[Mina_Shahi]

Originally Posted by ghorrocks

What effect do you expect the solid region to have on the pressure waves? Is this an FSI simulation?

Actually the aim is to get the temperature gradient within the solid.
By modeling the solid you may have slightly less pressure oscilations since you are modeling the heat loss. But as i told slightly not 10 times less!!!!
No it is not fsi simulation, i am solving both fluid and solid in cfx and solving them in the same time!

[ghorrocks]

Obviously there is some problem with the way the solid is being coupled to the flow. I would check that the heat transfer between the fluid and solid is correctly modelled. But it is hard to be sure without details of the simulation - and even if you provided details of the simulation I doubt I will have time to look at them. This is a development problem which you are going to have to sort out.

But playing devil's advocate: The time scales of the fluid flow and the pressure variations are very fast compared to the soid time scales in most cases. Then you can replace the coupled simulation with independant fluid and solid simulations, with the coupling replaced with time averaged conditions at the interface. This would greatly simplify this simulation. Can this be done in this case?

[Mina_Shahi]

Originally Posted by ghorrocks

Obviously there is some problem with the way the solid is being coupled to the flow. I would check that the heat transfer between the fluid and solid is correctly modelled. But it is hard to be sure without details of the simulation - and even if you provided details of the simulation I doubt I will have time to look at them. This is a development problem which you are going to have to sort out.

But playing devil's advocate: The time scales of the fluid flow and the pressure variations are very fast compared to the soid time scales in most cases. Then you can replace the coupled simulation with independant fluid and solid simulations, with the coupling replaced with time averaged conditions at the interface. This would greatly simplify this simulation. Can this be done in this case?

thanks for your answer!
in the interface i used conservative heat flux and that is it !!! nothing else to change!
however i tried another way. usually i first do steady state simulation and then initilized transient case with the steady result.
as i told you before with modeling Just fluid domain i had resoanable pressure signal, so this time i initialized the CHT with this transient result and for the solid i just set 400 K. now suprizely after 1400 iteration pressure amplitude is around 2500 which is good! the only problem is the solid part which shows just 1 degree change within the thickness; i think this value was not ok for initilization. and i am not clear what to set becasue whatever value i set it remaines the same for exxample if i set 700 it changes between 700-710
if i set 400 it changes 400-40.5 !
so i am confused, i suppose even if 400 k is not the finite answer it should increases to reach to other value like 700! but i doesn't.

[ghorrocks]

Are you trying to model the periodic steady state result (ie the result after many pressure cycles) or do you want the start up transient?

[Mina_Shahi]

Originally Posted by ghorrocks

Are you trying to model the periodic steady state result (ie the result after many pressure cycles) or do you want the start up transient?

I didn't get what you said; i want to initilized a transient calculation so i used the transient data (last time step saved data) from the model without solid to intilized the CHT case.

[ghorrocks]

Is what happens in the first cycle important? Or just what the flow settles down to after a lot of cycles?

But have you considered my comment in post #22? What dos the solid make any difference anyway? If it makes no difference then why bother modelling it?

[mortazavi]

Hi,
I am simulating the heat transfer between a flow and thin solid. As you know mesh in the solid is problematic in the case of very thin wall. I would like to know if there is any way in CFX to treat the interface as a wall, as in Fluent where we can apply the solid properties (thickness and material) on a surface without the solid domain. thanks