[halowine]

Hi everybody!

I have a question about the activation of density effect in Fluent.
I've made a simple test case to show the issue that I have. (https://ufile.io/x7zzb)

My test case is just a square domain of water in 2D with a circle (wall) in the middle. I just want to see the density effect without any flow happening. The gravity should change the static pressure using the well known 10m=1bar. My domain is 100m in 'y', so the pressure should change of 10bars...

When I compute the case, the static pressure changes a bit (a gradient is visible) but the values are completely wrong (~1e-9 Pa). (and the pressure is higher at the top...)

I don't understand how fluent works with gravity... It should be a source term on y-mom of the form: S=rho*g
You can even integrate it by hand: dp/dy=rho*g => p=rho*g*y

There is also 'operating condition' where I could change the 'operating density' but when I do that the solver diverges...

Does someone have made similar simulations ?

Thanks!

[LuckyTran]

Originally Posted by halowine

When I compute the case, the static pressure changes a bit (a gradient is visible) but the values are completely wrong (~1e-9 Pa). (and the pressure is higher at the top...)

All pressures in Fluent are gauge pressures. The values can be completely wrong or completely right depending on what you set the reference pressure to be and what the operating density is set to. You should of course be able to see a 10bar increase/decrease in pressure. However....

If you make a plot of the "static pressure" in Fluent, what you see is not the actual mechanical pressure. The plot that you can has the hydrostatic pressure part removed (the rho*g*h part). The reason for this is because all pressures in Fluent are gauge pressures and when gravitational acceleration is enabled, a modified pressure is solved that includes the hydrostatic contribution.

So, in order to visualize the 10bar change in pressure, you are at a slight loss. What you can do is defined your own custom field function pnew = pstatic + rho*g*h and then make a plot of this new variable and not the built-in Fluent static pressure.

[halowine]

So the pressure during the calculation is correct and this is just a problem of visualization?

If you add a source term in y-mom eq., the pressure should change accordingly... I still don't understand why Fluent does not show the correct pressure ?

And this is not a problem of static or total pressure. In the static pressure, I should be able to see the gradient...

The problem I try to solve uses the buoyancy force. I want to compute my buoyancy force directly from the pressure field (by integrating it at the wall). I know that I can directly compute it from the theory (F=rho*G*V) but I wanted to make the calculation directly from the static pressure field.

Does someone know how to use the "operating density" ? If I set it to 1000 (water) my calculation diverges...

Sorry for these strange questions but it is strange that Fluent is not capable of giving me that solution, no?