CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

Can anyone help me to mark interfaces and solve the problem in Fluent

Register Blogs Community New Posts Updated Threads Search

Like Tree1Likes
  • 1 Post By -mAx-

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   August 28, 2016, 06:20
Default Can anyone help me to mark interfaces and solve the problem in Fluent
  #1
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Dear friends,
I am analyzing a hydrodynamic journal bearing which is nothing but has a flow of fluid through eccentric annulus formed by one rotating cylinder (termed as shaft) and stationary cylinder (termed as bearing). The oil comes in the annular space through a inlet supply hole (termed as pressure inlet ) and comes out from the sides of annular space (termed as pressure outlet). One layer of fluid is attached to the shaft (interface 1) and second oil layer is attached to the upper layer.(interface 2) and there are other interfaces as shown in attached figure. The 3D model image is also attached. The fluid layer moves due to rotation of the shaft and pressures are built.

In interface figure, both magenta as well as green are part of fluid which are created due to slice option in design modeler while forming pressure inlet part. Other parts are i think clearly defined.
I am understanding the problem but facing problem in marking the interfaces and frame motions in fluent and analyze problem. Kindly help.
If needed, i can share model file also.
Thanks in advance.


Regards.
Attached Images
File Type: jpg int.jpg (56.5 KB, 23 views)
File Type: png 3D.png (140.3 KB, 19 views)
vedant123 is offline   Reply With Quote

Old   August 28, 2016, 11:15
Default
  #2
New Member
 
Daniel Riveros
Join Date: Nov 2015
Location: Genoa, Italy
Posts: 14
Rep Power: 10
DRiver is on a distinguished road
Hi Vedant

I hope I understand your problem.

As you may know, you only need to model the fluid zone (magenta, green and yellow zone), solid parts, if your are not modeling heat transfer, especially conduction, can be treated as walls. While you are defining boundary conditions, use stationary wall for bearing surface and rotationary wall for shaft surface.

Based on the given information, I think that you only need one fluid zone (magenta, green and yellow together) that way you don't need interfaces, only walls.

Be careful, simulations that only have pressure inlet and pressure outlet tend to be unstable.

Hope that I helped you.
DRiver is offline   Reply With Quote

Old   August 29, 2016, 02:59
Default
  #3
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
Let's say your bearing has Db diameter, and your shaft has Ds diameter.
Assuming you already have a fluid domain (annulus), then you have to split this annulus with a cylinder surface with diameter 0.5*(Db+Ds) for instance.
It will separate your domain into 2 annulus concentric domains.
Now you have to disconnect both annulus for enabling rotation of inner annulus (rotor).
Once they are disconnected, you need to define interfaces:
Interface 1 is the Surface 0.5*(Db+Ds) which belongs to rotor domain, and interface 2 is the one which belongs to the other domain.
Note that both interfaces are superimposed.
In the solver you need to create grid interfaces by picking both interfaces.
Once it is done, you only have to create a rigid motion of you annulus domain (rotor)
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 29, 2016, 05:50
Default
  #4
Member
 
Mustafa
Join Date: May 2013
Posts: 54
Rep Power: 12
Mohawk is on a distinguished road
Dear Max

You need to give names to the interface surfaces as ( interface ) in mesh . For example interface 1 and interface 2 .the in fluent you join them as coupled wall .
Also check this
http://adf.ly/1dWr7A
Mohawk is offline   Reply With Quote

Old   August 29, 2016, 12:24
Default
  #5
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Dear -mAx- and Mohawk, thanks for ur reply. But -mAx-, if you see, as per in fig 1, the shaft is eccentric to the bearing. Hence it is not possible to split the annular space exactly into two equal parts. Again what about magenta and green coloured oil, as the oil enters through inlet port provided in bearing (oil shown as magenta in this region) and green coloured oil is part of yellow coloured oil. So do i need to make a single part of all these three coloured oil and make a part as 'oil' ? Is my thinking correct? I am attaching the fig 2 which shows my understanding to suggestion by -mAx-. But again the earlier question remains.....it is not possible to split the oil into two equal parts. So how to do it.
Kindly help.
Again do i need to provide MRF in Fluent BC to solid shaft? Kindly elaborate the BC's i need to provide in Fluent.

Thanks in advance
Attached Images
File Type: jpg actual bearing.jpg (53.1 KB, 13 views)
File Type: jpg Split suggested.jpg (65.8 KB, 12 views)
vedant123 is offline   Reply With Quote

Old   August 29, 2016, 12:28
Default
  #6
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Thanks to Daniel Riveros also. Sorry i forgot to mention ur name. Actually i am modelling heat transfer too. So do i need to consider only surfaces or solid parts (shaft and bearing)? How to marks BC's in this case? Kindly elaborate. Thanks in advance.
vedant123 is offline   Reply With Quote

Old   August 30, 2016, 01:41
Default
  #7
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
Quote:
Originally Posted by vedant123 View Post
I am attaching the fig 2 which shows my understanding to suggestion by -mAx-. But again the earlier question remains.....it is not possible to split the oil into two equal parts. So how to do it.
If it is eccentric, then you may split as you displayed. Split surface in the middle of smallest gap between rotor and stator.

If I understand the magenta region (inlet), I cannot understand the green one.
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 30, 2016, 12:37
Default
  #8
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Dear -mAx- ,
the green coloured is also part of oil and it is supposed to be one part consisting of magenta, green and yellow colour to be called as fluid domain. It is created due to slice option which i have used while creating pressure inlet. Plz refer attached figure.
Attached Images
File Type: jpg journal bearing.jpg (60.9 KB, 8 views)
vedant123 is offline   Reply With Quote

Old   August 31, 2016, 01:43
Default
  #9
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
As you don't know how to set up a moving mesh which will involve interfaces, I would not start splitting your stator region.
Let's say you start with the yellow region which is the fluid domain between stator and rotor.
You inlet is already defined. Thus it is the top surface of the small cylinder which represents the hole in the stator housing.
Now for enabling the rotation of the rotor, you need to create a region nearby the rotor, and you also have to disconnect it from the fluid domain.
For that create the surface as you previously mentionned (blue line), use it for splitting the fluid domain into 2 separated fluid domains (rotor/stator).
Once it is done you have 2 fluid domains which are still connected.
You need to disconnect them.
For checking if they are disconnected, try to move the rotor somewhere with a translation vector.
If it fails it mean, that they are still connected.
If you can move the rotor, then it is ok and you may take the advantage of this configuration for picking and setting each interface: one from the rotor domain, and one from the stator domain.
Once it is done, move the rotor domain back at its original position.
Note that both interfaces are superimposed.
Now your model is ready, and you can impose a rotation on the rotor. Your rotor mesh will rotate, but stator one will not.
This is sliding mesh
Catabay likes this.
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 31, 2016, 14:23
Default
  #10
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Thanks -mAx- for ur detail reply. I have tried to split the lubricant fluid volume as shown in fig 1. Can you now tell me the next procedure in fluent for sliding mesh as well as moving mesh in detail?

Thanks for ur valuable time.
Attached Images
File Type: jpg splitted volume of lubricant.jpg (68.3 KB, 11 views)
File Type: png 3D splitted body.png (32.3 KB, 10 views)
vedant123 is offline   Reply With Quote

Old   September 1, 2016, 01:20
Default
  #11
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
*sliding mesh is a moving mesh technics.
*Are both domains 1 & 2 disconnected?
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   September 1, 2016, 01:57
Default
  #12
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
yes -mAx- both are spitted with the help of surface and disconnected from each other. That is why we see different colours for region 1 and 2.
vedant123 is offline   Reply With Quote

Old   September 1, 2016, 02:01
Default
  #13
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
alright then if interfaces are already set, you need to define the grid interfaces in the solver by calling both interfaces.
Once it is done, you can enable the motion or rotor by picking the right fluid domain and affecting rigid body motion.
Then test mesh motion.
The rotor should rotate but not the stator domain
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   September 1, 2016, 04:15
Default
  #14
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
do u mean to say, i have to define surface of region 1 as interface 1 and suraface of region2 as interface2? and create interface of both in solver? then should i give motion to shaft or the bottom region of interface 1? and stationary wall to outer surface of interface2?
vedant123 is offline   Reply With Quote

Old   September 1, 2016, 04:26
Default
  #15
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
yes: interface 1 for region1 and interface 2 for region 2
Rigid body motion only on region 2(rotor). Interface 2 belongs to region2, so its motion will be automatically set
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   September 1, 2016, 12:32
Default
  #16
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Dear -mAx, i tried to do as per ur guidelines. But i m getting error: "The mesh file exporter does not support overlapping geometry in Contact Regions. Please resolve the issue and try again."

Can i share my 3D model with you...Will u plz share ur email ID? Plz..i need help....
vedant123 is offline   Reply With Quote

Old   September 2, 2016, 01:20
Default
  #17
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
unfortunately I don't work with ansys/fluent anymore.
So I can't don anything with your files.
I just explained how it worked with gambit and fluent.

In your case maybe you have to export rotor and stator mesh separaretely.
Check the tutorials, for sure it is explained
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   September 3, 2016, 14:30
Default
  #18
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Thanks -mAx- for ur inputs... Can anyone else help me on this issue?

Thanks in advance.
vedant123 is offline   Reply With Quote

Old   September 4, 2016, 22:27
Default
  #19
Senior Member
 
Join Date: Apr 2014
Location: Melbourne
Posts: 584
Rep Power: 14
Kapi is on a distinguished road
mate i dont use Fluent, I use CFX!
with interface are you doing mesh in fluent or in Meshing?
you need to go one by one thru your named interface and see which one you have doubled up!
Kapi is offline   Reply With Quote

Old   September 17, 2016, 13:27
Default
  #20
New Member
 
Vedant
Join Date: Jul 2015
Posts: 12
Rep Power: 10
vedant123 is on a distinguished road
Thanks Kapi for ur reply. Can anyone else help me out on this issue? I can share my geometry if any one is ready to help....Plz guys help.....!
vedant123 is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
[ANSYS Meshing] Creating several components with indivdual meshes to solve fluent problem PeterPete ANSYS Meshing & Geometry 2 March 8, 2016 04:01
CFX does not solve heat transfer at fluid-solid interfaces Ivan Corgozinho CFX 2 April 7, 2015 00:08
how to solve rotor-stator interaction problem? ashish pandit FLUENT 8 November 7, 2008 06:37
problem with GGI interfaces strider CFX 6 May 29, 2006 14:20
Error on defining grid interfaces at FLUENT Ridwan FLUENT 1 July 27, 2005 05:41


All times are GMT -4. The time now is 22:54.