Initialization of two-phase simulation using single-phase result

vjo · October 11, 2023, 04:34

Hello to all.!!
I was reading some research papers on multi-phase flow in micro-channels for cooling electrical devices, when i came across this line by authors,

"Another important aspect of the present study is that, the two-phase flow calculation requires being initialized with the single-phase isothermal flow field solution for the specified mass flux. This data is then loaded into the two-phase case field and transient flow boiling simulation is started..."

My question is,
1) why simulation for two-phase flow into the microchannel has to be initiated using solution for single phase flow??
2) If we directly simulate the two phase flow, using at inlet boundary condition saturated fluid temperature which will boil immediately, will it be not right??

I have really searched for this answers, but cannot get a satisfactory answer.
I request all to please help me, will be very grateful.

Thanks in advance.

Vidzz22 · November 15, 2023, 03:55

I think the main reason is for better convergence. Numerically multiphase simulations are complex.
I am facing a convergence issue inspite of initialising with singlephase results.

I am working on a multiphase mixing tank simulation with turbulence. I have got a converged simulation for a single phase (water) mixed with an impeller. Now I am trying to use the results to initialise a multiphase simulation. In the multiphase setup, additionally air is introduced through a porous sparger (modeled as a surface with an effective diameter of bubbles coming out). Drag and surface tension forces are considered significant. But when the simulation runs the multiphase case it diverges.

Can someone help me out with possible reasons for the same? I can give more details as required.

PS: Mesh is unstructured and mesh quality, orthogonality and aspect ratio is well within limits

October 11, 2023, 04:34	Initialization of two-phase simulation using single-phase result	#1
vjo New Member Join Date: Apr 2022 Posts: 3 Rep Power: 4	Hello to all.!! I was reading some research papers on multi-phase flow in micro-channels for cooling electrical devices, when i came across this line by authors, "Another important aspect of the present study is that, the two-phase flow calculation requires being initialized with the single-phase isothermal flow field solution for the specified mass flux. This data is then loaded into the two-phase case field and transient flow boiling simulation is started..." My question is, 1) why simulation for two-phase flow into the microchannel has to be initiated using solution for single phase flow?? 2) If we directly simulate the two phase flow, using at inlet boundary condition saturated fluid temperature which will boil immediately, will it be not right?? I have really searched for this answers, but cannot get a satisfactory answer. I request all to please help me, will be very grateful. Thanks in advance.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to edit a single phase UDF for a multiphase case?	mataus	Fluent UDF and Scheme Programming	0	January 8, 2020 05:02
Simulation of Phase change material (PCM) and nanoparticles together	farah	Main CFD Forum	0	November 2, 2015 15:30
Difference in flow distribution of single phase and multiphase analysis	shivasluzz	CFX	0	April 22, 2015 23:54
coupling vof with single phase flow and gravity term	alame005	Main CFD Forum	7	August 6, 2013 12:02
two Phase column simulation	chemeng	OpenFOAM	3	August 18, 2010 13:53

November 15, 2023, 03:55		#2
Vidzz22 New Member Vidya Muthulakshmi M Join Date: Aug 2020 Posts: 7 Rep Power: 6	I think the main reason is for better convergence. Numerically multiphase simulations are complex. I am facing a convergence issue inspite of initialising with singlephase results. I am working on a multiphase mixing tank simulation with turbulence. I have got a converged simulation for a single phase (water) mixed with an impeller. Now I am trying to use the results to initialise a multiphase simulation. In the multiphase setup, additionally air is introduced through a porous sparger (modeled as a surface with an effective diameter of bubbles coming out). Drag and surface tension forces are considered significant. But when the simulation runs the multiphase case it diverges. Can someone help me out with possible reasons for the same? I can give more details as required. PS: Mesh is unstructured and mesh quality, orthogonality and aspect ratio is well within limits