Jump to content

MvdM

Moderators
  • Content count

    56
  • Joined

  • Last visited

  • Days Won

    1

MvdM last won the day on November 9

MvdM had the most liked content!

1 Follower

About MvdM

  • Rank
    Advanced User

Profile Information

  • Gender
    Not Telling
  • Country
    South Africa
  • Are you University user?
    No

Recent Profile Visitors

835 profile views
  1. MvdM

    Long time to change the geometry

    Hi @briantvd Unfortunately, this regression was introduced in FEKO 2018.1. It is fixed in the latest version (Feko 2018.2). I just confirmed this using your model (sq_param_model_11arr_2G4_75.cfx). I can reproduce the assertion failure when deleting the union from the model in FEKO 2018.1.1, but not in Feko 2018.2. Kind regards Madelé
  2. MvdM

    STR file currents

    Hello Shane I trust Mel's answer provided the answers you were looking for. If you need the support information in future, you can open the Feko or HyperWorks documentation and look for "Technical Support". You will find a list of support e-mail addresses. Alternatively, visit Altair Connect (from the HyperWorks website, https://altairhyperworks.com/, Support & Training menu or directly at https://connect.altair.com/), go to Contact Support and select your country:
  3. MvdM

    Long time to change the geometry

    Hello @briantvd Thanks for attaching your model. I can reproduce the slow behaviour and we are looking into the problem. Unfortunately, I don't have any helpful feedback for you yet.
  4. MvdM

    Long time to change the geometry

    Hello @briantvd Can you please attach your model (or an example model that shows this behaviour)? Alternatively, you can send it to Altair Feko support so that the performance problem can be investigated by the development team. Kind regards Madelé
  5. MvdM

    Feko solver ERROR 20103

    Great! I trust your simulation is working now.
  6. It looks like a performance bug when exporting Y and Z parameters - we'll ask our developers to investigate. I didn't see any mistake in your model. Exporting the S-parameters (during simulation or from POSTFEKO) seems to work and will have to be used as a workaround for now (like you mentioned).
  7. MvdM

    Feko solver ERROR 20103

    Hello @bastien0298 Non-ASCII characters in the configuration files are not currently supported. Can you please attach the file mentioned in the error message (C:\Program Files\Altair\2018-edu\feko\FEKOenvironment.lua) so that we can take a look? I don't expect the file to contain so many lines - it's normally a much shorter file. Sorry, I was thinking of the file in the same folder (FEKOenvironmentFromSetup.lua). The line that is problematic in your case is likely where the license location is set and it should not contain a non-ASCII character, but we can confirm once you attach the file. There have been other forum posts where others encountered error 20103, but it might have been due to different reasons:
  8. Hi @Meerten It would be great if you could provide the files (attach or send to support) so that we can take a look at what is going wrong. Thanks!
  9. On the Home tab in CADFEKO or POSTFEKO, click on "Script editor". You can open, view, edit and run Lua script files in the script editor. Refer to the "Scripting" section under the "Quick Tour of the CADFEKO Interface" in the Feko 2018 user manual. For more detailed information, see the chapter on "Scripts and Application Programming Interface (API)". For demonstrations of how to use scripting, you can watch the webinars that are available from the Altair HyperWorks Resource Library:
  10. Hello @mgmv7 Just browsing the forum, you would find some helpful advice, for example: You should enter the wire radius value that matches the dimension of the structure you are modelling. What you need to simulate really depends on the problem that you are trying to solve. If you know the length of the dipole antenna, then you do not need to do a parameter sweep to determine its length. If you need to determine the optimum length of the antenna, you could follow the steps in the example. To get started, you can use automatic meshing (for example, select "Standard" for the mesh size on the meshing dialog). Please refer to the user guide for more information on meshing and automatic meshing. If you'd like to familiarise yourself with the basic concepts, you can work through the Altair Feko Getting Started Guide and the section(s) of interest like "Antenna Synthesis and Analysis" in the Altair Feko Examples Guide.
  11. MvdM

    microstrip antenna array simulation

    To answer the second question in your original post: At the moment, for touching elements you need to build the geometry without using the array tool. Union the parts to ensure mesh connectivity. You can use copy and translate, but you would have to define the individual ports and voltage sources for each element. See this example for 2 elements: E_antenna_arra_2_elements_shared_dielectric.cfx. It would be easy to automate this process in a more current version of Feko.
  12. MvdM

    microstrip antenna array simulation

    The problem, for the version you are using, is the dielectric. The Domain Green's function method for fast array synthesis initially did not support dielectrics. This method was extended in Feko 14.0.430 to support dielectrics modelled with the Method of Moments using the surface equivalence principle (SEP).
  13. MvdM

    Optimise a parameter for bandwidth

    Glad you got it working!
  14. MvdM

    microstrip antenna array simulation

    No error appears in my simulation of your model, E_antenna_arra_1.cfx. I used version 2018.1.2. To solve this model in an older version, you cannot use the fast array method. On the ribbon in CADFEKO, go to the Solve/Run tab and select "Solver settings". Deactivate "Solver model with the Domain Green's Function Method (DGFM)".
  15. Hello @mgmv7 In the simple parameter sweep example where the mutual impedances of two of the same dipoles are plotted, the calculation is done on the trace itself (using "Enable maths") where the dipole's input impedance (read from a graph in the model with the single dipole, or obtained from the exported Z-parameters) gets subtracted. You would also be able to do this using Lua scripting, especially if you have more complex calculations in mind. Kind regards Madelé
×