Simulation fails after adding multi-component gas

ConorH · January 18, 2017, 09:26

Hi all,

I have an exhaust manifold that I am undertaking flow analysis on. when using a single gas model, the simulation runs without fault.
However, when I change to a multicomponent gas model, the following error message appears:

Error for Segregated Flow:

Bi-Conjugate Gradient Stabilized solver did not converge !
Reversed flow on 378 faces on xxxxxxxxxxxxxxxx

Error: AMG solver diverged.

Error: AMG solver diverged.

Error: AMG solver diverged.

Error: AMG solver diverged.

Error: AMG solver diverged.

Error: AMG solver diverged.
A floating point exception has occurred: floating point exception [Divide by zero]. The specific cause cannot be identified. Please refer to the troubleshooting section of the User's Guide.
Context: star.segregatedflow.SegregatedFlowSolver
Command: RunSimulation
error: Server Error

I have tried refining the mesh, changing from a polyhedral to a tetrahedral, segregated flow and coupled flow with both meshers.

Coupled Flow:
Coupled Energy, Coupled Flow, Coupled Species, Exact Wall Distance, Gradients, Ideal Gas, K-Epsilon Turbulence, Multi-Component Gas, Non-reacting, Realizable K-Epsilon Two-Layer, Reynolds-Averaged Navier-Stokes, Steady, Three Dimensional, Turbulent, Two-Layer All y+ Wall Treatment

Segregated Flow:
Exact Wall Distance, Gradients, Ideal Gas, K-Epsilon Turbulence, Multi-Component Gas, Non-reacting, Realizable K-Epsilon Two-Layer, Reynolds-Averaged Navier-Stokes, Segregated Flow, Segregated Fluid Isothermal, Segregated Species, Steady, Three Dimensional, Turbulent, Two-Layer All y+ Wall Treatment

Thanks

me3840 · January 20, 2017, 02:29

Your case is diverging. Did you have a lot of reversed flow on the single gas model?

ConorH · January 20, 2017, 06:48

Early iteration did have a lot of reversed flow but after 200 iterations there was no reversed flow.
Do you propose I change the turbulence model?

Thanks for the reply

me3840 · January 20, 2017, 10:50

No, I would say you should fix your initial conditions to remove so much reversed flow at the start.

Your BCs do different things when reversed flow occurs, which is probably why your mutli-component one explodes.

You could converge the single-component one and then just switch to mutli-component.

January 18, 2017, 09:26	Simulation fails after adding multi-component gas	#1
ConorH New Member Conor Hardiman Join Date: Jan 2017 Posts: 2 Rep Power: 0	Hi all, I have an exhaust manifold that I am undertaking flow analysis on. when using a single gas model, the simulation runs without fault. However, when I change to a multicomponent gas model, the following error message appears: Error for Segregated Flow: Bi-Conjugate Gradient Stabilized solver did not converge ! Reversed flow on 378 faces on xxxxxxxxxxxxxxxx Error: AMG solver diverged. Error: AMG solver diverged. Error: AMG solver diverged. Error: AMG solver diverged. Error: AMG solver diverged. Error: AMG solver diverged. A floating point exception has occurred: floating point exception [Divide by zero]. The specific cause cannot be identified. Please refer to the troubleshooting section of the User's Guide. Context: star.segregatedflow.SegregatedFlowSolver Command: RunSimulation error: Server Error I have tried refining the mesh, changing from a polyhedral to a tetrahedral, segregated flow and coupled flow with both meshers. Coupled Flow: Coupled Energy, Coupled Flow, Coupled Species, Exact Wall Distance, Gradients, Ideal Gas, K-Epsilon Turbulence, Multi-Component Gas, Non-reacting, Realizable K-Epsilon Two-Layer, Reynolds-Averaged Navier-Stokes, Steady, Three Dimensional, Turbulent, Two-Layer All y+ Wall Treatment Segregated Flow: Exact Wall Distance, Gradients, Ideal Gas, K-Epsilon Turbulence, Multi-Component Gas, Non-reacting, Realizable K-Epsilon Two-Layer, Reynolds-Averaged Navier-Stokes, Segregated Flow, Segregated Fluid Isothermal, Segregated Species, Steady, Three Dimensional, Turbulent, Two-Layer All y+ Wall Treatment Thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Simulation of radial gas turbine, Inlet velocity components	dmaz	CFX	17	May 19, 2019 10:45
2 Phase Flow, Gas Jet into Water Simulation Method	im_lenny	OpenFOAM	3	May 4, 2017 04:20
conjugateHeatFoam + interFoam	farhagim	OpenFOAM Programming & Development	15	July 19, 2016 08:55
multi component diffusivity	mahut	FLUENT	0	October 10, 2007 05:05
Gas pressure question	Dan Moskal	Main CFD Forum	0	October 24, 2002 23:02

January 20, 2017, 02:29		#2
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 25	Your case is diverging. Did you have a lot of reversed flow on the single gas model?

January 20, 2017, 06:48		#3
ConorH New Member Conor Hardiman Join Date: Jan 2017 Posts: 2 Rep Power: 0	Early iteration did have a lot of reversed flow but after 200 iterations there was no reversed flow. Do you propose I change the turbulence model? Thanks for the reply

January 20, 2017, 10:50		#4
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 25	No, I would say you should fix your initial conditions to remove so much reversed flow at the start. Your BCs do different things when reversed flow occurs, which is probably why your mutli-component one explodes. You could converge the single-component one and then just switch to mutli-component.