measured using a hot-wire anemometry. The temperature distribution on the heated flat surface is measured with thermocouples. The smoke-wire flow visualization technique was employed to understand the near-field flow structure qualitatively for different mesh screens. Large-scale toroidal vortices and high turbulence intensity enhance the heat transfer rate in the stagnation region. For a higher solidity, turbulence intensity become higher which increases the local heat transfer at small nozzle-to-plate spacings such as L/D<6. The local and average Nusselt numbers of impinging jet from the $\sigma$(sub)s=0.83 screen at L/D=2 are about 5.6∼7.5% and 7.1% larger than those for the case of no screen, respectively. For the nozzle-to-plate spacings larger than 6, however, the turbulence intensities for all tested screens approach to an asymptotic curve and the mean velocity along the jet centerline decreases monotonically. As the nozzle-to-plat spacing increases for high solidity screens, the heat transfer rate decreases due to the reduction in turbulence intensity and jet momentum.

• Journal of computational fluids engineering
• /
• v.18 no.1
• /
• pp.77-82
• /
• 2013

The error in measuring temperature profiles by thermocouple inside boundary layer mostly comes from the conduction heat transfer of the thermocouple. The error is not negligible when the conductivity of the thermocouple is very high. In this study, the effect of conduction heat transfer of the thermocouple on the temperature profile inside boundary layer was examined by considering both free-stream velocity and a thermocouple position. The conduction error of an E-type thermocouple was investigated by numerical analysis of three-dimensional conjugate heat transfer for various velocity profiles of boundary layer and thermocouple positions.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.10 no.5
• /
• pp.45-57
• /
• 2002

In this paper, the results gained by applying many impingement models to the cylinder and flat plate were analyzed in comparison with the experimental data to study a spray/wall interaction phenomena. To begin with, the behavior of spray injected normal to the wall was analysed using three different impingement models ; Naber and Reitz model(NR model), Watkins and Wang model(WW model) and Park and Watkins model(PW model) in the present calculation. The results obtained from these models were compared with experimental data of Katsura et. al. The results indicated that PW model was in better agreement with experimental data than the NR and WW model. Also f3r spray injected at 30DEG , the result of three models were compared with experimental data of Fujimoto et. al. The results showed that m model overpredicted the penetration in the radial direction because this model was based on the inviscid jet analogy. WW model did not predicted the radius and height of the wall spray effectively. It might be thought that this discrepancy was due to the lack of consideration of spray film velocity occurred at impingement site. The result of PW model agrees with the experimental data as time goes on. In particular, a height of the spray droplets was predicted more closely to the experimental data than the other two models. The results of PW model in which the spray droplets were distributed densely around the edge of droplet distribution shaped in a circle had an agreement with the experimental data of Fujimoto et. al. Therefore, it was concluded that PW model performed better than M and WW model for prediction of spray behavior. The numerical calculation using PW model performed to the cylinder similar to the real shape of DI engine. The results showed that vortex strength near the wall in the cylinder was stronger than that in the case of flat plate. Contrary to the flat plat, an existence of the side wall in the cylinder caused the tangential velocity component to be reduced and the normal velocity component to be increased. The flow tends to rotate to the inside of cylinder going upward to the right side wall of cylinder gradually as time passes. Also, the results showed that as the spray angle increases, the gas velocity distribution and the tumble flow seemed to be formed widely.