A problem: reversed flow in ... on outflow.

vandadt · November 6, 2009, 01:57

Hi everybody

I am modeling Taylor flow in microchannels ( gas-liquid two-phase flow) and i have a problem. during the solution there is a message like this:
"reversed flow in ... faces on outflow"
I choosed outflow for outlet boundary condition and guess this may be the problem but i am not sure.
Is there anyone who may experienced the simillar case and have a suggestion?

Thanks,
Vandad

-mAx- · November 6, 2009, 02:42

this is just a warning, it should dissapear while iterating and converging.
If not then, maybe you have an vortex across your outlet, and you may extend it

vandadt · November 6, 2009, 10:47

Thank you.

rr123 · November 9, 2009, 11:25

Hello both,

I have a geomtry which has an extended outlet much more than required.
I am still getting reversed flow in XX faces on pressure-outlet 2 or 3. the geometry is that ofa vertical tube almost 27 mtrs in length and 0.1 mtrs in dia.
The momentum URF is presently reduced to 0.3.
I am running unsteady, so what should i do ? Do you wish to have a look at my case/data file ?

Many thanks,
R.

vandadt · November 9, 2009, 12:19

Hi Rana
Are you doing single phase or two-phase? In the case of single phase reversed flow warning should be disapeared after the unsteady time duration but in the case of two-phase it would probably be a good idea to use pressure-outlet instead of outflow.

rr123 · November 9, 2009, 15:01

dear sir,

i am using 2 phase flow gas-solids, in which some kinetic reaction occur in the gas-phase with the help of thse solids.
the reversed flow problem is coming even using the extended outlet.
i think now that i have done a mistake and probably for this i am getting this reversed flow. i write now what i have done.
i have simulated a pipe with gas-solids in steady state, disabling the species equations. then i simulated for convergence which i got.
then i switched to unsteady wherein i enabled back these equations and then simulated and i started getting the reversed flow problem.

i think the reversed flow problem comes from the fact that the kinetics of the reaction has been wrongly implemented as i observed that even though after the disabling the species, the mass fractions was not zero in the reactor. so something went wrong there itself.
also when i switched on the unsteady state solver, it was very obvious then some of the mass fractions in the reactor had reacted instead of starting from the inlet. Now my feeling is that this has caused the reversed flow phenomenon. I may be wrong, so I would like to get your suggestions in this how to avoid the reversed flow.
I think i have well enough explained the case, if there is any thing to know pls ask me.

vandadt · November 10, 2009, 10:45

Hi Rana
Sorry, no idea yet!

teguhtf · November 10, 2009, 15:59

I dont think that pressure outlet can be used for 2 phase. Because there is different fluid in calculating back flow. especially for extreme different viscosity of fluid.

rr123 · November 10, 2009, 16:08

hello sir,

could you please explain what exactly do you mean by "In pressure outlet, you can determine your backflow without lengthen the BC" ?

hoping for a detailed reply

kind regards,
R.

teguhtf · November 10, 2009, 16:29

I'm sorry. I change my post because i think Pressure outlet has been explained by other user. maybe your UDF is wrong. see this example
http://www.fluent.com/software/sf_me...orial_wave.htm

rr123 · November 10, 2009, 17:40

hi,

yes, i use pressure-outlet as also suggested by vanadt.
the UDF is not important here because although i have the UDF i disable the macros and what i am trying is just a simple KTGF-enabled tubulent gas-solids flow to converge.
the UDF is not going to influence here.
note that the energy equation is ON.
i am obsevring now that all the residuals have converged, except the residual for the TDR, TKE, w-phase-2 and vof-phase-2. these four parameters are fluctuating slightly above the convergence limit.
URF for momentum, TDE and TKR had been lowered to 0.3 and rest all are default.
would you like to see my file ?

kind regards,
R.

November 6, 2009, 01:57	A problem: reversed flow in ... on outflow.	#1
vandadt New Member Vandad Talimi Join Date: Oct 2009 Posts: 12 Rep Power: 17	Hi everybody I am modeling Taylor flow in microchannels ( gas-liquid two-phase flow) and i have a problem. during the solution there is a message like this: "reversed flow in ... faces on outflow" I choosed outflow for outlet boundary condition and guess this may be the problem but i am not sure. Is there anyone who may experienced the simillar case and have a suggestion? Thanks, Vandad

November 6, 2009, 02:42		#2
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	this is just a warning, it should dissapear while iterating and converging. If not then, maybe you have an vortex across your outlet, and you may extend it __________________ In memory of my friend Hervé: CFD engineer & freerider

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Reversed flow at pressure outlet	Seeker Phil	FLUENT	9	January 2, 2010 06:21
Reversed flow on external aero problem	Ben	FLUENT	2	August 22, 2006 11:05
Combatting reversed flow	tucker	FLUENT	1	July 27, 2005 02:32
Inviscid Drag at subsonic, subcritical Mach #	Axel Rohde	Main CFD Forum	1	November 19, 2001 13:19
fluid flow fundas	ram	Main CFD Forum	5	June 17, 2000 22:31

November 6, 2009, 10:47		#3
vandadt New Member Vandad Talimi Join Date: Oct 2009 Posts: 12 Rep Power: 17	Thank you.

November 9, 2009, 11:25		#4
rr123 Member R. Roy Join Date: Mar 2009 Location: India Posts: 52 Rep Power: 17	Hello both, I have a geomtry which has an extended outlet much more than required. I am still getting reversed flow in XX faces on pressure-outlet 2 or 3. the geometry is that ofa vertical tube almost 27 mtrs in length and 0.1 mtrs in dia. The momentum URF is presently reduced to 0.3. I am running unsteady, so what should i do ? Do you wish to have a look at my case/data file ? Many thanks, R.

November 9, 2009, 12:19		#5
vandadt New Member Vandad Talimi Join Date: Oct 2009 Posts: 12 Rep Power: 17	Hi Rana Are you doing single phase or two-phase? In the case of single phase reversed flow warning should be disapeared after the unsteady time duration but in the case of two-phase it would probably be a good idea to use pressure-outlet instead of outflow.

November 9, 2009, 15:01		#6
rr123 Member R. Roy Join Date: Mar 2009 Location: India Posts: 52 Rep Power: 17	dear sir, i am using 2 phase flow gas-solids, in which some kinetic reaction occur in the gas-phase with the help of thse solids. the reversed flow problem is coming even using the extended outlet. i think now that i have done a mistake and probably for this i am getting this reversed flow. i write now what i have done. i have simulated a pipe with gas-solids in steady state, disabling the species equations. then i simulated for convergence which i got. then i switched to unsteady wherein i enabled back these equations and then simulated and i started getting the reversed flow problem. i think the reversed flow problem comes from the fact that the kinetics of the reaction has been wrongly implemented as i observed that even though after the disabling the species, the mass fractions was not zero in the reactor. so something went wrong there itself. also when i switched on the unsteady state solver, it was very obvious then some of the mass fractions in the reactor had reacted instead of starting from the inlet. Now my feeling is that this has caused the reversed flow phenomenon. I may be wrong, so I would like to get your suggestions in this how to avoid the reversed flow. I think i have well enough explained the case, if there is any thing to know pls ask me.

November 10, 2009, 10:45		#7
vandadt New Member Vandad Talimi Join Date: Oct 2009 Posts: 12 Rep Power: 17	Hi Rana Sorry, no idea yet!

November 10, 2009, 15:59		#8
teguhtf Senior Member teguh hady Join Date: Aug 2009 Location: Saga, Japan Posts: 222 Rep Power: 18	I dont think that pressure outlet can be used for 2 phase. Because there is different fluid in calculating back flow. especially for extreme different viscosity of fluid.

November 10, 2009, 16:08		#9
rr123 Member R. Roy Join Date: Mar 2009 Location: India Posts: 52 Rep Power: 17	hello sir, could you please explain what exactly do you mean by "In pressure outlet, you can determine your backflow without lengthen the BC" ? hoping for a detailed reply kind regards, R.

November 10, 2009, 16:29		#10
teguhtf Senior Member teguh hady Join Date: Aug 2009 Location: Saga, Japan Posts: 222 Rep Power: 18	I'm sorry. I change my post because i think Pressure outlet has been explained by other user. maybe your UDF is wrong. see this example http://www.fluent.com/software/sf_me...orial_wave.htm

November 10, 2009, 17:40		#11
rr123 Member R. Roy Join Date: Mar 2009 Location: India Posts: 52 Rep Power: 17	hi, yes, i use pressure-outlet as also suggested by vanadt. the UDF is not important here because although i have the UDF i disable the macros and what i am trying is just a simple KTGF-enabled tubulent gas-solids flow to converge. the UDF is not going to influence here. note that the energy equation is ON. i am obsevring now that all the residuals have converged, except the residual for the TDR, TKE, w-phase-2 and vof-phase-2. these four parameters are fluctuating slightly above the convergence limit. URF for momentum, TDE and TKR had been lowered to 0.3 and rest all are default. would you like to see my file ? kind regards, R.