CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Dynamic mesh motion using CFX (https://www.cfd-online.com/Forums/cfx/25887-dynamic-mesh-motion-using-cfx.html)

Anant June 2, 2008 11:49

Dynamic mesh motion using CFX
 
Hello All,

I would like to know how to set up the dynamic mesh motion in CFX (for hexa and/or tetra elements, with birth /death of mesh cells)

Thanking you

Anant

Mehul June 3, 2008 08:01

Re: Dynamic mesh motion using CFX
 
Anant,

Are you setting up for rotor stator application?

Anant June 3, 2008 08:35

Re: Dynamic mesh motion using CFX
 
Hello Mehul,

thanks for your reply, no , its not for the rotor stator application,

its like valve movement, I need to use adding and deleting cells (cell birth and death type) on the either side of the valve as it moves near & away from wall.

Do you have any method to use the dynamic mesh motion..? for the problem stated--?

please let me know

Thanking you Anant


Benjamon June 4, 2008 04:24

Re: Dynamic mesh motion using CFX
 
As far as I know, CFX do not allow you to use remeshing (add and delete cells). The only option in dynamic mesh is referred as "smoothing": moving the cell nodes as if there were connected with springs. This is only valid for small displacsments (if not high skew cells are obtained).

I was said that CFX will get the Fluent remeshing capability in the release 13. We will have to wait 2 or 3 years...

mohammad February 27, 2014 07:28

How to simulate a piston???
 
Hi everyone.
I am going to simulate a piston-cylinder assembly. the Diameter of the piston is 47 mm, stroke 27 mm and the frequency is 120Hz.

How can I simulate the model? Because I heard that dynamic mesh is only for small amplitude vibration, because a dense mesh is required inside the zone in which the movement takes place.

Thanks in advance

ghorrocks February 27, 2014 08:17

Dynamic mesh in CFX can most certainly handle piston-cylinder motion. I have used it in hundreds of models. A hint: the motion is easier if you start the model at TDC and expand the mesh, rather than compressing it from BDC. But with a bit of tweaking of the mesh motion parameters you can make it work either way.

Also dynamic remeshing can help too.

Musaab Kadem November 4, 2016 00:36

Hello everybody
I'm working on rotating disk apparatus to simulate the external flow using Ansys software. This apparatus is consist of a fluid container and a rotating disk immersed in the fluid. I have two questions, I appreciate if anyone can answer.
1. In my simulation is it better to use two domains (one is fluid domain and the other is immersed solid domain), or I have to use one fluid domain and I subtract the solid disk from it.
2. Is it important to use dynamic mesh for the rotating disk or the fluid mesh.

Thank you very much

ghorrocks November 4, 2016 00:41

There are many different ways to model this and the best method depends on exactly what you are modelling. You will need to show an image of what you are modelling and a more complete description for us to help you.

Musaab Kadem November 4, 2016 01:48

1 Attachment(s)
Thank you very much dear for your replying.
The attached picture shows the simulation geometry with one structured disk.
my working is to find the torque values of smooth disk and compare it with different structured disks to investigate the drag reduction behavior with riblets. These structured disks consist of four different groove types. After that the velocity distribution above the smooth disk will be investigated and compared with the structured one.

ghorrocks November 4, 2016 01:53

Labelling the image would help.....

Which axis does the disc spin about?

It appears you can model this with a simple tangential velocity on the wall, with no need for immersed solids or anything complex.

Musaab Kadem November 4, 2016 04:07

1 Attachment(s)
The disk spin about Z axis as shown in attached picture.

ghorrocks November 4, 2016 05:15

I assume the axis of rotation is the centre of it. In that case then my previous comment is correct - you do not need rigid bodies, immersed solids, rotating frames of reference or anything like that for this. This model can be done with a single fixed domain and a tangential velocity on the rotating component. This is far easier and much quicker and more accurate.

Musaab Kadem November 5, 2016 09:36

Thank you very much sir for your help I really appreciat.
I did both cases one with two domain and with one domain as you advice me. But in both case I cant get the valid results with my expermental work. Where i did all the runs experimently and the first part in my simulation is to valide the experimental torque with the simulation one.
So please can you advice me how to make this validation or what are the parameters which effect the torque value in CFX simulation except the meshing.

ghorrocks November 6, 2016 05:39

FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F


All times are GMT -4. The time now is 10:58.