How can I get my simulation results to match my experimental results?

Yolanda Gonzalez
Yolanda Gonzalez Altair Community Member
edited January 29 in Community Q&A

Hello, 

I have been using CADFEKO and POSTFEKO to run simulations on a device to measure its reflection coefficient. However, my results from the simulation do not match my experimental results. Below I have provided images of these results (Top=my results, Bottom=experimental results). 

imageimage

I run my simulation using a start frequency of 1GHz and an end frequency of 3GHz. The number of frequencies is 100 with frequency increments of 20.202MHz (using Linearly Spaced Discrete Points). I have also attached a copy of my model. 

What adjustments, changes, or additions should I make to my model so that the results will match/be similar? 

 

Tagged:

Best Answer

  • Jaehoon
    Jaehoon
    Altair Employee
    edited January 29 Answer ✓

    Hello Yolanda, 

    Based on your simulation result (reflection coefficients are larger than 1), a PIFA antenna and its substrate were not tightly coupled due to not-enough meshes.

    Since the number of the unknown of the (MoM) geometry becomes bigger (by finer meshing), I recommend you to change the substract part to the infinite multi-layer structure. (Then you don't have to worry about longer simulation time caused by the finer meshed .)

    I ran the simulation with the attached file by using the multilayer as below.

    image

    image

    Please check if there is any variation in antenna length or geometry or try more meshes with the attached model.

    Regards,

    Jaehoon

Answers

  • Jaehoon
    Jaehoon
    Altair Employee
    edited January 29 Answer ✓

    Hello Yolanda, 

    Based on your simulation result (reflection coefficients are larger than 1), a PIFA antenna and its substrate were not tightly coupled due to not-enough meshes.

    Since the number of the unknown of the (MoM) geometry becomes bigger (by finer meshing), I recommend you to change the substract part to the infinite multi-layer structure. (Then you don't have to worry about longer simulation time caused by the finer meshed .)

    I ran the simulation with the attached file by using the multilayer as below.

    image

    image

    Please check if there is any variation in antenna length or geometry or try more meshes with the attached model.

    Regards,

    Jaehoon

  • Yolanda Gonzalez
    Yolanda Gonzalez Altair Community Member
    edited January 22
    Jaehoon Kim_22316 said:

    Hello Yolanda, 

    Based on your simulation result (reflection coefficients are larger than 1), a PIFA antenna and its substrate were not tightly coupled due to not-enough meshes.

    Since the number of the unknown of the (MoM) geometry becomes bigger (by finer meshing), I recommend you to change the substract part to the infinite multi-layer structure. (Then you don't have to worry about longer simulation time caused by the finer meshed .)

    I ran the simulation with the attached file by using the multilayer as below.

    image

    image

    Please check if there is any variation in antenna length or geometry or try more meshes with the attached model.

    Regards,

    Jaehoon

    Hello @Jaehoon Kim , 

    Thank you for your help. I am making the changes to my model (changing the substrate to an infinite multilayer structure) however I have run into an error with the mesh. Could you please explain what this error means:

    image

    My computer currently has about 25.0 GB free of storage. Do I need more storage to complete the mesh? 

    Best, 

    Yolanda Gonzalez

     

     

Categories