Learning Forum

[Sushant P]

I am trying to simulate hydrulic jump for laminar as well as turbulent condition in 2D. I have different nozzle sizes and different spacings from the target plate. The size of the target plate is 20 times the nozzle diameter. I also have enough entry length to get a fully developed flow at the nozzle exit.
I have turned on VOF model for particle and using k-w SST with LRC for turbulent models. I am doing a transient simulation with a time step of 5e-5.

For the laminar models, I get a floating point exception when I turn on Surface Tension modeling with constant 0.072 for water (I am using air and water).
On turning off the Surface Tension I do not I get a jump but instead a thin film that goes all the way to the outlet.

My question is about Surface tension. Is there any special thing that I need to do with the Surface Tension. I have seen in some literature the mention adding phase angle but I am not sure how to do that.

Also, if you have any experience with modeling turbulent jets please let me know. 

[Rob]

How many cells have you got through the film thickness?  The contact angle is set in the Wall bc panel once you turn on adhesion. I assume you did turn everything on?

 

 

 

[Sushant P]

Thank you Rob, I did not had wall adhesion on. I turned it on and I am now able to add the contact angle.

I have about 10 cells going through the thickness. The height of the smallest cell in 1e-7. I have seen some model in my hw divergening due to too many elements, is this a similar case.

[Rob]

OK, main problem is cell aspect ratio. Boundary layer meshes are designed for flows where nothing changes along the cell. With multiphase we can't make that assumption, and especially when the film forms & you get waves. Aim for a mesh with a much lower aspect ratio. Then look at adaption (refinement) for the near wall region. 

The other thing to consider. Should you be using 2d or 2d axi-symmetric to model a jet?