CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Periodic boundary condition in 2 directions (https://www.cfd-online.com/Forums/fluent/48557-periodic-boundary-condition-2-directions.html)

 arash June 20, 2008 18:50

Periodic boundary condition in 2 directions

Hi! can anyone anwser me please? i am modeling a rotating channel and i have to define periodic boundary condition in two directions ,one in x and another in z direction , the channel is rotaing around y axes, i can make the 2 different periodic zones , but how can i set the periodic conditions for the both directions? because at the menu Defin/periodic condition i can just set the condition for one direction. thanks

 Phil June 21, 2008 13:18

Re: Periodic boundary condition in 2 directions

You should make the 2 seperate periodic boundaries first in the pre-processor, then define each seperate in fluent. You don't supply a direction in fluent just whether it is rotational or translational.

From what I gather your x is rotational, and your z translational.

Hope this helps Phil

 arash June 22, 2008 06:13

Re: Periodic boundary condition in 2 directions

any idea?

 arash June 22, 2008 06:18

Re: Periodic boundary condition in 2 directions

tnx ,but my problem is that , i should give a pressure gradiant in x direction ,but zero pressure gradiant in z direction , i can make the two periodic zones , but in periodic condition theres is nothing to choose one of this zones !and i think when i give a pressure gradiant it will assigned to both zones.

 Phil June 22, 2008 08:04

Re: Periodic boundary condition in 2 directions

To be honest I'm unsure what you mean by pressure gradient but it sounds like maybe one of them should be symmetry instead of periodic?

 Paolo Lampitella June 23, 2008 04:37

Re: Periodic boundary condition in 2 directions

if you carefully check the panel periodic boundary conditions under define (after you defined the two couples of periodic boundaries) you will see a blank for the specification of the pressure gradient and something (i don't remember actually) for specifying the direction of the pressure gradient

 arash June 24, 2008 09:29

Re: Periodic boundary condition in 2 directions

thanks for your reply , but as i have mentioned before i have to define two seperate periodic boundary conditions one with pressure gradient and another with zero pressure gradient, but in pannel which you mentioned there is no option for choosing the appropriate boundary condition zone ,then i think when i give a pressure gradient it will be set for both periodic zones which is not correct.

 Paolo Lampitella June 24, 2008 13:10

Re: Periodic boundary condition in 2 directions

sorry, i did not understood (my poor english) very well, but maybe could works also in your case.

For example, in my turbulent channel flow i have the flow along the positive x axis (streamwise) and the z axis is the transverse direction (the y axis is perpendicular to the walls). If i want the flow goes along the x axis i just put this as direction. The faces perpendicular to the z axis will have just the same gradient that will be parallel to those faces.

However this works just in non-rotating channel with the two couples of sides being perpendicular to each other.

If your domain is a rectangular channel you could apply a source term to simulate the rotation and use this method to apply periodic boundary conditions.

Hope this helps

 ccy329 August 6, 2015 03:31

Hi,

I was wondering if anyone knows how to solve this problem as I encounter the same problem.

I would like to set translational periodic boundary conditions in the streamwise direction with specified flow rate as well as in the spanwise direction with zero pressure gradient. If I set the periodic conditions for the streamwise direction with a mass flow rate, it sets the same periodic conditions to the other periodic boundary which in this case, the spanwise direction.

How do I define two different periodic boundary conditions separately in Fluent?

I am relatively new to Fluent. Any advise will be greatly appreciated.

 LuckyTran August 7, 2015 13:05

You need to specify the periodic boundary pair with the zero pressure gradient as an interface.

 Ethan_Sparkle December 9, 2016 00:51

Hi Derek, have you solved your problem? I just come across the same problem.

 Antanas December 9, 2016 01:21

In Fluent 17 to define periodicity you just need to create interfaces, activate checkbox that they are periodic, check whether they are rotational or translational and that's all.

 Ethan_Sparkle December 9, 2016 04:37

Quote:
 Originally Posted by Antanas (Post 628877) In Fluent 17 to define periodicity you just need to create interfaces, activate checkbox that they are periodic, check whether they are rotational or translational and that's all.
Thank you for your reply Antanas, I want one periodic BC where mass flow rate is specified to model fully developed flow, and another periodic BC with no pressure drop. But once I define the mass flow rate for one periodic BC, another periodic BC is defined with the same value. It seems they are related to each other. Do you know how to deal with that?

 oozcan December 9, 2016 04:51

could you share with your model and show onto where periodic bc's have been defined by yourself ?

 Ethan_Sparkle December 9, 2016 07:16

Quote:
 Originally Posted by oozcan (Post 628894) could you share with your model and show onto where periodic bc's have been defined by yourself ?
https://www.cfd-online.com/Forums/me...ure840-p-1.jpg
https://www.cfd-online.com/Forums/me...re841-p-11.jpg
like the picture show, when fully developed, the flow is periodic in two directions, which there is a pressure drop along the flow direction, and no pressure perpendicular to the flow direction.
https://www.cfd-online.com/Forums/me...ure842-p-3.png
I use the TUI make-periodic to set these periodic BC, but you can see there is only one periodic condition, it seems once I specified the mass flow rate, the value is given to the periodic BC in both directions.

 oozcan December 9, 2016 07:40

Quote:
 Originally Posted by Ethan_Sparkle (Post 628924) https://www.cfd-online.com/Forums/me...ure840-p-1.jpg https://www.cfd-online.com/Forums/me...re841-p-11.jpg like the picture show, when fully developed, the flow is periodic in two directions, which there is a pressure drop along the flow direction, and no pressure perpendicular to the flow direction. https://www.cfd-online.com/Forums/me...ure842-p-3.png I use the TUI make-periodic to set these periodic BC, but you can see there is only one periodic condition, it seems once I specified the mass flow rate, the value is given to the periodic BC in both directions.
I couldnt see what bc's have been given by yourself.As far as I know, inlet and outlet boundary conditions of flow domain should be both ''wall''. Then you can change with their ID numbers from wall to periodic and its shadow

Let me put it explicitly,

''c'' (from your screenshot) is inlet of fluid domain, then we need to define as ''wall'' like ''c' ''. we assume that their ID numbers are 9, 12 ,respectively,

with TUI command,

mesh>modify-zones>make-periodic

periodic zone :9
then question is : rotational or translational (yes or no)

Dont forget that 9 and 12 are not bc. They must be ID

you should do other one in this way

 Antanas December 9, 2016 07:47

Quote:
 Originally Posted by Ethan_Sparkle (Post 628924) like the picture show, when fully developed, the flow is periodic in two directions, which there is a pressure drop along the flow direction, and no pressure perpendicular to the flow direction. I use the TUI make-periodic to set these periodic BC, but you can see there is only one periodic condition, it seems once I specified the mass flow rate, the value is given to the periodic BC in both directions.
Obviously you should make two separate periodic interfaces (conditions)

 Ethan_Sparkle December 10, 2016 05:17

Quote:
 Originally Posted by Antanas (Post 628932) Obviously you should make two separate periodic interfaces (conditions)
Thanks Antanas, I understand both "interface" and the TUI command "make periodic" can be used to set the periodic boundary. I had successfully defined the periodic BC in both flow direction which is c-c', d-d', e-e' and the direction perpendicular to the flow direction a-a', b-b'. All together five periodic BCs were defined in both directions.
Now my problem is there is only one "Periodic Condition" dialog box in the Boundary Conditions panel, no metter which periodic BC you are editing. So when I define the mass flow rate for one periodic BC, I think it was attached to all the other periodic BCs.

 Ethan_Sparkle December 10, 2016 05:20

Quote:
 Originally Posted by oozcan (Post 628930) I couldnt see what bc's have been given by yourself.As far as I know, inlet and outlet boundary conditions of flow domain should be both ''wall''. Then you can change with their ID numbers from wall to periodic and its shadow Let me put it explicitly, ''c'' (from your screenshot) is inlet of fluid domain, then we need to define as ''wall'' like ''c' ''. we assume that their ID numbers are 9, 12 ,respectively, with TUI command, mesh>modify-zones>make-periodic periodic zone :9 its shadow zone : 12 then question is : rotational or translational (yes or no) make a periodic : your answer must be ''yes'' Dont forget that 9 and 12 are not bc. They must be ID you should do other one in this way
In the screenshot, the BCs inlet1 and inlet2 are c and d, wall-left is e, top-1 and top-2 are a and b, I had already successfully defined the periodic BC in both flow direction which is c-c', d-d', e-e' and the direction perpendicular to the flow direction a-a', b-b'. All together five periodic BCs were defined in both directions.
Now my problem is there is only one "Periodic Condition" dialog box in the Boundary Conditions panel, no metter which periodic BC you are editing. So when I define the mass flow rate for one periodic BC, I think it was attached to all the other periodic BCs. Do you have any idea about that? Thanks for your reply.

 Ethan_Sparkle December 21, 2016 22:03

I guess this problem i.e. model contains periodic boundaries in 2 directions with pressure drop, couldn't be done by Fluent.

 All times are GMT -4. The time now is 22:56.