Learning Forum

[ehsansadeghi]

Hi

I am trying to simulate the flow of a temperature dependent viscous material into a container. (pic below)

[subhamdas]

Scheme: I would prefer the implicit scheme over the explicit onenTransient Formulation: Increase the number of iterations per time step. Also check the Courant Number(should be less than 2) and reduce the time step further if requirednDiscretization: Start with first order schemes and then proceed with second ordernMesh: Since your geometry is straightforward, I assume that the mesh is not a problem herenUnder-Relaxation Factors: You may also consider starting with slightly low values and increase it as you acheive convergencenAlso check if the boundary conditions are correctn

[Rob]

Pretty much. Patch a small amount of plastic into the domain first too, it'll help the solver around the boundary surface. The manual will cover the VOF settings, you'll want the scheme designed for transient flow. nUR factors can be left alone in most cases, control model stability with the time step. Similarly retain higher order discretisation schemes and turn on Higher Order Term Relaxation for all equations. n

[ehsansadeghi]

I still have problems with this model. I can get the convergence, but the results don't make any sense. The plastic will lose heat and become more viscous and eventually solidify. But it does not happen. nTwo important things:n1- The model is in micrometer order. The inlet has 1e-5 m of radius and the 5e-5m length. n2- The model has only pressure inlet of 100Bars. It is expected that the plastic fill the cavity in 10-15 seconds but the total flow time is 1e-6 seconds.nAny suggestions? n

[YasserSelima]

So, you have three phases, not two. nYou should define the solid material solid plastic and enable melting/solidification model to get solidification.nnWhat is you time step size? n

[ehsansadeghi]

So, you have three phases, not two. You should define the solid material "solid plastic" and enable melting/solidification model to get solidification.What is you time step size?/forum/discussion/comment/112955#Comment_112955

Why three? The first phase is air (initially the domain is filled with the air) and the second phase is the plastic. Plastic's viscosity is temperature-dependent. As it cools down it should loose heat and get solid. Do I need to enable solidification and melting?Time steps: 1e-10n

[ehsansadeghi]

Why 3 phase? The temperature-dependent viscosity is used. nTime step size: 1e-10n

[YasserSelima]

I understand, but Fluent will continue treating the plastic as fluid even with high viscosity ... Fluid polymer will still behave different than solid polymer in addition to the latent heat required for solidification. n

[Rob]

It is two phase, but the liquid also uses the solidification/melting model. Check you have enough mesh in the domain, but also that all of the material properties are correct. n