Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Problems about Periodic Conditions in the tutorial model “Periodic Boundary Condition”

Hi all,

Recently I confront one confusion about periodic conditions in the tutorial model - Periodic Boundary Condition. In Periodic Condition, it selects only boundaries 1 and 6 which are parallel with the yz-plane. As you know in the model, the cube should be a repeated cell and certainly repeated on the xy-plane both in the x direction and in the y direction. However, boundaries 1 and 6 only correspond to x direction. So what about repetition in the y direction?

Then I try to select both boundary 1, 6 parallel with yz-plane and boundary 2, 5 parallel with xz-plane in the Periodic Condition. But some problem occurs just as follows:

Errors during multiphysics.
Failed to find destination boundaries or destination selection empty
- Feature: Periodic Condition 1 (pc1)

I don't know why this happens and how to set Periodic Condition that means a repeated cell in the xy-plane both in the x direction and in the y direction.

Thank you so much.

8 Replies Last Post Apr 10, 2015, 8:29 AM EDT
Alexander Kuznetsov
Posted: 5 years ago Feb 11, 2013, 2:17 PM EST
As far as I remember, this example considers particular set of angles of incidence and polarization, at which having periodic BC in y-direction is not required. For more general case, you will need to have periodic BC in both x and y directions. They should be defined as two different sets of BC, one for each of the directions. Then if you define your k-vectors properly, everything should work just fine.

Also make sure your mesh is identical on symmetric BC.
As far as I remember, this example considers particular set of angles of incidence and polarization, at which having periodic BC in y-direction is not required. For more general case, you will need to have periodic BC in both x and y directions. They should be defined as two different sets of BC, one for each of the directions. Then if you define your k-vectors properly, everything should work just fine. Also make sure your mesh is identical on symmetric BC.

Posted: 5 years ago Feb 11, 2013, 2:51 PM EST
Thank you so much for your reply. But I'm still confused about two possible situations when I set periodic BC in both x and y directions.


OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO


O
O
O
OOOOOOOOOO
O
O
O


I hope you can understand what I mean for the above pictures. Could you please help me to explain that?

Thank you.
Thank you so much for your reply. But I'm still confused about two possible situations when I set periodic BC in both x and y directions. OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO O O O OOOOOOOOOO O O O I hope you can understand what I mean for the above pictures. Could you please help me to explain that? Thank you.

Posted: 5 years ago Feb 11, 2013, 4:41 PM EST
There are some mistakes for the display of the second picture. The "O" in the first three lines and last three lines should be exactly above and under the "O". That's what I really mean.
There are some mistakes for the display of the second picture. The "O" in the first three lines and last three lines should be exactly above and under the "O". That's what I really mean.

Alexander Kuznetsov
Posted: 5 years ago Feb 12, 2013, 12:48 PM EST
Sorry, I don't get what those O-s mean.
Sorry, I don't get what those O-s mean.

Posted: 5 years ago Feb 12, 2013, 1:31 PM EST
Sorry to confuse you. I mean "O" is the original single cell.

When I set the x direction periodic BC, it will become like this

OOOOOOOOOOO

When I set the y direction periodic BC, it will become like this

O
O
O
O

O
O
O
O

But when I set both x and y direction periodic BC, what will it be? Probably it can be periodic on the x-y plane.

OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO
OOOOOOOOOO

However, there exists another probability that it may become a "cross" periodic model just like this

O
O
O
O
OOOOOOO
O
O
O
O

So I'm confused about these two probabilities and can't decide which it should be.

Thank you.
Sorry to confuse you. I mean "O" is the original single cell. When I set the x direction periodic BC, it will become like this OOOOOOOOOOO When I set the y direction periodic BC, it will become like this O O O O O O O O O But when I set both x and y direction periodic BC, what will it be? Probably it can be periodic on the x-y plane. OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO OOOOOOOOOO However, there exists another probability that it may become a "cross" periodic model just like this O O O O OOOOOOO O O O O So I'm confused about these two probabilities and can't decide which it should be. Thank you.

Alexander Kuznetsov
Posted: 5 years ago Feb 12, 2013, 5:34 PM EST
It's the first one with square of O's
It's the first one with square of O's

Posted: 5 years ago Feb 12, 2013, 11:26 PM EST
Thank you so much.
Thank you so much.

Duy Duc Nguyen
Posted: 3 years ago Apr 10, 2015, 8:29 AM EDT
Hi Nan Zhang

I would like to study Periodic condition in Comsol too. However, I can not locate where are the tutorial files of field in Comsol's library. Can you help me ? (Or models in www.comsol.com/models)

Thank you very much
Best regards
Duc
Hi Nan Zhang I would like to study Periodic condition in Comsol too. However, I can not locate where are the tutorial files of field in Comsol's library. Can you help me ? (Or models in http://www.comsol.com/models) Thank you very much Best regards Duc

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.