Searching free solvers for electromagnetic field simulation

[andre_teprom]

Hi there,


I've downloaded the student version of the QuickField tool in order to evaluate its features, but felt the lack of some useful features, such as the ability to measure the mechanical force resulted from the proximity of 2 magnets.

Does somebody has any recommendation of another freeware tool ?

[pardo-bsso]

Hi, haven't used it in a while but last time FEMM worked fine for me.

[dkozel]

OpenEMS is pretty well regarded, though has a learning curve I believe.
http://openems.de/start/index.php

[daqq]

OpenEMS is great for antenna design, RF problems, waveguides etc., in the time domain but to the best of my knowledge there's no way to simulate force effects and just magnets.

As was recommended, FEMM is more similar to quick field.

[andre_teprom]

@pardo/daqq,
Many thanks for this nice recommendatio, it's really amazing.

@dkozel,
Thanks anyway.

[cdev]

Are there any fully graphical free simulators that directly read in 3D models from whatever 3D package?
I'm also interested in an inductance and capacitance simulator that can do at least self and mutual L/C, or in other words, S parameters.

RFSim99 probably isnt exactly what you want, but if you're not familiar with it its a useful program. It can handle S-parameters for some simple use cases. Also, its free!

[Kelbit]

I have been searching for a good free 3D field simulator for a while. It's getting to the point where I'm seriously considering writing my own FDTD engine.

A quick summary of the options that *are* free:

NEC-2 is a public domain Method-of-Moments (MoM) simulator suitable mainly for wire antennas and some planar antennas which do not rely on dielectric structures (so no patch antenna simulations). It was originally written in the 1970s at Lawrence Livermore National Lab, so the interface leaves much to be desired. It's a command line program that takes a text file input, where each line of text represents an 80 column punched card in a card stack. There are a few graphical interfaces available (4NEC2 is the most popular free one) but they are all pretty old. There's definitely a demand for a modern 3D CAD type frontend, if anyone's looking for a weekend project.

OpenEMS is a free FDTD (Finite Difference Time Domain) simulator suitable for a wide range of computational EM problems. As an FDTD package, it will be a lot more memory heavy than a FEM or MoM simulator, but RAM is cheap these days. The main drawback is the interface - there's no UI, rather, the software is used programmatically through a MATLAB/Octave frontend. Again - making a UI would be a great weekend project if anyone's looking. If you can handle the lack of UI, this package is the best truly free option right now.

MEEP is another FDTD package. Looks more like it's geared towards photonics applications, although the physics are the same. Again, no UI. The main drawback I can see is that as it's more of a physics research package it doesn't include postprocessing features helpful for EE, namely s-parameter output and far-field radiation pattern computation - you'd need to write your own postprocessor.

EmGine is another FDTD package. Unlike the others, it has a UI. Unfortunately, development has stalled and apparently it's difficult to run on Windows on the latest Python. Also, the license prohibits commercial use.

Sonnet Lite is the free version of Sonnet, which is a planar EM simulator based on the Method of Moments. Very similar product to Agilent Momentum that's included in ADS. It's good for simple printed transmission line structures like microstrip, hybrid couplers, and filters. You can also do patch antennas with it, but you can't view the far-field patterns unless you buy the full version, so it'd be good for tuning return loss only.

FEKO Lite is the free version of FEKO (registration required). FEKO is a professional EM field sim comparable to HFSS or CST which has a big list of solvers: MoM, FEM, FDTD, and UTD, as well as several hybrid MoM-FEM solvers. It's got most of the features of the full version but the mesh size is heavily restricted. Still, you can solve small problems with it, like basic microstrip antennas.

Basically, there's a demonstrable need for a user-friendly 3D high frequency electromagnetics simulator in the open-source space, but nobody's picked it up yet. Of course, if this is for personal use, you could "acquire" a "time unlimited evaluation" of HFSS.

[andre_teprom]

A quick summary of the options that are free:

Hi Kelbit,

The list that you presented above has very interesting considerations, and brings tools that I even had not assessed before. However I noticed that the FEMM tool is not ranked there, but this happened because you do not consider it a good tool if comparing to the others ? Since that it can handle very high mesh grids, up to this time became my preferred option to carry out some simulations that I'm planning to do.

[Kelbit]

A quick summary of the options that are free:

Hi Kelbit,

The list that you presented above has very interesting considerations, and brings tools that I even had not assessed before. However I noticed that the FEMM tool is not ranked there, but this happened because you do not consider it a good tool if comparing to the others ? Since that it can handle very high mesh grids, up to this time became my preferred option to carry out some simulations that I'm planning to do.

FEMM is a great tool, but it's a low-frequency electromagnetics tool for electromachinery/transformer design, not a high-frequency tool for RF & microwave. Since this is the RF & Microwave forum I focused my post on high frequency solvers.

Generally different simplifications are made in the low vs. high frequency tools to allow the simulation to run in a reasonable amount of time. That's why there's different classes of tools for this. For example, ANSYS sells both Maxwell (low frequency, electromachinery design) and HFSS (high frequency, RF/microwave design).

[andre_teprom]

FEMM is a great tool, but it's a low-frequency electromagnetics tool for electromachinery/transformer design, not a high-frequency tool for RF & microwave. Since this is the RF & Microwave forum I focused my post on high frequency solvers.

You're totally right, I was not observant on this detail.
And once again, thank you for the kind support.

[Alex Eisenhut]

Maybe?

http://www.electronicspoint.com/threads/ibm-electromagnetic-field-solver-suite-of-tools.38548/

[andre_teprom]

Humm...IBM Electromagnetic Field Solver
Never heard talk about this tool, but it looks promising, but there is only a restriction in my particular case, which is that there is no option for Windows.

Platform requirements

Operating system: L32-bit Linux, Red Hat 9.1 (recommended)

[xygor]

"MMTL, the Multilayer Multiconductor Transmission Line 2-D and 2.5-D electromagnetic modeling tool suite, generates transmission parameters and SPICE models from descriptions of electronics interconnect dimensions and materials properties."

http://mmtl.sourceforge.net/

[femtohertz15]

Sorry I wondered in late.

4NEC (http://www.qsl.net/4nec2), GUI front end to a very old, old school command line driven wire and wire approximation antenna sim.

[andre_teprom]

4NEC (http://www.qsl.net/4nec2), GUI front end to a very old, old school command line driven wire and wire approximation antenna sim.

Nice tool
The last revision is not too old, what suggests that are persons giving support for improvements and bug fixes.

Tahnks