Fluid flows upwards despite the fact that gravity is enabled

• 95 Views
• Last Post 12 November 2017
• Topic Is Solved
Ivy.s posted this 06 November 2017

I want to simulate a screw through which cement (with a viscosity of 1000 Pa*s) flows. In the following pdf document you can see what i want to simulate.

My problem is, that the cement flows through the holes of the screw (which is correct) but after that it flows upwards on the outside of the screw. I enabled gravity, but that did not help. I don't know if i used the wrong model for that or what's wrong with my simulation. Please help me.

In the pictures you can see the settings of my simulation, i hope you can help me with this.

phase 1 = air

phase 2 = high viscosity liquid

Attached Files

Ivy.s posted this 06 November 2017

raul.raghav posted this 06 November 2017

Based on your coordinate system, the gravity should be +9.81 m/s^2. And edit gravity in the "General" tab. See screenshot below.

Rahul

peteroznewman posted this 06 November 2017

@Rahul, in the Structural Mechanics side of ANSYS, I use an acceleration of x = -9.81 m/s^2 to create the effect of a gravity force that pulls mass in the +x direction as shown in the last snapshot. I'm not an expert at Fluent, but I would assume it would be the same.

raul.raghav posted this 06 November 2017

@peteroznewman, I'm quite sure in Fluent and CFX its not the same way. The definition of gravity depends on the coordinate system.

Please see the tutorial below and the attached screenshot to get an idea of how gravity is accounted for (under Step 2: Models).

Fluent Tutorial: Using the Eulerian Granular Multiphase Model with Heat Transfer

Rahul

Ivy.s posted this 07 November 2017

I've already tried to Change the gravity into +9.81 but the result is the same. And I also edited it in the "General" tab. I just forgot to make a screenshot of it.

raul.raghav posted this 07 November 2017

Not sure what could have gone wrong. Would you be able to share your mesh file by any chance?

Rahul

Ivy.s posted this 09 November 2017

I can not upload it here, so that's the link to the mesh file.

https://www.dropbox.com/mesh_file

Raef.Kobeissi posted this 10 November 2017

Just a quick comment. You need to have your acceleration as +9.81m/s2 in the X dirextion otherwise your gravitational acceleration is wrong. From the picture you posted it seems the flow is reaching a wall and being squeezed upward or seeping through a hole. You need to check your geometery and mesh for any anomalies.

Raef Kobeissi

Ivy.s posted this 10 November 2017

I changed the gravity now to +9.81 m/s²

The result is the same. The liquid flows upwards outside of the screw.

It is correct, that the liquid is being squeezed through the holes of the screw, as you can see in the following picture.

But I don't understand why it flows upwards after that. In reality it flows a bit upwards as well, because of the pressure that squeezes the liquid out of the holes, but not that much. Most of the liquid must flow downwards because of the gravity.

Raef.Kobeissi posted this 10 November 2017

It depends on your geometry, when you rectified your gravity acceleration have you restarted the simulation? Can you share the pressure and velocity contour in your simulation please. Cheers

Raef Kobeissi

Ivy.s posted this 12 November 2017

Yes I restarted the simulation after it.

Raef.Kobeissi posted this 12 November 2017

Hello, from the contours you just showed it looks like you have high differential pressure and I believe the flow behaviour upward despite gravity is perfectly normal due to high pressure and flow deflection. Regards

Raef Kobeissi

Ivy.s posted this 12 November 2017

Do you know how I can set the pressure down, to test if the result is better with a lower pressure?

Raef.Kobeissi posted this 12 November 2017

You can use inlet pressure instead of velocity inlet, this way you can set the pressure at whatever value you want but you wont be able to control the velocity.

Raef Kobeissi

Recent Activity
maliuzair is a new member in the forum
giampiero.sindoni@uniroma1.it is a new member in the forum
maisan is a new member in the forum
Steklobanki-nut is a new member in the forum
monty.nmbu is a new member in the forum