Continue to Site

Welcome to

Welcome to our site! is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

setting CPW ports in IE3D

Not open for further replies.


Newbie level 1
May 21, 2005
Reaction score
Trophy points
Activity points
cpw fed in ie3d

Does any one know how to set up CPW ports in IE3D. So far I have a dielectric layer and a metal layer with a simple CPW line with finite grounds. but i dont know how to set the ports. Has anyone done this.? I am using IE3D 10.


cpw ground

There are examples in the manual. Here is one. Please try to limit a negative port's width to some size similar to the positive port. The results should be consistent and the width of the negative port does not affect it much. Some older versions have an example with a negative port defined on the whole width of the ground metal. It is normally ok for straight structure. For a bend structure, it may affect accuracy. Therefore, I suggest you to use a negative port on an edge comparable to the center conductor. Anyway, it is closer to reality for a finite ground CPW structure.
  • Like
Reactions: jadanz


    Points: 2
    Helpful Answer Positive Rating
ie3d can be for lower frequency

i'm having problems with ie3d cpw ports too. the examples are quite unusefull cause they show the result and i have difficulties with the steps i have to make to get the port. so could you jian describe it step by step how to get a cpw port??


infinite ground cpw

maybe you can via to ground for simulate the GCPW

ie3d differential port

Via are not making much difference in simulation results

ie3d differential ports

I'm sure Jian will be able to provide a good solution for you. In the meantime feel free to try SonnetLite (I work for Sonnet, and we do have some degree of competition with IE3D). It is free with no time-out, however with SonnetLite you can do no more than 4 ports.

There are two ways to do CPW ports with finite grounds. First, use three ports, one each for the ground, signal, and ground. Then number them -1, 1, and -1. However, after your first CPW port, you now only have one port left to use.

The other way is to note that the edge of the Sonnet substrate is perfect ground. So just number the middle port 1, and leave the ground strips touching the substrate edge (which is perfect ground) directly with no ports. If you do this, be careful to make sure the ground plane on the bottom side of your substrate is far far away. This makes the impedance of the microstrip mode very high so current will not flow in the microstrip mode, it flows only in the CPW mode. It is very important to keep the bottom side ground far from your CPW circuit when you build it too. The microstrip mode must not be allowed if you want good results.

It is all described in detail in the Sonnet documentation, which comes with SonnetLite. I'm not sure, but you might also be able to use some of these ideas in IE3D too. Worth a try anyway. IE3D does have a mode for analysis inside a box, which provides the perfect ground at the substrate edge, so at the very least, you should be able to use the second idea above.

ie3d ports

To All People Interested in IE3D Modeling of GCPW Structures:

Normally, modeling grounded CPW should be quite similar to modeling regular CPW which is well documented in IE3D User's Manual. However, you may need to be careful on modeling the grounds. You need to understand how the current flow so that you can define the port correctly. If your ground is infinite ground and your CPW is above the substrate and which is above the infinite ground, you can treat it as regular CPW structure. You should get good results. If your ground is a finite size of metallic piece, you may leave it as it is or you may want to build some vias to connect the CPW co-planar ground to the finite ground below. Otherwise, you may get some resonances. If your structure has no vias, the resonances are true phenomenon.

I think the reason you may get positive S(1,1) might be due to the fact that you are using extension ports while the port's extensions are coupled with the finite ground (see below). The port extension is coupled with the cpw ground and there is some positive feed back from it. IE3D/MGRID has some detection on it. It should issue you a warning on it even it still allows you to go ahead. In such a situation, you should try to use localized ports for it. We see users encounter this problem from time to time. We suggest users to pay attention to the warning issued by IE3D/MGRID when they do the simulation setup.

cross section view



As you can see, even for CPW structures, you may have many options. IE3D is an extremely powerful tool excellent in almost all aspects of high frequency simulations. In the manual, we try to document examples for all different kinds of applications (MMIC, RFIC, filters, microstrip antennas, CPW, wire antennas, handset antennas, modeling thin metal, modeling lossy semi-conductors, modeling structures inside a box, modeling periodical structures, modeling Tx and Rx antenna transfer functions, modeling time transient analysis of a structure, modeling BGA structures, ...). There are too many of them and we are fitting it into a 600 page manual. Still, we can not give all different examples for a specific topic. I think we should try to document more on the .geo files for different applications so that all the userse can benefit from it. Thank you very much.

Best regards,

cpw ie3d

Hi Jian,

I know that the following question in inappropriate in this context, I would like to ask you a question which I had been working on since 3 weeks but no luck. I hope you be able to advice me on this:

I am trying to simulate a coupled microstrip lines on ground plane. I am applying differential port excitation at one end of the cps.At the other end I am putting a complex load on the top of the strips. I am interested to measure the impedance. Is it possible?....
Also I tried checking my answer with short case and open case, to fing Z0, then teh complex propagation constant. I put a known load of say 100 ohms at the end, and tried the telegrapher equation for lossy lines, to calculate ZL. I cannot get the right answer?. Could you be kind enough to provide me with a hint.

ie3d cpw feed

Hi, maverick09:

Thank you for your question. You mentioned that you define a 100-ohm load at the 2nd end. I would like to know how you implement the load. If you define it as a differential port first. Then, you use a circuit simualtor (MODUA from IE3D) to connect the 100-ohms load or other load on it, it should be good. If you try to use a distributed (thin film resistor) load for it, it is hard to say because a distributed load is like what you specify at low frequency only. If you can, please send me the .geo file and I will check it for you. My e-mail is: Thanks!

Best regards,

using cpw port to simulate coupled microstrip

hi jian ,i am new to hfss.i have never used it before.can you plz help me with CPW and lumped ports?i am quite good at it when it comes to theoretical knowledge of antennas but need help with using hfss. I have to design a microstrip patch antenna with CPW at 10 GHz

cpw structure in ie3d

Hi, saqib:

I only know how to simulate it on IE3D not HFSS. Thanks!


Not open for further replies.

Part and Inventory Search

Welcome to