Discussion Forum

Hi Manu,

good to see that you are working on Acoustic metamaterial and trying to reproduce the photonic band structure tutorial. Some time back, I had been tried to reproduce the photonic band gap tutorial in comsol 4.3 but unfortunately I could not succeeded to do that. what have I been concluded that the model builder in COMSOL 4.3 is very different from the COMSOL 3.5a. The problems which you are facing is most probably due to model builder difference.

Hello
Check this model it was done in pressure acoustic.
I hope it helps you.

hello friend,

Thank you for your reply
I am trying to open this file in COMSOL 4.3 but it is showing that the file has created in later version.
plz, Can you tell me in which version of COMSOL you have been created this file.




cheers,
SANTOSH

It is in 4.3b. You should upgrade your comsol.

Dear Manu,

Things are done a little differently between v3.5 and 4.x. You can still calculate bandgaps but you have to recreate the model. The Floquet boundary condition for solids/structures was added I believe in v4.3 which simplifies the definition of the periodic boundary. Here is the description of a bandgap example I created using v4.2: www.veryst.com/project/Band-Gap-Phononic-Structure .

Nagi Elabbasi
Veryst Engineering

Dear Hussein,

Thank you for the reply, I am trying to access your model, but am unable to do so as it was made in 4.3b, and I am using 4.3. Can you attach the complete report of the model for my reference? It would be a great help!

Regards,
Manu Sahay

Dear Nagi,

I had gone through the page that you have mentioned when I was looking through the forums ( I had found the link in another discussion thread). I have been trying to recreate the model, but I haven't been able to do it successfully. As I understand, an eigenfrequency corresponds to a particular band. I have done a parametric sweep with k, varying k from 0 to 2pi/a in the symmetry direction, and I have done a point evaluation to extract the 'freq' . For plotting the band structure, what should I plot on the Y axis? Should it be the lambda value, or the frequency value? And how many eigenfrequencies should I solve for?

Your help is greatly appreciated.

Regards,
Manu

Dear Hussein,

thanks for your rply
Please can you attach the report file of the "acoustic_dispersion(1).mph" .


regards
santosh

Here is the report file.

I sent it.

Dear Hussein,

Thanks for your rply
I was looking and trying to understand the result file which you have been send but I did not find the terms for parametric sweep, global and scalar variable and eigenfrequency terms and how did you get the eigen fequency values.

Plz, If you could emphasis all the above point then it would be great help for us.

best regards
santosh

Dear Hussein

I have been finally able to access your file ( acoustic_dispersion.mph). I am trying to adapt the same for my case. I had a few queries about your file-

1. Why have you chosen to solve for 6 eigenfrequencies? As I understand, a given eigenfrequency corresponds to a particular band. Is this understanding correct? So, for getting the band structure, we solve in the irreducible brillouin zone. If we choose to solve for 6 bands, we will get 6 bands, right? So how can we know that we have got all the correct bands.

2. For band structure, we plot frequency vs k, in a 1-D plot (global) . In your case, you have provided the X-axis data as 'lambda' and the parameter as 'Solution number' . For getting k, I was thinking that the X-axis data source would be 'Outer Solutions' and the Parameter would be 'Parameter Value' . Is this thinking correct?

3. The value of kx should be varied from 0 to 2pi/a, where 'a' is the dimension of the unit cell, right? While varying 'kx' , we set floquet periodicity condition for the boundaries in the X direction, and set the Y-direction boundaries as Sound-hard boundaries. So, for varying 'ky', should we set the periodic condition in Y direction, and sound hard in X direction? And while varying both kx and ky ( for the TX case) should we set all 4 boundaries of the square unit cell as periodic?

4. How de we know what is the correct frequency to set in the 'search for eigenfrequencies around' field?

I know that for most of you these would be very basic queries. I have been trying to get the correct band structure for a long time without much success. I would be extremely indebted if you could help me out with the same.

Regards,
Manu Sahay

structure

structure

?

how can i do this (www.veryst.com/project/Band-Gap-Phononic-Structure )? can you give me this model !thank you very much