Best Antenna Modeling Software

Discussion in 'Wireless & RF Design' started by KL7AJ, Dec 10, 2008.

  1. KL7AJ

    Thread Starter AAC Fanatic!

    Nov 4, 2008
    2,040
    287
    My very first exposure to computing of any sort was modeling antenna patterns. We had an old DOS version of Mininec3 at Hipas Observatory that I used to design a 16 x17 element Yagi array for an ionospheric radar. The measured pattern on the finished array turned out textbook perfect. Needless to say, this greatly boosted my confidence in computer modeling...at least of antennas.

    Mininec3, by the way, was written in BASIC, but borrowing most of its algorithms from the original Lawrence Livermore NEC2 (Numeric Electric Code). Mininec3 was the first antenna modeling program that didn't take a mainframe computer to run. (Golly...that was only 20 years ago...how times have changed!)

    Since that time, a plethora of antenna modeling packages have shown up, most based on the original NEC2 code.....for good reason...nobody's been able to improve significantly on NEC2. The real improvements have been on the user interface, and 3D plotting capacity. I've found that 4nec2x beats all the other available modeling packages in this regard. If you've been daunted by antenna modeling till now, 4nec2 is a great way to start...and it's FREE, too!

    I have a lot of years of NEC experience...anyone here interested in a few pointers, just give me a scream.

    Eric
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,650
    2,348
  3. KL7AJ

    Thread Starter AAC Fanatic!

    Nov 4, 2008
    2,040
    287
  4. bertus

    Administrator

    Apr 5, 2008
    15,650
    2,348
  5. Dave

    Retired Moderator

    Nov 17, 2003
    6,960
    145
    Although I personally don't use them to model antennas, two of the main packages we use (and used in the wider industry) is HFSS, from Ansoft, and Microwave Studio, from CST. I am a big fan of Ansoft's products, and HFSS is a very powerful tool in the areas I work in; I use it almost as much as Maxwell, also from Ansoft.

    Quite a few colleagues have a very high opinion of CST MWS, citing the finitie-integral technique implemented in CST MWS as having the geometric flexibility of FE modellers but without the memory and processing overheads associated with the FE technique. Personally, I hate CST MWS, it has a clumsy interface (ironic since they say CST is "intuitive"), and in terms of usability it doesn't come close to HFSS. I guess its what your used too.

    Furthermore, we have developed our own in-house Transmission-Line Matrix modeller for modelling all manner of EM and thermal problems.

    Dave
     
  6. KL7AJ

    Thread Starter AAC Fanatic!

    Nov 4, 2008
    2,040
    287
    Interesting stuff. I downloaded the demo of Maxwell, and am playing around with it a bit.

    NEC has been around a long time though, and I don't think it's going away any time soon. As with any purely numerical methods, the accuracy is whatever you choose to make it, (commensurate with processing power, of course!) I remember modeling some large antenna arrays with NEC on a 386 (with math co-processor, mind you), where you'd start the model, go off and make some coffee, come back, and it would still be grinding away. Now, all but the most complex 3D structures come back in a few seconds! (Optimizing of some large arrays can still take some time, though...but it's still minutes instead of weeks!)

    Eric
     
  7. Dave

    Retired Moderator

    Nov 17, 2003
    6,960
    145
    It is certainly worth a look. Naturally the free version of Maxwell is limited compared to the full version (I should hope so when you look at the cost a licenses!). It should give you a good feel for Ansoft software. And note, once you get the general hang of one Ansoft program you have pretty much the hang of all of them - excepting specifics to the problem you are investigating, of course.

    Yes, I have just had a look around at NEC codes - I am surprised how many there are out there, most for free. I do note there are some limitations for NEC2 which have in part been addressed in NEC3 and NEC4.

    I will have a look into NEC more, seems quite an interesting development project. I'm surprised I haven't looked into it in more detail before, although that may be to do with the fact I don't work specifically on antennas, and because general EM modelling nowadays centres around the main 3-D full-wave solvers.

    If you like EM modelling, then might I suggest you have a look at Elmer (see http://www.csc.fi/english/pages/elmer); again it is free, and is a very powerful multiphysics solver (similar to Ansys MP).

    Dave
     
Loading...