[MK1000]

Hi,

i am look to create a simulation of ice formation in a pipe and would really appreciate any help i can get. looking for any example files. and possibly an explaination of what eqautions were used and variables.

thank you

[MK1000]

Would appreciate any step by step calculations, to predict the formation of ice

[michujo]

Hi. Are you referring to ice accretion inside a pipe or over the exterior surface of a pipe in crosswind?

Cheers,
Michujo.

[MK1000]

indeed i am

[MK1000]

but not in a cross wind, it will be in the direction of the flow inside a pipe section

[michujo]

Hi, I do not have experience in such problem but in icing predictions for external flow the adopted strategy is usually as follows:

1) Compute fluid flow (using either a panel method with boundary layer solver or Euler+ boundary layer or directly Navier-Stokes).

2) Compute water droplets trajectories. Either Lagrangian or Eulerian approaches are used depending on the problem (for instance Lagrangian transport of a finite number of droplets is widely used when computing icing over 2D airfoils, but an Eulerian framework is preferred when computing icing over 3D complex geometries)

3) Solve for the liquid film formation on the walls: From the liquid phase solution you can determine how much liquid mass impacts on the walls. You then solve for the liquid mass transport on the wall (surface mesh). The mass balance of the liquid film is coupled to a energy balance.
Determine how much ice is formed per unit surface and time by performing an energy balance on each cell of the surface grid used for the liquid film. The contributions are from the energy of the impinging liquid mass, the incoming and outgoing liquid film mass flow rate, the energy loss due to evaporation and heat loss to the gas phase, heat released in the liquid-to-ice transition, ...

4) Depending on how much ice per unit surface and time is accreted, the geometry is modified.


Those are the building blocks. Now, since ice accretion is going to modify your geometry, this will in turn affectthe fluid flow and so on. The procedure usually employed is to compute blocks 1 to 4 sequentially, "growing" an ice layer every time.

You might want to have a look at the LEWICE code manual (NASA), it's all there. There are also other similar codes such as the one from ONERA, or DLR.
As for commercial software I know of FENSAP-ICE, which can do 3D (newer versions of the LEWICE code can do it as well).

Cheers,
Michujo.

[MK1000]

this is very helpful. i appreciate it greatly