# Numerical oscillations in CFX with water boiling problem.

 Register Blogs Members List Search Today's Posts Mark Forums Read

December 13, 2011, 03:59
Numerical oscillations in CFX with water boiling problem.
#1
Senior Member

Join Date: Dec 2011
Posts: 134
Rep Power: 15
Hi everyone, this is my first post here so nice to meet you.

I'm a CFX user and I'm trying to calculate a water boiling problem, but I'm getting weird results. The problem looks like this:

- Water enters an annular tube (space between two coaxial cylinders) where it is heated up through the inner cylinder wall until it starts to evaporate (see attached picture).

- The cylinder is 340 mm long, diameters of inner and outer cylinders are 12.7 mm and 25.4 mm. Water enters 31 K below the saturation temperature at a velocity of 0.4 m/s. Inlet Reynolds number is Re_inlet = 5071.

- A heat flux is applied through the inner cylinder wall.

- I believe the computational mesh is pretty good, what do you think? (see attached picture)

- Computational parameters: I'm using upwind scheme for the advection terms. I did not set the Wall Boling model, phase change is expected to occur when the liquid temperature reaches the saturation temperature.

THE PROBLEM: when water starts to evaporate I get spurious oscillations! I tried different things to try and suppress them but they're still there! (attached picture shows water vapour volume fraction on several cut planes, saturated at volume_fraction_vapour = 0.1)

THE QUESTION: Is there anyone among you who got this same problem in the past? Can you give me any hint as to what might be the reason of this problem?

I thank you guys for your help, this is driving me crazy...
Cheers.
Attached Images
 foto1.jpg (16.7 KB, 40 views) foto2.jpg (98.9 KB, 43 views) foto3.jpg (34.8 KB, 54 views)

 December 13, 2011, 16:15 #2 Super Moderator   Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143 Why do you say this is spurious? Boiling is a very lumpy process and often ends up in slug flows and things like that. Upwinding for the advection terms will cause excessive damping. Use a second order scheme. Are you doing this steady state or transient?

 December 15, 2011, 03:42 #3 Senior Member   Join Date: Dec 2011 Location: Madrid, Spain Posts: 134 Rep Power: 15 Dear Glenn, I'm running it as a steady state-calculation. I also tried the high order scheme available in CFX with similar results ... I was trying to perform this simulation in order to validate this boiling model, so I took this case from the literature. In the experiment (and their simulations with CFX!) they get smooth solutions where vapour volume fraction grows along the tube so I was hoping to get similar results. Thanks for your reply anyway. Any other hint you might think of? Thanks! Cheers.

 December 15, 2011, 16:18 #4 Super Moderator   Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143 This FAQ covers some issues of relevance here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

 December 16, 2011, 09:00 #5 Senior Member   Join Date: Dec 2011 Location: Madrid, Spain Posts: 134 Rep Power: 15 Dear Glenn, thanks for your reply. I will have a look at the document. Cheers.