Divergence in calculation: water-emulsion-injection

Dennniem · October 8, 2018, 12:30

Hello,

I am simulating a water-emulsion-injection in an 4 stroke gas engine.
The Mesh was build with Fame Engine Plus.

srokes:
- 315°-526°: exhaust valves open
- 526-534°: valve overlapping
- 534°-772: intake valves open
- 772°-1035: all valves open

The problem is, that my calculation diverges around 360°CA (exhaust valves not full open).
The false report are such as:
- Problem diverges: TKE equation, variable: TKE, material: 1
or
- Problem diverges: continuity equation, variable: Pressure correction, material: 1

The following modules are activatated: Species Transport, Combustion, Spray, Wallfilm, Thin Walls.

When the modules Spray, Wallfilm and Thin Walls are deactivated, the problem are the same. So I think the problem is an incorrect setting in the solver control.

I have decreased the underrelaxation factors (pressure: 0.1, Momentum: 0.6, TKE: 0.3, Turb. diss. rate: 0.3. I also have tried to change the linear solver of continuity (GSTB/AMG and variated th max iterations), but the problem was still the same.

Here are other settings, which I used:
- Convergence Criteria Pressure: 0.01
- Boundary Condition Inlet: total pressure 1.5bar
- BC outlet: static pressure 1.07bar

It is possible, that the problem for the divergence is the mesh itself?
I use high refinements in valve seats and valve heads.

I am thankful for every recommendation.

Best regards

swerner0711 · October 9, 2018, 04:30

Hello,
did You do a dummy run after meshing with FE+? What were the results?

Also a refinement of cells might be necessary, when the valves are almost closed.

Dennniem · October 9, 2018, 06:24

The result of the dummy run were 2 twisted faces.

5°CA after the valves opened and before the valves will close I have used a refinement level 3 at valve heads and the valve seats.

My settings of the refinements are a recommendation of an example case in AVL Fire (case 920)

October 8, 2018, 12:30	Divergence in calculation: water-emulsion-injection	#1
Dennniem New Member Dennis Niemann Join Date: Oct 2018 Location: Germany Posts: 4 Rep Power: 8	Hello, I am simulating a water-emulsion-injection in an 4 stroke gas engine. The Mesh was build with Fame Engine Plus. srokes: - 315°-526°: exhaust valves open - 526-534°: valve overlapping - 534°-772: intake valves open - 772°-1035: all valves open The problem is, that my calculation diverges around 360°CA (exhaust valves not full open). The false report are such as: - Problem diverges: TKE equation, variable: TKE, material: 1 or - Problem diverges: continuity equation, variable: Pressure correction, material: 1 The following modules are activatated: Species Transport, Combustion, Spray, Wallfilm, Thin Walls. When the modules Spray, Wallfilm and Thin Walls are deactivated, the problem are the same. So I think the problem is an incorrect setting in the solver control. I have decreased the underrelaxation factors (pressure: 0.1, Momentum: 0.6, TKE: 0.3, Turb. diss. rate: 0.3. I also have tried to change the linear solver of continuity (GSTB/AMG and variated th max iterations), but the problem was still the same. Here are other settings, which I used: - Convergence Criteria Pressure: 0.01 - Boundary Condition Inlet: total pressure 1.5bar - BC outlet: static pressure 1.07bar It is possible, that the problem for the divergence is the mesh itself? I use high refinements in valve seats and valve heads. I am thankful for every recommendation. Best regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
I want to model injection of hot water in cold water dilution	ahmed_ashmawy	FLUENT	1	December 20, 2017 04:40
Injection of water droplets using DPM	Mona	FLUENT	6	March 13, 2013 01:12
Steam injection in to hot water	SEDL	Main CFD Forum	0	August 4, 2010 05:55
Injection of water in vapor	SimoneTecnoil	FLUENT	0	November 10, 2009 10:32
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 10:11

October 9, 2018, 04:30		#2
swerner0711 New Member Stephan Werner Join Date: Sep 2018 Posts: 6 Rep Power: 8	Hello, did You do a dummy run after meshing with FE+? What were the results? Also a refinement of cells might be necessary, when the valves are almost closed.

October 9, 2018, 06:24		#3
Dennniem New Member Dennis Niemann Join Date: Oct 2018 Location: Germany Posts: 4 Rep Power: 8	The result of the dummy run were 2 twisted faces. 5°CA after the valves opened and before the valves will close I have used a refinement level 3 at valve heads and the valve seats. My settings of the refinements are a recommendation of an example case in AVL Fire (case 920)