Jump to content

mel

Moderators
  • Content Count

    278
  • Joined

  • Last visited

  • Days Won

    13

Reputation Activity

  1. lostarmour liked a post in a topic by mel in Too many modes on the circular waveguide port   
    I would have to check with the developers on this specific error message, limitations, etc.
    This will probably only be in the 2nd week of January.
  2. lostarmour liked a post in a topic by mel in Too many modes on the circular waveguide port   
    This happens when the waveguide is over-moded. For example if a horn antenna is designed for 3 GHz and you solve it for 10 GHz there will be many propagating modes, where, usually, you would want only 1 mode to propagate.
  3. oussama1 liked a post in a topic by mel in The terminals of shielded cable   
    You can add the inductance to a port on the wire that connects to the ground.
    For a port on the cylinder, you can use the edge port. See e.g here below an edge port for a cylinder dipole. (Or you can make a small gap and use a short wire with a wire port.)

  4. oussama1 liked a post in a topic by mel in The terminals of shielded cable   
    >> However, how can we mesh the cylinder into triangles with the edge lengths the same as the spacing between the inner and outer conductor
    Just apply a local mesh size on the inner cylinder face.
     
    >> what about the outer conductor connection to the ground? should i use wire approximation or cylindrical?
    You can use either. You will be adding stray inductance to ground with either of them. But at low frequencies this inductance will be very small.
  5. oussama1 liked a post in a topic by mel in The terminals of shielded cable   
    Feko uses the thin wire approximation. You will not necessarily get accurate results by just specifying the wire radius. It only modifies the self impedance of the wires and will not necessarily accurately model the coupling to the shield, which is critically based on the physical spacing between the inner and outer conductors.
    It would be better, although computationally more expensive, to model the center conductor as a cylinder meshed into triangles. You must ensure that the meshed triangle edge lengths are roughly the same as the spacing between the inner and outer conductor.
     
    Secondly the radiation from the terminations can become significant if the the electrical length of the termination segments become large (so at high frequencies). This electrical length is not modeled in the MTL terminations. At higher frequencies you can compensate for the inductive effect by adding an equivalent capacitance.
  6. MOHAMED HAMDALLA liked a post in a topic by mel in Coupling to wire at different orientations   
    Hi Mohamed
     
    You should be using the numerical green's function (NGF). Have a look at the "H.4 Optimise Waveguide Pin Feed Location" model in the Examples Guide.
     
     
     
     
  7. lostarmour liked a post in a topic by mel in FEM mesh size when a finite conductivity is introduced   
    Yes, see "Automatic Meshing for Regions" in the Altair_Feko_User_Guide.pdf
  8. lostarmour liked a post in a topic by mel in FEM mesh size when a finite conductivity is introduced   
    Feko will mesh the FEM tets according to the wavelength inside the medium, so definitely defining a conductivity will reduce the mesh size.
  9. lostarmour liked a post in a topic by mel in Medium of the faces in FEKO   
    >> the meaning of a face with a medium as layered dielectric as I showed in the question means that it is a thin dielectric sheet only
    Yes!
  10. Feko user liked a post in a topic by mel in Warning issue   
    Your mesh could be too coarse, or you are using the SEP method at a too low frequency.
    For low frequency dielectric modelling, the VEP is more stable.
  11. lostarmour liked a post in a topic by mel in Medium of a polygon in Feko   
    PEC is treated as infinite conductivity.

    The "gaps" inside the wire model is of the order of the wavelength so I would not say these models are equivalent. 
     
  12. lostarmour liked a post in a topic by mel in Explicitly define UTD and PO solvers   
    This model is not suitable for UTD.
    UTD is for electrically very very large surfaces. There are small details in the model not suitable for UTD assumptions.
    You would have to use either MoM or MLFMM. (With multiple incident angles, the MoM might actually be the fastest, if you have enough memory for an in-core solution)
    PO would be faster, but again there are small geometrical details and possible resonances there would not always be modeled by the PO.
  13. Eliz liked a post in a topic by mel in Error 37389: DGFM   
    Hi Eliz
     
    With 2019 using the DGFM it will take 133 GByte. With a full solution (DGFM off) this will take 330 GByte.
    The default solver for dielectrics, Surface Equivalence, is computationally expensive when such fine meshing is used.
     
    A FEM solution would be much more efficient. The concept is demonstrated in the 2 videos on altairuniversity.com at:
    https://altairuniversity.com/learning-library/?type=learninglibraryitem&filter_resource_type=Training+Videos&filter_discipline=Electromagnetic&pg=3&search&filter_language&filter_source
    Note there are 2 parts, as shown in the image. (Not sure why the 2nd part is being listed twice).

     
    Instead of edge ports you would use a FEM line port with a FEM current source.
     
    Also, since the meshing is quite fine, the basis functions for the FEM should be set to first order:

    I estimate a few Gigabytes for this solution.
  14. Eliz liked a post in a topic by mel in Error 37389: DGFM   
    Hi Eliz
     
    1. The idea is to add as little as possible "air" tetrahedra to the model while still "absorbing" the fine elements inside the added tetrahedra. It will always depend on the model. With the original model and spacing required I would create two identical elements each with it's own "air" dielectric. For example:

    2. FEM/MoM and FEM/MLFMM is essentially full wave solutions, so all coupling effects are taken into account.
     
    In the v2 and v3 models you attached I don't see the feed lines you had in the original models. I cannot comment on the design/thinking behind the antenna. The modal port is not suitable for "microstrip" type feeds. Attached an example from the original model where I added one line port only. No air has been added.
    Viv_2el_TX_patt_v6_lineport.cfx
  15. pmcardle liked a post in a topic by mel in MoM effective exitation   
    Hi Pmcardle
     
    Without the model it's difficult to comment.
    But you are on the right track. The plane wave is not "seeing" any geometry. 
     
    For example, you would get this error if you create a flat plate in the XY plane and set the plane wave in the plane of the plate with vertical pol.
  16. pmcardle liked a post in a topic by mel in Half Planar Green's Function   
    Hi Pmcardle
     
    The structure is quite likely electrically too thick.
     
    Sometimes the warning message is conservative. To check it, calculate the near field inside the layer and run it over a frequency range. 
    The warning should disappear at higher frequencies. If the trend in the near field curve is consistent down to the frequency you require (maybe a straight line and does not pick up noise at low frequencies) then the results are quite likely reliable.
  17. pmcardle liked a post in a topic by mel in Phase Calculation   
    Hi
     
    The reference plane for plane waves is the origin.
  18. pmcardle liked a post in a topic by mel in Solution Accuracy   
    Hi Pmcardle
     
    1) Do a mesh convergence check -> rerun with 50% more mesh elements and see if the solution changes
    2) Compare with another (3rd party) solver if you have access to one.
    3) Compare a similar model with a published result 
    4) Solve a similar model for which there are measurements available and compare to gain confidence in your modelling.
  19. pmcardle liked a post in a topic by mel in Near vs Far Field Half Planar Green's Function   
    Hi pmcardle
     
    The incident field is indeed excluded in the far field.
  20. pmcardle liked a post in a topic by mel in Scattered Electric/Incident Electric Field Ratio/ Units   
    Hi pmcardle
     
    You could just request a scattered near field sufficiently far away from the object.
  21. pmcardle liked a post in a topic by mel in Scattering from an anisotropic body   
    Hi Pmcardle

    Your setup looks good.
     
     
     
  22. John1989 liked a post in a topic by mel in Phase difference between MOM and FEM solution   
    Hi John1989
     
    The reference direction for the ports are different. This causes the 180 degree phase shift.
    You can change it for the MoM port as shown.

    For the FEM port, it cannot be set, and is assigned automatically internally.
  23. Elmien liked a post in a topic by mel in Feko solver Error : struggling to execute runfeko   
    Hi Elmien
     
    You are running parallel FEKO.

    Try local run only:

  24. pmcardle liked a post in a topic by mel in Air Buffer for anisotropic FEM   
    For isotropic materials FEKO has an already implemented radiation boundary condition on the surface of the FEM region. Surface elements on the FEM boundary are used to couple with the Method of Moments in a hybrid FEM/MoM solution. Or in a FEM only solution, an absorbing boundary condition is used.
    While it is theoretically possible to implement a radiation boundary condition for anisotropic FEM dielectrics, it is just not implemented in FEKO yet.
×
×
  • Create New...