How to use openEMS. Discussion on examples, tutorials etc

Moderator: thorsten

Post Reply
Posts: 10
Joined: Tue 10 Apr 2012, 22:14


Post by linuxmill » Wed 11 Apr 2012, 23:28

I have a general question, which I will follow with specific questions for clarification:

General Question: How is excitation done in openEMS?

I see several approaches: a Lumped Port Node (as shown in the Helical_antenna.m model) for which a direction vector must be applied. There is also direct excitation applied along the antenna wire (as done in the infDipole example) and there is no Port.

I want to understand and document how the excitation gets onto the antenna wires. For example, in the case of a dipole, how is the notion of a positive and ground side distinguished? Physically in the model, it's a length of wire. In reality, it is two lengths of wire with a signal and impedance between them. I do not understand how the model simulates reality for either the lumped node (has no "AddExcitation" method applied) or the non-lumped node approach (where "AddExcitation is apparently needed).



Posts: 1438
Joined: Mon 27 Jun 2011, 12:26

Re: Excitation

Post by thorsten » Thu 12 Apr 2012, 10:36

Hello Jim,

first off: you always need an excitation!
But there are a couple of functions that encapsulate the excitation and combine it with other properties (like a resistor). For example the Lumped-Port is a combination of an excitation, voltage and current probes and a resistance... You can use AppCSXCAD for example to see all these different properties included by this one "advanced" function "AddLumpedPort". With other words, you could create your own lumped port using all this individual properties, but that would be less convenient.
But this topic about ports is not a very easy one and I think Sebastian (who wrote most of them) would be ideally suited to document them in more detail.
Especially the MSLPort etc. are quite complex to understand and document. Which makes a good documentation even more important of course!

I hope that clarifies it a bit?

I'm not sure I understood you completely about the inf dipole example. But in this example the entire infinite dipole is modeled using a hard electric source (type 1), which is kind of a metal with an electric field applied onto it. Of course in reality there is no dipole antenna that is infinitesimal short.

I have also create a lambda-half dipole (which is more realistic) but I have not yet committed it into the official openEMS version. But I will attach it, so that you can have a look.
It uses metal for the dipole and a lumped port as excitation. Have a look at it in AppCSXCAD.

Lambda-Half Dipole antenna example
(5.68 KiB) Downloaded 656 times

Posts: 10
Joined: Tue 10 Apr 2012, 22:14

Re: Excitation

Post by linuxmill » Fri 13 Apr 2012, 07:04

Thanks Thorsten. That does help.

Post Reply