Which Module should I choose?

500 views
Skip to first unread message

Arin Immo

unread,
May 6, 2013, 4:50:05 AM5/6/13
to comsol...@googlegroups.com
I want to deal with the problem:  A Guassian beam transports into a fibre from one end ,I want to know how it behaves at the other end  ,which  physical module should I choose?The " Mathmatics"  module  or the "  RF" moulde,or  another one?

Pavel Dergachev

unread,
May 6, 2013, 4:54:41 AM5/6/13
to comsol...@googlegroups.com
Лучше, для начала выбрать "RF". (там уже введены уравнения и граничные условия)
В "Mathmatics" все придется задавать самому - этот модуль нужен для случаев, когда не хватает функциональности стандартных модулей.

Arin Immo

unread,
May 6, 2013, 11:08:02 AM5/6/13
to comsol...@googlegroups.com
But how to define the incoming field as  a Guassian beam  in RF module? It seems impossible! Can you give me your email address? I will send you my file about it( because  to upload my file directly is very slow)! something wrong with it!

Pavel Dergachev

unread,
May 6, 2013, 2:38:51 PM5/6/13
to
Посмотрите на стандартные примеры из COMSOL:

UPD. Если проблемы с присоединением файла, попробуйте выложить на любой файлообменник.

Arin Immo

unread,
May 7, 2013, 4:40:07 AM5/7/13
to comsol...@googlegroups.com
I have met the error!  in the "Guassian .mph"->"Electromagnetic  Waves,Tansistent2"->"Scattering Boundary Condition1"->"Scattering Boundary Condition",  when I set E0,the yellow text tells: Unexpected unit of input.   What does it mean?

And you  go  "study 2" ->"compute",there will be another error!
Guassian.mph

Arin Immo

unread,
May 7, 2013, 4:45:12 AM5/7/13
to comsol...@googlegroups.com
I follow the file  in the attachment step by step!
models.rf.second_harmonic_generation (1).pdf

Arin Immo

unread,
May 7, 2013, 4:48:40 AM5/7/13
to comsol...@googlegroups.com
The second error lies in "Electromagnetic  Waves,Tansistent2"  ->"Wave Equation ,Electric 1"->"Electric Displacement  Field "

Pavel Dergachev

unread,
May 7, 2013, 5:52:47 AM5/7/13
to comsol...@googlegroups.com
1. В "Wave Equation, Electric 1" должно быть <d33*temw2.Ez^2>
2. В "Time Dependent" должно быть что-то похожее на <0 61[fs] 90[fs] 120[fs]>
3. В "Scattering Boundary Condition 1" Warning можно смело проигнорировать.

Arin Immo

unread,
May 7, 2013, 9:03:27 AM5/7/13
to comsol...@googlegroups.com
http://www.youtube.com/watch?v=DkY4vaKo72wYes! It  works!  what dose the term "perfectly   matched layer"  mean?And in which condition I may use it?

Arin Immo

unread,
May 8, 2013, 11:18:25 AM5/8/13
to comsol...@googlegroups.com
  I have met the error "There is no feature using these setting" when i go "mesh" -> "size 1", why does it happen?      And if you go "mesh  1"->"FreeTetrahedral  ",there will be another error,why?
Gussian_simulation.mph

Pavel Dergachev

unread,
May 12, 2013, 8:51:16 AM5/12/13
to comsol...@googlegroups.com
У меня вроде все работает - я перешел на версию 4.3b.

Arin Immo

unread,
May 12, 2013, 9:32:24 PM5/12/13
to comsol...@googlegroups.com
Welcome back! You must be very busy these days! I have met a lot of troubles and I finally overcome them!  However ,I don't know why the Electric field was always 0  whichever  study I use in 'RF'  module!  Maybe because I'm confused with"  wave equation    ,electic"   "Electric  field"  , it seems that they can't coexist  ,or they will override each other!


By the way,  how  did you switch it to 4.3b?  I  know  there is a  new photonic  module in it which may solve my stuff better!  I need it!
Message has been deleted

Pavel Dergachev

unread,
May 13, 2013, 3:32:05 AM5/13/13
to comsol...@googlegroups.com
У Вас геометрия разделена на 2 части и я не вижу условий сочленения геометрии в уравнениях.
Вероятно это и есть причина проблем.

http://www.comsol.com/products/4.3b/

Arin Immo

unread,
May 13, 2013, 9:48:45 AM5/13/13
to comsol...@googlegroups.com
I have been with the problem for five days! I still can't work it out !What conditions should I add to the joint geometry equations?

Pavel Dergachev

unread,
May 13, 2013, 9:59:45 AM5/13/13
to comsol...@googlegroups.com
Попробуйте геометрию собрать в виде "From Union", вместо текущего "From Assembly".

Arin Immo

unread,
May 13, 2013, 10:51:37 PM5/13/13
to comsol...@googlegroups.com
It doesn't work  either!  O h,God! Where is the way  out?

Pavel Dergachev

unread,
May 14, 2013, 2:59:59 AM5/14/13
to comsol...@googlegroups.com
Попробуйте использовать "Electromagnetic Waves, Transient".

Arin Immo

unread,
May 15, 2013, 10:52:14 AM5/15/13
to comsol...@googlegroups.com
It works !  I have met some small   problems
QQ截图20130515224614.jpg
QQ截图20130515224641.jpg

Arin Immo

unread,
May 15, 2013, 10:55:19 AM5/15/13
to comsol...@googlegroups.com
what does "From parent"(Picture 2)mean?  And when  I  plot something , why there is  the  only option"None"(Picture 1)?

Pavel Dergachev

unread,
May 15, 2013, 11:48:15 AM5/15/13
to comsol...@googlegroups.com
1. Данные берутся такие как у его родительского элемента в дереве.
2. Видимо нет данных нужного типа.

Arin Immo

unread,
May 16, 2013, 8:15:17 AM5/16/13
to comsol...@googlegroups.com
What do they mean of the options in boundary probe ?I  mean,for  example,the maximum  option,does   the solver  record the maximum value of Ez field in the boundary every time point?
QQ截图20130516195507.jpg

Pavel Dergachev

unread,
May 16, 2013, 8:20:16 AM5/16/13
to comsol...@googlegroups.com
Максимальное, минимальное и среднее значение указанной величины на выбранной границе, в каждый момент времени.

Arin Immo

unread,
May 16, 2013, 9:07:41 AM5/16/13
to comsol...@googlegroups.com
for the output surface, I want to define it without reflection,   which boundary   condition   should I use,the scattered one or the initial one or anther one ?  These days I'm considering the appropriate boundary condition for the   solid!

Pavel Dergachev

unread,
May 16, 2013, 9:33:16 AM5/16/13
to comsol...@googlegroups.com
Мне сложно ответить на вопрос - я практически не занимаюсь задачами такого типа.
Надо разбираться, что бы сказать точно, а на это нет времени.

Arin Immo

unread,
May 17, 2013, 12:39:15 PM5/17/13
to comsol...@googlegroups.com
You must be  very tired! It seems that you are the only one answering questions  for us!  Thanks for all you have done for the forum!
For the last question, I mean this:  It's  a periodic boundary condition of  Bragg fiber structure!
 From the innermost boundary to the exterior boundaries 
at the first innermost boundary ,the transverse  electric field is 0,at the second boundary ,the electric field is minimum,  the other boundaries  follow  the two boundaries on order
so,what boundary conditions should I define  for the two boundary respectively?
bragg_fiber.mph

Pavel Dergachev

unread,
May 20, 2013, 3:24:04 PM5/20/13
to comsol...@googlegroups.com
Геометрия должна быть "From Union".

Arin Immo

unread,
May 22, 2013, 9:49:12 AM5/22/13
to comsol...@googlegroups.com
I don't understand why there is negative  value of (emw.Ez) ^2 in the figure! There may be imaginary part in "emw.Ez" . But what is the physical meaning of  the imaginary part   of Ez? And if Ez is a complex number,  how to get the module of Ez  ?
QQ截图20130522210437.jpg

Arin Immo

unread,
May 23, 2013, 5:00:45 AM5/23/13
to comsol...@googlegroups.com
What does (-nz)mean? Does it indicate the direction of the surface? 
未命名.jpg

Pavel Dergachev

unread,
May 24, 2013, 1:44:35 AM5/24/13
to comsol...@googlegroups.com
Модуль -> "abs()"
Мнимая часть -> "imag()"
Мнимая часть -> "real()"

"nz" - z компонента вектора нормали к поверхности

Arin Immo

unread,
May 25, 2013, 12:41:15 AM5/25/13
to comsol...@googlegroups.com
for Nz ,If I want to  indicate  a certain direction ,should I put a specific value for Nz  or just let it be Nz?

Pavel Dergachev

unread,
May 26, 2013, 7:19:03 AM5/26/13
to comsol...@googlegroups.com
Задается направление, наверное просто "nz".

Arin Immo

unread,
May 29, 2013, 6:24:09 AM5/29/13
to comsol...@googlegroups.com
How should I deal with the Error:
Errors during multiphysics.
Automatic scaled regions cannot treat selections without exterior and interior boundaries.
- Feature: Perfectly Matched Layer 1 (pml1)
 

Pavel Dergachev

unread,
May 29, 2013, 1:30:54 PM5/29/13
to comsol...@googlegroups.com
Неправильно задан PML, я не могу сказать большего, чем уже написано.
Для него видимо надо выбрать внешнюю и внутреннюю границы.

Arin Immo

unread,
May 30, 2013, 7:10:10 AM5/30/13
to comsol...@googlegroups.com
Please help me with PML settings  of the file! I don't know where my settings are wrong.
TWO.mph

Pavel Dergachev

unread,
May 30, 2013, 8:02:38 AM5/30/13
to comsol...@googlegroups.com
Пожалуйста:


TWO_fix.mph

Arin Immo

unread,
May 31, 2013, 7:39:56 AM5/31/13
to comsol...@googlegroups.com
First, I can open the file because of lower   comsol  version. Can you cut some pictures to show where you have made corrections ? 

Second, In 3-D situations ,I just can't use PML under the condition that I make the geometry an "Assembly"  ,It shows me the same error as above .Please help me with it ! I need the "Assembly",in that I can must the boundary condition "continuity",how to achieve that?
Ratio3_1.mph

Pavel Dergachev

unread,
May 31, 2013, 8:46:40 AM5/31/13
to comsol...@googlegroups.com
1. Не надо использовать сборку и "continuity". Это необходимо когда геометрия двигается.
2. PML не должен содержать границ с "continuity" и все будет работать. Я просто добавил внутреннюю границу в последний слой (без "continuity").
3. Тоже и в 3D.

Зачем Вы постоянно пытаетесь использовать "continuity"?

Arin Immo

unread,
May 31, 2013, 10:42:30 PM5/31/13
to comsol...@googlegroups.com
I just want to make the boundary consistent  for electric field at each layer ,I wonder if a "union" satisfies the requirement  as default ??

Arin Immo

unread,
Jun 1, 2013, 10:33:21 PM6/1/13
to comsol...@googlegroups.com
And which  kind  of material  should I use  for PML ? the same with the outermost layer ,or the air core or another arbitrary  one? 

Pavel Dergachev

unread,
Jun 3, 2013, 6:25:40 AM6/3/13
to
Правильные граничные условия будут и без "union".
Если просто сделать геометрию как union - все будет работать.

Только надо сильно уменьшить сетку - она была слишком грубая.

Pavel Dergachev

unread,
Jun 3, 2013, 6:26:23 AM6/3/13
to comsol...@googlegroups.com
PML должен быть из какого-то линейного материала (с постоянными коэффициентами) - вроде это главное условие.
Reply all
Reply to author
Forward
0 new messages