Solving a problem
 

Hello everyone. My professor assigned me a problem. Therefore, here is the problem:

http://imageshack.us/a/img89/1725/ge...squareprob.png

Flow flows from -x to x. The simulation runs in 2d. The dimension are indicated above. The grid size = 0.5 cm. There are five parts on this geometry which were INLET (the left side), OUTLET (the right side), WALL (the top and bottom sides), FLUID, and SQUARE (the geometric shape in the middle)

Fluent Setup:
General: The simulation was started as steady first, and after 20 iterations it was run as transient.
Models: Laminar flow, since the Reynolds number was 5000.
Materials: Water
Boundary Conditions: WALL and SQUARE were set as wall. INLET was set as velocity-inlet. The OUTLET was set as pressure-outlet.
Monitors: Residual Monitors: Residual Criteria was set to 10E-6.
Solution Methods: Transient Formulation first was set as Order Implicit, then was set as Second Order Implicit.
Run Calculation: Time Step Size was set as 0.01. Number of Time Steps were set as 100. Reporting Interval was set as 0.2. Max Iterations were set as 5000.
The rest of the results were set as default.

This is the residual plot that I received.

http://imageshack.us/a/img594/4240/r...orsquarepr.png

The vector plot that I was received.

http://imageshack.us/a/img546/5533/v...yplotforsq.png

As it can be seen from the vector plot, there was no wake behind the square. This proves that result is not valid. Therefore, could you help me to get a better result on this assignment? Thanks.

http://www.cfd-online.com/Forums/dat...BJRU5ErkJggg==

As I think, you had mainly a problem with the computational grid. The computational cell size is too large (0.5 cm), it must be at least in 10 times smaller, especially in the region near to the wall surface.

My prof. said that grid resolutions are fine.