[smhk]

Hello,
I have some questions about how (hydrostatic) pressure is calculated in Ansys CFX.
In Ansys CFX theory guide momentum equation is written without the gravitational term (but there is a source term). I can activate buoyancy in domain settings but I'm not sure how that works.

• Why we don't always turn gravity term on (like what it is in Navier-Stokes equation)?

• Is the buoyancy model work only for multi-phase flows, buoyancy driven flows or you can use it for every other simulation like a Kaplan turbine or a hydrostatic case?

• Do I need to activate buoyancy model in solving a water turbine model?(I know by using mass flow rate I somehow apply the effect of gravitational driving force for the turbine but I'm not sure shall I use buoyancy or not).

I also tried a simple case. I got these results which are weird to me.
Here is a simple 1m*1m*3m channel and blue arrows show mass flow at the inlet:
In the first experiment I didn't activate buoyancy. The relative pressure is symmetric. you can't see the hydrostatic pressure in relative pressure contour. it's symmetric:




The absolute pressure is also symmetric with no difference on any axis:








I wanted to see how buoyancy model works with this case so I activated buoyancy model in the domain setting:




when buoyancy reference temperature is the same as fluid reference temperature (in materials), I get this results for relative pressure and absolute pressure respectively:
you can't see the hydrostatic pressure in the relative pressure:





But in absolute pressure, hydrostatic pressure share is obvious the difference in pressure along Y axis must be about 10000Pa (g=10 m/s^2 & rho=1000 kg/m^3)







when buoyancy reference temperature is different from fluid reference temperature , I get this results for relative pressure and absolute pressure respectively:
In relative pressure contour there is pressure gradient in along Y but it's not 10000Pa it's only about 700Pa



but in absolute pressure contours, hydrostatic pressure seems correct and it's difference is about 10000Pa

why this happens?


Thank you

[Opaque]

Originally Posted by smhk

Hello,
I have some questions about how (hydrostatic) pressure is calculated in Ansys CFX.
In Ansys CFX theory guide momentum equation is written without the gravitational term (but there is a source term). I can activate buoyancy in domain settings but I'm not sure how that works.

• Why we don't always turn gravity term on (like what it is in Navier-Stokes equation)?

• Is the buoyancy model work only for multi-phase flows, buoyancy driven flows or you can use it for every other simulation like a Kaplan turbine or a hydrostatic case?

• Do I need to activate buoyancy model in solving a water turbine model?(I know by using mass flow rate I somehow apply the effect of gravitational driving force for the turbine but I'm not sure shall I use buoyancy or not).

I also tried a simple case. I got these results which are weird to me.
Here is a simple 1m*1m*3m channel and blue arrows show mass flow at the inlet:
In the first experiment I didn't activate buoyancy. The relative pressure is symmetric. you can't see the hydrostatic pressure in relative pressure contour. it's symmetric:





The absolute pressure is also symmetric with no difference on any axis:





I wanted to see how buoyancy model works with this case so I activated buoyancy model in the domain setting:






when buoyancy reference temperature is the same as fluid reference temperature (in materials), I get this results for relative pressure and absolute pressure respectively:
you can't see the hydrostatic pressure in the relative pressure:






But in absolute pressure, hydrostatic pressure share is obvious the difference in pressure along Y axis must be about 10000Pa (g=10 m/s^2 & rho=1000 kg/m^3)







when buoyancy reference temperature is different from fluid reference temperature , I get this results for relative pressure and absolute pressure respectively:
In relative pressure contour there is pressure gradient in along Y but it's not 10000Pa it's only about 700Pa



but in absolute pressure contours, hydrostatic pressure seems correct and it's difference is about 10000Pa

why this happens?


Thank you

Absolute Pressure = (Pressure + Reference Pressure) + Reference Density * gravity * (g dot local_position vector - local reference position)

Absolute pressure gradient = pressure gradient + Reference density * gravity * gradient *( g dot local position - local reference position)

Hope that helps

[smhk]

Thank you for replying.

Absolute Pressure = Pressure + density * gravity * (g dot local_position vector - local reference position)

Is this the case when I use buoyancy model?
Why when I change the reference temperature (in buoyancy model) to some value other than temperature in my material properties the relative pressure becomes incorrect.
When Should I active buoyancy model?

[Opaque]

Originally Posted by smhk

Thank you for replying.



Is this the case when I use buoyancy model?
Why when I change the reference temperature (in buoyancy model) to some value other than temperature in my material properties the relative pressure becomes incorrect.
When Should I active buoyancy model?

That was the simplified version (which I updated to match the documentation)

For compressible, multiphase flow, the full buoyancy model is used. That is, the "(Density - Reference Density) * gravity is evaluated directly.

Have you read the documentation? Both sections: Theory Guide, and Solver Modeling Guide