CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS Meshing & Geometry (
-   -   [ICEM] problem with meshing in icem cfd (

saurabhjoshi2011 February 1, 2013 00:14

problem with meshing in icem cfd
dear friends
i am meshing a gas turbine combustor using icem cfd. my geometry has a outer and inner surface and volume between the problem is that when i am meshing the geometry , only the outer surfaces and the volume is getting meshed. the inner surfaces which have a very important role to play in the combustor are not getting recognised. as a result my simulation is not giving the results as on expected lines. Dear friends what could be the reason. i am using patch independent method for surface meshing and octree method for volume meshing. plz help . i am chennai based and my no is 9790702545
saurabh joshi

Far February 1, 2013 05:29

Pics would certainly help to understand problem. Are these zero thickness surfaces?

saurabhjoshi2011 February 1, 2013 06:31

dear FAR
yes they have zero thickness.the inner as well as outer surfaces all have zero thickness.the inner surfaces have holes by design. the outer surface is a closed one. the holes created in the geometry are for metering purpose. for metering of the fluid. do i close these inner holes and make a separate part for them and then give them as some boundary condition(velocity etc). what could be the problem.if you can mail me your address i can send some pics
saurabh joshi

Far February 1, 2013 06:50

You need to turn on the int wall in part mesh setup, after that ICEM will recogise them as internal wall. My email address can be found on left side of this post :rolleyes:

diamondx February 1, 2013 09:37

someone correct me if i'm wrong, but if you have an inner and an outer surface, and if you don't mesh the inside so you declare later on fluent that it's a coupled wall with a thickness. then you need to only have on mid-surface between the two

saurabhjoshi2011 February 4, 2013 00:01

dear FAR
i think it should work ( the thanks is in anticipation)
let the lab open today and let me try
thanks again
saurabh joshi

Far February 4, 2013 00:26


Originally Posted by diamondx (Post 405501)
someone correct me if i'm wrong, but if you have an inner and an outer surface, and if you don't mesh the inside so you declare later on fluent that it's a coupled wall with a thickness. then you need to only have on mid-surface between the two

You mean internal wall?

saurabhjoshi2011 February 4, 2013 01:48

inner surfaces recognised... thanks
Dear FAR
thanks for the correct advise , the inner surfaces have been recognised.
howeever, when the mesh is being imported to FLUENT 14, it is creating shadow of all the inner surfaces.
when i am dispalying the inner surface mesh , only one mesh gets displayed for the shadow and the component;however this shadow is coming in the boundary condition drop down list
what could be the problem
is the shadow ok
plz guide
saurabh joshi

Far February 4, 2013 02:24

Yes it is ok.

saurabhjoshi2011 February 4, 2013 03:21

dear FAR
so i will now run my simulation.
lets hope these shadows which are created will not create problem
saurabh joshi

saurabhjoshi2011 March 11, 2013 01:08

a query regarding setting up the Boundary conditions
dear friend
my computation involves high pressure air at the inlet. i know the velocity, temperature, pressure, mass flow rate of the air at the inlet. at the outlet i need to compute these things(specially the pressure)
now the problem what i am facing is
when i select velocity inlet and outflow as my boundary condition for inlet and exit, during hybrid initialization fluent does not knows the pressure at the BC and my pressure results come unrealisitc
when i select pressure inlet and outflow, they are incompatible.
my flow speed is 130 m/s, and the inlet static pressure is 810600 pascal and the inlet temperature is 572 k.
i am setting this pressure(810600) as the the pressure in the velocity inlet bc. why is the system does not recognising this pressure during hybrid initialization.
choosing pressure inlet and pressure outlet bc for inlet and exit must be wrong as i dont know the oulet pressure.
ideally the pressure should not drop. but it is the pressure drop which is significant in our case, which needs to be calculated.
what is going wrong with my setting up of BC's

stuart23 March 12, 2013 07:39

You need to specify some sort of outlet BC. Try mass-flow if you are unsure of the pressure.


saurabhjoshi2011 March 12, 2013 12:14

Dear Mr Stuart
So are you suggesting mass flow inlet as the inlet BC and may be outflow as exit BC?
actually setting up pressure inlet and pressure outlet is causing problem.
actually for my problem i am interested in knowing the pressure drop.
so for this the system must calculate the pressure at the exit
and in pressure BC i am setting up the outlet pressure which is wrong as per physics.(as i dont know the outlet condition)
what should i do
velocity inlet and outflow has the problem that during initialisation fluent does not knows pressure at BC.
so plz guide

saurabhjoshi2011 March 12, 2013 12:27

dear friends
if the surface integral of mass flow at a certain surface comes +ve than it means mass is flowing in, if it comes negative does it means mass is flowing out.
if i know only inlet conditions(static pressure,tempertaure,axialvelocity) and outlet conditions of the system are unknown i.e to be calculated what could be chosen as BC
velocity------------------ outflow
massflow inlet-------------outflow
plz guide

All times are GMT -4. The time now is 19:03.