I am currently calculating a flow in a heart valve. The flow is considered unsteady, turbulent(k-e model), 3d, compressible(perfect gas law). In inlet, mass flow rate is the function of time. The heart valve(plate) located in flow field(duct) moves(from close to open or from open to close)periodically with time by force actived to valve(plate). I want to regenerate mesh(grid) with time. How can i do it? or what is other method? I needs help. Thank you.

It was done several times using StarCD also with Fluid-Structure-Interaction (www.pallas.de/cispar). Fluent don't has the ability to handle this kind of moving mesh problems as far as i know. Sorry

I don't know about the grid problem but I noticed one other thing... I thought that blood was a non-newtonian liquid and not a perfect gas! How can you use the perfect gas law for blood?

I am sorry. The flow is considered unsteady, turbulent(k-e model), 3d, compressible(perfect gas law)------>It is mistake.

Thank you for reply.

Hi,

As Jonas said, the flow in blood is considered as non-Newtonian INCOMPRESSIBLE flow and in no way a perfect compressible flow.

Farid.

(1). In principle, it is not difficult to create a sequence of mesh as a function of time. Like the computer animation, you do it one at a time for each time step. (2). So, at any time step, when you re-position the valve, you can create a brand new mesh. (3). But in reality, the code must be able to handle the solutions at two different time steps on two different meshes. In other words, the solution is no longer a function of time only. The boundary conditions are also re-located at different places. This makes the problem of moving boundary difficult to handle. (4). You need to know whether the code you are using can handle this type of problems or not. You may want to contact the code vendor first to find out the answer to this problem. (5). I really hope that you are dealing with a simulated flow through a moving valve problem and the related 3-D animation. ( There is no law against that , right? just a joke! )

Hy,

in fluent,a sliding mesh can be defined, but only in these 3 ways:

1. rotational: you give the rotational axis and the rotational speed.

2. translational: you give the translational direction

3. you combine 1. and 2., but in this case, the rotational and translational vector must have the same direction

tom

Hi Choeng-Ryul,

FLUENT 4.5 has a moving/deforming mesh capability. It employs a method that is similar to the ones used in other commercial codes. There's a tutorial problem in the manual.

And, moving/deforming mesh is coming soon in FLUENT 5+. We are making a fast progress, and I believe you will like it when it gets delivered. It's not a good idea to sell something we don't have right now. But I can tell you one thing. With the moving/deforming mesh in FLUENT 5+, you won't need to worry about the mesh. The code will take all the mesh motion for you.

Thank you for help. everyone

If you have some advice, tell me.

Thank you.

Hello! Joern Beilke.

I can't to contact you through your e-mail(Joern Beilke <joern@beilke-cfd.de>). Confirm your address and tell me it.

I am Choeng-Ryul, Choi and live in seoul, south korea. I am candidate to Ph. Dr. in Dept. of mech. eng., Kyung-Hee Univ.

I needs help.

I am currently calculating a flow in a heart valve.

I posted about "grid regeneration in FLUENT 5.0" in "http://www.cfd-online.com".

I read your answer. Thank you for your advice.

I visited in "http://www.pallas.de/cispar" and saw "Bending Flap Test Case" example.

I am interesting it because it is similar to my problem.

If you performed it, i want to know method that you used.

I want to discuss and to have advice about my problem.

Tell me your advice by means of e-mail.

Thank you for your consideration again.

I will wait your mail.

I hope to see you again in e-mail.