How to draw 2 Inlets in 3D pipe

mariam.sara · December 12, 2011, 05:48

Hi I am tried to draw 3D pipe with two inlets in gambit, one for water (which flows on the bottom) and the second for air (which flow on the top of the pipe) as illustrated in picture the outlet face is shared for both phases may somebody help me with it.

Mariam.sara

mariam.sara · January 4, 2012, 07:57

max here is my question I resend it I want draw the same pipe as shown on picture exactly how to do it ?
thanks
mariam

mariam.sara · January 4, 2012, 08:00

max here is my question I resend it I want draw the same pipe as shown on picture exactly how to do it ? the problem the phases not distributed equally the water level is 11.5 mm, pipe diameter is 60 mm and length is 16 m?
thanks
mariam

-mAx- · January 4, 2012, 08:47

On inlet face, you can split the surface according to your dimensions.
Then in BC define your 2 inlets separately

mariam.sara · January 4, 2012, 11:15

Thaks max for the reply I did several attempts but failed to draw the same mesh succesfully I will be grateful if you send me a JOU file for the pipe with meshing it. Furthermore if i want draw the interface between the phases on the the same geometry how i can do that ?

Many thanks
Mariam

-mAx- · January 5, 2012, 02:46

*create first your pipe (60x16000)
*Then split the volume for having just one half
*Then split the half pipe with a plane (normal to symmetry plane but parallel to pipe axis) which will cut the volume at 11.5mm (water level)
Now you pipe is divided in 2 subdomains, one for water and one for air (hydrostatic)
Finally mesh both volumes

Here you can find basic tutorials: http://my.fit.edu/itresources/manual...uide/tgtoc.htm

mariam.sara · January 6, 2012, 16:58

Thanks max for the tutorials. I draw half of the pipe but I think I failed with drawing the same boundary layers at the interface as seen from the picture (this picture from a paper I want to compare my results with) the mesh is concentrated above and down the interface as seen, furthermore I want to give interfacial roughness Ki for the interface to take the effect of waves ? how I can do this?

Thanks again

-mAx- · January 9, 2012, 02:15

did you split the half volume for having 2 volumes (water and air)?
In this way you can create an interior surface which represents the "interface" between your 2 mediums

mariam.sara · January 10, 2012, 07:31

max i sent you the JOU file yes i was split the volume I am afraid there is a wrong with my interface representation I define it as a wall in gambit then give it roughness in fluent to set it as a rough interface is this idea true?

see attached JOU file pls

-mAx- · January 10, 2012, 08:12

I don't exactly what you mean with interface roughness, but setting it as wall is wrong.
You also set your interface as interface, which is wrong, since interface in numerics doesn't mean the same as in the physics.
Hence in Gambit, Interface means you want to interpolate data from a surface mesh (interface) to another surface mesh (interface), where both surface meshes aren't connected. >> check sliding mesh for better understanding.
In your case, splitting your volumes into 2 subvolumes, gives you the advantage to patch your vof values on both domains for initialization.
Now for a vof calculation, I would only split the volume, define fluid_region on both volumes, your inlet and outlet.
Interaction between both phases should be available in fluent (surface tension)
check tutorial: http://my.fit.edu/itresources/manual...tg/node290.htm
And finally you don t need to set wall condition, since all surfaces without BC are set to wall

mariam.sara · January 10, 2012, 16:13

Thanks a lot max for the valuable reply. I mean with interface the free surface between the two phases (air & water) and I do not mean with it interface BC between two different mesh regions I aware this thing. The free surface is wavy in nature and rough hence I tried to define it as a rough moving wall (I set velocity at interface equal to liquid velocity) I am really not sure if this idea is true and please if there are other ideas you can suggest me about it let me know.

takin · January 11, 2012, 01:21

Dear Maryam!
do you have a problem with creating the geometry in gambit or with CFD simulation of it?

mariam.sara · January 11, 2012, 06:55

Hello Takin I think I am succeeded with drawing the mesh but not sure from my free surface representation and the results not seems too close to experiments cause the velocity of water region appeared as zero may something wrong with the free surface boundary which i represent it as moving rough wall?

takin · January 12, 2012, 02:06

when you see your boundaries in fluent, you should see surface_name and surface_name_shadow. for this problem, you should have one surface between two fluids, not two surfaces. it is better to make the geometry with SPLIT function.

mariam.sara · January 12, 2012, 08:09

Yes that's true I have two surfaces one named shadow surface when exporting the msh file to fluent. I not understand what you mean with SPLIT function is it the same as split option in gambit?

mariam.sara · January 14, 2012, 15:18

max may you tell me what takin mean by split function which he mentioned it in his past reply to me? do you know how I can use it in Gambit?

Thanks a lot
mariam

-mAx- · January 16, 2012, 02:14

http://my.fit.edu/itresources/manual...eom_volm_split

January 4, 2012, 08:47		#4
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	On inlet face, you can split the surface according to your dimensions. Then in BC define your 2 inlets separately __________________ In memory of my friend Hervé: CFD engineer & freerider

January 5, 2012, 02:46		#6
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	create first your pipe (60x16000) Then split the volume for having just one half *Then split the half pipe with a plane (normal to symmetry plane but parallel to pipe axis) which will cut the volume at 11.5mm (water level) Now you pipe is divided in 2 subdomains, one for water and one for air (hydrostatic) Finally mesh both volumes Here you can find basic tutorials: http://my.fit.edu/itresources/manual...uide/tgtoc.htm __________________ In memory of my friend Hervé: CFD engineer & freerider

January 9, 2012, 02:15		#8
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	did you split the half volume for having 2 volumes (water and air)? In this way you can create an interior surface which represents the "interface" between your 2 mediums __________________ In memory of my friend Hervé: CFD engineer & freerider

January 10, 2012, 08:12		#10
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	I don't exactly what you mean with interface roughness, but setting it as wall is wrong. You also set your interface as interface, which is wrong, since interface in numerics doesn't mean the same as in the physics. Hence in Gambit, Interface means you want to interpolate data from a surface mesh (interface) to another surface mesh (interface), where both surface meshes aren't connected. >> check sliding mesh for better understanding. In your case, splitting your volumes into 2 subvolumes, gives you the advantage to patch your vof values on both domains for initialization. Now for a vof calculation, I would only split the volume, define fluid_region on both volumes, your inlet and outlet. Interaction between both phases should be available in fluent (surface tension) check tutorial: http://my.fit.edu/itresources/manual...tg/node290.htm And finally you don t need to set wall condition, since all surfaces without BC are set to wall __________________ In memory of my friend Hervé: CFD engineer & freerider

January 10, 2012, 16:13		#11
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Thanks a lot max for the valuable reply. I mean with interface the free surface between the two phases (air & water) and I do not mean with it interface BC between two different mesh regions I aware this thing. The free surface is wavy in nature and rough hence I tried to define it as a rough moving wall (I set velocity at interface equal to liquid velocity) I am really not sure if this idea is true and please if there are other ideas you can suggest me about it let me know. Last edited by mariam.sara; January 11, 2012 at 06:46.

January 4, 2012, 07:57		#2
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	max here is my question I resend it I want draw the same pipe as shown on picture exactly how to do it ? thanks mariam

January 4, 2012, 08:00		#3
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	max here is my question I resend it I want draw the same pipe as shown on picture exactly how to do it ? the problem the phases not distributed equally the water level is 11.5 mm, pipe diameter is 60 mm and length is 16 m? thanks mariam

January 4, 2012, 11:15		#5
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Thaks max for the reply I did several attempts but failed to draw the same mesh succesfully I will be grateful if you send me a JOU file for the pipe with meshing it. Furthermore if i want draw the interface between the phases on the the same geometry how i can do that ? Many thanks Mariam

January 6, 2012, 16:58		#7
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Thanks max for the tutorials. I draw half of the pipe but I think I failed with drawing the same boundary layers at the interface as seen from the picture (this picture from a paper I want to compare my results with) the mesh is concentrated above and down the interface as seen, furthermore I want to give interfacial roughness Ki for the interface to take the effect of waves ? how I can do this? Thanks again

January 11, 2012, 01:21		#12
takin New Member MK Join Date: Aug 2009 Posts: 22 Rep Power: 17	Dear Maryam! do you have a problem with creating the geometry in gambit or with CFD simulation of it?

January 11, 2012, 06:55		#13
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Hello Takin I think I am succeeded with drawing the mesh but not sure from my free surface representation and the results not seems too close to experiments cause the velocity of water region appeared as zero may something wrong with the free surface boundary which i represent it as moving rough wall?

January 12, 2012, 02:06		#14
takin New Member MK Join Date: Aug 2009 Posts: 22 Rep Power: 17	when you see your boundaries in fluent, you should see surface_name and surface_name_shadow. for this problem, you should have one surface between two fluids, not two surfaces. it is better to make the geometry with SPLIT function.

January 12, 2012, 08:09		#15
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Yes that's true I have two surfaces one named shadow surface when exporting the msh file to fluent. I not understand what you mean with SPLIT function is it the same as split option in gambit?

January 14, 2012, 15:18		#16
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	max may you tell me what takin mean by split function which he mentioned it in his past reply to me? do you know how I can use it in Gambit? Thanks a lot mariam

January 16, 2012, 02:14		#17
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	http://my.fit.edu/itresources/manual...eom_volm_split __________________ In memory of my friend Hervé: CFD engineer & freerider

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[DesignModeler] DesignModeler Pipe within pipe	shields	ANSYS Meshing & Geometry	13	November 25, 2018 23:14
How to draw pipe mesh	Ismael	FLUENT	2	March 3, 2007 17:27
how to draw velocity profile(in unstructure pipe )	lenson	Siemens	8	November 19, 2005 07:13
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 10:11
fluid flow fundas	ram	Main CFD Forum	5	June 17, 2000 22:31