Learning Forum

[1993saurav]

So, I am modeling the FSI in an Aortic valve which serves the unidirectional blood flow. This valve has leaflets, which open when the inlet pressure is more than the outlet pressure and closes at other times. I don't think that more details regarding the valve are actually required. However, I am using 1/6th symmetry model for convenience and pressure inlet and pressure outlet boundary conditions with all the FSI as a wall. I am also using the contact detection feature in dynamic mesh to detect the contact zone in between the fluid contact(which is a wall with specified shear) and leaflet as soon as set tolerance is reached. I believe the contact detection works. I have been using the same value for tolerance as I am using for contact offset in mechanical in transient structural. However after the new cell zone that is created after detection has to basically stop the flow i.e the backflow(just to give a brief into to this backflow- At the start of the simulation, the outlet pressure is higher than the inlet ,hence flow is backwards and since there is a  contact offset given in mechanical that has led to a small gap for that backflow to go. Basically, the valve that's supposed to control the unidirectional flow of blood is acting oppositely).

Coming back to the main topic. the new cell zone that is created is assigned as a porous zone with default viscous resistances(2.11e+ so that it does not allow the flow (backflow). But the problem is that it does not look like the porous zone is stopping the backflow. I may have got the porous zone inputs wrong.  please, I need some guidance with this issue.

[Rob]

Porous zones will add a pressure drop in any/all directions so won't prevent backflow. If backflow is affecting the area of interest move the outlet further away. Also, you don't say whether this is a problem on the first or later cycles: never take data off the first some cycles, wait until the transient monitors are truly periodic otherwise the results may still retain artefacts of the initialisation process. 

[1993saurav]

Thanks for responding, and for the detailed description. Perhaps using the word "backflow" was a bit misleading. Essentially, I'm trying to simulate contact between two surfaces which will stop fluid flow in the newly created porous region. I know this has been done before using a similar method (shown below).

Although not detailed, from what I gathered the users implemented contact detection which created a porous zone after the proximity threshold was reached. This zone served to obstruct flow in this region, which is exactly what I'm trying to do. 

Furthermore, I've seen others use porous zones to block fluid flow as well, although they implemented a UDF to do so (

). I can get the contact detection technique to work correctly, however the porous zone that is created does not obstruct flow as intended. I've already tried changing the porosity coefficient from 1 to 0, which the manual says should have simulated a "solid" zone. 

For reference, here is a picture of my porous zone settings:

Any insight would be greatly appreciated.

 

[Rob]

Porosity is used for solids content for heat transfer: it doesn't do anything to the flow resistance. 

Set the inertial resistance too. Also note, the porous zone will mostly divert flow around the region, depending on the flow field it's not going to completely stop it. 

[1993saurav]

Diverting flow might work for my application , but I'll have to run a test case to be sure. As far as setting the inertial resistance, I'm assuming using excessively large values would divert the flow well enough? 

 

[wcj1n18]

Hi, this is some time ago now. Did you manage to find a solution to your problem of completely shutting off the flow through the section you referred to? Was setting very large inertial resistance values sufficient? Thanksn