Effect on microstructure of boundary layer by disturbing the flow field such as imposing suction, blowing and employing variety forms of surface roughnesses are subjects of most interested research field in heat transfer enhancement. In this context, some experimental techniques are employed by research workers. While the experimental techniques explore overall fluid behavior, the numerical methods unveil details of flow structure, exploring causes to manage advanced techniques for enhancement. In present work, effect of a rectangular rod, it’s aspect ratio and distance from a flat plate on local heat transfer enhancement are studied. The study focuses more on details of flow structure and contribution that each of flow fields may make in enhancement. As a result of insertion of the rod into flow field, a stagnation point, a jet and recirculation zone form. More study, reveals that the stagnation point, while being a cause of local drop in heat transfer, forms a high pressure point driving part of boundary layer flow through the gap between the rod and the flat plate, causing wash away of local turbulent boundary layer and instead grows a new boundary layer inside the gap. As an effect, local heat transfer coefficient (LHTC) increases abruptly to a maximum but falls as new boundary layer thickenes along the gap. After the gap, a vortex forms. Its size is a function of obstacle height as well as strength of the jet flow. The vortex, thougth less enhancement imposes into the heat transfer coeficient compared to the jet, its effects widens along the flat plate. If compared to single flate plate, an overall heat transfer enhancement of more than 40% over the affected area is achieved. In the study, Navier- Stocks equations together with energy equation are solved. To handel turbulence development into the flow field, low Reynolds number Launder-Sharma turbulence model is employed. The model is expected to deal with wake generating flow with good accuracy. All equations are expanded over physical domain using SIMPLE procedure in 2 dimentional physical domain. To maintain accuracy for low Reynolds number turbulent models, fine meshing near the wall is necessary. Four grids are set inside the laminar sublayer, i.e. one on the wall, one at y+ =11 and two more in between these two. Therefore each grid point has a distace of Dy= 1´10-4mm from the other grid in y direction.