Learning Forum

[abdullahhassan]

Dear Sir,

I am new to ANSYS and doing an Implicit VOF transient simulation on heat exchanger. 

Viscous Model:

K-Epsilon with realizable and standard wall function

Material:

Tube Domain = One fluid is water that converts from vapor to liquid with VOF function.

Shell Domain = Second fluid is ammonia liquid. (no phase change in Ammonia).

No air is used in the domain.

]

Following are the UR Factors I am using to enhance the convergence.

 

Boundry conditions:

Velocity inlet and pressure outlet.

Velocity inlet for Water vapor = 0.5m/s

 

Lee’s Model Frequency:

From phase = 5 / s

To phase = 5 / s

Method:

I am using PISO method with first order upwind for some iterations around 2000 to stabilize and then change to second order upwind.

 

Problem:

The problem I am continuously facing is the reversible flow. You can see the images. Around all the Water liquid that was formed is flowing backwards.

I have done around 25,000 iterations twice.

 

 

[abdullahhassan]

Hello, Can you please suggest me some solution?????

[Rob]

Where is the free surface level? Ie how much water have you got? What is the backflow condition on the outlet? 

[abdullahhassan]

Regarding backflow conditions: I didn't put any backflow except the temperature, like at inlet, temperature is 100C and at the pressure outlet, the backflow temp is also 100C. Condensation happens at a constant temperature. Also, I tried to reduce backflow temperature to the default setting (25C), still got the reversed flow. 

Let me show you the stats: 

Volume fraction of water vapor at inlet is 1 and water liquid at outlet is 1.

In domain, it is shown as below, water vapor 0.24 and water liquid 0.75.

 

Now my question is, if VF = 1 is showing for water liquid at outlet, it means all the liquid is flowing out of the domain. but still If I see the vectors of flow then the majority of flow is showing as a backflow. and I am also receiving the BACKFLOW warning from start of simulation. 

Below is the scaled residual. 

 

[Rob]

What's the mass flux at the outlet for each phase?  Whilst trouble shooting I suggest losing the shell side part of the model to simplify things. 

[abdullahhassan]

The mass flux or mass flow rate inlet/outlet of shell side is equal as only NH3 liquid is passing through it.  

but the mass flux for tube domain is not equal. because water vapor is at inlet and water liquid is at outlet. 

 the only issue i am facing is in tube side, as condensation happening has more reversible flow which i cant understand.

 

[Rob]

Mass flux should about balance as mass entering has to go somewhere. However as you're losing a lot of volume you will be drawing material in from the outlet.  Please can you post the velocity contour from Fluent, along with phase contour to try and understand what's happening.