reversed flow at outlet for bubble column

melj · March 2, 2020, 17:55

Hello

I am modeling a multiphase bubble column with turbulence using the Eulerian model.

1) The simulation converged initially for a few steps and then it stopped converging. It just runs for the specified no. of iterations. The residual value remains constant.

2) For every iteration, a message shows "Reversed flow on #### faces of pressure-outlet #."

Could anyone please let me know what these observations mean?

thank you

vinerm · March 3, 2020, 05:22

The residuals are not good indicators of convergence in multiphase flows. As long as those are below 1.0, those are good. What you should monitor are some averaged values for the field variables. If those stabilize, then, solution can be considered as converged.

As far as flow reversal is concerned, that could be just temporary or real. Until unless the flow-field is converged, you may observe some flow reversal. However, if the number of faces with flow reversal keep on reducing, then there is nothing to be worried about. If the flow reversal is maintained, then you may have to consider the appropriateness of the boundary. Either it is located too close to some internal obstruction or the pressure applied is too high.

melj · March 3, 2020, 18:17

About the reversal of flow: I have modeled the bubble column to the height of liquid present in the column. So my outlet is essentially at the surface where liquid ends. To change my outlet I'll have to redo my geometry right?

I reduced the mixture gauge pressure to zero but the flow reversal still exists.

vinerm · March 4, 2020, 03:24

If you are modeling only up to the freeboard region, then you will always get flow reversal. Either extend the domain beyond freeboard region or use degassing boundary condition instead of outlet.

Freddy08 · August 30, 2021, 01:37

Hi, I have been trying to simulate a bubble column. I find no problem if I don't include the freeboard region above my static liquid level. But if I include the freeboard region, by patching it with air it is not converging. I get messages like "Reversed flow on #### faces" and "Turbulent viscosity ratio limited to 1e5". Can someone help me with this?

I am using the outflow boundary condition at the outlet. If I fill the entire column with liquid (no freeboard region), I don't get any of these messages and my residuals are properly converging.

Thanks & Regards,
Frederick

March 2, 2020, 17:55	reversed flow at outlet for bubble column	#1
melj Member mln Join Date: Dec 2019 Posts: 39 Rep Power: 7	Hello I am modeling a multiphase bubble column with turbulence using the Eulerian model. 1) The simulation converged initially for a few steps and then it stopped converging. It just runs for the specified no. of iterations. The residual value remains constant. 2) For every iteration, a message shows "Reversed flow on #### faces of pressure-outlet #." Could anyone please let me know what these observations mean? thank you

March 3, 2020, 05:22	Convergence in multiphase	#2
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	The residuals are not good indicators of convergence in multiphase flows. As long as those are below 1.0, those are good. What you should monitor are some averaged values for the field variables. If those stabilize, then, solution can be considered as converged. As far as flow reversal is concerned, that could be just temporary or real. Until unless the flow-field is converged, you may observe some flow reversal. However, if the number of faces with flow reversal keep on reducing, then there is nothing to be worried about. If the flow reversal is maintained, then you may have to consider the appropriateness of the boundary. Either it is located too close to some internal obstruction or the pressure applied is too high. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

March 4, 2020, 03:24	Degassing	#4
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	If you are modeling only up to the freeboard region, then you will always get flow reversal. Either extend the domain beyond freeboard region or use degassing boundary condition instead of outlet. melj likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

August 30, 2021, 01:37	Diverging if a freeboard region is included	#5
Freddy08 New Member Fred J Join Date: Aug 2021 Posts: 1 Rep Power: 0	Hi, I have been trying to simulate a bubble column. I find no problem if I don't include the freeboard region above my static liquid level. But if I include the freeboard region, by patching it with air it is not converging. I get messages like "Reversed flow on #### faces" and "Turbulent viscosity ratio limited to 1e5". Can someone help me with this? I am using the outflow boundary condition at the outlet. If I fill the entire column with liquid (no freeboard region), I don't get any of these messages and my residuals are properly converging. Thanks & Regards, Frederick

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
reversed flow in xxxx faces on pressure outlet	franzdrs	FLUENT	8	January 4, 2020 21:12
Reversed Flow in Outlet	JSped	STAR-CCM+	2	January 20, 2016 11:59
reversed flow when using Exhaust Fan as outlet	Dziyad	FLUENT	0	February 20, 2014 21:13
Reversed Flow in Outlet	Steve	FLUENT	18	March 30, 2013 08:50
Reversed flow at the outlet	swati_mohanty	FLUENT	0	January 5, 2011 02:00

March 3, 2020, 18:17		#3
melj Member mln Join Date: Dec 2019 Posts: 39 Rep Power: 7	About the reversal of flow: I have modeled the bubble column to the height of liquid present in the column. So my outlet is essentially at the surface where liquid ends. To change my outlet I'll have to redo my geometry right? I reduced the mixture gauge pressure to zero but the flow reversal still exists.