| Author |
Message |
albede
Joined: 03 Feb 2006
Posts: 16
Helped: 1
|
 25 May 2006 20:34 FDTD easy question |
|
|
|
Hi
I inserted a (narrow-waisted) gaussian beam (not temporal gaussian) to my FDTD program using TF/SF method.
In this case, the beam is diverging.
But I have no idea about making the converging or collimating beam.
Does anybody have any idea or experience ?
Hope any reply 
|
|
| Back to top |
|
 |
tmalc
Joined: 18 Apr 2007
Posts: 25
|
| 19 Jun 2008 13:24 FDTD easy question |
|
|
|
| ask the same question, anyone give any advice?
|
|
| Back to top |
|
|
rrumpf
Joined: 06 Jul 2007
Posts: 80
Helped: 10
|
| 23 Aug 2008 21:13 Re: FDTD easy question |
|
|
|
There are two important things you need to consider. First, the source you inject into your model must be a rigorous solution to Maxwell's equations or things can get a little crazy. Second, the phase part of your source will need to be considered to get a collimated beam, or converging beam.
You can try this. Setup a model with a simple plane wave. Pass the plane wave through a PEC aperture and perhaps even place a lens in the aperture. Play with the geometry until you get a good collimated beam on the other side of the lens. When you get this, record the source on the other side of the lens. Simply import this source into your "real" simulation. The best way to do this would be to record the complex amplitude and phase of the beam. This way you will not have to record a large set of data.
You can also try playing the phase portion of your Gaussian beam to give it some curvature as if it just passed through a lens.
Hope this helps!
-tip
|
|
| Back to top |
|
|
