Learning Forum

[dougs]

I'm using Fluent for the first time, and although I have a strong background in CFD, I've never run any commercial CFD software. I'm trying to run a transient simulation of the filling of a vessel with a gas. At the inlet boundary I'm specifying a velocity, and I'm running a single time step with a tiny time step (to keep the Courant number small). I initially tried the pressure-based solver, but it blew up. The density-based solver completed a time step, but had lots of stability problems along the way, and the result is garbage. I plotted the resulting velocity field, and it does not seem to be holding the specified inlet velocity boundary condition. Is there a way to plot the initial condition without running a time step? I want to make sure all the boundary conditions (inlet, no-slip on walls) are being implemented correctly.

[Karthik Remella]

Hello You will need to provide us with additional details about the problem - is this a compressible flow? What are the velocity scales (Mach number, if you are dealing with a high-speed flow). Is this a single-phase flow?
In the meantime, you might be interested in these learning resources. There are several simulation examples.
Fundamentals of Compressible Flows Learning Track | Ansys Courses
Real Viscous Flows - ANSYS Innovation Courses
Karthik

[dougs]

My eventual goal is to run a compressible problem with single-phase hydrogen gas, but for my first run I'm trying low-speed air (also single-phase). I initially tried an inlet velocity of 1 m/s. The command window showed that stability issues were occurring, and it kept trying smaller and smaller Courant numbers, al the way down to 5e-5. I tried lowering the inlet velocity to 0.001 m/s, but that didn't seem to matter. Thanks for the links; I'll look them over.

[Karthik Remella]

Can you share some details about the boundary conditions, solver settings, your geometry / mesh, initial conditions, and the residual and monitor plots?

[dougs]

The domain is an axisymmetric tank being filled through a tube at one end:
I'm starting with a coarse grid that will not resolve the boundary layers just to try and get something running. Here's a close-up of the grid near the inlet:
The boundary conditions are specified inlet velocity, axisymmetric along bottom line, adiabatic walls. I believe it's no-slip on walls, but I'm not sure how to verify this. I specified laminar flow for starters to avoid and complications from solving the turbulence transport equations. I tried various inlet velocities; it's currently 0.01 m/s.
I don't know how to specify or verify the initial conditions. That's what the initial question was asking. I don't have residual plots because it won't run one time step. Here's the console output:

[dougs]

I think I've got it working. I hadn't initialized the computation!