A ghost like outlet appears in my model

minecraftgp · August 25, 2018, 11:27

Hi guys,

I was modeling a ventilated room with a inlet and outlet at the top. When I was using pimpleFoam, everything went well. The air went in and out through inlet and outlet.

However, when I changed the solver to buoyantPimpleFoam, things were totally out of control: There was totally no air went out through the preset outlet, but instead an unknown outlet was found at the bottom corner wall of the room.

I call it a ghost like outlet since I have no clue how it shows up in my model, I have checked all of my settings including the geomtry mesh setup and there is no geomtry or code about "generating a outlet at the corner".

Anyone could help plz??? This almost drives me mad!

ghost like outlet.png
the ghost like outlet

normal1.png
the normal inlet and outlet

normal.png
the normal inlet and outlet

minecraftgp · August 25, 2018, 13:33

Anybody could help? There seems to be another similar problem occured when others using buoyantPimpleFoam. buoyantBoussinesqPimpleFoam makes a random cell an outlet

should this be a bug of the solver?

wyldckat · August 25, 2018, 13:58

Quick answers:

How to give enough info to get help
It's due to wrong boundary conditions.
My guess is that you are ignoring how the pressure field is defined with the buoyantPimpleFoam solver. If you take a look at the pressure fields at 0s and at that time step, you will see a monstrous change in values.
buoyantPimpleFoam solver uses gravity, while pimpleFoam does not.

minecraftgp · August 26, 2018, 18:55

Hi Bruno

Thanks for your reply and tips on how to get help! Also I want to correct my post that the solver I used is the buoyantBousinesqPimepleFoam rather than the buoyantPimpleFoam.

Yes I did find that this unknown outlet is caused by boundry setting.The orignal U for inlet is (0 0 -0.1) and for outlet is zero gradient. I found that if I change the outlet U to the fixed value (0 0 0.1), then the unknown outlet disappeared and the U and T field in the room looked reasonable.

In this study I am focusing on the CO2 concentration under different air flow rather than the ventilation itself, so it seems that fixed velocity for inlet and outlet can totally satisfied the need of this study.

But it is approparite to use such settings? Since there seems no openfoam tutorial using a fixed velocity outlet. And considering buoyantBousinesqPimepleFoam is not a completely incompressible solver, should I use a "pressure based inlet" rather than a "velocity based inlet" ?

Cheers

piu58 · August 27, 2018, 00:54

My 2c:

It is not possible to get a well thought answer if you don't describe your problem in a complete way. May be, such a description helps yourself understanding your problem.

I assume that you know that the buoyant solvers requires p_rgh instead of p?

The solver tends to numerical errors in recirculation areas. I don't know whether this is your problem. But if, you may split the calculation in a flow part and a diffusion part - making two independend simulations. This is possible because your flow does not depend on the diffusion (as far as I understand your case).

minecraftgp · August 31, 2018, 11:37

Hi Uwe

Thanks for your reply.

Yes, I found that the problem was totally caused by p_rgh setting, and it had been solved by change the inlet and outlet p_rgh from fixedFlux to fixed value. Split the simulation into two parts seems worth to try!

August 25, 2018, 11:27	A ghost like outlet appears in my model	#1
minecraftgp New Member Pu Gong Join Date: Jul 2018 Location: London Posts: 22 Rep Power: 7	Hi guys, I was modeling a ventilated room with a inlet and outlet at the top. When I was using pimpleFoam, everything went well. The air went in and out through inlet and outlet. However, when I changed the solver to buoyantPimpleFoam, things were totally out of control: There was totally no air went out through the preset outlet, but instead an unknown outlet was found at the bottom corner wall of the room. I call it a ghost like outlet since I have no clue how it shows up in my model, I have checked all of my settings including the geomtry mesh setup and there is no geomtry or code about "generating a outlet at the corner". Anyone could help plz??? This almost drives me mad! ghost like outlet.png the ghost like outlet normal1.png the normal inlet and outlet normal.png the normal inlet and outlet

August 25, 2018, 13:58		#3
wyldckat Retired Super Moderator Bruno Santos Join Date: Mar 2009 Location: Lisbon, Portugal Posts: 10,975 Blog Entries: 45 Rep Power: 128	Quick answers: How to give enough info to get help It's due to wrong boundary conditions. My guess is that you are ignoring how the pressure field is defined with the buoyantPimpleFoam solver. If you take a look at the pressure fields at 0s and at that time step, you will see a monstrous change in values. buoyantPimpleFoam solver uses gravity, while pimpleFoam does not. __________________ OpenFOAM: FAQ \| Getting started Forum: How to get help, to post code/output and forum guide Read this before sending me PM

August 27, 2018, 00:54		#5
piu58 Senior Member Uwe Pilz Join Date: Feb 2017 Location: Leipzig, Germany Posts: 744 Rep Power: 15	My 2c: It is not possible to get a well thought answer if you don't describe your problem in a complete way. May be, such a description helps yourself understanding your problem. I assume that you know that the buoyant solvers requires p_rgh instead of p? The solver tends to numerical errors in recirculation areas. I don't know whether this is your problem. But if, you may split the calculation in a flow part and a diffusion part - making two independend simulations. This is possible because your flow does not depend on the diffusion (as far as I understand your case). __________________ Uwe Pilz -- Die der Hauptbewegung überlagerte Schwankungsbewegung ist in ihren Einzelheiten so hoffnungslos kompliziert, daß ihre theoretische Berechnung aussichtslos erscheint. (Hermann Schlichting, 1950)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Inconsistencies in reading .dat file during run time in new injection model	Scram_1	OpenFOAM	0	March 23, 2018 22:29
How to calculate particle size at outlet of a 3D nozzle model in fluent?	nishu3210	Main CFD Forum	0	January 21, 2015 13:54
manualInjection model in sprayFoam	Mentalo	OpenFOAM Running, Solving & CFD	1	April 2, 2014 09:29
one phase outlet VOF - Model	Jim87	FLUENT	6	November 13, 2013 02:15
Question about outlet boundary condtion settting for transitional model	Anna Tian	CFX	2	March 3, 2013 18:35

August 25, 2018, 13:33		#2
minecraftgp New Member Pu Gong Join Date: Jul 2018 Location: London Posts: 22 Rep Power: 7	Anybody could help? There seems to be another similar problem occured when others using buoyantPimpleFoam. buoyantBoussinesqPimpleFoam makes a random cell an outlet should this be a bug of the solver?

August 26, 2018, 18:55		#4
minecraftgp New Member Pu Gong Join Date: Jul 2018 Location: London Posts: 22 Rep Power: 7	Hi Bruno Thanks for your reply and tips on how to get help! Also I want to correct my post that the solver I used is the buoyantBousinesqPimepleFoam rather than the buoyantPimpleFoam. Yes I did find that this unknown outlet is caused by boundry setting.The orignal U for inlet is (0 0 -0.1) and for outlet is zero gradient. I found that if I change the outlet U to the fixed value (0 0 0.1), then the unknown outlet disappeared and the U and T field in the room looked reasonable. In this study I am focusing on the CO2 concentration under different air flow rather than the ventilation itself, so it seems that fixed velocity for inlet and outlet can totally satisfied the need of this study. But it is approparite to use such settings? Since there seems no openfoam tutorial using a fixed velocity outlet. And considering buoyantBousinesqPimepleFoam is not a completely incompressible solver, should I use a "pressure based inlet" rather than a "velocity based inlet" ? Cheers

August 31, 2018, 11:37		#6
minecraftgp New Member Pu Gong Join Date: Jul 2018 Location: London Posts: 22 Rep Power: 7	Hi Uwe Thanks for your reply. Yes, I found that the problem was totally caused by p_rgh setting, and it had been solved by change the inlet and outlet p_rgh from fixedFlux to fixed value. Split the simulation into two parts seems worth to try!