How big should be the airbox ?

Haydar - I am answering your question here since I think it could interest others.

Q: How to define the size of the radiation boundary airbox in HFSS ? I am getting different results with different airbox sizes.

A:
1) Since you are using lambda refinment , bigger boxes means bigger mesh and requires more time to converge. Check the accuracy of your convergence in both designs and see if the different results are within the convergence accuracy.


2) In order to decide the airbox size you will have to plot the CmplxMag(E) on a plane. Set the axis in the middle of your model and use one of the planes (XZ or YZ) to plot the fields (use logarithmic scale for the colors). The radiation boundary intersects with the field lines. You should see that the magnitude of these field lines are much smaller than the maximum at the center (for example 100 or 1000 time smaller).


Regards,
Itai

itaifrenkel said:

....
1) Since you are using lambda refinment , bigger boxes means bigger mesh and requires more time to converge. Check the accuracy of your convergence in both designs and see if the different results are within the convergence accuracy.
.....

Itai

Click to expand...

Thanks Itai
can you explain how to check and compare accuracy of the convergence from the solution data?
thanks a lot

Hi,

Also you should check that the incidence angle is as near to normal as possible to get an accurate results, otherwise the radiation boundary behaves poorly e.g. for incidence angle near 90 deg (i.e. nearly parallel to the boundary) the refelection coefficient is near to 1.

BR

adel_48 said:

Hi,

Also you should check that the incidence angle is as near to normal as possible to get an accurate results, otherwise the radiation boundary behaves poorly e.g. for incidence angle near 90 deg (i.e. nearly parallel to the boundary) the refelection coefficient is near to 1.

BR

Click to expand...

hi
As I asked above, can you anyone tell how to check those parameters, such as incidence angle, accuracy of the convergence, etc. For example, in the solution report there is interpolation error. What is the acceptable limits?
thanks a lot

1. In order to check convergence look at the convergence tab in the solution data window. From the project menu, right click results, click solution data and select the convergence tab. This convergence describes the accuracy of the mesh/solution in the setup frequency.

2. If you are using interpolating sweep, you can check the profile tab and see that the interpolating sweep converged. The default values for interpolating sweep are ussually good enough.

3. The incidence angle is a parameter of a plane wave excitations.

Itai