Hello everyone, I am working on a flow simulation at the moment, but I am unable to make sense to the results that I am getting.

I want to simulate the flow between two circular discs with an approximate diameter of 1000 mm. The discs are approximately 3 mm apart. The inlet is a 200 mm cylinder in the center of the discs. The outlet is at the very outer edge of the circular discs. Ultimately, the goal of this simulation is to determine the best locations for spacers between the two discs, such that there is minimal variation in fluid velocity along the  outer diameter of the discs. 

In this simulation there are no spacers in the simulation.

Due to the very high aspect ratio between the discs diameter and their spacing, I want to simulate this using a 2D computational domain. 

I am defining a inlet boundary condition using an inlet volume flow. The outlet is defined as a static pressure outlet.

In the image below, the vertical, cylindrical face (parallel to the view in this case) in the center of the model is the inlet. The outlet is the vertical, cylindrical face at the very outer edge.

My expectation is that without any obstacles, the flow should distribute symmetrically about the central axis and the velocity at the outlet should be even. 

As can be seen in the image, in the center this is the case. However, in the bottom left and right corners vortecies appear that significantly impact the velocity distribution.

There are no input parameters that could have an effect on this. All turbulence settings are left at their default. 

I have run similar variations of this simulation multiple times, and always encounter the same issue in this simulation setup. The location of the vorticies changes, but in principle it is the same result. I noticed that increasing the gap between the discs resolves the issue, however this makes little sense to me because the discs both lie outside of the computational domain.

In the past, I have also done some similar simulations using a standard 3D computational domain. In those simulations, I have never encountered such behaviors. This behaviour was also never observed in experimental testing in the past. 

Something I have seen in SW Flow Simulation before, is that the radial flow such as in this scenario results in strange patterns that typically align with the cartesian mesh (e.g. identical radial flow forms square velocity profiles in the cut plot). 

I am using SW Flow Simulation 2021

Does anyone know what the cause of this might be? Is it common for SW Flow Simulation to struggle with radial flow such as this? 

Any help would be appreciated

Flow SimulationCFD SOLIDWORKSSimulation