CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   query about domain and mesh motion of tube carrying fluid (https://www.cfd-online.com/Forums/cfx/93759-query-about-domain-mesh-motion-tube-carrying-fluid.html)

hasan56 October 25, 2011 23:22

query about domain and mesh motion of tube carrying fluid
 
1 Attachment(s)
Hello ,

I am trying to find the fluid flow in coronary artery which have a periodic motion also . I have attached the model designed in ICEM CFD . The basic concept is I want to keep the inlet boundary stationary while the Outlet and cylindrical wall will move in a periodic motion that will bend the artery back and forth with respect to time just like bending a tube having one end fixed . The motion is in XY plane only .
My question is .....does ansys allow this kind of simulation? as i have meshed only the volume inside the artery and all the boundary surfaces but nothing outside the model was meshed as fluid domain is inside the artery volume . So as the model is moving along with the mesh ( I have defined mesh motion specified displacement with CEL expression for WALL and OUTLET) so does ansys automatically support this kind of mesh movement fro flow occuring inside the tube?or do i need to define another meshed domain outside the artery for flow solution?
I tried one simulation but it gave negative volume element found and terminated. was that because of the inlet stationary while wall was moving which could have generated negative volume element?

For convenience I have attached the picture

Thanx for your attention.

Hasan56

ghorrocks October 26, 2011 07:24

This sort of model should be possible. Does not sound too difficult. You will need to specify the motion of the thing, and as long as the motion is not too severe you should not have neg vol elements.

Are you sure you want to move the outlet? It is difficult to specify a physically realistic boundary condition on a moving surface. You probably want to move the boundary downstream and make it fixed.

hasan56 October 26, 2011 12:35

Thanx a lot for your reply.
So , it is not required to mesh anything outside the domain for this problem though the tube is moving and whole domain is changing geometry, right ?
The frequency of motion is high (1 Hz apprx.) and radius of curvature reduces 4cm from initial one ( around 7 cm) . But problem is i have to give this motion to the whole domain and boundaries except the inlet . Do u think is there any alternative idea to get this kind of motion to the geometry with the displacement function that I have with respect to time?

ghorrocks October 26, 2011 16:53

Quote:

it is not required to mesh anything outside the domain for this problem though the tube is moving and whole domain is changing geometry, right ?
Correct.

Quote:

The frequency of motion is high (1 Hz apprx.)
:) it is interesting to see people's value judgement on things like this. I work in MEMS and our firing event is all finished by about 2 microseconds. 1Hz is an eternity.

Quote:

Do u think is there any alternative idea to get this kind of motion to the geometry with the displacement function that I have with respect to time?
I do not understand this question.


All times are GMT -4. The time now is 16:46.