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.

PML/Port configuration for power flow calculation

Please login with a confirmed email address before reporting spam

Hi,
I have a problem in the PML/port configuration for the calculation of Transmittance.
I don't understand why in my model (attached), despite the unitary imput power at port 1, the integration on the middle line of the sample turns in a value different than one, and the integration on the port 1 line gives me NaN.

Anyone one can help me?

Thanks,
G,.P.


8 Replies Last Post 2010年10月13日 GMT-4 07:03

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月8日 GMT-4 01:47
Hello,

When using port BC, you may need to use "Boundary Pair" between PML and Air subdomains. Otherwise, you can use "scattering BC" on the top and bottom BC outside the PML domains to calculate transmittance. There are some examples in the example models that you can check.

Cheers,

David
Hello, When using port BC, you may need to use "Boundary Pair" between PML and Air subdomains. Otherwise, you can use "scattering BC" on the top and bottom BC outside the PML domains to calculate transmittance. There are some examples in the example models that you can check. Cheers, David

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月8日 GMT-4 05:17
Hello,
your proposed configuration works well on comsol 35a but it fails to run well in version >= 4.0 (the model i attached is performed in version 4.0a) .
Thanks,
G.P.
Hello, your proposed configuration works well on comsol 35a but it fails to run well in version >= 4.0 (the model i attached is performed in version 4.0a) . Thanks, G.P.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月11日 GMT-4 04:30
Hello,

I am suffering this issue by COMSOL 4.0a that I have reported it to the local agency but still no response. As I know, COMSOL is demonstrating many new modules except for the RF module.

David
Hello, I am suffering this issue by COMSOL 4.0a that I have reported it to the local agency but still no response. As I know, COMSOL is demonstrating many new modules except for the RF module. David

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月11日 GMT-4 08:20
Hi David,
thank you for your answer.Now I know that it's a general problem to fix in the 4.0 and 4.0a versions.
I'll send this issue to my local technical support as you just did ,in order to try to fix the problem.

Giuseppe
Hi David, thank you for your answer.Now I know that it's a general problem to fix in the 4.0 and 4.0a versions. I'll send this issue to my local technical support as you just did ,in order to try to fix the problem. Giuseppe

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月12日 GMT-4 01:14
Hi G. P.,

I just slightly modified your previous model by changing port BC to two scattering BC (on the top and bottom) so that you can excite a plane wave from top to bottom through two PMLs. Particularly, I put a dielectric layer in the air domain with n=2 (SiN) and probe the powerout flow along one boundary near bottom PML. It looks like this model now works well. FYR.

Cheers,

David
Hi G. P., I just slightly modified your previous model by changing port BC to two scattering BC (on the top and bottom) so that you can excite a plane wave from top to bottom through two PMLs. Particularly, I put a dielectric layer in the air domain with n=2 (SiN) and probe the powerout flow along one boundary near bottom PML. It looks like this model now works well. FYR. Cheers, David


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月12日 GMT-4 06:38
Hi david,
I see your modified configuration to my model.As result I see a surface plot where the field is overall zero except close the top PML (becouse it has been absorbed, i suppose).It don't seem to work well, I mean i don't see the field inside the domains.I want a configuration (PML+Port was the best case in previous versions) for witch i can excite a plane wave in the interior domains and it can be scattered by any possible scattering object inside them.I need the PML on top and above side in order to absorbe any scattered component of the wave.

With regards ,
Giuseppe.
Hi david, I see your modified configuration to my model.As result I see a surface plot where the field is overall zero except close the top PML (becouse it has been absorbed, i suppose).It don't seem to work well, I mean i don't see the field inside the domains.I want a configuration (PML+Port was the best case in previous versions) for witch i can excite a plane wave in the interior domains and it can be scattered by any possible scattering object inside them.I need the PML on top and above side in order to absorbe any scattered component of the wave. With regards , Giuseppe.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月12日 GMT-4 12:21
Hi G. P.,

I still have no idea how I can set "Boundary Pair" using 4.0a. However, the model I modified really works. You need to do another run but with no dielectric layer in the air domain to get background field. Then, you just divide the two integration values along the bottom BC ( one is with dielectric layer, the other is without dielectric layer) so that you can get the transmittance. You don't need to care about the field in the air domain since you only need the normalized value. That's what I did.

Regards,

David
Hi G. P., I still have no idea how I can set "Boundary Pair" using 4.0a. However, the model I modified really works. You need to do another run but with no dielectric layer in the air domain to get background field. Then, you just divide the two integration values along the bottom BC ( one is with dielectric layer, the other is without dielectric layer) so that you can get the transmittance. You don't need to care about the field in the air domain since you only need the normalized value. That's what I did. Regards, David

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年10月13日 GMT-4 07:03
Hi David,
Thank you very much.It works well now also if you have to run two times your model for using the background solution as imput and it's more expensive in terms of time and cpu.I hope that they will switch to pair port configuration like in version 35a.
With regards,
Giuseppe.
Hi David, Thank you very much.It works well now also if you have to run two times your model for using the background solution as imput and it's more expensive in terms of time and cpu.I hope that they will switch to pair port configuration like in version 35a. With regards, Giuseppe.

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.