|
[Sponsors] |
October 8, 2014, 02:44 |
grid interfaces
|
#1 |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
hi everyone
im working on a project and want to simulate a vawt. im really new to cfd.. i have made my plan in gambit but i think i have problem with grid interfaces in the first step, i want to check it to be correct. in the first picture u see my model in gambit. u see there is a ring that i want to use sliding mesh for. in the second image u see my zone defining page. i have defined interface for inner and outer circles. but i dont know whether my grid defining is okay or not. which edges shall i choose for grid interfaces defining? can u help me? |
|
October 8, 2014, 04:03 |
|
#2 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
will you work in 2d or 3d?
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
October 8, 2014, 04:14 |
|
#3 |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
im working in 2d
|
|
October 8, 2014, 04:29 |
|
#4 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
ok
Basically, you create a plane, and a disk. You move the disk at its place, and you split the plane with the disk. Now copy the disk anywhere, and delete the original disk. If you enable shaded mode on, you should see your plane with a hole. Set the circle edge from hollow as interface_1. Set the circle edge from copy also as interface_2. Finally move back (not copy) the disk where hollow is. Now your sliding mesh geometry is ready for meshing
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
October 8, 2014, 04:43 |
|
#5 | |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
Quote:
i have 3 faces, the wind tunnel face that is a square and a hole in it. the second is the ring that i will use sliding mesh for this area and the last face that is actually the inner circle. as u see in gambit i have defined 2 edges as interface, inner & outer circles. now i want to define grid interface in fluent, i choose these 2 interface edges. is it okay? or i should change interfaces in gambit? |
||
October 8, 2014, 04:57 |
|
#6 |
Senior Member
Rick
Join Date: Oct 2010
Posts: 1,016
Rep Power: 27 |
Hi!
see attachment for interfaces bc and zone definition. You have to create 4 interfaces bc in gambit and 2 zones. In fluent you will create 2 interfaces: - interface a) with interface outer 1 and interface outer 2 - interface b) with interface inner 1 and interface inner 2 Daniele |
|
October 8, 2014, 05:17 |
|
#7 | |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
Quote:
about the zones i have already done this but now a question arises, u said that to choose 2 interfaces for 1 edge. that means 4 interfaces for 2 edges. i can define them but is that okay to choose each edges 2 times to introduce 2 interfaces? is it okay? then the second question is that in fluent in grid interface define section how should define grid interfaces with these 4 interfaces? cuz before i had only 2 interfaces sorry im really a beginner |
||
October 8, 2014, 05:22 |
|
#8 |
Senior Member
Rick
Join Date: Oct 2010
Posts: 1,016
Rep Power: 27 |
No, it isn't the same edge for outer interface (for example); you have 2 overlapping edges there.
So, following Maxime's advice, create the ring, then move the ring somewhere else, and set correct bc. |
|
October 8, 2014, 05:25 |
|
#9 |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
sorry i didnt read ur text, i just checked the pic
okay grid interfaces will be okay in fluent but the first question still remains cuz i hav only 2 edges as inner and outer circle. i remember that i could choose which side of an edge i want to work with in ABAQUS but there's nothing like this in gambit. |
|
October 8, 2014, 05:30 |
|
#10 |
Senior Member
Rick
Join Date: Oct 2010
Posts: 1,016
Rep Power: 27 |
This is because the edges are shared between the rotor and the stator domains, and this is not right (try to move the ring face, if an error occurs this means edges are shared).
I suggest to delete faces (uncheck delete lower geometry): you will have only edges; then copy both the ring edges (do not move them). Then move the 2 new created edges somewhere else. Create the faces: outer wind tunnel, ring and face inside the ring; create 4 interfaces in gambit, then move back the ring to its original position. Daniele |
|
October 8, 2014, 05:31 |
|
#11 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
you will have to set also 2 fluid domains (one for stator and one for rotor)
Doing that, your interfaces will have 2 adjacent zones (stator and rotor) https://www.sharcnet.ca/Software/Gam...zone_cont_type
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
October 8, 2014, 07:06 |
|
#12 | |
New Member
Mohyeddin Gholami
Join Date: Oct 2014
Posts: 6
Rep Power: 12 |
Quote:
|
||
October 8, 2014, 07:16 |
|
#13 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
you don't have to move back the circles, but the ring (surface)
And the ring contains the circles (interfaces) Anyway, if you try to move the circles, it should fail because they are supposed to be attached to the surface
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
October 8, 2014, 07:53 |
|
#14 |
Senior Member
Rick
Join Date: Oct 2010
Posts: 1,016
Rep Power: 27 |
See attached.
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
grid interfaces in unsteady RANS | Koen | FLUENT | 0 | September 19, 2007 10:55 |
Non-Conformal Grid Interfaces | Sridhar | FLUENT | 0 | May 3, 2002 04:04 |
Fluent - discontinuous grid interfaces | Simon Archen | FLUENT | 2 | November 16, 2001 10:52 |
Grid Independent Solution | Chuck Leakeas | Main CFD Forum | 2 | May 26, 2000 12:18 |
Numerical methods for discontinuous grid interfaces? | Hansong Hang | Main CFD Forum | 12 | September 16, 1998 23:26 |