# Two way FSI Aortic Valve / Blood Flow

 Register Blogs Members List Search Today's Posts Mark Forums Read

 July 22, 2015, 10:24 Two way FSI Aortic Valve / Blood Flow #1 New Member   Philipp Schmidt Join Date: Apr 2015 Posts: 3 Rep Power: 10 Hey all, I want to simulate the blood flow through an aortic valve. I started with modeling a simple tube with a velocity profile at the inlet and a pressure function, depending on the volumetric flow rate and a resistance coefficient at the outlet. The wall could deform in reaction to the pressure changes caused by the bc's. This worked fine. In the next step I tried to implement some simple flaps into the tube. Parameters for the flap material are: Density: 500 kg/m³ Young's Modulus: 0.5 MPa Poisson's Ratio: 0.4 Thickness: 0.5 mm Fluid parameters: Density: 1105 kg/m³ Non-newtonian fluid When only the flaps are allowed to deform (the vessel wall is rigid) the solution also worked. I put an Under Relaxation Factor of 0.005 for the fluid --> solid data transfer in the system coupling tab. If now the vessel wall is also allowed to deform, the solution won't converge under any circumstances i've tried. I modified the URF, tried Solution Stabilization (Dynamic Mesh tab) within fluent, but nothing helped. All I know is, that the model needs to be under relaxated, but the parameters that do this are not clear to me. Also for me it is hard to understand why each problem itself works, but the complete package does not. It would help a lot, if somebody who dealt with similar problems could give me a hint. Best regards

 July 27, 2015, 04:59 #2 New Member   Join Date: Jul 2015 Posts: 4 Rep Power: 9 Hey , I am working on similar type of problem. In my case the valve has following physical properties rho=1000kg/m^3 E=1.5 Mpa poisson`s ratio=0.49 thickness=0.5 mm and blood rho=1000 kg/m^3 viscosity=0.0043 Now when I am using air instead of blood the setup is running without any problems but when I am using blood as a fluid I am getting error saying : Excessive distortion of elements . In my case only the valve is allowed to deform not the vessel.At inlet I have given a sinusoidal velocity pulse having amplitude 0.2m/s and outlet a constant 0 Pa pressure. Do you have any idea how do I solve this problem ? Also how did you come up with value of under relaxation factor ?

 July 27, 2015, 05:26 #3 New Member   Philipp Schmidt Join Date: Apr 2015 Posts: 3 Rep Power: 10 Hi, I got the same error, because the pressure at the interface walls diverged and therefor the forces passed by fluent were way too high. It worked for me when I monitored the pressure at the interface walls and adjusted the factor until it didn't diverge any more. But due to the factor you need many coupling steps within a time step. You should monitor the deformation at the valve tip, so you can see if the deformation is converged.

 October 18, 2016, 14:24 Similar problem #4 New Member   Mazen Abou Gamrah Join Date: Oct 2016 Posts: 4 Rep Power: 8 Hey guys, I am trying to work on a similar problem as you but unfortunately I am stuck with a problem. The leaflets (flaps) in my model are not visible in the cfx setup since they are surface bodies. I gave them a thickness of 1mm in the structural setup and suppressed the fluid volume (tube). In the cfx mesh I suppressed the leaflets and I meshed the fluid volume. After linking the blocks when I open the cfx setup I can not pick the leaflets as interfaces. In fact they just disappear from the model. If you guys can give me any tips on how the CAD should be modified (CATIA) or what I should do in the setup I would be very grateful. Thanks

 Tags blood, fsi, fsi 2-way coupling, valve