I came across the soft and hard sources in FDTD formulation. What is the difference between them and which one is more helpful? While writing a code if I use these sources, it is a two way source. How can I get an one way source? Please provide any attachments with mathematical explanation. Thanks.
There really is not much use for a "hard source," although they are the simplest to implement. They are called hard because other waves cannot pass through the region of the hard source. They reflect and scatter instead leading to unrealistic simulations. The soft source is transparent to other waves allowing them to pass through unaffected. This is most probably the type of source you want to use.
To get a "one-way" source there are a few techniques. I think the most popular is the the total-field/scattered-field technique for doing this. I teach a course in FDTD and have provided all the notes and even recorded lectures here:
http://emlab.utep.edu/ee5390fdtd.htm
For a description and movies showing the difference between the sources, see Lecture 6. For the TF/SF source in two-dimensions, see Lecture 15. For 3D, see Lecture 23. Also, if you are writing your own code, I highly recommend developing it in stages. I talk about these stages in Lectures 6 and 11. You will have to watch the recorded lectures to see the movies.
Hope this helps!
There really is not much use for a "hard source," although they are the simplest to implement. They are called hard because other waves cannot pass through the region of the hard source. They reflect and scatter instead leading to unrealistic simulations. The soft source is transparent to other waves allowing them to pass through unaffected. This is most probably the type of source you want to use.
To get a "one-way" source there are a few techniques. I think the most popular is the the total-field/scattered-field technique for doing this. I teach a course in FDTD and have provided all the notes and even recorded lectures here:
http://emlab.utep.edu/ee5390fdtd.htm
For a description and movies showing the difference between the sources, see Lecture 6. For the TF/SF source in two-dimensions, see Lecture 15. For 3D, see Lecture 23. Also, if you are writing your own code, I highly recommend developing it in stages. I talk about these stages in Lectures 6 and 11. You will have to watch the recorded lectures to see the movies.
Hope this helps!
In hard source you are enforcing a field at certain location; of course, you are doing this by enforcing excitation function e.g. a Gaussian pulse at this location. This will lead to a PEC (or PMC if you are exciting a H component) after the excitation pulse goes to zero. So you can describe the hard source as a source + PEC. Thus, any wave reflected from the boundary may face a PEC (or PMC) in the source location. It depends on the simulation duration and the grid size. Obviously, the hard source has no well known physical explanation. Unlike Hard source, soft source can be considered as transparent to the incoming waves and it has a well known explanation. In hard source you are enforcing a current density at a required location.
In summary, it depends on you application. If you have a simple application and long enough duration with respect to the domain size, go for hard source. If you want to know how to implement them, please inform me on the language that you are using and I may give u a hand.
The hard source (not penetrable) means that the effect of the source metal material is considered in the simulation. The wave cannot propagate (penetrate) inside this material. It behaves as a PEC. The soft source (penetrable) is vice versa.
- Badges
- Science topic