Characteristic Mode Analysis Problem

ynathan
ynathan Altair Community Member
edited October 2020 in Community Q&A

Hi,

When I try to solve my model given in attachment, I can't track any mode between 4-6 GHz. The frequency step size is 20 MHz and mesh size is 1/30 lamda for center frequency. The solver works on double precision still I can't get any eigenvalue, modal significance or excitation coefficient. Thanks for the help. 

Unable to find an attachment - read this blog

Tagged:

Answers

  • JIF
    JIF
    Altair Employee
    edited September 2018

    Hello ynathan,

     

    I started running the model, but if you could include the output files (.fek, .bof and .out), that would make it much easier for me to investigate.

  • ynathan
    ynathan Altair Community Member
    edited September 2018

    Apart from this case I also have some problem about meshing a similar structure. Does anyone know why I can't mesh the given model

    I get the following errors:

    Error 16561: Mesher aborted - Unable to mesh face 'Union1.Face4464', probable model problem at unknown.

    Error 17580: The mesher aborted abnormally while meshing Union1.

    Unable to find an attachment - read this blog

  • Madelé
    Madelé
    Altair Employee
    edited September 2018

    Hi ynathan

     

    Please attach your CADFEKO model file (.cfx) if you'd like others on the forum to take a look. A .cfs file contains the
    view settings in your .cfx model at the time of saving. It cannot be opened by itself.

     

    Kind regards

    Madelé

  • ynathan
    ynathan Altair Community Member
    edited September 2018

    Really sorry for the missing file, I didn't know the cfx file was necessary. Here you can see the both .cfs and .cfx files

    Unable to find an attachment - read this blog

  • Madelé
    Madelé
    Altair Employee
    edited September 2018

    No problem. Which version of FEKO are you using? I quickly checked in the latest version of CADFEKO (FEKO 2018.1.1) and it meshes the model (no meshing error). I do see suspect geometry items and don't know the reason for those off-hand. That might be the cause of your meshing problems...

  • ynathan
    ynathan Altair Community Member
    edited September 2018

    I use 7.0  version but I can't see any suspect geometry how can you determine which parts are suspicous? Did you use standard meshing or fine, custom etc. ?

  • Madelé
    Madelé
    Altair Employee
    edited September 2018

    You can see where the geometry is a problem for CADFEKO from FEKO Suite 7.0 in the 3D view by zooming in where the swept edge connects to the top or bottom face:

    image.png.433bfc7aebad8b14bf318d1b5c94a6eb.png

     

    Meshing fails because it does not know how to handle the geometry correctly. The same geometry is not a problem in FEKO 2017 or later (after geometry re-evaluation if opening the older model), so I would strongly suggest that you upgrade to the latest version. It can mesh regardless of the mesh setting (standard, fine, custom). Using the older CADFEKO, I would suggest that you try to build this face in some other way than sweeping a copied edge and hopefully that would resolve the problem.

     

    Regarding the suspect items, I don't think that is a big problem in this case, more a nuisance. You can see suspect geometry by re-evaluating the geometry (Transform tab, Alter group, Re-evaluate). In the attached model, the suspicious settings are caused by the cylinder media settings. Somehow the geometry evaluator fails to automatically determine the correct user settings. One way to avoid the suspicious geometry is to first subtract the cylinder shapes from the rest of the geometry, then add the cylinders again (you can use copy/duplicate), set their media properties and then perform the union between the cylinders and the rest of the geometry.