Learning Forum

[amirmohammad]

Dear friends,

I'm working on the validation of an FIV case, which for making phase difference between force and displacement, I used an initial velocity in the structural, but after several coupling step, the drag force becomes constant, therefore the structure doesn't fluctuate, the drag force is more than the amount mentioned in the original case which shows vibration in the structure.

any suggestion will be appreciated

 

 

[Rob]

What does the flow field & CFD mesh look like?

[amirmohammad]

Dear Sir,

thank you for the response, as you can see in the attached photos I used PRISM element for making structured mesh in the coupled face, in this case, we have cross-flow and fluid is water, the turbulence model is K-epsilon and my Y+ is 54.

thanks in advance for your help

 

 

[Rob]

And the flow field?

[amirmohammad]

hope it would be useful for you

I

[Rob]

Focus around the centre obstruction: what does the velocity field look like on a plane?

[amirmohammad]

if your mean is meshing I doubled it but nothing changes.

in the viscous model, I checked EWT, do you think checking pressure gradient effects can affect anyhow. as you can see pressure difference around tube is high.      

[Rob]

OK, so the geometry is as expected. Now look at a plane of constant z through the inlet/outlet centreline. Focus on eddy shedding. 

As an aside, what licence are you using? Ie home or Uni? 

[amirmohammad]

do you think the turbulence model might be the cause???

I'm using the university CFD center.

[Rob]

Not sure it's the turbulence model to blame: read up on eddy shedding as I suggested. I think you've found an equilibrium condition. 

[amirmohammad]

in the article which I'm working on it, the elements which have been transferred from structural are velocity and displacement, as you know in system coupling only the displacement can be transferred, what do you think about that ????

[Rob]

It's a displacement with a time, that should give velocity?

[amirmohammad]

I understand what you're saying, but the point is, the velocity in two mediums is not the same in one step.

in several articles I reviewed, all used CFX for solving the CFD part, is there any reason for that ???

 

[Rob]

No reason to use CFX over Fluent, the preference may be down to the user . 

Velocity in the tube may not have much effect on the tube shape if the outer flow dominates: why you aren't seeing eddy shedding is a bigger question. 

[amirmohammad]

are you saying the mesh is the problem???

the issue I just come up with is at the first I run my structural separately and I see major displacement, acceleration, ... but in system coupling, I run the same condition and the fluid doesn't strike the tube yet but the solution is not the same.