Menu

Discussion Forum

New Discussion
查找

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

failed to evalute variable Jacobian"

Posted 2012年10月24日 GMT-4 16:49 RF & Microwave Engineering Version 4.3 6 Replies

Naema Othman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi:
I am designing a structure which is gold stripe that is put above Silicon. I am trying to plot the Electric field (y-component) of the resulted mode. The problem is that I did not get the mode. I do have an error saying that
" failed to evalute variable Jacobian"
-variable: material .n13
-geometry :1
-domain: 3
I guess it is an error in my Geometry 1 which is the Silicon.

I attached the file in case some one will have time to see my problem.
i will appreciate any help

Naema

Attachments:

6 Replies Last Post 2012年10月30日 GMT-4 04:43
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月24日 GMT-4 16:58
Hi
indeed you have mised to define any material on the "small" domain 3 which is your Au I assume, there is a red "X" on the material showing something not OK ;)

--
Good luck
Ivar
Hi indeed you have mised to define any material on the "small" domain 3 which is your Au I assume, there is a red "X" on the material showing something not OK ;) -- Good luck Ivar
Naema Othman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月24日 GMT-4 17:26
Hi Ivar:
thanks so much for the quick reply. It works now. However, the solution does not seem right. There is another thing that has to be fixed.
I did use the same parameters and material constants in version 3.5a and I was able to get the right mode.
My question is why I am not able to get the right mode with this version of comsol (4.3)?

I am attaching the 3.5a comsol file in which I got the right mode.

Hi Ivar: thanks so much for the quick reply. It works now. However, the solution does not seem right. There is another thing that has to be fixed. I did use the same parameters and material constants in version 3.5a and I was able to get the right mode. My question is why I am not able to get the right mode with this version of comsol (4.3)? I am attaching the 3.5a comsol file in which I got the right mode.

Attachments:

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月25日 GMT-4 01:22
Hi

when I solve your 3.5a model as is in my 4.3a I get the same value for the effective mode index, but another field value. Now in eigenmode analysis, normally the field absolute value is arbitrary, its the mode shape that is important and the shapes looks similar.

So it could be that there is a slightly different normalisation factor between the two COMSOL versions, check the release notes,

But I cannot really say more, I haven't been working in optics/RF for some time so I do not have many references in changes over versions lately, but you could also ask "support" they must know ;)

But there are also users working in the optics field out here that might have some good remarks

--
Good luck
Ivar
Hi when I solve your 3.5a model as is in my 4.3a I get the same value for the effective mode index, but another field value. Now in eigenmode analysis, normally the field absolute value is arbitrary, its the mode shape that is important and the shapes looks similar. So it could be that there is a slightly different normalisation factor between the two COMSOL versions, check the release notes, But I cannot really say more, I haven't been working in optics/RF for some time so I do not have many references in changes over versions lately, but you could also ask "support" they must know ;) But there are also users working in the optics field out here that might have some good remarks -- Good luck Ivar
Naema Othman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月29日 GMT-4 16:27
Hi Ivar:
thanks for the reply. I actually did not get the same refractive index for both comsol files(3.5a and 3.4) as you did.
I suppose to have the same index in both of them because I am solving the same problem and the solution is only one. I know you got the same index for both(3.5a and 3.4) comsol files but I do not know how.

to better explain what I need:
I attaced a comsol file called 50t_100w_2( done with version 3.5a) in which I got the right solution. Please open it again because I have modified it.

I also attached another comsol file called spp2( done with version 4.3) in which I have tried to get the same solution as in ( 50t_100w_2) . However, I did not get succees in doing this.

As you will see in the file (50t_100w_2) , I am trying to plot the y-component of the electric field.


I will look forward for your reply,
Naema

Hi Ivar: thanks for the reply. I actually did not get the same refractive index for both comsol files(3.5a and 3.4) as you did. I suppose to have the same index in both of them because I am solving the same problem and the solution is only one. I know you got the same index for both(3.5a and 3.4) comsol files but I do not know how. to better explain what I need: I attaced a comsol file called 50t_100w_2( done with version 3.5a) in which I got the right solution. Please open it again because I have modified it. I also attached another comsol file called spp2( done with version 4.3) in which I have tried to get the same solution as in ( 50t_100w_2) . However, I did not get succees in doing this. As you will see in the file (50t_100w_2) , I am trying to plot the y-component of the electric field. I will look forward for your reply, Naema
Naema Othman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月29日 GMT-4 16:28
Hi:
the attachments are here
sorry for forgetting them,

Naema
Hi: the attachments are here sorry for forgetting them, Naema

Attachments:

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年10月30日 GMT-4 04:43
Hi

the main error you are doing is that your total domain is less than a wavelength across, you need much more "free space to avoid that the exteral boundaries makes a "box" and change completely the behaviour of your model. Even using "infinite elements" to exend the model would be of interest.

Then you need a much finer mesh, it should be at least 5-10 elements across a wavelength in the material, hence 1300 nm/3.5/10 or some max 40 nm

This wil lend upt with quite some RAM requirements, you might need to consider to model only the half of your model and use a symmetry condition.

Check the model library,and the doc there should be some RF examples on how to adapt the max mesh size with the wavelength

--
Good luck
Ivar
Hi the main error you are doing is that your total domain is less than a wavelength across, you need much more "free space to avoid that the exteral boundaries makes a "box" and change completely the behaviour of your model. Even using "infinite elements" to exend the model would be of interest. Then you need a much finer mesh, it should be at least 5-10 elements across a wavelength in the material, hence 1300 nm/3.5/10 or some max 40 nm This wil lend upt with quite some RAM requirements, you might need to consider to model only the half of your model and use a symmetry condition. Check the model library,and the doc there should be some RF examples on how to adapt the max mesh size with the wavelength -- Good luck Ivar

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.

Suggested Content