Learning Forum

[sjohn]

I am using a multiphase VOF model to simulate water spray through an atomizer nozzle. I have meshed the volume extract of the atomizer nozzle. Both fluids are from the Fluent material database (water and air). I am using Viscous (Laminar) physics with multphase (VOF) and Energy is On. The walls are set to lose heat 5 W/m2.K.

As the sim is running I am seeing messages that indicate large variation in temperature which is unnatural for a simple water atomization.

Example:

Stabilizing temperature to enhance linear solver robustness.

temperature limited to 1.000000e+00 in 802733 cells on zone 5424 in domain 1

temperature limited to 5.000000e+03 in 220188 cells on zone 5424 in domain 1

temperature limited to 1.000000e+00 in 322 cells on zone 5423 in domain 1

temperature limited to 5.000000e+03 in 2034 cells on zone 5423 in domain 1

temperature limited to 1.000000e+00 in 5334 cells on zone 5421 in domain 1

temperature limited to 1.000000e+00 in 5064 cells on zone 5420 in domain 1

temperature limited to 1.000000e+00 on 1729 faces

temperature limited to 5.000000e+03 on 185 faces

[Rob]

How well resolved is the nozzle region? Ie as the liquid breaks up have you resolved the droplets? Which phase requires the energy equation? 

[sjohn]

There are three meshing regions in the nozzle region - body size 1 (target: 0.040 mm), body size 2 (target 0.080 mm), and general (0.6 mm). I tried to figure where do the zones in the error message belong but can't find them.

[Rob]

Try creating an isosurface of temperature, 10K and 4800K may be good starting points. 

[sjohn]

 

The iso surfaces form at the walls. The boundary condition I have is Convection (htc 5 W/m2.K and free stream temperature 26.85 C).

The solution limits under Solution/Controls are -100 C and 100 C.

[Rob]

What temperature are the droplets, and what other physics (density models & phase change) are active?

[sjohn]

Water is entering at 26.85 C. Outlet backflow total temp is 26.85 C. Contour plot the mid-plane (cross section) ranges from -272 C to 4725 C, which are at the edges. The mass-average temperature at the mid-plane is -37.14 C.

The following physics are enabled - Multiphase (VOF, Implicit, Interface Modeling - Sharp, 2 phases with constant surface tension and surface tension force modeling enabled), Viscous(Laminar with Viscous Heating disabled)

[sjohn]

Boundary condition:

pressure-inlet 615 kPa

pressure-outlet Gauge Pressure 0 Pa (Backflow - From Neigbhoring Cell)

[Rob]

OK, so why do you need energy?  How well resolved are the droplets? 

[sjohn]

 

When I switch to a different user defined material, it has properties that are temperature dependent and that causes the energy to turn on by default. I am starting with water (Fluent material) to make sure the model is working well and then will switch to my user defined material.

The droplets should have an average diameter of 16-18 micron (theoretically) and the target cell size is 80 micron in the zone where the droplets are formed. Below this zone, where the fluid escapes the cell size is 600 micron

 

[Rob]

With VOF you need to resolve the droplets, so you'll need 2-4micron cells. It's DPM where droplets must be smaller than the cells. What density settings have you got? 

[sjohn]

I read that the cell size should be 1/10th of the droplet diameter so I have been creating a very fine mesh but running into stair-step warning and another warning - "Skipping synchronization with compute node 1 as size field is already synchronized." The mehsing process gets stuck at 46%. Tried on a computer with 48 cores.

I don't understand "density settings"