Immersed solid, vs solid-fluid interaction to determine forces on a submerged solid
 

Hello all,

I've been using Ansys 13 CFX to try to determine the forces exerted on a fixed, stationary device that's inserted into a pipe with water flowing through. Up till now, I've been building the (steady-state) simulations as follows:

• Modelling the pipe's interior volume and the device in Solidworks.
• Subtracting the device's volume from the pipe model, thus giving the actual physical space occupied by the fluid.
• Creating an assembly from those two parts and imported it into CFX.
• Defining the fluid domain by specifying the location and material.
• Defining the boundary conditions for the fluid domain (inlet = 8bar and outlet = 25kg/s).
• Defining the solid domain by specifying the location, domain type (solid domain) and material.
• Creating the fluid solid interface by highlighting the faces that are common between the fluid and solid domains for both domains.

I've then run the simulation and used the function calculator to work out the force on the fluid-solid interface side that's in the fluid domain, which gives me a force of about 12.5N in the direction of the fluid flow. This I am taking to be the force exerted on the device due to the flow.

I've since discovered that rather than defining the device as a solid domain, I can define it as an immersed solid and perform the simulation that way. So, I've rebuilt the model as follows:

• Created an assembly consisting of the device inside the pipe interior volume. However, unlike the previous simulation, I did not hollow out the shape of the device from the pipe volume.
• Defined the fluid domain by specifying the location and material
• Defined the boundary conditions for the fluid domain (inlet = 8bar and outlet = 25kg/s).
• Defined the solid domain by specifying the location and setting the domain type to Immersed Solid, with the domain motion being stationary.

I've since run the simulation and used the function calculator to work out the force on the solid domain, but this now gives me a force of 2.5N as opposed to 12.5N

I'm not really sure which value is the correct one, or whether I'm using the right simulation method and building the simulations right.

Any thoughts?

Thanks,