Jump to content

Marcus Chang

Members
  • Content Count

    18
  • Joined

  • Last visited

About Marcus Chang

  • Rank
    Beginner

Profile Information

  • Country
    Taiwan
  • Are you University user?
    Yes
  1. Hi @Torben Voigt, Now I have a question that, is it possible to change the size of corner reflector (s=3*c0/(fmin+f_inc*i), i=0 to 2) for each simualtion (per frequency, such as 9GHz, 10GHz and 11GHz in this case), if it can, can you help me modify this .lua file? BR/Marcus MacroRecording_multifreq91011GHz.lua antenna_only_NF.hfe antenna_only_NF.efe
  2. Hi @Torben Voigt, I've tried to use this .lua file to run the simulation, however, it still doesn't work, and the error message is shown as below, please check: ERROR 30968 in line 40 of the file test2.pre: Error in opening the file 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1 To my understanding, the input characters should be like this: 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1.efe not like this: 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1 So I would like to check with you, how can I write a propreiate code to generate the character string shown as above, such as: 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1.efe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_2.efe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_3.efe BR/Marcus test2.cfx temp.lua
  3. Hi @Torben Voigt, Here is another question for scripting related to this case. If I have three E-field and H-field data, and I would like to use it as my source in next simulation. So I write a script as attached, however, it seemed like I must to input file named including .efe or .hfe these characters. I want to let the input looked like this: but I don't know how to modify this script, such as how to add additional .efe or .hfe these characters to the code as below: properties.EFieldFilename = [[02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_]]..i properties.HFieldFilename = [[02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_]]..i If I use the script as shown as above, it will be looked like: BR/Marcus 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1.efe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_1.hfe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_2.efe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_2.hfe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_3.efe 02_NF source_PEC corner reflector_190406_sphere placed 72 lambda_alt_NearField_3.hfe temp.lua
  4. Hi @Torben Voigt, Due the the version problem, could you please help us provide the modified file in previous version (2018.2-338937 (x64)) in this transition period? BR/Marcus
  5. Hi @Torben Voigt, If I would like to do multi-frequencies simulation, such as with a bandwidth, how to use script to import near field data for every single frequency in this case. Because when I try to run a multi-frequencies problem, I use the first .cfx file and add multi-frequencies setup, I can get a NF data with multi-frequencies. But when I use this NF data as my equivalent source (NF source), the error message was shown as below: ERROR 3392: The near field source cannot be used inside an implicit frequency loop (frequency dependent data) And I check there was a member had similar problem, it seemed like I need to solve it with EDITFEKO. Can you provide us an example (EDITFEKO), for this three files, such as how to write a loop then I can finally get a plot with RCS values in a bandwidth, such as 9-11 GHz. BR/Marcus
  6. Dear Mel, Is there any update for the modified .cfx file? Looking forward receiving your response. BR/Marcus
  7. Dear Mel, Thanks for your help, it works now! BR/Marcus
  8. Dear Mel, I have tried to union all the parts, and it still didn't work. And also, due to there is an error if I use waveguide port, so I use lump port alike, to excite it. It would be grateful if you can help us modify the settings for this simulation. BR/Marcus
  9. Dear support team, Recently I try to simulate a very simple patch antenna working at 2.45 GHz, but it show an error message as below, please check it: ERROR 3849: Wrong specification of the medium for a metallic triangle and I try to open POSTFEKO and use "Find Element" method, to see the problems occurs by which triangle, but two meshes seemed like not overlap. can you help us check what's going on in my simulation? Thanks a lot! BR/Marcus patch.cfx
  10. Dear Support Team, Recently we had an easy exercise of RCS simulation in FEKO with large element object. Due to the size of the file is exceed 48.83MB, so please check below link: https://drive.google.com/file/d/1oee6-vL0pyZJTeJ-w8NQA7ZYRi1ycWM6/view?usp=sharing There is two problems that: First, if you use this file to run simulation, "the triangular mesh overlap..." error message will show up in the end. Then I try to union these separated dielectric objects to an object and exported to a new .stp file then import it again to replace original object. The first problem is solved, but there is a second problem occurs, out of memory by using our workstation with 256 GB RAM. And also I have try the other method, such as replacing the free space region between two objects in the file by using a dielectric constant equal to 1 with FEM solver, and modify the mesh quality to FAST, refinement factor to coarse ,minimum element size to medium in "Create mesh", but it seemed not better than before. Please help us check what's happening to our simulation file, and provide us some guidelines if we face this kind of problem next time. Thanks a lot. BR/Marcus
  11. Hi @Torben Voigt, It works now, the received power for RX antenna in situation of corner reflector is also 10 times larger than metallic sphere. Thanks for your suggestion! BR/Marcus
  12. Dear @Torben Voigt @mel @JIF, Recently, I try to run some simulations for metallic sphere and corner reflector respectively, the results for these simulation is weird, the received power bounced by metallic sphere is larger than corner reflector. But actually, the RCS for metallic sphere should almost 10 times smaller than corner reflector, no matter it's calculated by formula or simulated in FEKO (I also upload these simulations as attached, please check) So I would like to check with you again, to see what's happened to the setting in FEKO. BR/Marcus 190406.7z
  13. Dear @Torben Voigt, For the suggestion of the contribution of the surface currents of the sphere you provided, it seemed not work as below, there is no differences when I use it, please check: BR/Marcus 02_FF source_PEC sphere_sphere surface NF source_190325_sphere placed 72 lambda_alt.cfx
  14. Dear @Torben Voigt @mel @JIF MATLAB code is just a easy way to calculate it. But it is a simple and fundamental formula as below: while the RCS value I use is: here Gt=18.3=Gr, lambda=0.03(m), r=3*lambda, R=72*lambda (=2D^2/lambda, D=2*r) and Pt=1W (in Feko), then I can get the value of Pr is equal to 177nW, however, the results simulated in Feko is less than 100 nW. And thanks again for your suggestion of receiving far field pattern of reflected signal, I will try it and get back to you if there is any further questions. BR/Marcus
  15. Hi @Torben Voigt, Thanks for you support. About Mistake #2, the location of receiving antenna (RX) should be the same as transmitting antenna (TX). So I place the near field data at z=0 rather than z=0.2138m. (horn aperture is placed at z=0). Then I increase the distance between horn and metallic sphere to 72 times lengths of wavelength. (2*D^2/lambda=72 lambda) to satisfy far filed requirement, so that warning 39270 is not shown again. Finally I can get the results shown as below, please check: But... this results do not accord with the number calculated by MATLAB, the coding is shown as below, please check: f=10*10^9; c0=3*10^8; lambda=c0/f; %units: m lambda Pt=1; Gt=18.3; % simulated in FEKO Gr=18.3; % simulated in FEKO r=3*lambda %units: m, radius for metallic sphere RCS=pi*r^2 R=72*lambda %units: m, distance between sensor and object Pr=Pt*1*Gt*Gr*RCS*(lambda^2)/((4*pi*R^2)*(4*pi*R^2)*(4*pi)) so that I can get Pr=177.56e-09 (W), 177nW approximately, almost twice the power of the results (100nW) simulated in FEKO. Second question is, is it possible to get far field pattern of the reflected field of sphere? BR/Marcus onlyonechancy@gmail.com 190326.zip
×
×
  • Create New...