What is the best 3D EM simulator?

Hi coniglioe,
I am 8 months into proper EM simulations.I started out with NEC for wire antennas and now I am using IE3D 10.2 for my project.It simply rocks.Period.Many here swear about CST MWS 5. IE3D is not as pricey as MWS. I haven't checked it out with any complex structure.May be veterans in this could throw some light.

-Arun

It really depends on what you're doing; both in terms of the technology domain (antennas, filters, optics, ... ) and if you're doing pure research or not.

If the subject you're working in is well treated, you're probably safe with the a 3d solver which claims to be suited for the task. e.g. some of the first things you'll be looking for is:

- is it a full-wave solver? What assumptions are made about the wavelength size relative to the structure?

- does it solve in a bounded or unbounded domain? what type of boundary conditions are possible?

- is it a time- or frequency domain solver?

- what kind of excitations can be applied to the structure (e.g. lumped ports, waveguide ports, plane waves?)

- is the interface intuitive and clear? is there access to good tutorials/examples/help files?

To see if a solver is good for your topic it is probably best to take some published results with measurement data and try to reproduce these.


If you're doing research in unexplored areas of science or engineering, then you'll probably need at least 2 different solvers that are based on different numerical techniques, to compare results between them.


Some well know solvers are CST Microwave Studio (FIT method), Ansoft HFSS (FE method), Ansoft Designer 2D (MoM method), Zeland IE3D (MoM Method)

There was a useful discussion on this board about a comparison between CST MWS and Ansoft HFSS; you could check out that too.

Greetings,

Tony

On my experience:
CST MWS - for wideband and antennas problems without strong resonances.
HFSS - for arbitrary resonant structures.
WASP-NET and uWaveWizard - the best for waveguide structures, but have the geometry limitations.

Bet regards,
Kit-the-great

For the best results and efficiency use HFSS.

See Ya.

Hello coniglioe:

I really agree with Tony (tboloney). The "best em solver" depends completely on the application. There is no one EM software application or approach that solves all problems equally well. They all have strengths, they all have weaknesses.

HFSS is great for many applications, but certainly not for all. I wouldn't touch it for planar RFIC applications that involve, for instance, 0.2 um oxide layers; the mesher goes crazy with stuff with ultra thin dielectrics if I am also trying to solver large matching circuits (spiral inductors, MIM caps, transmission lines, etc.)

On the other hands, I would prefer HFSS over IE3D for BGA or detailed packaging design because I think IE3D has to make too many compromises in mesh selection (for speed) to provide the consistently accurate results that I need, where HFSS more accurately models the complex field behavior around the interconnects (just my opinion).

I like HFSS for smaller, complex problems where the analysis space doesn't get too large. I also like it when I am looking at relatively narrowband simulations.

I like the time domain 3D codes (CST Microwave Studio, REMCOM XFDTD, Vector Fields, Flomerics Micro-Stripes) for larger space problems, antenna array analysis, or for when I need very broadband simulations. They are especially well suited to broadband simulation, and they can also simulation and show direct transient field behaviour, which the frequency domain codes (like HFSS, etc.) can't do.

However, when you get to planar problems (microstrip, stripline, etc.) you find that you have to push the 3D meshing tools pretty hard to get consistent agreement between measured and calculated, and at some point you find that the planar tools give you a better value in terms of simulation time, accuracy, and general grief required to get a model together and simulate it. The planar EM tools pretty much break down into open domain (IE3D, Ansoft 2D which used to be called Ensemble, EMpicasso, and Agilent Momentum) and shielded domain (Sonnet, AWR EMSight).

The open domain planar tools handle antenna systems with ease, while the shielded domain planar tools usually show better error convergence for highly tuned circuits or for error sensitive Q analyses for on-chip spiral inductors.

The planar tools are usually Method of Moments, though there are starting to appear a few tools based on Partial-Element Equivalent Circuit (PEEC) method. I don't think the PEEC code developers have figured out how to de-embed ports well yet, so their frequency range for accurate simulation seems to be restricted to the low microwave frequencies.

There are also very large-space problems (like analysis of radiation scattering off of a car or a ship, or analysis of a very large reflector antenna system) that none of the above will handle with any degree of what could be called efficiency. For these cases, you probably need a wire-plate MoM code like FEKO, or you need a hybrid approach that combines a local fine analysis where current or field activity is detailed, and an optical approach for "large space" propagation.

So you can see that the answer to your question is "it depends."

I also find that the answer also depends a lot on your error requirements. If you are working in a commercial company, you will find that the ability to understand the sources of error in your simulations will drive your choice a lot of the time. If you are in an academic setting, this may not be as important to you as you may be looking more for trends. Beware the error issues; with some tools they can pop up and bite you without warning if you don't investigate the limits of the tool that you are using. And this can affect your career in a big way!

--Max

You can also take a look on EMC Studio with TriD EM Solver integrated. It is like FEKO, but with additional EMC possibilities and sometimes more faster.

I like HFSS, but it is expensive and memory hungry.

I think HFSS is faster in setting up a model,and the model's filesize is much smaller than MWS's.
MWS is easier to use than the HFSS.

i think semcad and xfdtd are much powerful but it talke much time.i simulate with HFSS and CST but not give aresult as measured
regards

I like CST MWS. One of the reasons is the following: I can create the structure and simulation setup very close to what I am doing in experiments. Usually it requires a lot of memory. But it gives quite good results. I wish it was more memory efficient. But I can live with that. Has anyone tried the 64 bit version of MWS? Or has anyone gone beyond 4 GB RAM usage in MWS? I am quite interested in hearing your experiences on this issue.

I think IE3D is best for planar circuits.

I think this book or article gives clear or best picture on planar EM simulators!

http://wes.feec.vutbr.cz/~raida/mult...tml.iso-8859-1

hfss v9.2 and ie3d

I have tried so many EM softwares. So far, may conclusion is:

FEKO: too much memory requirement. Can not handle so many different type of EM problems

HFSS: could be much more user friendly

Fullwave: bad CAD layout

CST: so far the best. can handle different type of EM problems.

as far as i am know ansoft hfss is really good for a 3d simulator.
The ansoft designer is another s/w which uses some rf approximations to compute the results about 10 times faster than hfss.
Although i havnt worked with CST, i heard that it has some additional features
Some 2D and 2.5D simulations could be done using ADS and microwave office.

Hello,

Very nice article on EM Simulators presently available in the industry...

http://www.mwrf.com/Articles/ArticleID/10919/10919.html

I hope this clearly tells One Simulator cannot handle every problem, EM simulators of various forms and functions are enhancing their ability to analyze and solve design problems...


EM Simulators Reveal Contents Of Crystal Ball
Although they cannot handle every problem, EM simulators of various forms and functions are enhancing their ability to analyze and solve design problems.

Electromagnetic (EM) simulators are fundamentally computer software tools that are used for microwave analysis, design, and optimization. Their existence has provided RF and high-speed digital designers with the resources needed to confront very difficult design problems. The primary objective of an EM simulator is to analyze electromagnetic fields. Field solvers apply this capability in applications like antennas, active devices, electromagnetic interference (EMI), and RF and microwave circuits.

Field-solver software breaks down into a few broad categories: two-dimensional (2D) cross-section solvers, 2.5D planar solvers, and 3D arbitrary geometry solvers. Although much work is now being done on 3D solvers, many EM software companies realize how critical their 2D field solvers are to certain areas of design. As a result, they continue to invest time and resources into these solvers.

An example is the Ansoft Designer integrated 2D field solver. Last month, Ansoft Corp. (Pittsburgh, PA) announced the latest version of Ansoft Designer SV. This free microwave and RF circuit-design software tool is based on the commercial version of Ansoft Designer.

The goal of Ansoft Designer SV is to give students and professionals an easy-to-use tool for applying basic circuit theories and techniques (Fig. 1). The new version flaunts a problem-size-restricted planar EM solver, a complete set of linear-component electrical models, and proprietary physics-based distributed and discontinuity models. It also offers a fully integrated schematic/layout editor, filter and transmission-line synthesis, a frequency-domain linear simulator, and a Smith Tool matching utility. The tool's single-database software architecture supports fully synchronized design entry based on schematic, netlist, and/or layout editing.

A similar free-software offering targets the 3D-planar-circuit space. It hails from Sonnet Software, Inc. (Syracuse, NY). Sonnet Lite is a feature-limited version of the company's professional Sonnet Suite. It provides a full-wave EM solution for 3D planar circuits. Sonnet Lite also can be used to analyze planar structures like microstrip matching networks, lossy spiral inductors with bridges, coupled transmission-line analysis, and microwave-circuit discontinuities.

This past spring, Sonnet released version 10 of Sonnet Suites. This version includes conformal-meshing technology that accommodates true thickness modeling for thick metal transmission lines. It also boasts an interface to Cadence Virtuoso. With broadband SPICE-model extraction, a circuit model is created that is valid over a broad band of frequencies.

Rockwell Collins (Cedar Rapids, IA) just selected Sonnet Suites Professional Release 10 as part of its complete set of electronic-design-automation (EDA) design tools. Rockwell Collins is counting on Sonnet Professional to aid its designers in their development of advanced packaging technology for government applications. The company is particularly focused on low-temperature co-fired ceramic (LTCC) technology. Because it uses the Fast Fourier Transform (FFT) technique, Sonnet Suites Professional should be able to efficiently analyze multilayer problems like LTCC packages.

Last month, Sonnet also made news through its work with Applied Wave Research (www.appwave.com). AWR began offering open access to its proprietary Xmodels technology to third-party EM-analysis software vendors that want to integrate with its Microwave Office circuit-design software suite. AWR's Xmodels are a group of discontinuity models. By using the results of full-wave EM solutions of parameterized discontinuity, they estimate the electrical performance of that discontinuity. With AWR's EM Socket open-standard interface, circuit designers will now be able to perform EM analysis using Sonnet Suites professional for the first set of Xmodels within the AWR Microwave Office suite.

The recently released EM software from Agilent Technologies (Palo Alto, CA) is a planar solver like the Sonnet Suites. Yet this software is a 64-b version of Momentum, the company's 3D-planar electromagnetic software (Fig. 2). With this software, memory limitations are eliminated and EM simulation and verification time is supposedly halved.

Until this point, 3D-planar EM simulators were only available for 32-b processing. But EM tools for 32-b computers are limited to a few gigabytes of memory, which limits the size and complexity of the problems that they can solve. In addition, designers have had to stay within memory limitations, which meant sacrificing accuracy to simplify their designs. The Momentum 3D-planar EM simulator accepts arbitrary design geometries including multilayer structures. It vows to accurately simulate complex electromagnetic effects like coupling and parasitics.

According to Sonnet, most structures fall into the categories of either planar?as exemplified by our last two examples?or fully three-dimensional. The fully 3D space includes Sonnet's CST Microwave Studio, which combines high-frequency 3D EM analysis, simulation in the time domain, a solid modeling interface, and vibrant graphics.

The full-3D EM simulator from Zeland Software (Fremont, CA) is dubbed Fidelity. It is a finite-difference time-domain (FDTD)?based simulator for modeling microwave circuits, components, antennas, EMC and EMI structures, and other high-speed and high-frequency circuitry. The simulator offers non-uniform mesh for modeling planar and 3D structures with a complicated dielectric configuration. Users are not limited to this mesh, which can be adjusted to fit a geometry. In Fidelity, radiating boundary conditions are modeled as various absorbing boundary conditions including PML.

The XFDTD 3D EM solver from Remcom (State College, PA) also is based on the FDTD method. Version 6.2 of XFDTD flaunts features like automatic convergence, adaptive background mesh, calculation of system efficiency, and loss tangent specification. In addition, "surface" conductivity allows XFDTD to account for the effective conductivity of good conductors at a specific frequency. In doing so, it does not have to use high cell resolution to resolve the relatively short wavelength within conductors.

This solver also provides support for the new female body mesh. Should greater precision in the human body be required, Remcom recently developed high-fidelity human meshes for both full-body (male and female) and head/shoulders regions. All three meshes can automatically adjust the permittivity and conductivity of the tissues for any specified frequency between 1 MHz and 20 GHz.

The Empire 3D EM field software from RTS Scientific (Thornhill, Ontario, Canada) also is based on the 3D FDTD method. It includes specially modified algorithms and methods for the efficient utilization of multiple floating points and caches. The software's newest graphical user interface, known as Ganymede, provides tools for the construction and modification of the objects that are being analyzed. Instead of restarting problems from scratch, the polygon editor, priority modeling modes, and built-in script enable the fast setup of solutions and modifications on the go.

This past June, Computer Simulation Technology or CST (Wellesley Hills, MA) previewed CST Microwave Studio (CST MWS) 2006. The next release of this time-domain 3D EM simulator will be accessible through the CST Design Environment, which will enable a 3D and schematic view of models. Other key features of CST Microwave Studio include a completed tetrahedral frequency-domain solver that incorporates the following: advanced true surface meshing, absorbing boundary conditions, farfields, gyrotropic media, lumped elements, arbitrarily shaped unit cells, adaptive meshing, and adaptive broadband-frequency sweep. A new PBA meshing algorithm targets hexahedral solvers. In addition, the simulator boasts automated co-simulation with Agilent's Advanced Design System (ADS) and an improved interface with Cadence Allegro.

The integration of CST Microwave Studio with Agilent's ADS electronic-design-automation software also was announced in June. The automated co-simulation strives to advance workflow integration for the design engineer who is seeking to improve passive-circuit performance. Users of Agilent ADS can manage the parameterization of CST MWS models without leaving the ADS interface, performing optimizations, or parameter tuning. If results for a desired parameter set are not available, ADS automatically launches a CST MWS simulation to create and store the missing data.

The EDA and 3D EM simulation markets also are bridged by Ansoft. HFSS is the company's 3D EM simulation software tool for RF, wireless, packaging, and optoelectronic design. The newest release, HFSS v9, boasts enhancements like Ansoft Desktop. This new design architecture enables familiar Microsoft Windows-based processes and superior EM-based design-flow automation. In addition, the new version offers design capture, analysis, and post-processing. This summer also welcomed Ansoft's Turbo Package Analyzer (TPA) v4.2. It combines new bidirectional integration with Synopys' Encore package-design software and Ansoft's 3D EM simulation.

Although the previous examples by no means form a complete list of EM simulators, they do provide a snapshot of the recent trends in this industry segment. As always, the industry favorites continue to improve themselves as new application areas emerge. Yet some rather unconventional approaches also are emerging. An example is EMtoSPICE from EMWonder (Norcross, GA). This software tool converts S-parameters to SPICE macromodels. It thereby allows the simulation of any device using S-parameters in SPICE. Because EMtoSPICE interpolates data, only a limited data sample is needed. Although the approach of this software tool is novel, its value is obvious. Should innovations like EMtoSPICE continue to appear, they may succeed in changing the face of the EM-simulator/EDA space before next year.

I have felt that HFSS is much more user friendly than that of CST. but i have seen the CST results to be much accurate when compared to HFSS.

Moreover since CST does analysis in time domain , wider frequency ranges take lesser time than when compared to HFSS.


so really depends on what we are doing.

Hi Manjunatha -- Nice write-up! Do want to add one thing about the EMtoSPICE, this kind of capbabiliy has been availble for a long time all serisous EM tools, including Sonnet. Most of the work involves use of the Pade rational polynomial and identifying poles and zeros in the transfer function by one or another approach. The poles and zeros are then used to create a lumped model. I'm not sure, but I wonder if I might have started interest in this field in the microwave area with my paper:

J. C. Rautio, "Synthesis of Lumped Models from N-Port Scattering Parameter Data," IEEE Tran. Microwave Theory Tech., Vol. 42, No. 3, March 1994, pp. 535-537.

If anyone knows of eariler work in the microwave area, please feel free to post the references. The above work does not use Pade rational polynomials, but the input (S-parameters) and output (lumped model) are the same, and these lumped models (when valid) actually correspond physically to the structure being modeled. The Pade rational polynomial models are more broad band and general, but have no physical coorespondence to the structure being modeled. As far as I know, Sonnet is the only one that offers both.

A good test for model synthesis is the step-in-width for a microstrip line (de-embedded to zero physical length). It is a decpetively simple discontinuity, but it is exceptionally difficult to get lumped models whose lumped element values do not change with frequency due to EM numerical noise issues.

Dear Mr. Rautio,


Yes you are correct...presentlymost of the tools provide this capability...
But when it comes with compatiblity between the RF & MW EDA tools then each tool has problems

some doesn't export in HSPICE format (*.sp),
some does not export in standard PSPICE format (*.cir)...& again some tools have import problems...
like so many issues...

In sonnet it is good that it has a broadband SPICE output capability, but it generates in *.lib/*.olb
format which compatible with Spectre or ADS but what about HSPICE or other time doain tools compatibility...

Hence engg. will look for some standalone tools to convert S or any port parameters to Spice models

which helps them to quickly verify for transient analysis...


I heard that Optimal Corporation SIEDA is some what does good job
I attached that manual for your reference...


Regards,
manju