## Fluids

#### Convergence problem with multiphase model?

Tagged:

• kailash
Subscriber
Hi. I am running multiphase mixture model and I see after 5000 iterations, there is a sudden spike in residual values for continuity, momentum, k and epsilon. I also see that my y+ value on the wall is very high in one particular point in the wall which i haven't seen before when the iterations started? May I know why this is happening?n
• DrAmine
Ansys Employee
What are you modeling? What are you solution settings? Mesh quality, Are the Sudden spikes occurring after certain event,? At restart?...n
• kailash
Subscriber
Hi ,nI am modeling a large water tank about 100,000 gallons using symmetry boundary condition. I am injector air with water through an injector and studying the overall mixing performance. I am using simple scheme, Body force weighted for pressure, Second order upwind for kinetic energy, dissipation energy,momentum and QUICK for volume fraction.nMy under-relaxation factors are 0.5 for kinetic energy and dissipation rate, 1 for turbulent viscosity, density, body forces, 0.1 for slip velocity and 0.95 for volume fraction.Min orthogonal Quality = 1.9e-2 and max aspect ratio = 1.07e02nnYes the spike occurs after 3500 iterations. Until then my momentum is at 1e-3 and continuity is at 1 e-5 residual.
• DrAmine
Ansys Employee
Steady? Accumulation of secondary phase occurring? Symmetry where?n
• kailash
Subscriber
Yeah I am running on steady state solution. The symmetry is right across the central plane of the tank. nI am not sure if i understood you question on accumulation of secondary phase. When i see the volume fraction of phase 2 , i see that air is getting accumulated on the top of the tank as it should be. In the residual the volume fraction of phase 2 is kind of oscillating up and down.nnThanks and Regards,nKailash R.nn
• DrAmine
Ansys Employee
Please use coupled solver with pseudo transient option. The top boundary is defined as?n
• kailash
Subscriber
oh sure let me try that, is there any specific reason why you think this would be better? Also, should I change the under relaxation factors or leave it to default?. The top boundary is shear free wall.nRegards,nKailash R.nn
• DrAmine
Ansys Employee
Recommended for all steady state multiphase runs as it adds a time marching under relaxation.nDo you have an outlet or are you just filling.nStick to default urfs n
• kailash
Subscriber
Got it. Yes I have pressure outlet and I am using mass flow inlet. Also, any recommended timescale factor for fluid time scale?.Regards,nKailash R.n